You are on page 1of 149





The best part about writing this book was renew- tion. Special thanks to owner-builders Peter and
ing old acquaintances (and making new) with all Eileen Allen (for help with the r a d o n c o m m e n
the slightly out-of-step eccentrics who d a r e to do tary), Richard and Lisa Guay; Elaine Rieiiy Cos-
something radically sensible. Thanks to architects grove, Becky Gillette and Roger Danley. and Linda
Malcolm Wells, Don Metz, and Les Boyer for the and Bay Hurst f o r s h a r i n g their e x p e r i e n c e s
helpful chats, and to Ray Sterling at the Under- Extra-special thanks go to Geoff Huggins,
ground Space Center, Susan Nelson at the Ameri- Siegfried Blum, and John a n d Edith Rylander lor
can Underground Space Association, and Peter actually writing their o w n excellent case studies.
Carpenter at the British Earth Sheltering Associa-

Lovingly dedicated to Juki, my building partner at Log End Cottage,

Log End Cave, Earthwood. Mushwood, and who knows what in the
Future. Thanks for the memories past and all those yet to come. It's
been fun.

Library of Congress Cataloging-in-Publication Data

Roy, Robert L.
Complete book of underground houses : how to build a low-cost home/
by Rob Roy.
p. cm.
Includes bibliographical references and index.
ISBN 0-8069-0728-2
1 Earth sheltered houses Design and construction. I, Title.
IH4819.E27R678 1994
690.837—dc20 94-16840

10 9 8 7 6 5 4 3 21

Published by Sterling Publishing Company, Inc

387 Park Avenue South. New York. N Y. 10016
© 1994 by Robert L. Roy
Distributed in Canada by Sterling Publishing
% Canadian Manda Group, One Atlantic Avenue, Suite 105
Toronto, Ontario, Canada M6K 3E7
Distributed in Great Britain and Europe bv Cassell PLC
Villiers House, 41/47 Strand, London WC2N 5JE, England
Distributed in Australia bv Capricorn Link (Australia) Pty l td.
P.O. Box 6651, Baulkham Hills, Business Centre, NSW 2153, Australia
Manufactured in the United States of America
All rights reserved

Sterling ISBN 0-80H9-0728-2

Introduction 4
Preface 5
1. Past & Present 7
2. Design 10
3. Siting & Excavation 17
4. The Footing 28
5. The Floor 40
6. External Walls 55
7. Timber Framework 63
Color section follows page 64.
8. Roof Deck 73
9. Waterproofing 80
10. Insulation, Drainage & Backfilling 86
11. E a r t h Roof 91
12. Closing In 97
13. Retaining Walls & Landscaping 103
14. The Interior 107
15. Performance 113
16. Case Studies in the North 118
17. C a s e Studies in the Southern U.S 127
Appendix 1: Radon 135
Appendix 2: Sources 137
Footnotes 138
Bibliography 139
Metric Equivalents 140
Index 141
by Malcolm Wells
Hob Roy is a m a n you can trust, and I don't say it yet he a s s u m e s that I've d o n e f a r more than I have.
just because he has returned the books I lent him. Rob's not only talked about u n d e r g r o u n d archi
I lis honesty is obvious. Glance at any p a g e in this tecture, and written about it. he's actually gotten
book if you w a n t proof. He's so open about the few out there and d o n e it—and taught a lot of others to
errors he's m a d e during his " u n d e r g r o u n d " ca do it. too.
reer it's almost e m b a r r a s s i n g to read about them. If you have even a slight interest in building a
But then, as I read on. I find his openness reassur gentle, unobtrusive kind of house, read on Rob
my, his solutions simple and convincing. will take all the mystery out of it I le'll even lei his
Hob is one of a g r o w i n g n u m b e r of people clients tell of their a d v e n t u r e s with his designs.
who've built far more u n d e r g r o u n d houses than I,
Wells is an architect in Brewster. Mass. His Intent
book is Infra Structures.
Over the past 20 years, my career has consisted of 1979) sold over 100,000 copies in twelve printings,
building, writing and teaching about two rather but has been out of print for several years. This
distinct building styles: cord w o o d - m a s o n r y con- present volume contains much material from that
struction and u n d e r g r o u n d housing. Cordwood work, now greatly revised and u|xiated.
masonry, wherein walls are constructed of short U n d e r g r o u n d housing is a broad subject, much
logs laid transversely like a stack of firewood, will b r o a d e r than c o r d w o o d masonry, for example.
be only briefly touched U[M>n in this volume. I low- I've f o u n d it n e c e s s a r y to speciali/.e within t h e
ever. many of the builders featured in the case- larger Held, and accent methods and techniques
studies section did i n c o r p o r a t e c o r d w o o d ma- that make u n d e r g r o u n d housing both doable by
sonry into their homes. Those interested in more and affordable to the owner-builder w h o hasn't
information on cordwood masonry should refer won the state lottery. The m e t h o d s accented here
to the Complete Hook of Cordwood Masonry are: the slab f o u n d a t i o n , s u r f a c e - b o n d e d block
Housebuilding: The Earthwood Method (Sterling walls, post-and-beam and plank-and-beam fram-
Publishing Co.. Inc., 1992). That book also gives a ing, and the self-sealing rolled-out waterproofing
thorough step-by-step account of the construction membrane. These are the methods that my wife.
o r o u r own r o u n d c o r d w o o d - m a s o n r y e a r t h - Jaki, and I have used successfully at Log End
sheltered home, Earthwood. Cave, Earthwood, and several earth-covered out-
U n d e r g r o u n d (also k n o w n a s " e a r t h - s h e l - buildings.
tered") housing is a b r o a d subject. In the late 1970s This book uses Log End Cave as a model, and it
and early 1980s, there was a national mini-craze takes the r e a d e r through the step-by-step con-
about u n d e r g r o u n d h o u s i n g , p r o b a b l y in re- struction of a single-storey rectilinear low-cost un-
sponse to heightened awareness of energy con d e r g r o u n d home. W h e r e improvements to the de-
servation. Several new books on the subject were sign a r e s u g g e s t e d , a t t e n t i o n is called to t h e
being published each year. 40' x 40' Log End Cave plans on p a g e 15. The
Some of the best of these books w e r e produced case-studies section e x p a n d s on the basic theme
by the Underground Space Center at the Univer- somewhat by including houses of other shapes,
sity of Minnesota. The Center is still in existence, and shows examples f r o m the "Deep South" of the
but it has de-emphasized housebuilding, probably U.S. to Ontario, Canada.
in response to diminished national interest Its The reader, if he intends to build his own house,
attention has recently been focused on other uses should design it himself (or herself) too. This is o n e
for u n d e r g r o u n d space: storage, industrial and of only t w o ways to get the house you want The
commercial use and tunnelling. In a recent conver- other way is to hire a good architect.
sation with Hay Sterling, USC's director, who also There's no d a n g e r of my putting contractors or
co-authored many of the Center's books, 1 learned architects out of business by advising thai folks
that interest in e a r t h - s h e l t e r i n g is picking up design and build their o w n homes. While the
again. methods and dimensions described in this book
Malcolm Wells, architect, writer and visionary have worked well for us, it's important to have
in the u n d e r g r o u n d - h o u s i n g m o v e m e n t , con- one's plans checked by a competent architect or
tinues his work in Brewster, Mass. Wells has been structural engineer. After all, conditions of soil,
at it since 19t>4! climate and materials vary from site to site. But
My own Underground Houses: How to Build a having plans checked or critiqued is much less
Low-Cost Home (Sterling Publishing Co., Inc., expensive than handing an architect a blank sheet


and saying, "Design a house that will be right for and getting discounts at the local building-supply
me." Sometimes it's possible to have stress-load yards. Some r e a d e r s may subcontract almost all of
calculations checked by a college's structural- the work, and still derive substantial savings over a
engineering class as a study project. contractor-built home.
As for construction, even the most resourceful Finally, our own E a r t h w o o d Building School
owner builders may feel that there are building has since 1980 acted as a clearinghouse f o r infor-
trades that are simply beyond their confidence or mation about earth-sheltered housing and cord-
ability. Even here, though, it's possible to save wood masonry.
money by acting as one's own contractor, sending —Rob Roy
work write-ups out to subcontractors for bidding, West Chazy, N.Y.
Past & Present
U n d e r g r o u n d h o u s i n g , even low cost u n d e r - c a v e d w e l l i n g s t h r o u g h t o t h e p r e s e n t day. I n m a n y
ground housing, is nothing new. After all, what locations, f a r m i n g c o n t i n u e s o n the s u r f a c e a b o v e the
did it cost the first caveman to walk into his new b e l o w - g r a d e h o u s e s . A p p r o x i m a t e l y 2 0 million p e o p l e
shelter? He certainly didn't have a thirty-year live m c a v e d w e l l i n g s in C h i n a t o d a y . 1
Although there's s o m e very posh housing con- On ' l e n e r i f e , in the C a n a r y I s l a n d s , cave-
verted f r o m true caves in the Loire Valley in dwelling is f o u n d in t h e agricultural uplands.
France, m o d e r n earth shelters have little in com- When I visited this a r e a , I thought these homes to
mon with caves. be most uncavelike. It w a s as if houses had been
Man has lived u n d e r g r o u n d at many times and set into the hillside, with only o n e wall left ex-
places throughout history. In Cappadocia, Tbrkey posed. T h e r e w e r e even curtains on the windows.
people have been living in u n d e r g r o u n d towns The houses weren't brightly lit (as it is possible to
and cities for thousands of years. Hie settlements, make them today with the advent of the insulated
some of which extend eight stories below ground, skylight) but they w e r e cheery and will no doubt
are hewn out of soft stone. These settlements are a be much the s a m e h u n d r e d s of years from now.
response to a hostile surface environment, and the How many g e n e r a t i o n s of little troglodytes re-
microclimate Of the u n d e r g r o u n d villages remains turned to their cool h o m e s a f t e r a hot day on the
constant a n d comfortable, despite harsh varia- terraces?
tions of heat and cold on the g r o u n d ' s surface. Why would anyone w a n t to build u n d e r g r o u n d
Folk arc hitecture has always made Intelligent in the modern world? Malcolm Wells says:
use of available materials. When these consist of
no more than the soft strata itself, it's only natural Every s q u a r e foot of this planet's s u r f a c e — l a n d and
that u n d e r g r o u n d dwellings evolve. In China, this s e a — i s s u p p o s e d t o b e r o b u s t l y alive. I t i s n o t s u p -
type of development has persisted for 1>U0U years. p o s e d t o b e s h o p p i n g c e n t e r e d , p a r k i n g - l o t t e d , as-
In Earth Sheltered Housing Design, we learn: p h a l t e d , c o n c r e t e d , c o n d o ' d . h o u s e d , m o w e d , pol-
luted, p o i s o n e d , t r a m p l e d , or in any way strangled in
I n 4 0 0 0 B.C.. v i l l a g e r s a t t h e r e c e n t l y u n e a r t h e d B a n p o o r d e r that we just o n e o f a million s p e c i e s — c a n keep
site i n C h i n a lived i n s e m i u n d e r g r o u n d pit d w e l l i n g s on making the s a m e mistakes.2
w i t h a n A - f r a m e r o o f s u p p o r t i n g a t h i n l a y e r o f soil
a n d v e g e t a t i o n I n Inter c e n t u r i e s r a v e d w e l l i n g b e - Every house which we can design and build has
came a w i d e l y a d o p t e d practice in the arid r e g i o n s of an impact of s o m e kind on the planet, usually neg-
e a s t e r n c e n t r a l C h i n a , w h e r e a d e e p l o e s s soil p r o -
ative. The type of h o m e with the potential for the
v i d e d ideal c o n d i t i o n s f o r s e l f - s u p p o r t i n g e x c a v a t i o n s
least negative impact is the u n d e r g r o u n d home.
in soil M a r c o Polo . . n o t e d s e v e r a l t r i b e s w h o lived
i n e x c a v a t e d h o m e s . T h e t h e r m a l a d v a n t a g e s o f lhe
It's the only kind of h o u s i n g that allows a return to
e a r t h s t e m p e r a t u r e m o d e r a t i o n in a c o n t i n e n t a l cli something a p p r o x i m a t i n g the original landscape
m a t e w i t h c o l d w i n t e r s a n d hot s u m m e r s , t o g e t h e r A great example of this housing is a beautiful
w i t h t h e a b i l i t y t o p r o v i d e s h e l t e r w i t h o n l y a pick a n d home in Yorkshire called "Underbill." It was de-
shovel, h a v e led t o t h e c o n s t r u c t i o n o f m i l l i o n s o f t h e s e signed and built by A r t h u r Quarmby, architect


and plastics engineer. Having lived in Britain for In Architerra's Nice project, the density was
seven years. I knew that one of the primary con- about 10 houses per acre, similar in density to the
cerns of the county and regional planning b o a r d s a f o r e m e n t i o n e d a b o v e - g r o u n d project in Scot-
was to retain the "visual amenity" of the villages land. Architerra tried to launch similar projects in
and countryside. A much better job of retaining the United States in the 1980s, but their great ideas
this has been d o n e in the U.K. than has been d o n e went unfunded.
in the U.S. Q u a r m b / s house, at that time, w a s the Many r e a d e r s who've followed (or tried to Fol-
only fully e a r t h sheltered h o m e in England (a low) the u n d e r g r o u n d movement since the heady
handful have b e e n built since 1980). largely be days of the late 1970s may w o n d e r what has h a p
cause of resistance by the planners! This is very pened in the past ten or fifteen years. Malcolm
difficult to understand. At "Underhill," the s a m e Wells, us stated, is still active, even after three
sheep which had been grazing the Q u a r m b y lot on decades in the movement. Ray Sterling and John
the edge of the village w e r e able to come back to Carmody, authors of Earth Sheltered I lousing De-
continue their grazing after the home was Fin sign, still work at the U n d e r g r o u n d Space Center
ished. I-rum the mam road, the home is hardly at the University of M i n n e s o t a , although their
seen, t hat's visual amenity. . . not to mention agri- work h a s m o r e t o d o n o w with u n d e r g r o u n d
cultural and environmental integrity (IIIus. 1 1). space for uses other than housing. Architect Don
laki and I o n c e lived in an old stone cottage on Metz, w h o designed and supervised the construe
the outskirts of Dingwall in Scotland. We were in tion of some of the most beautiful earth-sheltered
homes I've seen, is still active, as is Les Boyer,
the middle of a 6<X)-acre sheep-and-catUe farm on
Professor of Architecture at Tfcxas A&M Univer-
the south-facing slope of a long hill. Over the
sity Andy Davis, w h o publicized his "Davis Caves"
years, the town of Dingwall gradually m a d e its
so well in the 1970s is still g o i n g strong. The British
way towards the farm where we lived, one field at
Earth Sheltering Assoc iation is probably more ac-
a time falling to tract housing. The tall (almost
tive in promoting residential u s e s of u n d e r g r o u n d
A frame) houses that were built were crowded 10 housing than any c o m p a r a b l e organization in
to the acre in the development, the net effect being America, although only a handful of earth shelters
that the "view" from any house was the gable e n d s have actually been built in Britain.
of the houses facing it. I lad a development o f " U n
derhills" been approved instead, the view from The bad news: The magazine Earth Shelter Di-
each home would have been of the grassy roof of gest was a < asualty of the 1980s, along with Archi
the home facing it. leaving an unobstructed view terra and several other construction-and-design
ac ross the C r o m a r t y Firth. The sheep could have companies which tried to m a k e a living at spe-
returned. cializing in earth-sheltered housing.
Architerra. an international architectural The g o o d news: There's a sense of optimism
g r o u p specializing in earth-sheltered housing, ac a m o n g the old g u a r d that u n d e r g r o u n d is on the
tually built several complexes in France and Spain way back. Interest has picked up in Malcolm Wells'
that would have been perfect in place of the impos books, and, a f t e r a period of slow going, at the
ing homes that were built n e a r Dingwall. Underground Space C e n t e r as well Interest is
increasing again at E a r t h w o o d Building School,
w h e r e we've c o n d u c t e d u n d e r g r o u n d - h o u s i n g
Bv virtue of the stepped design oi the Architerra com- workshops continuously since 1980.
plexes. and the curved glass front walls oft he units, all What c a u s e d the bust w h i c h followed the
of the residents enjoy a 180 view of the surrounding
b o o m ? Why is interest r e t u r n i n g ?
landscape, fhese south facing glass walls provide
natural light and view and passive solar gain and help Publicity in the 1970s followed a new awareness
the units appear naturally integrated with the hillside. that e n e r g y s o u r c e s w e r e n ' t inexhaustible The
Stacking the units stair fashion in slots cut into the public took a fancy to the "new' idea of under-
hillside imparts a sense of privacy to each unit g r o u n d houses. Articles in m a g a z i n e s such as
because of the unobstructed view of the horizon New Shelter a n d The Mother Earth News prolif-
while permitting .i high unit density.1' erated. Everyone seemed to be publishing a book

on underground housing. I've got dozens of differ- pared with other industrial nations with a similar
ent volumes on my reference shell'. standard of living. We continue to encroach fur-
Then came the public perception that the en- ther and further upon the very ecological systems
ergy crisis was either over, or had been a fraud in that support life on this planet open water and
the First place; neither view turned out to be true. wetlands, air, forests, lopsoil.
President Keagan appealed to upward mobility
and unbridled economic optimism. While the rich For the first time in decade, young people
got richer, the poor got poorer. The middle class is seem to be concerned again with the environ-
only now realizing that the same old realities are ment, with quality of life instead of standard of
still with us, to wit: We still waste inordinate living. Underground housing is very much in tune
amounts of energy in the United States when com with this thinking.

Illus. 1-1. Arthur Quarmby's "Underbill" home, in Holme, Yorkshire, U.KV maintains the visual
amenity of the village and countryside.
window locations. An example of this is the fanci-
BERMED VERSUS TRUE ful "hobbit" door (lllus. 2-1) in Arthur Q u a r m b / s
UNDERGROUND "Underhill" h o m e in Yorkshire, in the United
There are two different approaches to earth- Kingdom.
sheltered housing: the b e r m e d house and the
chambered (or truly "underground") house. The
bermed house involves building the structure at or
close to original grade and "berming" (mounding
earth against) the side walls, Very often, an earth
roof is chosen to complete the harmony of the
building. In the chambered house, the entire struc-
ture is below original grade.
There aren't many "true" underground homes
in the United States, although there's a great deal
of underground commercial space. The only such
home I've visited is architect John Barnard's first
Ecology House in Marstons Mills, Mass. A below-
grade central courtyard provides access to several
underground rooms opening onto it. The court-
yard, in turn, is accessed by a single stairway-
down from ground level. The approach is very
similar to that taken in China, where individual
homes are carved out of the loess subsoil, all ac-
cessed to a central courtyard.
In later designs, John B a r n a r d ' s c o u r t y a r d
evolved into a covered atrium. The below-grade
aspect of the original home was tempered some-
what by more of a bermed approach, with a south-
facing elevation providing access and light onto
In addition to the courtyard and the covered
atrium, leaving one or more sides of the home Many different techniques have evolved over
exposed to grade (this is called the "elevational" the past 30 y e a r s w h i c h m a k e u n d e r g r o u n d
approach to earth-sheltered housing) is another houses as light and bright and airy as those homes
way to provide ingress to the home, natural light, built above ground. Underground designers seem
and ventilation. Yet another way is to use sidewall to go out of their way in this respect, cognizant not
penetrations through the earth berm as door and only of the code-enforcement officer's strict ad-

herence to building regulations, but also of the second means of escape. Now, this isn't purely a
common citizen's belief that underground hous design fault of the south-elevational wall alone.
ing equals dark, damp, dingy basements. Mike There are other ways to incorporate the code-
Oehler says: mandated (and sensible) second entrance be-
sides placing a second door on the elevational
An underground house has no more in common with wall: a penetrational doorway through the berm.
a basement than a penthouse apartment has in com for example.
mon with a hot, dark, dusty attic.
But, the elevational plan did fit in well with the
I remember my visit to "Baldtop Dugout," ar- site, despite the e r r o r s cited above. People have
chitect Don Metz's earth-sheltered home. The 270° commented that the Cave seemed to fit the natural
panoramic view of the surrounding New I lamp terrain better than most other earth-sheltered
shire mountains and Connecticut River Valley dis- homes. We felt that the atrium and sidewall pene-
pelled once and for all the notion that under- tration design techniques weren't as well suited to
ground houses are lacking in views. Our own Log providing the external view that was so important.
End Cave had a wonderful close view into the These techniques would also involve expenses that
woods, where the activities of the local wildlife we weren't prepared to make at the time.
seemed to be almost a part of the living space. The Log End Cave which we actually built in
1977 was about 30' x 35'. I include these plans
(Illustrations 2-2, 2 - 3 , and 2-4) as well, although
THE LOG END CAVE DESIGN they are inferior, to illustrate certain points in the
narrative. The dimensions were a function of site
At Log End Cave, we decided to compromise be-
considerations, afford ability, availability of mate-
tween the bermed and chambered styles. We
rials, and certain goals for heating and cooling
would use material excavated for the foundation
efficiency. The lloor plan featured an open living/
to build up the east and west berms for the home.
kitchen/dining area, and smaller rooms on the east
We would also use a semi-bermed south eleva-
and west sides, very much like the 40' x 40" plans
tion to provide access, lots of light, and a view. I
shown on page 15. For structural safety with 4 x 8
must confess that there are three glaring design
rafters, we limited ceiling spans to about 8'6". This
faults on this original Log End Cave design's south
made the perimeter rooms rather small, a short-
elevation. They are:
coming which w a s corrected in the 40' x 40'
• No thermal gain is accomplished by berming up plans, based on ten-foot-square sections. When
to the underside of the windows on the south we visit Richard and Lisa Quay's home, based on
side. It just looks nice. The drawback is that snow those plans and described in chapter 16, we're
starts to accumulate right there, and, in northern impressed by the comparatively roomy bedrooms
New York, it isn't long before somebody is out and bathroom with their ten-foot spans.
there with a snow shovel. Also, every square fool Although every owner-builder likes to design
of a south-facing elevation given over to double- his own home, incorporating the features which
pane insulated glass will actual provide a net he finds important, 1 offer a 40' x 40' Log End
energy gain in northern climes. Cave plan (Illustrations 2-5. 2 6, and 2-71 to show
• We suffered from an energy "nosebleed" where certain construction techniques common to un-
the east- and west-side block walls conducted derground homes. The house can be built on a
the home's internal heat directly to the outside. gentle slope, as ours was, or on a flat site as a
We corrected the problem with some retrofitted bermed structure.
polystyrene insulation on the exterior, but this At the original Cave, our north-south dimen
should have been attended to at the design stage. sion (30') w a s limited by the availability of three 30'
• The south elevational entrance is the only means 10" x 10" barn beams. At the time, we didn't real-
of ingress and egress. Even a chipmunk knows ize that overall structural strength isn't compro-
better. One never knows when a fox (or a build- mised by joining shorter girders over the pillars.
ing inspector) might come to the door. Have a With ordinary light-frame construction, bending

lllus 2-3. The south-wall elevational plan of the original I og End (jive, a
bermed structure built on a gentle slope

strength generally comes into play before shear separate 10' girders joined over a post are actually
strength, but on heavy-timber construction used stronger on shear than a single 20' girder sup-
for heavy loads, the opposite is usually the case. ported halfway along by a post. While bending
(Bending failure is when a m e m b e r s n a p s some- strength is slightly d e c r e a s e d in this example,
where near the middle because of the load. Shear shear strength increases by about 20%, and shear
failure is the tendency of all the fibres of the w o o d strength is usually the weak link in the calcula
to "shear" through, usually right n e a r where the tions. Be happy that you only have to deal with 10'
member is supported by a wall or a post.) Two girders, not 20' or 30" behemoths.

lllus. 2-4. The original

Log l-nd Cave's floor plan

Of 2 3 V 5"

[ w | wooosrwe ( s ) SKYLI6-HT

The 40' x 40' floor plan (lllus. 2 7) will result in Our east-west dimension of about 35' w a s lim-
peripheral rooms that are a lew d e g r e e s cooler ited by the s t r e n g t h of 4" x 8" r a f t e r s . T h e
than the open-plan great room, assuming that you 40' x 40' plans call for 5 x 10s, as we used at o u r
have centrally located wood heat. This w a s true at round E a r t h w o o d house. In 10' to 12' lengths,
the original Cave, and w a s considered at the de they're heavy, but not unmanageable. You'll want
sign stage. Jaki is English, and l had lived seven s o m e help with the heavy timbers. The stress-load
years in Britain, so we were used to cool bed- engineering for t h e plans is b a s e d u p o n a 10' x 10'
rooms, and believe that they a r e healthier than module, repeated 16 times. Therefore, the plans
their o v e r h e a t e d A m e r i c a n c o u n t e r p a r t s . are easily adaptable to other sizes and shapes;
Peripheral room temperatures c a n be flirther reg- 30' x 60', 20' x 40', etc.
ulated by opening or closing the internal doors.
For nonwoodburners, perimeter b a s e b o a r d heat- MOW AN UNDERGROUND HOUSE
ing with zoned thermostatic control is always an
option. Finally. I always design floor plans to make
the joining of internal walls with exposed rafters Many people casually a c q u a i n t e d with u n d e r -
and girders both neat and easy to construct. ground housing think we build this way to take

lllus. 2-5. The new (40' x 40 )

Cave's block, rafter, post-and-
beam plan

lllus. 2-6. Below: The 40 x 40

Log End Cave plan improves upon
the original design. The home is
designed to support a roof load of
at least 150 lbs. per square foot.


O i 2 3 K4 ; 8 id
SLIte /V fff r

advantage of s o m e great insulative value of earth. perature just outside the h o u s e walls in winter is
This isn't true! In fact, earth is a pretty poor insula- about 40 F (lllus. 2-8). When the outside winter air
tor. and wet e a r t h is a terrible insulator. temperature is 20°E as is often the case, the
So how do u n d e r g r o u n d houses save so much u n d e r g r o u n d house starts o u t with a 60°F advan-
energy in heating or cooling if earth is poor insula- tage over the house on the surface. Put another
tion? The earth is a great capacitor. Just as an way, the u n d e r g r o u n d h o u s e need only be 30°F
electrical capacitor stores an electric charge, the w a r m e r than the ambient t e m p e r a t u r e to reach a
earth is a capacitor which stores heat. For us, c o m f o r t level of 70°F. M e a n w h i l e , the s u r f a c e
building near P i t t s b u r g h , N.Y., building the house dwelling needs to be 90f'F w a r m e r than its ambi-
6' to 8' below g r a d e would be like building it 1000 ent, the frigid o u t d o o r air.
miles to the south, with o winter climate more like While the earth is a g o o d capacitor, it's also a
that of Charleston, S.C. The ambient earth tem- good conductor. The 40°F e a r t h will try to rob the

lllus. 2-8. The heating and cooling advantages of an undergmund house


heat from the house, and an internal t e m p e r a t u r e Although both external d o o r s are on the eleva-
of 40°F won't be very comfortable. Thankfully, tional side, there's no reason why a penetrational
there's another thermal mass at our disposal, one d o o r couldn't be incorporated on one of the other
over which we can more easily exercise control. walls if access w e r e required in a different direc-
This is the mass of the fabric of the building itself: tion, or if you encountered building-code diffi-
the concrete floor, the walls, the footings, and any culties. Penetrational b e d r o o m windows, if they
internal mass such as a m a s o n r y stove. The best open, can also satisfy building codes egress. For
way to regulate this thermal mass is to separate it example, the National Building C o d e (NBC) allows
from the earth's thermal flywheel by a thermal the e g r e s s vvindowsill to be no m o r e than 48" off
break, typically rigid-foam insulation placed cor- the floor. Other codes may specify 44" rypically,
rectly on the exterior of the home's fabric. Now the rescue w i n d o w s f r o m sleeping r o o m s must have a
house itself can be brought up to t e m p e r a t u r e and minimum net clear o p e n i n g of 5.7 sq. ft The mini-
the advantage of the earth's favorable ambient mum net clear-opening height of 24" and the net
climate can be utilized. This is very important. The clear-opening width must be at least 20" S o m e
worst thing to do (and it's amazing how often it's building codes will allow the alternative of two
done) is to place insulation on the interior of the doors f r o m b e d r o o m s offering t w o separate paths
thermal mass. Not only has all control over the of escape. Hay Sterling and John C a r m o d y say, in
mass (perhaps 100 tons) of the home's fabric been Earth Sheltered Housing Design.
lost, but now the earth can freeze up against the
cold walls of the h o m e and c a u s e s t r u c t u r a l The intent of the egress requirement is clear If a fire-
damage. should start in any part of the house other than the
In the summer, there's a similar a d v a n t a g e , bedrooms, occupants should have a clear means of
fhink of the earth as storing "coolth." In northern escape directly to the outside without going into a
New York, the highest earth t e m p e r a t u r e s a r e smoke- or fire-filled part of the house.''
about 60°F at six feet deep. This peak occurs in
August, after m o n t h s of slow, steady rise after the While earth-sheltered houses are very much
40°F earth temperature in early March. Even if it's less likely to catch fire d u e to their use of massive-
95°F outside, no energy is required for cooling. materials such as concrete, concrete blocks, and
Residual heat in the home, sunlight, people heat heavy limbers, it's still possible for d a n g e r o u s
(98.6 I ). dogs, lighting, and cooking will all bring smoke fires to occur in furniture, or electrical fires
the house up to maybe 75°F. The house built on the in internal walls. Meeting the r e q u i r e m e n t of
surface has two choices: stilling heat or energy- building codes may sometimes involve overbuild-
draining air conditioning. ing and g r e a t e r expense, but it can also eliminate a
If an earth roof is included in the design, there's lot of unpleasant hassles. The b e d r o o m s in the
another great cooling advantage. Unlike the high 40' x 40' plans need either interconnected door-
surface t e m p e r a t u r e s of asphalt t a r s c a p e s , the ways or penetrational code-worthy windows to
earth roof is cool just a few inches below the sur- pass an inspection. Check with your local code-
face. Respiration by plants and e v a p o r a t i o n of enforcement officer. Or, if you're building in one of
moisture off the earth roof both help to cool the those few remaining rural a r e a s w h e r e building
building, just as a wet towel draped over a 5-gallon codes are rare, use your o w n best-informed judg-
bucket will help to keep your drinks cold. ment. Err on the side of safety.
When designing an u n d e r g r o u n d house, keep
in mind three important words: strength (because
of heavy earth loads on the roof and sidewalls),
livability, a n d waterproof. You s h o u l d h a v e
enough light and ventilation to a s s u r e a pleasant,
open, nonclaustrophobic atmosphere. As for
The 40' x 40' Log End Cave plan is an example of keeping the house dry, d r a i n a g e is the better part
elevational t e r r a t e c t u r e with b e r m e d sidewalls. of waterproofing.
Siting & Excavation
tion must be paid to lateral stress on the walls.
SOLAR ORIENTATION Positive d r a i n a g e is particularly important, espe-
If you haven't bough l y o u r p r o p e r t y yet, there are cially if the site is on the side of a long hill. In this
a few c o n s i d e r a t i o n s special to u n d e r g r o u n d case, a s u r f a c e d r a i n a g e ditch with p e r f o r a t e d
houses which might be less important with on- drain tile n e a r the bottom and filled with 3" to
g r a d e homes. 4"-diameter stone is installed on the uphill side of
In the northern U.S. a n d Canada, the ideal site the h o m e to carry runoff away from it Typically,
would feature a gentle south-facing slope that these surface drains are 12" to 18" wide and 36" to
takes advantage of solar gain. Stu Campbell in The 48" deep, depending on frost depth and other con
Underground House Book says: siderations peculiar to the site. On any steep-slope
projects, consult with a soil or structural engi-
The perfect exposure for a window meant to collect neer. Flat sites can also work, provided they have
solar radiation is 15" west of true south, but 20° to good drainage characteristics.
either side of this point is still excellent." The Log End Cave homestead featured a knoll
higher than any point within a quarter-mile fhis
In the southern U.S.. w h e r e cooling the home knoll sloped gently in all directions, but, luckily, its
may be the m o r e i m p o r t a n t energy consideration, greatest slope w a s angled almost d u e south The
a north-facing slope might be preferable. Remem- site combined the a d v a n t a g e s of g o o d drainage,
ber. too, that north light is much less harsh than away from the building, with southern exposure
south light Hast and west-facing sites wouldn't be for solar gain in the winter. An added benefit was
so bad in the midsection of the U.S., but decide if that the site w a s closer to our windplant than Log
you're a morning or evening person, as the quan F.nd Cottage had been. With a 12-volt e n e r g y sys-
tity of light will vary a g r e a t deal in different parts tem. short cables are best, as there's considerable
of the house f r o m sunrise until sunset. The use of line loss at low voltage. Finally, driveway access
the rooms come into play here when you design
w a s excellent. Building materials and ready-mix
the floor plan. As the sun begins to set, a great deal
concrete w e r e easy to deliver to the site.
of heat comes in from west-facing windows.
The site had been a m e a d o w when o u r home-
stead had been part of a hilltop farm m a n y years
ago, but this particular c o r n e r of the m e a d o w was
overgrown with small apple, cherry, and poplar
trees. It took Jaki and me a couple of days to clear
e n o u g h of the g r o w t h to see the terrain clearly. But
we'd need a more accurate u n d e r s t a n d i n g of the
SLOPES contours than we could discern visually. Front-
Gentle slopes are good for single-storey homes, end loaders cost $22/hour back then, a great deal
which would include the majority of earth shelters of money for us in 1977. Plan ahead so that just the
that have been built. For a two-storey home, a right a m o u n t of earth is moved, and not moved
steeper slope would work, but much m o r e atten- twice.

FIT THE HOME TO THE SITE The easiest way I know to find south is to read the
paper or listen to the evening weather, and learn
Our method of siting the house worked well for us, the times of sunrise and sunset on the day you're
and I'd recommend it to anyone faced with a hill- working. Halfway between those times the sun
top or hillside. First, we set up a surveyor's transit will be at true south. Cloudy for a week? Use a
at the top of the knoll. You can rent a contractor's magnetic compass, but don't forget to take into
level (which will do the same job) quite inexpen- account the magnetic deflection for your area. A
sively at tool-rental stores. We plumbed and lev- good transit will have a magnetic compass built in,
elled the transit to a point on the ground within the but a rented contractor's level probably won't.
legs of the tripod and marked the spot with a half- At Log End Cave, we set the zero-degree mark
brick. An even better benchmark would be a large of the level's compass rose to a fixed point, the
nail driven into the ground, with a rod piece of corner of Log End Cottage. Then, with Jaki hold-
tape tied just under the head so the nail can be ing a calibrated g r a d e stick (you can rent or make
found again. one) and carrying the end of a 50-foot tape, and
Establish true south and drive in a stake along with me reading the level and marking the dis-
the true-south alignment. South would then corre- tance on a chart, we statistically mapped the area.
spond to 180' on the compass rose of the transit The procedure was to establish a slope for each

ray of the circle divisible by 15°: 0°. 15°, 30", 45°, and on the same ray until we read a 6" drop. Again, we
so on. I'd set the level or transit at 0", for example, measured the distance. Likewise, we recorded ev-
and Jaki would move away from me with the grade ery 6" drop along the ray until we were outside the
stick until we were able to discern a drop in the vicinity of the house site. Then we repeated the
land. The stick's calibrations began reading up- procedure along the 15° ray, and so on to 180 . We
wards in inches from a point on the slick equal to now had a statistical abstract (Table 1) of the large
the height of the level off the ground, say four feet. general area of the building site. This j o b took us
As s(xm as we could perceive an inch of drop, we all afternoon.
took a measurement from the benchmark and re- 1 took the transit inside for the evening and
corded the distance. Then Jaki moved further away transcribed the figu res onto a large piece of graph

lllus. 3-1. The north and south walls were almost even with the original ground level;
the terrain around the cast and west walls was built up using excavated material.

paper. I let 1 V (the size of the squares on the graph MARKING THE EXCAVATION
paper) equal one foot I drew a light pencil line at
every 15w of arc for the half-circle in which the
house would surely fall, and measured the scaled
distances from the center point, marking with a
dot each 6" drop in contour along the way. To avoid There's another big time-saver that comes out of
confusion later, I lightly marked the elevation next these little maps, now that the trouble has been
to the corresponding dot: -1", - 6 " , 12", and taken to do them: the remarkable ease of plat ing
so on. the four flags to guide the excavation contractor.
We considered 1" to be level for our purposes. Once we determined on paper the best location for
After all the figures from my chart were thus tran- the house, all we had to do was transpose the four
scribed, it was an easy task to connect the dots of corner points to the site itself. We used a simple
like elevation with gently curving lines. Lo and angle-and-distance system. Using a small protrac
behold an extremely accurate contour map of the tor and a ruler, it was easy to determine, for exam -
site emerged as the dots were connected (lllus. pie, that the northeast corner of the house should
3-1). be 18'3" from the half-brick used as a benchmark,
All this may sound like a lot of work, but it paid at an angle of 29°. It's important to be able to set up
off by eliminating a lot of guessing later on. From a the transit or contractor's level in exactly the same
piece of the same kind of graph paper, I cut out a spot a s it was when the o r i g i n a l f i g u r e s were
scale model of the foundation plan, using the out- taken. We set up over the brick using the corner of
side dimensions of the planned block wall. I noted the cottage as 0°, as before, and. at an angle of 29 ,
the location of the door on the south wall of the we measured out 18'3" and drove a white-birch
scaled drawing Now all we had to do w a s slide the stake into the ground (lllus. 3-2). We did the same
little square around the contour m a p until the with the other three corners. Then we checked the
most sensible location emerged. wall lengths and diagonals with our 50' tape, hold
We knew that the top of the block wall on the ing the tape level to account for the d r o p in terrain.
south, where there would be l a r g e windows, Every dimension was within 5". and after ten min-
would be 42" below the top of the other three walls. utes of juggling, we put out all the stakes. Walls
We knew that the i oncrete slab at door level would were the right length; diagonals checked.
be 78" below the tops of the three full-height walls. I happened to have a 4' x 4' sheet of plywood
As seen in lllus. 3 1, the north and south walls are lying around, which I used to establish the corners
about even with the original grade. The south wall of the excavation itself (lllus. 3 31. We drove in
required (>" additional excavation. The earth that another set of hefty birch stakes to mark these
comes out of the hole during excavation is used to points. Of course, the original four stakes marking
build up the terrain to the tops of the east and west the house corners would be eaten by the front-end
walls. By this plan, the new ground level melds loader, but I wanted the operator to be able to
nicely into the shallow-pitched, earth-covered visualize the project clearly before starting. It's
roof. important that the operator sees the site as the
Our job, then, was to plot the location of the owner builder does. With our contour map, it was
excavation and to formulate the most efficient ex- easy to explain just how d e e p to go at each corner
cavation depth to lake advantage of natural con of the excavation.
tours, keeping in mind that we'd have to do some- A 4' work space all around the walls may seem
thing with every cubic foot of material that came like a lot, but it's important to have plenty of room
out of the hole. Drawing a contour m a p made this to build a n d w a t e r p r o o f the walls. A certain
very much easier to estimate. It w a s worth the amount of erosion can be expected during the job,
effort: The landscaping around the Cave wasn't a and although we spaced our stakes 4' out f r o m the
big job and it came out very well. house corners in each direction, we didn't end up

lllus. 3 2. Transposing the four corner points from paper to the site,
using a protractor, ruler, and surveyor's level

with anything like 4' of space outside the walls. The ing in the g r o u n d , the excavation slope (called the
space was about 30" on average, at the base of the "angle of repose") will cut away into the four feet at
excavation. If the digger leaves the markers stand- the bottom (lllus. 3 4).

If you're fortunate enough to have very sandy or If in doubt about the quality of the subsoil, check
gravelly subsoil (which means that you have excel- with a percolation test and/or a deep-hole test.
lent drainage), then the angle of repose will proba- Your local cooperative extension office or county
bly be more like that shown in lllus. 3-5. Our soil health department may be able to help with soil
had a fair amount of clay, which allows steep exca- maps and information about septic systems. The
vation sidewalks, like those shown in lllus. 3-4. same tests used to determine percolation for the
With sandy soils, it's necessary to make a bigger design of septic systems will give you important
excavation, so the second set of stakes might have information on the drainage characteristics of lo-
to be 6' or more out from the house corner stakes. cal subsoils. Since you'll probably have to conduct
On the positive side, you'll probably be able to these tests anyway to satisfy the health depart-
backfill with the same material which came out of ment for approval of a septic-system design, you
the hole, because drainage is excellent in sandy might as well do the tests before you do the site
soils. plan. A deep-hole test (5'- or 6 - d e e p hole) may

reveal ledge or bedrock which might affect the If you've got poor drainage, read c h a p t e r 10
whole site plan, or you might find that that won- now. If you've got h o r r e n d o u s drainage, such as
derful sandy soil is only three feet deep, with hard- soils designated as "expansive clays," or if a deep-
pan or clay below: hole test breaks through to the water table, recon-
sider the site altogether, or build an above-ground
structure by methods which have proven to be
successful in y o u r a r e a . U n d e r g r o u n d h o u s i n g
isn't for everyone's building site.

Radon is a clear, tasteless, odorless gas, which, in THE FLAT SITE
sufficient quantity, can cause lung cancer. It enters If your site is flat, you'll be spared the tedium of
the home through cracks in the foundation. Un- creating the contour map. All you need to know is
derground houses can pose a higher risk than do the soil conditions, and I would definitely advise a
other types of housing by the n a t u r e of their con deep-hole test. The hole doesn't take long for a
strucUon. If you're in an area known for high ra-
backhoe to dig, and. you may be required to do
don concentrations, or if you're building in gravel
one anyway, as is the case in New York State.
over shale, granite, or phosphate deposits, or if
For flat terrain, 1 r e c o m m e n d the bermed style.
you simply want peace of mind on this matter,
now's the time to test the site for radon. Appendix 1 The Log End Cave design, in fact, is midway be-
treats radon in detail, and gives sources of addi- tween the b e r m e d and c h a m b e r e d styles of under-
tional information. g r o u n d home. With the bermed style, the builder
need only calculate the amount of earth needed to
mound up gracefully against the sidewalls. But
keep in mind that unless the site is entirely of sandy
soil or good-draining gravel, it isn't wise to backfill
with the excavated material. Clay and other soils
with p o o r percolation qualities should be kept
AN IMPORTANT QUESTION away f r o m foundations and earth-sheltered walls.
The drainage characteristics of the subsoil w h e r e Bring in sand or gravel, if necessary, to e n s u r e
you build must be considered carefully when cal- good drainage. Because o u r subsoils have such
culating the size of the excavation. You need to poor drainage, we had to backfill with 25 d u m p -
answer one very important question: Can the ex- truck loads of 5 cubic yards each If it's necessary
cavated material be used to backfill the walls of to bring in backfilling material, this task should be
the home? If percolation in the subsoil is good, figured in when calculating the depth of the exca-
then the a n s w e r is yes. If the soil holds water or vation. With a b e r m e d house, it's easy to get rid of
doesn't let it through, as with claylike soils, then a little extra material. The thicker the berm, the
the answer is no. If percolation is poor, you'll have better. In fact, if the earth piled up against the walls
to bring in backfilling material or use one of the is thick e n o u g h (5' to 6'). the heating advantages
various drainage products made for that purpose, are almost the same as they'd be for a house just
'fhese products will be discussed in chapter 10. but below original grade.
the builder should know about them early in the The two-storey earth-sheltered home w h e r e we
design stage, as the dimensions of the excavation live now; called E a r t h w o o d (see the photos in the
will depend on whether or not it will be necessary color section), w a s built on a perfectly Hat site, a
to bring in backfilling material. This will probably gravel pit. There was no topsoil. We built a pad of
be an economic decision, weighing the cost of good percolating sand, which w a s right on site,
many loads of coarse sand or gravel versus the and "floated" the f o u n d a t i o n slab on this p a d .
cost of manufactured drainage materials which Forty percent of the cylindrical h o m e was shel-
are designed to be laid up against the sidewalls of tered with an earth b e r m constructed of gravel
the home. pushed up f r o m the area in front of the home.

FLAT-SITE CALCULATIONS will be out of the way. l i t i s task will save time and
money later when the roof and final landscaping
Let's calculate the required depth of excavation for are done. This hypothetical site isn't blessed with a
the gabled berm-style house shown in lllus. 3-6. good depth of topsoil, Allow for additional topsoil
We'll assume poor soil percolation, making it nec- if you need it at your site. How much material of
essary to bring in backfilling material. If the site the kind taken from the excavation will be required
has good topsoil, the whole area should be on the berm? "Calculated guessing" will be our
sc raped by a bulldozer and the soil piled where it best approach to determine this amount.

lllus. 3-6. Calculating the required depths of excavation for this gabled, bcrm style, 30' * 30' house should take into
account poor percolation and the necessity for extra backfill.

The plan is for external wall dimensions of 30' 0.833), or 120 cubic yards (0.833 x 144 - 120). I"he
square, or 10 yards by 10 yards. Let's say we were sand for the drainage remains the same 64 cubic
to excavate one yard (36") deep over the whole area yards. The berm is (V higher now. The 3 in 15 pitch
within one yard of the walls. This square, 12 yards (3:15) established by the roof adds 30" to the width
on a side and I yard deep, will yield 144 cubic yards of the berm at original ground level.* Expressed in
of material (12 x 12 x 1 = 144). Because of poor yards, then, the berm is 1.167 yards high and 5.833
percolation, the yard of space right next to the yards wide. The volume of the berm. as we saw 111
house won't be tilled with the excavated earth. the first calculation, is &bhl (where I length
I low much earth will the rest of the berm require? t a k e n a l o n g the i n s i d e o f t h e b e r m ) plus
The three sides of the berm directly adjacent to the 2 x (V^hb2) for the two delta wing shapes marked
sand backfill have the cross-sectional shape of a " D . " S u b s t i t u t i n g : (.5M5.833M1.167M34) 4
right triangle.-1 yard high (h> and 5 yards wide (b); (2)(.25)(1.167)(5.833)(5.833) - 115.72 4 19.85
that is, a cross-sectional area of 2V2 square yards 135.57 cubic yards. We dug 120 compacted cubic
(A Vzbh - Vt x 5 x 1 = 2Yt). The total length of yards out of the hole. Not bad. A bit deeper than
2V> feet (30") should be perfect.
the berm is 34 yards (11 4 12 + 11) where it is
directly adjacent to the sand backfill, so the volume The example is a realistic one, except for the
is 2 ; ' times 34, or 85 cubic yards. Add to this the obvious lack of a second entrance. Such a bermed
volume of the two delta-wing Shapes where the house would be just slightly smaller than the Log
berms meet at the corners, marked "D" in lllus Hnd Cave we actually built. Perhaps a structure set
3 6 The volume formula for these corners is V-ihb2, only 30" deep shouldn't be called underground
so, by substitution, Vt x 1 x 5 x 5 = 6'/» cubic housing, but such a house is a closer relative to a
yards. In all, il will inquire 97V<r cubic yards of earth subterranean house than it is to a conventional
to build the berm (85 4 6V< f 6V* - 97VA But we've surface dwelling. The thick berm and e a r t h -
taken 144 cubic yards out of the hole! The 46'/z covered roof offer nearly the same advantages of
cubic-yard difference is quite a bit to haul away or heating a n d cooling as are enjoyed by an under-
to spread around the site. ground house. And the visual and environmental
Before we make a second guess, let's consider impact is about the same as that of the original Log
End Cave. From a distance, the house would look
the situation if the excavated material had been of
like little more than a knoll on a flat landscape.
good enough drainage to use for backfill. The
backfilling c an be considered as a rectangular vol
time 1 yard wide, 2 yards high, and 32 yards long
(10 + 1 + 10 + 1 + 10 - 32). The formula for WASTEWATER SYSTEMS
volume in a rectilinear solid (V - lwh) yields 6-1
cubic yards (1 x 2 x 32 = 64). In this case, the At the earliest stages of planning and siting, con-
total volume of the berm right up to the Walls is sideration must be given to the disposal of waste-
161 Vk cubic yards (97V? 4- 64 = 161VH a little more water. Wastewater disposal systems based 011
than the 144 cubic yards that came out of the hole. electric p u m p s are expensive beyond the budgets
As loose earth occupies more space than it did described in this book. Moreover, pumping sys-
originally, this isn't too bad. The berms could be tems are subject to ongoing maintenance, con-
stant consumption of power over their lifetime.
made a little steeper, if necessary, or the excavation
deepened very slightly. (As the hole gets deeper,
the berms above grade get smaller.) Remember, "For this example, I've d r a w n the gable o n e yard 13') hlflher
too. that we'll need about 7 cubic yards of backfill than the sidewalls, which a r e f» yards 115') away Irom the
along the front wall of this plan, which is similar to midline of the house. The roof pitch, therefore. is3" of rise for
the original Log End Gave. 1f> lateral foot, expressed 3.15. Roof pitch is usually measured
m terms n( rise per 12 lateral feet. Our example is equivalent
Let's assume that we have poor soil. Let's try lo a ZVzAZ pilch An earth roof should have a pitch ol'al least
excavating just ZVz feet instead of 3 feet. This time, 1.12 to p r o m o t e drainage, but not more than about 3.12. '.0
the volume of the excavation will be 0.833 of what avoid d o w n w a r d slumping of the earth The Log t n d Cave
it had been before (because 2'/? divided by 3 - pitch w a s 1-y«:12

reliability problems, and failure d u r i n g power out- system. A contractor will be inclined to give a
ages. Therefore, integrate the siting of a below- better price if justified by the volume of work. I
grade house very carefully with the location of the Stayed with the s a m e contractor throughout the
septic tank and drain field. On the side of a hill, work at Log End Cave, and liked his work so much
waste drainage hy gravity won't he a problem On that I retained him again at F a r t h w o o d . I get excel-
a flat site, it may be necessary to dig d e e p e r tracks lent service as a regular customer.
and drain fields than normal, keep the elevation of Price isn't the only consideration; ability is an-
the house higher (which might mean more land- other. If two contractors give similar estimates, but
scape sculpting to build the berms). or even raise one has a better reputation, go with the good
the floor level of the bathroom by a stop or two in reputation, even if it costs a few dollars more. I also
order to establish a correct gradient for a gravity prefer to pay by the hour rather than by the j o b
waste-disposal system. T h e r e a r e so m a n y i m p o n d e r a b l e s in u n d e r -
If the water table at the building site is ever likely ground housing. You may c h a n g e your plans d u r
to be higher than the drains, rule out the site imme- ing construction, about w h e r e to put a snakawuy
diately. Rule out lloodplains, too. Even if tile walls (dry well), for example. If you're paying by the job.
are built as watertight as a swimming pool, the the contractor will penalize you for c h a n g e s in
waste-disposal system would fail a n d probably plans, and rightfully so.
back up into the home.


EXCAVATION At the Cottage, o u r cellar hole w a s dug entirely
Ibgether with backfilling and landscaping, exca with a backhoe. The backhoe worked well, so natu-
vation represents one of the biggest (and costliest) rally I assumed that I needed one at the Cave, but
jobs connected with building u n d e r g r o u n d . Esti- the major excavation there was d o n e much more
mates should be obtained f r o m several heavy- efficiently with a front-end loader The loader was
equipment contractors. You can get an estimate S22 an hour (in 1977 dollars), the backhoe only $16.
for the whole job or you can pay by the hour for but the loader probably did the j o b in little m o r e
earth-moving equipment. If the contractor knows than half the time it would have taken the backhoe.
his business, the j o b estimate will be pretty close to The excavation for the Cave w a s much b i g g e r
the cost on a per-hour basis. Some contractors than the o n e for the Cottage and, because it w a s
may tack on what a p p e a r s to be a hefty profit for cut into a hillside, it w a s easy for the front-end
the job You can't hlame them. They also have to loader, with its 6' bucket, to maneuver. The e n -
cover unforeseen circumstances when bidding by trance on the south side w a s the natural way for
the job. such as overfastidious clients. 1 always get the loader to c o m e in and out of the hole easily. It's
the per-hour charge, as well, and this is almost true that the loader has to back away with each
always the way I hire the equipment contractor. bucketful and d u m p it, while the b a c k h o e c a n
Make sure that the contractors a r e pricing for stand in one spot for a while and with its long
the same thing; otherwise, no intelligent compari- boom place the earth outside the excavation. But
son can be made. For example. Contractor Smith the loader moves nearly one cubic yard with each
gets $40 an hour for his backhoe, while Contractor scoop. Actually, to finish s o m e of the corners, we
Jones gets $45 an hour. But Smith uses an 18" hoe, did use a backhoe, instead; at that point, the loader
while Jones has a 24" bucket. For excavation, Jones had to travel too far with each load to d u m p it
will come out cheaper, other things being equal. where it would be useful later for lands< aping. It's
Ask the contractor if he charges for hauling the a big plus to hire a contractor w h o has a variety of
equipment to the site. Such c h a n g e s can make e q u i p m e n t , but m a k e s u r e y o u ' r e only b e i n g
quite a difference, especially when there's much charged for o n e machine at a time.
follow-up work. The loader would also do well for the excavation
Bear in mind that you'll need equipment later of the b e r m e d house used as an example in this
for backfilling, landscaping, and p e r h a p s a septic- chapter. In a relatively shallow excavation without

a lot of big rocks, a good operator on a bulldozer ished before embarking on these jobs. You want to
can do a remarkable job quickly. be sure that grades are right (actual, not theoreti-
The backhoe is the only machine to use for dig- cal elevations are important here) and, too, you
ging the septic lines, soakaways, and drainage don't want a lot of dangerous ditches and holes all
ditches, but wait until house construction is fin- over the site.
The Footing


The footing, generally made of reinforced con- The site should be excavated to a flat surface about
crete, is the foundation base of the wall. The foot- two feet beyond the outside of the planned footing
ing supports the entire structure and distributes (Illustrations 4-1, 4—2). Check the level with a con-
the weight of the walls and roof over a base that's tractor's level and g r a d e stick. Because of the slope
broader than the thickness of the walls. The foot- that will be left if the work is done with a front-end
ing is given tensile strength by the use of strong loader, the 2' figure is consistent with the place-
iron reinforcing bars, often called "rebars," so ment of the second set of white-birch pegs 4' out-
side those that mark the house corners.
that the foundation taker, on the characteristic of a
monolithic ring beam. For relatively lightweight On the flat excavation, mark the location of the
concrete-block construction, such as for a small four outside corners of the footing so that a back-
block building, the dimensions of the footing fol- hoe can draw the tracks within which the footing
low a simple rule. The depth of the footing should will be poured. There are two ways to plot these
be equal to the width of the wall. The width of the corners: batter boards and educated guesswork.
footing should be twice the width of the block wall Going by the book, Jaki and I built four batter
it will support. Following this rule at Log End boards way up on the surface so that we could
Cave, with its planned 12"-vvide block walls. I de- slide su ings back and forth to make sure the sides
cided on footings 12" thick and 24" wide. were the right length and square. These boards
A few years later, it was pointed out to me by can be seen in Illus. 4-2. A local contractor friend,
concrete-foundation experts that using a 12"-thick Jonathan Cross, came over on a Saturday morn-
footing is really "overkill." More than a 9" depth of ing and found us struggling with these grotesque
footing is a waste of money on concrete. We're not batter boards. We'd been at it for hours. We fig-
building a skyscraper! Properly reinforced con- ured that once we had the batter boards estab-
crete is phenomenally strong on both compres- lished, we could use them for the footing's inside
sion and tension (the resistance against settling at and outside dimensions as well as for the block
a weak point in the subsoil). Earthwood is a much walls. "Don't need 'em," said Jonathan. "They'll
heavier home than Log End Cave, and is sup- just get in the way of the backhoe." We moved to
ported quite happily on a footing 9" deep and 24" the educated-guesswork method then and there
wide. I still like the 24" width, because this de- and it w a s much easier.
creases the load per square foot on the earth, When you've got a levelled area with dimen-
resulting in a more stable building and less set- sions 4' greater than the dimensions of the footing,
tling. I lad I known in 1977 what I know now, I could drive a 2 x 4 stake at the northwest corner (for
have saved about a third of the money spent on example) 2' in from each side of the flat area. Put a
concrete for the footings at the Cave. Let this be the nail in the top of the stake, leaving the nail head
first of several mistakes that we made which the sticking out an inch for tying the mason's line. Buy
reader can avoid. a ball of good nylon mason's line; you'll be using it

frequently for this project. Measure the length of ner, describe an arc with a radius equal to the
the footing along the north wall to a point (3511" in shorter footing d i m e n s i o n , 30'8Vfe" in our case.
our case), keeping about 2' in from the sloped Next, hook the tape on the nail at the northeast
edge of the excavation. Drive a stake into the corner- a n d describe a second arc equal to the
ground and a nail into the stake. Now figure the diagonal m e a s u r e m e n t 47 3" in o u r example. The
hypotenuse (diagonal measurement) of your foot- point w h e r e the two arcs intersect is the southwest
ing figures. We'll use our own figures as an exam- corner. Drive in a stake and a nail there. Find the
ple. Our footing dimensions are 30'8'/z" by 35'11" s o u t h e a s t c o r n e r by intersecting the east side
(30.71 x 35.92). Thanks to Pythagoras, we can cal- m e a s u r e m e n t with the south-side measurement.
culate the hypotenuse (c): Check the work by m e a s u r i n g the other diagonal.
The diagonals must be the s a m e in o r d e r for the
rectangle to have four s q u a r e corners.
P e r h a p s the rectangle you've laid out doesn't
use the cleared space to the best advantage. It
might c r o w d one of the excavation slopes, but it
might have plenty of r o o m on the adjacent side. It
doesn't take long to rotate the rectangle slightly to
alleviate this problem. You might even have to do a
little pick-and-shovel work if o n e of the sides
These calculations t a n be d o n e in a jiffy by doesn't have enough room.
using a c a l c u l a t o r h a v i n g t h e s c j u a r e - r o o t "Calculated guessing" will take a few trials to
function. get all four sides and the two diagonals to check,
Now hook your tape to the nail on the northwest but it beats making batter b o a r d s that you're only
corner and. in the g r o u n d near the southwest cor- going to use for one job. We actually had our four

corners to within Vz" in twenty minutes, and that's forced-hot-water system in rubber tubes laid in
accurate enough for the footing. the slab itself. In this case, a 5Vz"-thick floor is
lb get ready for the backhoe, place flags or recommended by manufacturers of such systems.
white stakes on the various bankings for the oper- People who have in-slab heating speak very highly
ator to use as guides. You can set these guide of it. Get information on products from your local
stakes by eyeballing. Sight from one stake to an- plumbing-and-heating supply store Be sure to in-
other and instruct a helper to plant a third marker sulate with extruded polystyrene below the slab.
on the banking in line with the two stakes. In all, I low much? At least an inch, but check with the
you'll place eight guide markers. supplier for the requirements of your locality.
lllus. 4 - 3 shows that the concrete floor is de-
signed to resist lateral pressures on the base of the
wall and the possibility of the first course of blocks
breaking loose from the footing. Although the ac-
How deep should you dig the footing trench? Fig- tual Cave footings w e r e 12" x 24", the drawing
ure this depth carefully. The important relation- shows a footing with a sufficient 9" depth. TWo
ship is the one between the level of the top of the inches of the concrete floor and 4" of compacted
footing and the top of the floor. The cross-hatching sand will be below the top of the footing. This
in lllus. 4 3 represents undisturbed earth. For un- leaves just 3" of additional material to be excavated
derfloor drainage, bring in 4" of sand to lay the to accommodate the footing track. This footing
floor on. track should be square, level, and about 30" wide to
1 like a 4"-thick concrete floor, but if you've got a easily accommodate the 24" footing, two forming
good honest 3" at the thinnest portions, you've got boards, and 2" of extruded polystyrene insulation.
a gcx)d strong floor. The advantage of a full 4" is Digging the footing track is probably best done by
simply more thermal mass for heat storage. The hand, unless digging conditions are poor because
exception to the 4" floor would be if you intend to of hard soils or large boulders. In such cases, a
incorporate an in-slab heating system, such as a backhoe can save a backache.

KEYED JOINT below the tloor instead of 4". Note that there's also
1" of extruded polystyrene (such as Dow Chemi-
There's another way of "keying" the first course of cal's Styrofoam"' Blueboard®) u n d e r both the foot-
blocks to the foundation. A keyed joint can be ing and under the floor.
created by setting a piece of wood flush with the
surface of the wet concrete, and halfway between
the inner and outer forming boards. A good key-
wav can be made by rip-sawing a 2 x 4 down the
center, using a circular saw. If the blade angle of
the saw is set at about 75° instead of 90°, an excel- W r a p p i n g the footings with extruded polystyrene
lent draft angle will be created so that the board is very important. At Log End Cave, we didn't
can be easily removed from the partially set con- include this detail and suffered condensation at
crete. Oiling the key way board is also strongly the base of the wall and about 6" in from the floor
recommended. I"he resulting keyed joint will look during w a r m moist conditions in May, June, and
like the one shown in lllus. 4 - 4 . Note that the draft early July, when the footing w a s still conducting

angle is kept to the outside of the footing. Later, the "coolth" from the still-cold soils at 7' of depth. Not
first course of blocks can be firmly tied to the until the footings w a r m e d up near the end of July
footing by filling the block cores halfway with con - did the c o n d e n s a t i o n disappear. W r a p p i n g the
crete. If I w e r e doing a rectilinear earth-sheltered footings (and floor) with e x t r u d e d polystyrene
house again, this is the method I'd choose, because stops this condensation, as we've proven at Earth-
the blocks wouldn't be in the way of screeding w o o d , w h e r e there's no condensation at any time
(flattening the top of the concrete so that it's even of the year. The Blueboard* used at Earthwood
with the top of the forming boards) and finishing k e e p s the fooling t e m p e r a t u r e above the d e w
the floor, as they were at Log End Cave. Also, using point. lllus. 4 - 5 shows the difference.
this method, it's really pointless to excavate foot- Under the footings, it's important to use genu-
ing tracks at all. The forms can be built right on the ine Styrofoam® Blueboard* which has a com-
floor of the excavation, providing the floor is flat pression strength of 5600 p o u n d s per square foot
and level. with only 10% deflection. At most, the load per
Another method of eliminating footing-track s q u a r e foot on a Log End Cave type house will be
excavation is to set up footing forms right on the less than half that, so deflection (compression) of
flat excavated site and use 7" of compacted sand the Blueboard® will be considerably less than 10%.

I don't know of any other extruded polystyrene the door. The footing below this portion of wall
foam with sufficient compression strength. Do not doesn't have the protection of 3' of earth, as does
use expanded polystyrene (also known as bead- the rest of the south-facing wall. So we decided to
hoard) around the footings or under the floor. increase the footing depth to 24" over a 12' section
A final word about the footing tracks: If a large in the s o u t h w e s t c o r n e r . This e x t r a c o n c r e t e
boulder has to be removed, be sure to fill the hole worked line; the foundation has never heaved in 17
with well-compacted earth or sand. Be sure to wet years. In theory, the frost wall should have de
such material for maximum compaction. Concrete scended to 48". the depth of frost penetration in
should always be poured over undisturbed earth. our area. But this p r o t e c t e d south-facing wall
If the subsoil is disturbed, mechanical compaction doesn't experience that extent of frost penetra-
is imperative. tion. Extra c a r e w o u l d b e i m p e r a t i v e f o r un-
protected footings on any other wall.
There's another method of protecting against
frost heaving which is being used m o r e and more
frequently. The method uses insulation to take ad-
FROST WALL vantage of frost's tendency to p e r m e a t e into the
In northern areas, it's necessary to prevent foot- soil from above at a 45' angle, lllus. 4 6 shows how
ings close to the surface from frost heaving. Heav- this works. The advantages of this system are case
ing can occur when wet g r o u n d beneath the foot- of construction and savings on materials.
ing freezes and expands, pushing u p w a r d s on the Extend a m i n i m u m of 2" of e x t r u d e d poly-
footing. The massive weight of the wall is of little styrene (R-10) away f r o m the frost footing to a
help. The expansion forces of freezing don't care distance equal to the local frost depth. This pro-
how much weight they're called upon to lift. The tects the cross-hatched area shown in lllus. 4 - 6 .
only solution is to make s u r e that the ground be- The rigid foam is protected from ultraviolet radia-
low the footing won't freeze. This can be d o n e by tion by covering it with 6-mil black polyethylene
going deeper with the footing, protecting it with and 3" of #2 crushed stone. Pitch the foam and
extruded-polystyrene insulation, or both. poly away from the h o m e for d r a i n a g e purposes.
The frost wall at the original U>g End Cave is Alternative methods of frost protection, shown in
that portion of the south wall which is entirely the insets, a r e very good, but they involve more
above final grade, 8' of wall near to a n d including excavation and replacement of soils.

lllus. 4-6. Because frost penetrates on a 45° angle from the surface, footings can he protected by the
placement of rigid-foam iniiilation. The insets show alternate methods of placement.

the north and south sides are 3" longer than the
FORMING footing, to allow the 1V1-"-thick east- and west-side
We used 2" x 10" forming boards kindly lent to us forms to butt against them. Similarly, the east and
by Jonathan Cross. These worked out in forming west forms on the inner ring are 3" shorter than
for our 12"-deep footing (we just let the extra 2" the inner measurement of the footing because the
concrete at the bottom take the shape of the track thickness of the planks of the other two forms will
itself), but they would have been perfect for mak- make up the difference. All of the other forming
ing a 9"-deep footing with an inch of Styrofoam* boards are consistent with the actual footing mea-
at the bottom of the form. surements. (The footing measurements given are
Preparing the forms takes time. First, figure out based on the location of the 12" blocks laid up by
exactly what lengths your forms need to be. No the surface-bonding method.)
room for error here; you don't want to cut a good It will be necessary to cleat planks together (II
plank wrong. I drew a diagram of the whole form- lus. 4 8) to make each full length of the form. Cut
ing system, showing clearly how the corners were cleats at least 3' long and use 16-penny scaffolding
to be constructed (lllus. 4—7). The long forms on nails to fasten the cleats to the planks. Scatlblding

lllus. 4-7. farming diagram showing the lengths of all footing forms

lllus. 4 8. dealing together two planks


lor duplex) nails have two heads so that the nails PLACING THE FORMS
can be easily removed when the forms are dis-
mantled. I use ten nails on each cleat, five for each
plank. Butt the planks tightly together and eyeball This j o b would be difficult without a contractor's
them dead straight with o n e another before nail level. Beg, b o r r o w or rent one. The most impor-
ing the cleat. Note that the required measurement tant consideration in setting up the forms is that
of our north and south exterior forms is 362". To they be level with each other. Set up the level at
make g o o d use of Cross' 18' planks, we cheated a some point outside the foundation w h e r e you'll
little and left them 2" short of butting against each have a clear view of a g r a d e stick held at each of the
other This is okay as long as you install a little four corners (lllus. 4 9). Using the existing corner
spacer (lllus. 4—8) into the gap. The spacer will give stakes as guides, bring in the longer of the north-
almost the same stability as t w o planks butting side f o r m s and put it roughly in place. Drive n e w
directly against each other. stakes into the g r o u n d at the northwest and north-
You'll need a lot of strong 2 x 4 stakes, about east corners, positioning them so that t h e / II be on
24" long. Allow five or six for cnch length of form the outside of the forms. Don't drive stakes on the
ing and a few extra for the o n e s you'll smash to side of the forming b o a r d s w h e r e the concrete is
splinters with the sledgehammer. I was fortunate poured. Drive all the n e w corner stakes so that
in borrowing stakes from Jonathan, but you can they're at the s a m e elevation, as judged by the
make stakes from economy-grade studs. Fifteen contractor's level. This will be about 10" above the
8-foot 2 x 4s will make sixty 24" stakes. Put well- average g r a d e of the excavation or of the footing
tapered points on each stake. tracks, depending on the relationship of the floor

lllus. 4-9. The most important consideration when

placing the footing forms is that they he level with
one another.

level to the footing. This relationship, in turn, de- into the ground, so that the point of the nail is just
pends on which method you've chosen for keying barely showing through the stake. Use a rock or
the first course of blocks. Make a cross-sectional sledgehammer to help resist the pressure as you
sketch (similar to lllus. 4—3) of this relationship, but drive the nails all the way into the forms. Put at
including all of the details of your floor; footing least one stake between the corners and the mid-
dimensions, insulation, keying method, etc. point, maybe two over an exceptionally long span.
lb get the average grade of either the founda- Check for level along the whole form
tion area in general or of the footing tracks, lake 12 One down, seven to go. The rest are done in the
readings at equally spaced locations, and average same way as the first. Complete the outer form
the results. before proceeding to the inner ring Make sure the
Nail the long form to the new stakes so that the diagonals check! The inner ring is placed to leave a
top of the form is level with the top of the stake. space equal to the width of the footing, 2" at U*}
Hyeball the form straight and drive a third stake End Cave. Make 26"-wide tracks to allow for the
into the ground about halfway along. Only the inch of extruded polystyrene on each side of the
corner stakes need to be at the same grade as the concrete.
top of the form, so that they c a n s e r v e as Make the inner ring level with the outer ring.
benchmarks to work from. Other stakes can be Check the whole job with the contractor's level,
driven slightly lower than the level of the forms, so moving every 8' or so around the forms. Make
that they will be out of the way of screeding when slight adjustments by moving the stakes up or
the pour is made. down. Use a lever to move stakes upwards, a
Now level the form. This chore is easiest with sledge to move them downwards. Hit the stakes,
three people: one to hold the grade stick, one to not the forms. It may be necessary to clear some
read the contractor's level, and one to pound the earth from beneath the forming boards to allow
nails. Again, use scaffolding nails, coming in from them to settle into the proper grade. Finally, nail
the outside of the stake and into the forms. Drive 2 x 4 buttresses every 10' or so to resist the tre-
the nails into the stake before you pound the stake mendous outward pressure exerted on the forms

lllus. 4 10. Movable cleats will keep the forms (shutteringI intact during
the pour.

during the pour. These buttresses can be footed supports you use, they'll b e c o m e a part of the
back to the walls of the excavation. Another means concrete, so choose something clean, strong, and
of resisting this pressure is to construct about a of the right height, about 3". Keep the rebar at least
dozen movable cleats (lllus. 4 -10). 3" in from the e d g e s of the forming track
We used W-thick iron silo hoops for rebar at
both Log E n d Cave and E a r t h w o o d These hoops
were left over w h e n we recycled fallen or leaning
silos for their s p r u c e planking. If s o m e o n e has torn
down a silo in your area, try to work a deal for the
hoops; they make strong, cheap rebar. Old hoops
usually have to be cut apart with a hacksaw, and
THE POUR straightening them can be a job, but the savings
With the forms firmly in place, it's time to install the can be w o r t h the effort. Yet, construction rebar
very important rigid-foam insulation. As previ isn't very expensive, so weigh time and effort
ously stated, the rigid foam at the bottom of the against money.
footing should be Blueboard® for its compression As accurately as possible, calculate the amount
strength. Other extruded polystyrenes (not ex of concrete you need. It sells by the cubic yard and
paneled) can be used along the sides, such as the going rate in 1994 in northern New York was
Dow's Grayboard®, but you might as well use the $60/cubic yard for 3000-pound-test mix. which is
Blueboard 1 for all purposes, as you need it under what you should be ordering. Mere are the calcula-
the footing, unless there's a significant cost sav- tions for t he Log E n d Cave footings. If you follow
ings with one of the other extruded polystyrenes. this example, you'll be able to do your own calcula
An added advantage of using the rigid foam is that tions for lootings of different dimensions.
you won't have to oil the forms for r e m o v a l On The v o l u m e o f c o n c r e t e e q u a l s the c r o s s -
buildings with no floor dimension greater than sectional area of the footing times the perimeter.
40', expansion joints aren't needed, and you could V = V x 2' x 126' = 252 cubic feet. Dividing by
just put the insulation 4" up from the bottom on the 27 gives cubic yards: - 9.33 cubic yards. We
inside of the forms. Old engine oil painted on to the added an extra cubic yard for our d e e p frost wall
top 4" of the inner forms will facilitate form re- footings at the southeast corner, which gave us
moval II floor dimensions exceed 40', just leave 10.33 cubic yards. Actually, o u r computation was a
the rigid foam right to the top of the inner forms little m o r e involved than this, since the footings
and the foam becomes your expansion joint. actually m e a s u r e d a b o u t 11" x 23" in c r o s s -
Reinforce the footings with Vt" rebar, two b a r s section. This adjustment rcduccd the figure to 9.3
throughout the footing. (Some building inspectors cubic y a r d s . The cement truck had a maximum
may require three pieces, but two is really suffi capacity of 9.5 yards, very close to our estimate, so
cient for the tensile strength required.) The rule of I called for a full load. We ended up with about a
thumb lor overlapping reinforcing rod is that join- w h e e l b a r r o w of concrete left over.
ing pieces should be overlapped by 40 times the It's better not to cut your estimate quite that
diameter of the rebar. So the overlap with Vz" rebar close. H a v e a place to put any leftover concrete,
should be about 2tf'. Tie the overlapping pieces some small project. Form a sidewalk, a trashcan
together with forming wire. Bend right angles in pad, a playhouse foundation, anything to make
some of the pieces for use at the corners. use of a n y excess. You can even make concrete
The rebar should be placed in the bottom half of p a v i n g s l a b s , a l w a y s u s e f u l f o r a v a r i e t y of
the footing pour, in o r d e r for it to lend its maxi- purposes.
mum tensile strength against settling. This posi- On the morning of the pour, apply a coat of used
tioning is shown in lllus. 4 -3. The rebar can be engine oil to any part of the forms in contact with
supported during the pour by pieces of broken the concrete. This coating will make removing and
bricks or small 3"-thick flat stones. Special wire cleaning the f o r m s much easier. If you borrowed
supports m a d e for the purpose, called frogs or the forms, you'll want to return them in as g o o d a
chairs, a r e commercially available. Whichever condition as you found them.

POURING THE FOOTING work, especially stiff concrete. In o r d e r to prevent

Round up plenty of help on the day of the pour. If voids, try to vibrate the c o n c r e t e d o w n into the
they're fairly sturdy individuals, a crew of four or forms with your tools. After one side has been
five is sufficient to d r a w the concrete around the poured, one or two folks can start to screed the
ring (lllus. 4—11). Have a few strong garden rakes concrete b e f o r e it sets too m u c h . Screed with a
or hoes on hand. A metal rake is the best tool for 2 x 4 (lllus. 4 12), d r a w i n g the c o n c r e t e along
pulling concrete along between the forms. with a constant backward and forward sliding
You'll be charged for overtime if the cement- motion. You can use I l ie screed board on e d g e for
truck driver has to stand around and wait for last- the initial flattening, and then for a relatively
minute preparations. Have all the forms set and smooth surface give it a n o t h e r screeding with the
supported, and the insulation and rebar in place. Side of the 2 x 4 . If you intend to use the keyway
Make sure access to the site is good. Have a wheel- method shown in lllus. 4 - 4 , now's the time to in-
barrow and planks ready to wheel concrete to sert the oiled keyway pieces. Push them right into
awkward spots. Ideally, the driver should be able the concrete in the center of the track and screed
to chute the concrete to any place in the forming. over the pieces, leaving the top w o o d e n e d g e
Ask the driver f o r a stiff mix; a 3" slump is good, showing on the surface.
should he ask you for such a figure. Slump refers to Can you home-mix concrete for the footings or
how much the concrete sags when a test cone is the floor? I won't mix my o w n concrete. The sav-
removed from a f r e s h sample. The more water the ings aren't really very great, especially if any value
easier the work, b u t the weaker the concrete. at all is placed on your time. A n d a footing poured
Soupy concrete can't be the 3(XX)-pound-test you over several days won't be as strong, because of all
ordered and might be only half as strong. the "cold joints" between sections.
There's not much to say about drawing the con Clean all your tools at the conclusion of the
crete around the f o r m s except that it's d a r n e d hard pour. The cement-truck driver should have a spray

lllus. 4 11. A crew of tour or five, each with a good strong rake,
should be sufficient to draw the concrete around the looting ring.

hose right on board which he uses to clean his next day; and they should be easy to remove if
chutes, and he'll let you spray off your rakes, you've included the 1" rigid-foam buffer up against
shovels, hoes, and wheelbarrow: the forming boards. Clean the forms and stakes for
reuse, particularly if they belong to someone else.
It took six of us two hours to d r a w and level the
concrete. This is as it should be. You don't want to
pay overtime for the cement truck. Jaki and I re-
Concrete dries at varying rates depending on its moved the forms the next day, breaking one which
strength, stillness, the air temperature, and the was held fast at a place where the concrete had
humidity. You can usually remove the forms the leaked.

lllus. 4-12. The concrete can be screeded (made level with the top of
the forming) with 2 x 4s.
The Floor
then able to push and backblade the sand onto the
PREPARATION floor. Using the footing as a guide for levelling, the
The preparation for each j o b is usually m o r e work operator did a g o o d j o b of spreading the sand
than the j o b itself. We poured the footing on July evenly o v e r the area. He'd d o n e all he could in an
11th. Although we'd hoped to be ready for the floor hour, and we took our time finishing the job by
pour by the 16th. we weren't fully prepared until hand, checking with the contractor's level now and
July 23rd. It's incredible how many things there then. T h e next day 1 hired a power c o m p a c t o r and
are to do before the floor can be poured. tamped the sand, with Jaki watering the area be-
Contractors will often pour the footing and the tween tampings, using a watering can, since we
lloor together on the same day. This concrete mass had no r u n n i n g w a t e r on-site. This power com-
is known as a monolithic slab. The footing portion, pacting is very important to give the floor a solid
sometimes referred to as a thickened edge, is base. Tamping dry sand is an exercise in futility, so
merely a thicker part of the floor pour. A 4" floor be sure to soak the sand b e t w e e n tampings. I like
with a thickened e d g e of 8" or 9" deep and 16" to 24" to hit each s q u a r e foot at least t h r e e times with the
wide is c o m m o n . In fact, we p o u r e d both o u r compactor.
summer-cottage slab and my workshop slab at
L'arthwood monolithically. But I don't r e c o m m e n d
this method to first-time builders on a full-size
house foundation. The task should be broken
down into two more manageable bites, footing
one day. floor a different day. The kinds of prepara-
tion are different, too, and there's a great advan- Our 12" concrete block walls are laid up by the
tage to using the completed footing as a sturdy surface bonding method. In brief, the blocks are
and uniform e d g e for screeding the floor, an ad- dry-stacked without mortar, and a strong stress-
vantage you won't have if you pour a monolithic skin coating of surface-bonding cement is applied
slab. to each side of the wall, bonding block to block.
Because of our plan of footing the base of the This c e m e n t coating is saturated with millions of
block wall with the top 2" of the floor pour, it was tiny glass fibres, which impart a t r e m e n d o u s ten
necessary to lay the first course of blocks all the sile strength to the wall. The tensile strength of this
way around the building. If I w e r e to build a similar monolithic m e m b r a n e applied to the wall has been
building today, I'd choose the keyway system found to be six times stronger than the tensile
shown in lllus 4 4, but in this narrative, I'll follow strength of a conventionally mortared wall. De-
what we actually did. tails of construction are given in the next chapter,
For good under-floor drainage, and before do- but laying up the first c o u r s e will be discussed
ing the block work, we spread 4" of sand over the here, b e c a u s e a completed first course w a s neces-
floor. By July 14th. the footing was aged enough to sary to p o u r i n g the floor.
build a r a m p of sand over the frost-wall portion. With s u r f a c e - b o n d e d blocks, the first course is
We brought in two loads of sand and dumped laid on the footing in an almost conventional man-
them just outside the "door." The bulldozer was ner: on a 3/a" to Vz" m o r t a r joint. This mortar's

purpose isn't to "glue" the wall to the foundation, need for cutting any blocks. When using surface
but rather to accurately level the first course for the bonding, the true size of the blocks must be used in
later placement of dry-stacked blocks. The only calculations. Standard 16" blocks, for example,
difference between the t r e a t m e n t of the first area actually 15%" long by 75/s" high. lable 2 gives
course of surface-bonded blocks and the first true dry-stacked wall dimensions.
course of blocks in a "normal" wall is that in a Note that 1/i" is added for each 10' of wall to allow
surface-bonded wall no mortar is placed between for nonuniformity of blocks. Our block plans were
adjacent blocks; the blocks are butted tight against based on size and space requirements of o u r gen
one another. eral floor plan, as well as the availability and di
I'he footing dimensions of Log End Cave ac- mensions of certain materials, especially the roof-
commodate full courses of blocks without the framing timbers. For greater lateral stability, we

chose 12"-wide blocks, instead of 8" or 10" blocks. easy to cut 4" off a 4 x 8 x 16 solid. Mark the block
The east-west length of the Cave, then, is 26 whole with a pencil and score the pencil line (both sides
blocks and one block laid widthwise, shown in of the block) with hammer and chisel. Go around
lllus. 2 - 2 We can break 26 down to 13 blocks twice the block again, hitting the chisel a little harder.
and use the chart: 16'UVfe" x 2 = 33'10W. Add 12" Before long, the block will break right along the
for the block laid widthwise (this is the block that scored mark.
"turns the corner") and add W to allow for the Let's discuss blocks: the various sizes and mate-
nonuniformity of the blocks, and a total length of rials, how they fit together, cost, and so on. We
34' 11" is obtained. Similarly, the cast and west walls chose V<t"-wide concrete blocks: 12" Tor a stable
can be computed: 14'37/h" (11 blocks) + 14'3%T (11 wall, concrete for strength. Cinder blocks are
blocks) + 12" (the turned block) + 3A" (nonunifor- lighter, cheaper and easier to use, but they dion't
mity margin) - 29'8fc". The south wall has the compare with concrete for strength. The blocks
same number of blocks as the north wall, except we used have three full cores and two half-cores in
that five blocks are left out where the door panel is them, except for the corner blocks, which have the
located. space filled where the half-cores would normally
be (lllus. 5-1). The regular block weighs 60 to 65
pounds, corner blocks about 5 pounds more. You
BLOCKS could use 8" blocks, but with these ifs necessary to
The first task in laying the first course of blocks is build a pilaster every 10' around the perimeter of
to mark the corners. Use methods similar to those the wall (lllus. 5-2) to resist lateral pressure.
already described for establishing the corners of We used 8" blocks and pilasters in the basement
the footing. In theory, the outside e d g e of the block of our first home, Log End Cottage, and we g r e w
wall will be 6" in from each edge of the footing, but to dislike the pilasters, which got in the way of the
ifs advisable to obtain W accuracy, so fiddle with floor plan. They're also a pain to apply surface
the tape and your marks until all the walls are the bonding to. They slow production enormously,
right length and the diagonals check. Now snap a both w h e n laying t h e m a n d w h e n s u r f a c e -
chalkline on the footing between cornei^ marking bonding. We decided to use stable 12" blocks and
the outside of the wall. Stick to your block plan to we filled the cores with vertical rebar and concrete
get the first course of blocks in the right place. It's at every location that would normally take a
necessary to cut one 4" x 8" x 16" solid block pilaster.
down to a 12" length to complete this and every Ten-inch blocks are offered by most block com-
course. By using a hammer and cold chisel, it's panies, but I can't recommend them. Their awk-

lllus. 5-1. Standard 12" concrete blocks, weighing about

65 lbs. each, have half-cores at the ends; corner blocks are
filled at the ends and weigh about 70 lbs.

IIIus. 5-2. A typical pilaster tor strengthening a wall or supporting a beam.

Use this pilaster if you'll be using 8 concrete blocks.

ward dimensions mean constant cutting of blocks, used with a block plan similar to o u r own Cutting
and you still need those pesky pilasters. All blocks, blocks, especially cored blocks, is difficult. A little
by the way are 15%" long by 7%" high. adjustment to the dimensions of your house can
Uniform blocks m a d e especially for surface save time. We end each course with a block turned
bonding are available in some areas. The blocks widthwise (lllus. 5-3), instead of having a whole
are passed through surfacc grinders al the block number of blocks on a side.
plant, assuring uniformity. They are, of course, Ask for a volume discount. You could set your-
more expensive than "normal" blocks. self up as a contractor and open a charge account
To avoid cutting blocks, 12" blocks should be with the block plant; this approach could save you

lllus. 5-3. Adjust your plans a bit so that corners can be made without having to cut blocks

money You should get at least 10% off for 500 or bargain. In 1977, we paid $0.51 per block; in 1994
more blocks. Ask about haulage charges, too. the same block cost SI.10.
These charges usually vary with distance, but they
might also be a function of volume. If you're fortu-
nate enough to have two or more suppliers, get
comparative bids. Compare the blocks. With sur- Lay the first course of blocks in a rich mortar of 1
face bonding, it's worth a few cents extra for uni- part masonry cement, 1 part Portland cement, 5
form blocks. Poorly made blocks at $0.10 off are no parts sand, equal parts by volume. Make a stiff

mix, as for stone masonry, particularly if the foot- allowed quite a lean to develop in the first course,
ing is uneven and the mortar has to be thick in and it took considerable shimming later on to cor
places. rect it.
Start with the highest corner, but know how Keep the blocks tight to each other (lllus. 5 4)
much higher it is than other parts of the wall, so Make sure no mortar has crept in from below
you can estimate a practical thickness for the mor- between the blocks. A light tap with a hammer will
tar bed. Remember that the purpose of the first usually assure that two adjacent blocks are well
course is to establish a level base for the dry- butted against each other. When you lay the last
stacked courses, so the thickness of the mortar block, you may find that it doesn't fill the last g a p
bed may vary around the footing. Set up the four exactly You may have to make small adjustments,
corner blocks so that they're all precisely the same even opening tiny spaces (less than V«") between
level. Then tightly stretch a nylon mason's line blocks. I found it useful to check the length of the

lllus. 5-4. Keep the blocks tight against each other.

from the lop of one corner to the next. Stretch the first six blocks I laid, Lo find out how my work
line as light as you can and tie it to something that compared to the values given in Iable 2. Any ad-
won't move, like another 12" block. Eyeball the line. justments to the quality of the work could then l>e
There will probably be a Vis" dip. Keep that in mind made. Remember that we allowed an extra W for
as you proceed along the course, so that halfway each 1' of wall. I found this to be a generous al-
along, the block is, say a string's width above the lowance, but another company's blocks might Ix.-
line Check each block for level as you lay it. The slightly different. We set a 4" x 12" block (cut from
string will tell you the level along the length of the a 4" solid) inside the last full-size block on the south
wall, but use a two-foot level to check along the wall to establish the b a s e for a short pillar, which
wall's width. This checking is important. Laziness strengthens the end of that wall. This can be seen
here will cost you time later. On our north wall, I in the block plans in Illustrations 2-2 or 2-5.

I'd never laid blocks before, though I'd worked barn timbers. Each of these posts must have a
for masons, but it only took about ten hours to get pillar footing beneath it to support the heavy roof
a decent job. decent except for the aforementioned load without cracking the floor. We decided to
north wall, where I didn't use the level often make the footing under each post 24" square by 12"
enough. deep, although 9" deep would have been sufficient.
litis first course of blocks established the edge We measured for the location of each footing, scal-
of the concrete door at Log End Cave, and pro- ing off the rafter-framing plan, and with a stick
vided a flat level surface for sc reeding the floor. No marked the squares in the sand. Because the level
further blocks were laid until the floor was fin- of sand was already designed to carry a 4" floor,
ished. In fact the main part of the block order we only had to excavate an additional 8" to give us
wasn't even delivered until the floor was hard, so the 12" depth. Pick and shovel work. We hauled the
that the pallets of blocks could be set down on the material away with a wheelbarrow; although we
lloor for convenient access during wall building. used some of the sand to build up one or t w o low
We hauled the first 90 blocks for the first course spots on the sand pad. It was stiflingly hot the
with our little pickup truck, three heavy loads. week we prepared for the floor pour, and the dogs
Again, consider strongly the keyway detail al- found these pillar-footing holes (lllus. 5 5) to be
ready described. It's easier to pour the floor to the ideal spots for lapping up some of the cooler earth
same level as the footings, and all the blocks can be temperatures below the sand.
delivered to the site in one load. Although at the Cave we poured these pillar
footings at the same time as we poured the floor,
we varied a few details three years later when we
built Earthwood. There, we poured the lower half
PILLAR FOOTINGS of the 9"-deep pillar footings at the same time that
The three main girders at the Cave are well sup- we poured the perimeter footing, and we included
ported by posts cut from 8"-square and ^"-square an inch of Styrofoam® Blue-board* under the pour

lllus. 5-5. The bottom half of the pillar footings can be poured at the
same time as you pour the floor.

lllus. 5-6. The bottom half of the pillar footings lllus. 5-7. Styrofbam* and wire mesh is in place,
are poured and the concrete is left rough to make There's already Styrofbam• under the pillar
b good "cold joint." footing.

to stop much of the e n e r g y "nosebleed" to the away f r o m the site, either to a large soakaway hole
earth This meant that when we poured the floor a filled with stones or, better, to a point out above
few days later, we didn't have these awkward holes g r a d e . The exposed end of the d r a i n tubing should
underfoot to trip in (Illustrations 5 - 6 and 5-7). be covered with a vermin-proof screen, such as 1/a*
mesh h a r d w a r e cloth. The drain tubing I use is a
black, ribbed, flexible pipe with a white nylon or
fiberglass filtration sock to keep silts out of the
pipe. This tubing is cheap i n s u r a n c e against water
problems f r o m below. See lllus. 5 - 8 .
The sand under the floor provides drainage for Many people think that the 6-mil plastic often
water which might find its way under the floor for seen installed below a concrete floor is a water-
any reason. To carry the water away, install 4" proofing course, but its main p u r p o s e is to retard
perforated drain tubing (also called drain file) in the set of the concrete (the concrete won't d r y too
this drainage layer, a n d slope the tubing away to rapidly). When any cement p r o d u c t (mortar, plas-
some point d o w n g r a d e from the site. My method, ter, or concrete) dries too quickly, ifs p r o n e to
which we used again at Earthwood, is to use a hoe s h r i n k a g e cracking. Without such a moisture-
to draw a track in the sand for the drain tubing, proof layer, the d r y earth, sand in particular, will
spiralling around the sand pad in such a way that rob the moisture f r o m the c o n c r e t e so rapidly that
no point under the floor is more than a few feet it will be difficult to apply a g o o d trowelled finish
from the drain tile. T connectors are m a d e for join- before the concrete becomes unworkably stiff.
ing branch lines to a main line. I try to slope the In n o r t h e r n climates at least, the home should
drain ever so slightly towards the point w h e r e it have a thermal break to the soil below using 1" of
exits the floor under the footing. Outside the foun- extruded polystyrene beneath the pour. We didn't
dation, use nonperforated tubing, and continue it use the polystyrene at the Cave, but we should

lllus. 5-8. Perforated 4" tubing was laid to remove any water that
might accumulate under the floor:

have. I he floor could have been kept a few d e g r e e s

w a r m e r in the winter. Happily; e x t r u d e d poly- UNDER-STOVE VENTS
s t y r e n e ( S t y r o f o a m ® ) is close-celled, w h i c h
means that w a t e r won't p a s s t h r o u g h it. Sty- Wood stoves need air, or they'll choke from lack of
rofoam n can thus be used as flotation material for oxygen. We laid 4" n o n p e r f o r a t e d flexible tubing
rafts or floating docks. E x p a n d e d foam Ibead- under o u r floor and took it up through the floor
board) isn't r e c o m m e n d e d in horizontal applica- under the planned location of the t w o woodstoves.
tion"; underground, because such foam can be These t w o tubes connect with each other by a tec
come saturated. and travel u n d e r the footing to the lower of the two
Because of the impermeability of the extruded retaining walls near the door. The inlet is covered
foam, it's not necessary to use the 6-mil plastic as with W h a r d w a r e cloth. Supplying air directly to
well. In the southern U.S., if heat dissipation is the stoves from the outside prevents d r a f t s and
more important for comfort than heat retention, keeps the stoves f r o m robbing oxygen from the
eliminate the foam from under the floor, or confine interior air. Also, with the tube b u r i e d under-
its use to within 48" of the footing. In either case, in ground for 20' or more, the earth actually preheats
order to retard the concrete set, use 6-mil poly- the combustion air quite a bit on its way in.
ethylene over any exposed s a n d . Don't worry- 'IVy to avoid a "U" t r a p in this 4" intake line as it
about a few punctures in the plastic, as it really goes u n d e r the footing. Condensation can eventu-
isn't meant to function as waterproofing. ally fill the trap and stop the airflow. If you can't

avoid the " U " it might be better to use perforated ished floor level. Be s u r e to use the roughing in
pipe so that any condensate will leach into the distance (from the wall) for the actual toilet you
sand. We occasionally had to "bail out" the U -trap plan to use. Leave all o t h e r waste pipes sticking up
we'd e r r o n e o u s l y c r e a t e d u n d e r the south-wall a good foot above floor level and wire them to iron
footing by using a r o p e to snake a towel back and pipe or r e b a r to make sure they won't be upset
forth through the pipe. Avoid our mistake, but do during the pour. Cover exposed o p e n i n g s to pre-
provide direct air to wood stoves. vent sand or concrete f r o m getting into the waste-
plumbing system. Most of the waste plumbing at
the Cave is in the b a t h r o o m and consists of a
bathtub drain, a w a s h b u s i n drain, a vent stack, a
UNDER-FLOOR PLUMBING toilet receptacle, and a clean-out. There are all
Now's the time, the only time, to install the waste sorts of different connectors, elbows, expanders,
plumbing. Ours was a simple system. The main and clean outs on the market, but for ease of in-
waste line to the septic tank is a 4" Schedule-40 stallation by the o w n e r - b u i l d e r , I r e c o m m e n d
ABS plastic pipe. In the house, ifs c o m m o n prac- sticking with ABS or PVC Schedule 40 pipe. These
tice to tie all the exhaust lines from fixtures to a 3" plastic pipes are tough stuff, they'll probably out-
line (called the house drain) which expands to a 4" last cast iron, and they're easy to cut and glue. Be-
line outside the house. The toilet Hushes directly s u r e to use the c o r r e c t glue for the p i p e you
into the 3" pipe Other fixtures, such as bathtubs c h o o s e . Find a h a r d w a r e s t o r e or p l u m b i n g
and w a s h b a s i n s , should have V/z" outlet pipes, supply shop with a g o o d stock of parts and a
which join the 3" pipe with a Y-connector m a d e for knowledgeable clerk.
the purpose. A vent stack, necessary for p r o p e r The only other waste plumbing at Log Fnd Cave
Operation of the d r a i n a g e system, should be a 2" is the kitchen-sink outflow pipe. For convenience,
pipe. In large houses with lots of fixtures, there and to lessen the load on the septic system, we ran
may be a main vent stack of 3" or more going up this pipe u n d e r the footing and into the same soak-
through the roof. Check your plumbing design away that a b s o r b s w a t e r f r o m the under-floor
with the local building-code enforcement officer. drains. Check to see if such a layout is acceptable
I "he vent's purpose is that of a pressure equalizer. in your area by contacting the town, county or
Without the vent, sinks may drain slowly and toi- state health d e p a r t m e n t , whichever has jurisdic
lets won't flush properly. Odors can also find their lion. This simple g r e y - w a t e r system shouldn't be
way back to the house from the septic tank. used with g a r b a g e disposals.
We installed a clean-out just in front of the toilet Only waste plumbing is discussed here, as its
and hid the clean-out beneath the floor (lllus. 5-9), installation must be accomplished before the floor
enabling us to run snakes or r a m r o d s straight to is poured. Try to consolidate your waste plumb-
the septic tank from the bathroom, should that ing. If possible, the b a t h r o o m and kitchen plumb-
ever be necessary. One of Murphy's laws states ing should be on o p p o s i t e sides of a c o m m o n wall
that if you don't put it in. you'll wish you had. If you If you're fearful about all these connections being
do install the clean-out, you'll probably never need underneath 4" of reinforced concrete, then that
it An improvement, according to plumbing man portion of the floor containing the waste plumbing
uals and standard practice, would be for the vent could be boxed in a n d a w o o d e n floor installed
stack to be on the septic-tank side of the toilet. later, so that access c a n be gained by using a screw
All w a s t e plumbing must be set firmly in the gun instead of a jackhammer. On all the houses I've
compacted sand, so that no settling can occur. The built, I've taken the lime to install the waste plumb-
proper slope for all underfloor pipes in the waste ing carefully, c o m p a c t i n g a r o u n d each pipe, and
system is W p e r foot. This translates to a 1" d r o p testing the system b e f o r e pouring the concrete. So
over the length of a four-foot level, or Vz" on a two- far, no problems.
foot level. Supply plumbing is generally d o n e after the
For e a s e of connecting the toilet later on, the floor is poured. Before the fl<x>r pour, provide a
toilet receptacle should be placed exactly at fin- pipe into the h o m e f r o m your w a t e r source. A IVz"

1. toilet receptacle unit

1. to 3" right angle join
(from tub, vent, and basin)
3. 4" sewer pipe
4. 3 " t o 4" expander
5. 3" PVC pipe
6. 3" right angle join
7. 3" 45 angle
8. cleanout {3" threaded cap)
9. wooden cover
to. vent stack to outside
11. tub drain
12. PVC right angle
13. ^Y' PVC pipe to basin

lllus. 5-9. The under-floor plumbing

drinking-water-quality flexible pipe should be lation can be placed. To minimize energy "nose-
more than adequate. bleed," try to keep the foam sheets joined together
After the plumbing, under-floor drains, and as tight as possible. Use 4' x 8' sheets, easier to
stove-air inlets are installed, the 1" rigid-foam insu- use (and with fewer joins) than the more c o m m o n

2' x 8' sheets. Don't put insulation over the half- to use. At the Cave, we actually u s e d " g o a t
poured pillar footings. I .ook at lllus. 5 - 7 again. The fencing" of similar dimensions. Hundreds of feet
footings should already have Blueboard* beneath of the fencing ran through the w o o d s on our prop-
them. erty. Someone many years ago had made a corral
in the woods. All we had to do w a s cut the wire and
d r a g it out. Recycling cuts costs. This reinforcing
can be seen in lllus. 5-10. The regular stuff, used at
Earthwood, can be seen on top of the insulation
WIRE-MESH REINFORCING layer in lllus. 5 - 7 .

I'he purpose of the wire mesh is to hold the con-

crete together when it cracks. Most large floor
pours crack despite all the best efforts to prevent
it. Cracking is normal. We've had minor cracks in All that remains to p r e p a r e for the pour is to build
almost every floor we've poured, and so does ev- a f e u l v down the middle of the floor area to sup
eryone else. port the screeding planks. The fence cuts the nar-
The mesh is called 6 x 6 x 10 x 10 reinforcing row dimension of the pour in half, making screed
mesh. This means that the matrix is composed of ing possible. Drive 2 x 4 stakes firmly into the
6" x 6" squares, and the wire size is #10 in each ground and nail a row of 2 x 4s to these stakes so
direction. You can buy the mesh in rolls or in Hat that the top of the horizontal 2 x 4s is at the s a m e
sheets, usually 5' x 10'. The flat sheets are easier level as the top of the first course of blocks. Cut off

11 his. 5-10. Wire mesh holds the floor together if for any reason
cracks develop in the concrete.

any stakes sticking up above the top edge of the calculated 16—about 5% more. 1 think it's reason-
fence. P r e p a r e a c o u p l e of screed planks by able to allow about 5% extra c o n c r e t e in case of
notching out the a p p r o p r i a t e amount of a 2 x 10, m e a s u r i n g (depth) error. A n d r e m e m b e r : A 4"
6" in our case. We used the same 2 x 10s that were floor is plenty. If you've got an honest 3" every-
used a s footing f o r m s . The f e n c e a n d s c r e e d where, you've got a good floor. Use a depth of 5VsT
boards are seen in Illustration 5-11 and 5-12. With if in-slab heating pipes are to be installed.
the contractor's level, check the fence for elevation.


1 hope you don't experience the last-minute rush-
Finally, calculate the volume of the floor. With a 4" ing around that we had to go t h r o u g h . We were up
floor the cubic y a r d a g e will be one-ninth of the at 5:00 A.M. finishing off the plumbing, and we
square yardage (4736" = In our case, I knew by finished levelling the fence about noon, while the
taking several test m e a s u r e m e n t s that our floor cement truck was mixing on site.
would average close to 5" in depth, and figured Get the s a m e helpers you p r e s s - g a n g e d for the
accordingly: 5"/36u x 101 square yards - 14 cubic- footing pour, and toll them you've really got a job
yards. I added two cubic yards for the deep bottom for them this time. Experienced c o n c r e t e pushers
portion of our eight pillar footings, giving 16 cubic are hard to find. Again, for a s t r o n g floor, ask for a
yards. We knew the j o b would require two loads, stiff mix. With the tines of a rake, d r a w the wire
so we ordered 9 yards f o r the first load, in case our mesh up into the concrete as you p o u r (lllus. 5-11)
estimates were a little off. Nine yards did a little Keep the mesh within an inch or t w o of the Sty
more than half the floor. Eight yards on the second rofoam® to provide good strength and lessen the
load finished the job almost perfectly. Again, there d a n g e r of getting the power-trowel blades caught
was about a wheelbarrow-load left over. So we up in the mesh later on. 1 always employ a desig-
actually used about 163A cubic yards instead of the nated wire-puller-upper, to make s u r e that the j o b

lllus. 5 11. It should fx* someone's job to make sure that the wire mesh
is putled up into the concrete.

gets done. Mesh sitting on the bottom of the pour After the first pour, there was a g o o d hour's
won't do the job it's supposed to. wait for the truck to make its return. Wc* used the
Do one side of the fence at a time. As soon as time to catch up on screeding and to have lunch.
you've got a fair area poured, say, 20 percent of o n e After a while, though, the welcome rest became
side, stari screeding. We used a power screed, worrying. The first pour was setting rapidly and
which is nothing m o r e than a big vibrator to there w a s still no sign of the second load. We
which two wooden planks are attached. Using the covered the first p o u r with Styrofoam* sheets in
power screed w a s only moderately successful. I an attempt to retard the setting, w a t e r e d the e d g e
think our planks were too long. To d r a w the screed where the two pours would join, and inserted 16"
along required another m a n at each end of the iron reinforcing rods 8" into the e d g e of the first
planks helping to slide them along the fence and
pour to help knit the t w o p o u r s together. At last the
the first course of blocks. The power screed might
truck arrived and we w e r e back at work, refreshed
have been better if used on a shorter span or with
by the break. Still it w a s a little scary; if the second
smoother supporting r u n s Inot blocks), but we
found out later that t w o m e n with a single plank load had been delayed another half-hour, it might
could do as well as the power tool (lllus. 5-12), have affected the quality of the join. I'he moral:
saving a man's labor (and the cost of renting the Specify two trucks. Usually, the trucks are radio-
power screed, if we'd known). equipped, but if they're not, use the nearest phone
to shout for the second load as soon as you can
Water can be d r a w n to the surface and the floor
make a good estimate of w h a f s n e e d e d to finish
smoothed further with a "bull-lloat," also shown
the job. Using Styrofoam ' 1 or plastic beneath the
in lllus. 5-12. A bull-lloat is a large, perfectly fiat
flour (as wc- did at Earthwood), you won't have this
masonry lloat. usually m a d e of aluminum or mag-
nesium. Extension handles can be added to avoid problem, since the w a t e r won't be d r a w n into the
walking on the wet concrete. I borrowed a bull- sand as it w a s the day we poured the Cave floor.
float, but they can be r e n t e d inexpensively from a It w a s a long day. Drawing s o m e 17 yards of
tool rental store. concrete with rakes, screeding it. battling with

lllus. 5-12. A power served c a n be used on a short span of concrete, but

in this case two men screeding a single plank did just as well.

wire mesh, working around plumbing and under- set every few minutes to see if we could commence
stove vents it's tough. And you, the owner- power-trowelling. He impressed upon me the im
builder. are responsible for the success or failure portance of experience in the handling of the
of the project. You must coordinate the various power trowel, a machine which can run away with
jobs and make the decisions. I was extremely for- the operator if one of its four rotating blades digs
tunate to have Jonathan Cross on hand to keep us into the concrete. I w a s thankful that he was will-
right. His experience and takc-charge attitude ing to run the machine for us. I sprinkled water or
probably made the difference between a so-so cement as needed, and worked the edges with a
floor and the excellent floor we ended up with. If hand trowel. One trick in obtaining a smooth sur-
you can enlist the help of someone who has experi- face is to scatter a dusting o! dry Portland cement
ence with concrete slabs, j u m p at the chance. An on the floor. The power trowel then does an excel
other time, someone might ask you for the benefit lent job of producing a llat, smooth floor, and I
of your experience. highly recommend its use.
By five o'clock, we were ahead of the work and We finished after 10:00 P.M., illuminated by car
everyone went home except Cross. We had supper headlights. The next day w a s Sunday, and we
and took it easy for an hour, Jonathan checking the made sure it was a day of rest.
External Walls

WALL CHOICES volved in u n d e r g r o u n d - h o m e c o n s t r u c t i o n . I'm

reluctant to advise inexperienced builders to form
The external walls can be made of stone, rein and p o u r their o w n walls. The s t r e s s e s on the
forced concrete, or concrete blocks. Building with forms are incredible. O n e owner-builder of an un-
stone is labor-intensive. T h r e e families in our com- d e r g r o u n d house near us poured his own walls.
munity built stone b a s e m e n t s at about the same During the pour, there w a s a "blow-out" of the
time we built our u n d e r g r o u n d house. All three forms, and several cubic yards of concrete rup-
homes w e r e slightly smaller than the Cave. The tured onto the floor. This sort of disaster could
man w i t h m a s o n r y e x p e r i e n c e built with a spoil your whole day!
wooden form on the outside and free-form ma- We got bids f r o m three contractors to pour the
sonry on the inside. Working like a man pos- Cave walls, the low bid being $3000 in 1977. Triple
sessed, he completed his walls in three weeks. The that today. O u r 12" block walls as described herein
other two families spent the entire s u m m e r on cost us $911 in m a t e r i a l s , i n c l u d i n g s u r f a c e -
their walls, and I doubt if either of the basements bonding cement, and $150 for hired help. While
would meet waterproofing requirements for fin- poured concrete may be s t r o n g e r than surface-
ished u n d e r g r o u n d living space. bonded blocks, the extra strength doesn't justify
As much as 1 like s t o n e masonry, I have s o m e the much higher cost. We're v e r y happy with
doubt about the e c o n o m y of building with stone surface-bonded walls, and we used the technique
as opposed to blocks. O n e of the couples, using the again at Earth wood. Finally, a p o u r e d wall takes
slip-form method popularized by Helen and Scott quite a while to fully cure, adding to the home's
Nearing (authors of Living the G o o d Life), used humidity for a year or two.
over two hundred bags of cement. They hauled in People ask me about using c o r d w o o d masonry
sand from a local pit for free and had old stone for below-grade walls. I'm not enthusiastic. The
walls near their site. Using stone m a s o n r y is frac- secret to c o r d w o o d ' s longevity is its ability to
tionally cheaper than using concrete blocks only if "breathe" along the end grain. W h e n one side of
a zero dollar-value is placed on labor. Finally, the the c o r d w o o d is w a t e r p r o o f e d , this b r e a t h i n g
outer surface needs to be purged (coated) with a ability is lost. Also, c o r d w o o d is weak on tension.
smooth layer of cement plaster prior to applica- Its resistance against lateral pressure, then, is low.
tion of the waterproofing membrane. A curved wall a d d r e s s e s this problem, but the
Poured concrete is the c o m m o n material used Issue of lack of breathing still remains. Finally,
by most of the well-known advocates of under- c o r d w o o d - m a s o n r y walls, while beautiful, are
ground housing. They're good architects, for the light-absorbing. A bright white wall maximizes
most part, but they d o n ' t actually p e r f o r m the light in places w h e r e natural light is minimal. Mal-
physical labor. For the owner-builder, my view is colm Wells says:
that the e x t r a o r d i n a r y f o r m i n g involved with
poured concrete is hardly worth the gain in struc- One of the easiest ways to stop wasting energy comes
tural strength, particularly since the methods de- in buckets It's called white paint, whic h, when applied
scribed in this book a r e already stronger than to darker walls, makes rooms so bright you need only
vvhafs needed to bear the loads and stresses in- half the light bulbs you formerly used ;

This leaves blocks. A conventionally mortared without mortar, block tight against block. Each
and a surface-bonded wall cost about the same block should cover the join b e t w e e n blocks of the
when all factors are considered. The mortared- previous course. The tops and b o t t o m s of blocks
block wall requires m o r e skill and more labor than should be cleaned of b u r r s by rubbing blocks to-
does the s u r f a c e - b o n d e d wall, yet the surface- gether, or by rasping them with a piece of broken
bonded wall is much s t r o n g e r than is the mortared block (lllus. 6-1).
wall against lateral pressures. For a home with Stack the blocks three courses high at the cor
straight walls, use 12" concrete blocks in conjunc- ners (lllus. 6 2), and check for p l u m b using the
tion with surface bonding. This method is strong, plumb bubble of a 4' level. I"hen fill in between
m<xlerate in cost, easily learnable, and relatively corners with blocks (lllus. 6-3), always working to
fast, and it supplies an excellent base for water- a mason's line stretched tight b e t w e e n corners.
proofing outside a n d painting inside. These are Repeat the p r o c e d u r e with a n o t h e r three courses
the reasons I accent surface-bonded block walls of blocks, and so on.
here. If a little wobble is perceived in a block, shim
one of the c o r n e r s with a thin metal shim cut f r o m
a piece of aluminum offset-printing plate or alumi-
BUILDING SURFACE-BONDED num flashing. Use your level to d e t e r m i n e which
of two c o r n e r s is best to shim. Sometimes a dou-
bled or folded shim may be necessary.
My nephew Steve Roy and his friend Bruce Mayer Buy r e a d y - m i x e d s u r f a c e - b o n d i n g c e m e n t ,
arrived at Log End during the wall construction. available f r o m several different m a n u f a c t u r e r s
They were to become indispensable additions to (Conproco, W. R. Bonsai Co., Quikrete, Stone-
the work team d u r i n g the next three weeks. Mountain Manufacturing Co.). Ask at your local
With the first c o u r s e in place and level, succeed- m a s o n r y - s u p p l y yard for surface-bonding ce-
ing courses are dry-stacked; that is, they're laid ment. They may call it block bond, a holdover from
the days when there w a s a p r o d u c t by that name.
On a single-storey h o m e like U>g End Cave, you
can actually stack all of the blocks b e f o r e any
s u r f a c e - b o n d i n g c e m e n t is a p p l i e d . With 12"
blocks, you'll be impressed at h o w strong and sta-
ble the wall is, even in its dry-stac ked state.
Soak the wall with water (lllus. 6 - 4 ) before you
mix the bonding cement. This soaking prevents
overly rapid drying of the surface-bonding ce-
ment. Tlie wall should be saturated, but not drip-
ping. Now mix the cement according to the manu-
facturer's instructions. The c e m e n t should be quite
wet, like very thick paint, but not so wet that it falls
off the wall during application.
Apply the surface-bonding cement to the wall
using a flat trowel as shown in lllus. 6 - 5 . Load the
flat trowel using a pointed m a s o n ' s trowel. Apply
the mix with firm trowel pressure, pushing the
load upward and outward until a uniform cover-
age is attained. To even out the plastered area and
to s p r e a d e x c e s s c e m e n t , follow with longer,
lighter strokes, holding the trowel at about a 5°
angle to the wall surface. Don't over-trowel, as this
lllus. 6-1. Before slacking, a piece of broken block can cause cracking or crazing. It takes a little prac-
can be slid across other blocks to knock oil excess tice. but the t e c h n i q u e can be l e a r n e d fairly
materials and burrs. quickly.

Illustrations 6-2 and 6-3. A lter the first c o u r s e of blocks is in place and
level, build the surface-bonded wall by stacking the blocks three
courses high at each corner (above); then fill in between t h e c o r n e r s

f/tu.s. 6 - 4 . Befoiv applying the surface bonding,

upmy the wtdl w j i h wviter unfif it's wet, but nuf

If] us. G-5. WbHt the m i x on to the watt a p l a s t e r e r ' s frowc-J,


The inner surface of a rectilinear u n d e r g r o u n d

building should have an a v e r a g e s u r f a c e - b o n d
coating that's W thick, which d e p t h can be A good way to impart extra tensile strength to a
checked with the c o r n e r of your trowel (until you long, straight earth-sheltered wall is to incorpo-
get the knack). The side of the wall facing the earth rate a bond beam on the final c o u r s e (the 11th
need only have an honest Vio" coating of cement, as course on the 40 x 40 plans). The b o n d beam is
the backfilled side of the wall isn't in tension. The made by laying out a c o u r s e of special bond beam
wall can only collapse inward, not outward. Above blocks, shown in lllus. 6 - 6 (an important drawing,
grade, use a Vfe" coating on each side of the wall, containing much information. This illustration will
because you can't be s u r e in which direction fail- be referred to again during discussion of roofing,
ure might occur. w a t e r p r o o f i n g , and insulating). Bond-beam
After application, mist the surface-bonding ce- blocks have a U-shaped cross-section. The trough
ment frequently, at least twice within the first 24 formed by several consecutive blocks is filled with
hours, more frequently if the walls are in the sun. c o n c r e t e a n d V?" rebar, as s h o w n . This solid
Keeping the wall w e t will g r e a t l y r e d u c e the concrete bond beam is very s t r o n g in resisting
chance of shrinkage cracking or crazing. Thy to lateral pressure. Even m o r e tensile s t r e n g t h can be
work in the shade, if you can, paying attention to gained by incorporating a b o n d henm on another
the sun's location at different times of the day. course about halfway up the wall.

Concrete for filling the bond-beam course, or trough and hooked u n d e r the rebar. as shown in
for filling block cores w h e r e the vertical rebar is to lllus. 6 - 6 . Use a 12" a n c h o r bolt having o n e
be installed, can be made in a wheelbarrow; using threaded end and an ell bend at its opposite end
either a premixed concrete mix such as Sakrcte® The ell assures that the bolt can't be pulled out
in bags, or by mixing your own concrete, using a o n c e the concrete is set. Leave the bolt exposed
recipe of 3 parts #1 stone, 2 parts sand, and I part above the wall a distance equal to the thickness of
Portland cement. the plate to be used.
O n e way to make sure that the bolts are plumb
and at the right height is to make a template out of
a 12" piece of the s a m e material f r o m which the
plates are made, 2 x 12s, for example. Drill a hole
SURFACE BONDING AS the s a m e diameter as the bolt, or slightly larger.
WATERPROOFING Take care that the hole is perpendicular to the top
of the plate. This template can now be used to
While surface-bonding cement has been rated ex-
check both depth and plumb.
cellent for resistance to wind-driven rain at veloc-
Plan the location of the a n c h o r bolts based on
ities up to 100 and has withstood a 4' hy-
the lengths of the planks you'll use for plates. Two
drostatic head with no w a t e r p e n e t r a t i o n , it
bolts are sufficient for short plates; use three for
shouldn't be considered to be an absolutely water-
plates over 9' long. Keep the bolts about 12" in from
proof membrane for u n d e r g r o u n d housing. The
each end of the plate.
problem with any cement waterproofing is that its
1 like wooden plates that are at least 2" thick.
integrity d e p e n d s upon there being no break in
Rough-cut 2 x 12s are perfect for the Cave plan
the crystalline structure of the waterproofing ma-
being discussed. O r d i n a r y dimensional 2 > 12s
terial If t h e r e a r e a n y c r a / i n g , s e t t l i n g , or
are actually only ll'/V' x \W, which makes them
shrinkage cracks, the m e m b r a n e is broken. 1 low-
just a little n a r r o w at the top of a 12" block wall. At
ever. surface-bonding cement certainly provides
the Cave, we used 4 x 8 plates on the easi and
an excellent base for the application of a sheet
west walls, to a c c o m m o d a t e the 4 x » r a f t e r s
membrane, and may have some value as a second
(roof joists). The 40" x 40' plans, however, call for
line of defense under suc h a membrane.
5" x 10" joists. Moreover, I now place all rigid
foam insulation on the exterior of the roof- and
wall-waterproofing m e m b r a n e s , so another ad
justrnent has been m a d e from what we actually
did at Log End Cave. The n e w detailing is shown in
The changeover f r o m m a s o n r y to timber con- lllus. 6 - 6 .
struction is accomplished by the use of wide After the anchor bolts a r e permanently set in
wooden plates {sills) bolted to the top of the ma- the concrete (allow t w o days), locate the position
sonry wall. With a stone wall, the anchor bolts of the holes in the plate by laying the plate on the
must be built in with the last course of stones. With anchor bolts and whacking o n c e sharply with a
a poured wall, the bolts are set into the concrete heavy h a m m e r (lllus. 6-7). Have a helper hold the
while ifs still workable. There are two ways of plate steady and in the correct position while you
setting the anchor bolts into the blocks' cavities, do this. The indentations formed mark the location
and we employed both, depending on the location of the holes to be drilled. Tbrn the plate upside
of the bolts. In corners and other locations where d o w n and drill holes of the s a m e diameter as the
the cavities were completely filled, the bolts are set bolts, being careful to keep the b r a c e and bit (or
in the concrete while it's still workable. If an anchor drilli straight up and d o w n (lllus. 6-H). Next, chisel
bolt is required between filled-core locations, stuff out a depression to a c c o m m o d a t e a washer and a
a wad of crumpled n e w s p a p e r 18" into the hole and nut (Illustrations 6 - 9 and 6-10). Then lay a thin
fill it with concrete. If a b o n d - b e a m c o u r s e of layer of flexible polyethylene sill-sealer on the top
blocks is used to finish off the top of the wall, the of the wall w h e r e the plates will go, and bolt the
anchor bolts can be set right in the concrete-filled plates d o w n with a crescent- or socket wrench.

Illustrations 6 7. 6-8. 6 9, and 6-10. After the anchor holts have set in the concrete, locate the holes in
the plate by laying the plate on the anchor bolts and whacking once sharply with a sledgehammer. I'he
indentations in the plate mark the locations of the holes to be drilled. Turn the plate upside down and
drill holes the same diameter as the bolts. Chisel out a depression to accommodate a washer and n u t .

compressing the "sill seal." The sill seal, which you mortaring blocks as we did at the original Cave,
can buy at building-supply yards, protects against are that the inside looks better (you don't see bare
infiltration by drafts and insects. concrete blocks) a n d installing a riser piece is
To finish the wall, 1 recommend the installation much quicker than mortaring blocks.
of a riser piece, on edge, as shown in lllus. 6-6. The only drawback is that the wooden riser
This riser piece will be made of two or three pieces might be m o r e expensive than working out
lengths of 4" x 10" timbers butted up to each other a detail using blocks. If time is money, though, the
to match the length of the 30" or 40' sidewalls. The little extra cost, if any, is more than offset by the
riser pieces should be somewhat dry, so as to not advantages. Note the chamfered edge at the top of
trap a lot of moisture after waterproofing, and the riser piece, so that the waterproofing mem
they can be "toe screwed" to the top plate, or brane can more easily bend around this corner.
fastened with angle-iron braces. In this detail, the The planking on t o p of the 5" x 10" rafters comes
advantages of using a riser piece, as opposed to to the same level as the top of the riser.
Timber Framework
(morrises) that were chiselled out for niortise-
POST-AND-BEAM and tenon joints.
Post-and beam construction is my favorite fram- • Old b a r n t i m b e r s a r e no l o n g e r g r e e n and
ing technique, not only because it has the strength they've finished shrinking. Using dry wood may
to support the kinds of loads dealt with in under- not make a lot of ditlerence with s o m e kinds of
ground housing, but b e c a u s e one can feel the construction, but o u r intention was to use a lot of
strength in an almost ethereal way, particularly in cordwood m a s o n r y to fill many of the panels
structures w h e r e the f r a m i n g is exposed. The framed by the posts and beams. Green w o o d will
same holds true for plank-and-beam roofing, the shrink away f r o m the m a s o n r y infilling. The
natural companion to p o s t - a n d - b e a m framing. wood m a s o n r y units used with this style of con-
Most people like the rustic atmosphere of old inns, struction a r e called loq ends; thus "Log End
vvitli their exposed hand-hewn timbers enclosing Cave."
plaster panels. 1 think we pick up a sense of secu- • I like the character of the old beams: the mortises,
rity from seeing those massive beams overhead. the adze marks, the dowel holes. The w o o d ' s
We know that they aren't going to fall in on us. character is imparted into the atmosphere of the
finished room.
Which old barn b e a m s to use is a value judg-
OLD BARN BEAMS ment. and I'm fastidious in their selection. We find
You can have timbers milled f r o m selected logs, or plenty of nonstructural uses around the h o m e for
you can use old hand-hewn b a r n timbers. I chose "rejects." Our strategy, which we've used at three
barn timbers for the post-and-beam framework of the houses that Jaki and I have built, is lo care-
and had 'ho roof rafters and ? x f i roof decking fully examine and catalog each beam, noting its
milled from straight hemlock trees which I se- dimensions a n d quality. Using this list as a guide,
lected myself, rhere were three reasons I decided the timber framework is then designed on paper. If
on barn timbers: we're a beam or t w o short, we seek out replace-
ments; we don't try to make a questionable timber
• They're incredibly strong. Most of the main sup- serve a structural purpose.
port beams during the 19th century were hewn
to 8" x 8" or 10" x 10" dimensions, and the trees
were f i n e r - g r a i n e d in the old days, b e f o r e
planned forestation was introduced to encour- ROUGH-CUT TIMBERS
age rapid growth. Of course, any beams that If you're building in an area with building codes,
show rot, whether from insects or moisture, the use of recycled timbers for structure may not
should be rejected immediately. Choose care- be allowed by the code-enforcement officer. New
fully, and store all wood up off the ground while rough-cut posts and b e a m s from a local sawmill
it's waiting to be used. With barn beams, stack are very strong, economical, and can be quite at-
them so that water cannot collect in the hollows tractive, especially if bleached after a year or two

in the building. Juki bleached many of the rafters he's not using anymore. We bought a silo for $100
and floor joists at Earth w o o d using a toothbrush Figure the b o a r d feet and make an otter. And don't
(!) and a 50% bleach-and-water solution. Some forget: Those old silo hoops make g r e a t rebar for
building inspectors may require that wood from a footings and block wall cores.
sawmill be g r a d e d (inspected and evaluated I. Find A big asset to building economically is being a
out before you start. good scrounger. Old recycled timbers a r e often
We looked for suitable rafters among old b a r n s superior to what's lw?ing produced today. What's
and houses that were being torn down in our area, more, you'd be saving trees.
and found a few rafters that would have d o n e the We built the Cave a r o u n d the three major sup-
job, but not enough of them were of consistent port members, 30' 10 x 10 barn b e a m s , which
size, so we decided to use new 4" x 8" hemlock we'd bought 25 miles away for a dollar a running
rafters. S h r i n k a g e on e x p o s e d r a f t e r s doesn't foot. They're beautiful timbers, exposed through
really affect the construction in any way, and, as out the house, a real design feature of the con-
we were able to choose o u r own straight trees, we struction. We tried to plan o u r post location to lake
knew the t i m b e r s would be s t r o n g . We also into account both structural considerations and
bought the wood for the planking from the same practicality with regard to the floor plan. We chose
supplier, a local sawmill. the beam with the biggest cross-section as the
A year or two later. I learned that hemlock is one center-support girder. I didn't want to span more
of the weakest woods in terms of shear strength. than 10' with an old barn timber, although it was in
Shear doesn't normally c o m e into play with light- excellent condition and w a s a full-size 10 x 10, s o l
frame construction, but with heavy loads it does. planned for two interior support posts, one out of
the way on the stone hearth, and o n e b e t w e e n the
Our hemlock's low shear strength was the weak
kitchen and dining areas. All t h r e e s p a n s thus
link in the entire structural system at Log End
created a r e less than 10'. Nowadays, 1 actually pet-
Cave. In fact, calculations later revealed that the
form the stress-load calculations b e f o r e building,
rafter shear strength w a s technically less than that
but I didn't know how to do that back in 1977.
required. The building never seemed to suffer any
for this, and I credit safety margins in both formu-
lae and in wood g r a d i n g as the reason for our
having no problem. The new 40' x 40' plans are INTEGRATING THE STRUCTURAL
engineered correctly for both shear and bending
for the spans and loads required. Have your plans
c hecked by a structural engineer. The floor plan was designed so that the north
Green wood should be stacked on site (with south internal walls would fall under the other two
wooden laths or stickers between courses) as long 30' beams, so they're each s u p p o r t e d by three
as possible before using it. Although shrinkage on major internal posts, dividing the depth of the
the rafters doesn't matter much (they don't shrink house into approximate quarters. We placed the
at all on length), you don't want Vz spaces between posts at the intersections of the internal walls, so
the roofing planks. If I w e r e doing the Cave again. that the 4"-thick walls could butt a g a i n s t the
I'd use seasoned s p r u c e or pine t o n g u e - a n d - 8" x 8" squared posts, leaving the p o s t s exposed
groove 2 x 6s. In n o r t h e r n New York, they cost in the corners of the peripheral r o o m s . The two
$730 per thousand b o a r d feet in .January 1994, peripheral 30' girders are positioned exactly be-
which works out to be about $1460 per thousand tween the center-support g i r d e r and the east and
square feet. With two-by material, one square foot west sidewalls, so that the clear s p a n s <)f the rafters
contains two board feet. are all consistently 8'. A big a d v a n t a g e of this plan
If you're lucky e n o u g h to find an old silo that's is that the long north-south internal walls rise up
being pulled down, as we did when we built Log to meet the underside of the big girders, so there's
End Cottage, you'll have good roof planking: old, no tricky fitting of the walls around the exposed
dry, full-size, t o n g u e a n d - g r o o v e 2 x 6 planks. rafters. The exception is w h e r e Rohan's room jut-
Spruce, if you're really lucky. Ask around. Some ted out into the living-room area, necessary to
farmer may be willing to sell a leaning silo which give him a reasonably sized space (lllus. 7-1).

lllus. 7—1. One room juts out into the Jiving room, the only
place where the walls had to be specially fitted around the
exposed rafters.

All the east-west walls rise up to meet the un into the girder decreases its shear strength by
derside of 4" x 8" rafters, except for the wall be- 10%. Just toenail the top of the post into the girder
tween the office and the master bedroom, which Even easier than nailing is to drill a pilot hole on a
meets the ceiling halfway between two rafters. 45° angle and install 4" steel deck screws, using a
There are two good reasons for integrating the cordless drill instead of toenailing.
structural and floor plans: Integrating the floor In 1993, I llnally p u r c h a s e d a hand-held re-
plan with the support structure saves slightly on chargeable land reversible) drill It has become my
materials, because the 8"-high rafter is used as a most-used tool. I hardly nail anything anymore. It's
part of the internal wall; and it results in some- easier and stronger to use screws, and screws can
thing that looks better than just missing a rafter be removed easily. A great variety of screws is
with a wall twhich would also form a great space available today, m a d e for every purpose, from
for cobwebs). Because of space considerations, hanging sheetrock to fastening deck boards
we varied slightly from this plan with the office- Since the south-wall post was in two pieces, it
bedroom wall, splitting the difference between was a bit trickier to figure. The base post, the one-
rafters in this case. standing on the wooden plate, had to be to
a c c o m m o d a t e the 48"-high w i n d o w s we'd
planned for the south wall. The header for the
windows was to be a 4 x 10 laid on its side, the 4"
SETTING UP THE POSTS dimension to resist against sagging, 10" to estab
The building plan called for a 22" space between lish the width of the cordwood-masonry wall later
the top of the block wall and the underside of the on. Hie short post on the top, then, was all that was
center girder. Subtracting 2" for the plate, we were left of the 95" of height established at the north
left with ;i 20" post to support the north end of the end. The subtraction was: 95" minus 28'// (the
girder (lllus. 7-2). We called the floor level and short block wall! minus 2" (the plate! minus
based the heights of the two internal support posts (the base post between windows) minus 4" (the
on the total height of the girder off the floor at the window header), leaving a short "post" of 12". The
north end, 95" (7'1D to its underside. We cut the calculations checked in the field. A mason's line
two interior freestanding posts to 96" (8'), so that stretched from the top of the north-wall post to the
we could recess them an inch into the underside of south-wall post showed one of the interior posts
the girder. I wouldn't do this again, as that 1" check sticking up 1". the other Vs. This was just right for

our intended purpose of checking or keying the location is established on the plate such that B
posts into the girder. The floor was level after all! B\ Now consider the total height of the supports
Having removed all nails and marked the for the 4 x 8 as measured from t he top of the block
rough-hewn posts as best we could with a square, wall. At the lower end, the rafter is to be supported
we cut one end of the post with a chainsaw and directly by the 4"-high plate. The high end of the
stood the post upright to check for plumb. If the rafter is supported by a 2" plate, a 20" post, and a
post tilted in any direction, it was necessary to 10" beam 32" in all. The height of X, then, should
recut the bottom, compensating for the error. Only be midway between the 4" and the 32" figures, or
when the bottom of the post was flat did we mea- 18". Subtracting 2" for the plate and 10" for the
sure for the cut at the top. You can use the square support beam leaves a 6" post. I cannot recom
again on the top measurement, but I've found that mencl actually using a "post" as short as 6", be-
it's less risky to measure the length of the post cause a heavy load might cause it to split open, just
from the good end along each of the four corner as a thin slab of wood is easier to split with an axe
edges, and then connect the marks. The irregu- than is a thick one. Instead, I used two short pieces
larities of a barn beam can throw the square off, cut from a scrap of an old 3 x 10; stacked one
and there's little room for error on the second cut. upon the other. These two short pieces worked
You won't have such problems if you use milled perfectly: They were exactly the right dimensions
beams. and were easy to fasten to the plate with 20-penny
Don't stand wood directly on a concrete floor, nails. A measurement from the floor established
because of the possibility of trapping dampness. the height of the internal posts, as discussed previ-
Cut a square of 90-pound roll roofing or asphalt ously for the center beam.
shingles to stand each post on. W. R. Grace's Bit- The support structure of our house is ac tually a
uthene' waterproofing membrane, described in very simple one and I hope that by examining in
the next chapter, is also excellent for this purpose. some detail the ways in which we met our prob
Phis damp-proof shim also helps to steady the lems, the reader will gain a sense of the principles
post, lb lessen the chance of someone being hurt involved, enabling him to solve the structural
by a falling timber, we waited until all eight inter- problems of a design that's somewhat different
nal posts were cut and squared before standing from the one he might wish to employ. The mathe-
them up permanently. Calculating the heights of matics involved is never more than simple arith-
posts that support the two peripheral girders is metic and the most basic geometry Keep a set of
best explained with the aid of lllus. 7 2. plans nearby, make notes on them based on actual

lllus. 7 2. Calculating the heights of the internal posts that will support the two
peripheral beams

lo make the best structural use of the 4 x 8 (not theoretical) measurements, and refer to your
rafters, I wanted to make the spans A and A' notes constantly, double-checking your figures
equidistant. This will be accomplished if the post from two different approaches, if possible.

When all eight internal posts were squared and

cut. we stood them up on their squares of roll
roofing at the correct p r e m e a s u r e d locations,
shimming, if necessary, to assure perfect plumb.
We made the posts rigid with a rough framework
of 2 x 4s (lllus. 7-3). We didn't attempt to fasten
the posts to the concrete floor. The tons of weight
that these posts support, as well as the internal
wall tie-ins, assure that the posts won't move once
the roof is on. The rough framework's purpose is
to hold the posts upright while putting the massive
10 • 10s in place We didn't dismantle this tempor-
ary framework until after the rafters were fixed.
The window and door header (or lintel) on the
south side of the house had to be put into place
before the small post for the south end of the cen-
ter beam could be permanently located. Hiis lintel
consists of a pair of 17-foot 4 x 10s, cut for the
purpose at our local sawmill. It took four of us to
maneuver the things into place, f h e south ends of
the other 30-fOoters would rest directly on the
headers at the points of their greatest strength:
right over stout 8 x 8 barn posts (see south-wall
elevational plan, lllus. 2-3).
The plan k-and-beam roof of our house is car-
ried by the east and west block walls and the three
aforementioned 10 x 10s. The dimensions of the
beams may vary quite ;i bit from beam to beam lllus. 7-3 After the internal posts are squared and
and from end to end, but the smallest dimension at cut. stand them up on their squares of roll roofing
any of the six ends is a full 8" x 9", and the average at the premeasured locations and shim, it
size is very close to a full 10 x 10. necessary, to assure perfect plumb. The internal
"The beams were stacked almost a quarter-mile posts can be kept in place with a rough framework
from the Cave site. Steve, Bruce, and I look very of 2 x 4s.
careful measurements of the locations of the posts
and transposed the measurements onto the actual
timbers. We'd chosen the heaviest of the three
beams as the center beam, and labelled the others surfaces of the beams were later creosoted to a
"east" and "west." uniform color.
We squared the ends of the beams. The total I went out to recruit some able bodies while the
length of the center beam is 30'3", which allows the boys finished up the checks in the beams to at com
ends to extend 4" beyond the support posts (for the modate the internal posts. As usual, it was a hectic
sake of appearance). The west beam was cut at day of running around making last-minute prepa-
29'H" to give a 2" overhang at each end. We rations. By late afternoon. I'd gathered four strong
wanted to make the east beam the same, but we bodies to supplement Steve a n d Brut e. We backed
could only get a g o o d 29'8" out of it. We solved the the pickup t<> the western beam, the lightest ol the
problem later by adding a 3" piece from another three. We figured that if we couldn't move that one,
beam, which in no way affected the structural con- we'd have no chance with the MX)-pound center
siderations. because the last rafter still had good beam. With laki looking after the baby, I was left
solid wood to rest upon. This cosmetic surgery is with the easy job: driving the truck. Six men lifted
only noticeable if it's pointed out, as all the external the north end of the timber while I backed under.

The beam was now tucked right up to the back of by all, we hauled it up and into place. Again, the
the cab. Then, paired at each end of strong ash checks were beautifully positioned, but, alas, one
branches, the three teams lifted the south end off wasn't deep enough. "It's gotta come off." I said.
the ground and shouted the command "Go!" Like Groans from the crew. "Maybe the post will com-
some great crawling insect, we started up the hill. press," suggested one of the tired backs hopefully.
Once, halfway up the hill, I was chastised by six We took the beam otf the posts and laid it to one
screaming banshees, in no uncertain terms, to side while I used a chainsaw to remove VH" from the
quit speeding. We purposely overshot the drive- top of a post. Steve did a little chisel work on the
way and backed up to the house site. We backed check. Back went the beam, this time fitting almost
the beam so that almost half of it was cantilevered perfectly. We'd actually overcompensated slightly
over the edge of the north wall, and then man- and the beam didn't rest on the post by Vu". Good
handled it from below into position on the posts enough. We knew the beam would sag W under
I lllus. 7 4). Ib our great delight, the post lit per- the roof load and rest on the post, as intended.
fectly into the checks. So much for the easy one. The last beam, by comparison, was easy, espe-
T he boys voted to tacklc the biggie next, reasoning cially since 1 was still driving the truck!
not unwisely that they mightn't be able to manage
it after hauling another of the small 500-pounders.
Not only was the center beam considerably
heavier than the other two, but it had to be raised
almost 2' higher. We manhandled it halfway out Our 4 x 8 rafters are the "beams" o f plankand-
over the hole, as before, and then put our strongest beam construction. Why 4 x 8s? A 4 x 8 is proof
and tallest man. Ron Light, below the beam, cat- against twisting under a weight load. The base of
walking with Bruce along some makeshift scaf- the rafter is so wide that it sits unassisted on the
folding. Almost single-handedly. Ron passed the support beams. Toenails are sufficient to keep it in
south end up to where Paul and I could get our ash place. With two-bys. ifs necessary to cross-brace
branch under it. Then, with a monumental effort adjacent rafters with metal ties or wooden blocks

lllus. 7-4. The beam was then cantilevered over the north wall and
positioned on the posts.

recommended with the 4<)' x 4<)' plans, greatly

diminishing the weight of the timbers that need to
be manhandled.
We put o u r rafters 32" on c e n t e r (32" o.c.).
mainly for strength. "On center" refers to the dis-
tance from the centerline of one rafter to the cen-
terline of an adjacent rafter. We tailored our floor
plan and the east- and west-wall lengths to multi
pies of the 32" centers. Putting rafters 32" o.c also
worked in well with the blocks. Someone accus-
tomed to stick framing and plywood may think
that 32" centers are rather wide and. for that type
of building, they would be. But plank-and beam
construction permits up to eight foot centers 196"
o.c.l to carry a "normal" roof. An earth roof, ol
course, isn't "normal." so we went with 32" cen-
ters. The planking itself would be quite happy on
48" centers, as can be seen at Earthwood, taut there
wouldn't be enough rafters to carry the heavy
earth-roof load.

lllus. 7-5. The advantage of 4 X 8 Or 3 x 10 INSTALLING THE RAFTERS

rafters: 3 * 8s and 4 x 10s don't return the
maximum number of rafters from round logs. The three girders were in place, but the rafters
were still in the form of logs. At least they were at
the sawmill. To speed up progress and get the
to protect against twisting under a weight load. wood out, the boys and I had to spend two days at
The 8" measurement provides excellent strength the property where I bought the hemlock. The
on the relatively short 8' spans and has the advan- order was already seven weeks overdue. Bad
tage of being twice the 4" m e a s u r e m e n t . This enough that this had delayed the job, but it also
makes for economic use of the log and gives a well meant that we'd be putting the plank-and beam
proportioned rafter, pleasing to the eye if left ex- roof up green, causing considerable shrinkage
posed. The alternative size which we considered Have these materials lined up well in advance,
was a 3 x K) rafter, as we'd used at the Cottage. particularly if you're dealing with small-scale log-
These rafters, too, fit nicely into the cross section gers and sawmills, who sometimes take a relaxed
of a log, but they require larger trees, which may attitude towards things like delivery dates. Even
be difficult to find nowadays. The reader will two months' drying makes a huge difference in the
readily sec in lllus. 7 - 5 that measurements such as moisture content of lumber.
3 x 8 and 4 x 10 don't return the maximum num- At Log End Cave, we "bird's-mouthed" all the
ber of rafters from round logs. rafters so that they'd rest with stability on the east
In the new 40' x 40' Gave plans, the rooms are and west walls, and the girders. A "bird's-mouth"
bigger than they were in the original Gave. Rafters is the space left when a wedge of wood is taken out
span 10' instead of 8'. These longer rafters give a of the underside of the rafter so that rafter has a
lot more room in the bedrooms and bathroom, so flat bearing surface where it rests on the support
5" x 10" rafters are required instead of 4" x 8". structure below. I don't advise bird's-mouthing
But. because single spans are used instead of dou- anymore, because it can decrease shear strength
ble spans, the actual timbers are shorter, lighter, of the rafter by 10% or more. Shear strength is a
and easier to handle. Single-span girders are now function of the cross-sectional area of the rafter.

lllus. 7-6. Loll: Cross-sectional area is lost, along with some shear strength,
when a rafter is bird's-mouthed to rest solidly on a girder (or wall plate). Might:
Shimming doesn't decrease the shear strength of the rafter.

and shear failure occurs not in the middle of a Just place the rafter in position, and jam in the
span, but right where it's supported from below right amount of cedar roofing shingles to give
by a wall or beam. So the bird's-mouth is at the good bearing (lllus. 7-6).
worst possible place. At Earthwood. we shimmed We used three men to slide the rafters into their
the rafters with tapered cedar shingles instead of positions. After snapping the picture (lllus. 7-7), I
bird's-mouthing. Not only is shimming stronger moved into position on the center girder so that
than bird's-mouthing, it's very much easier to do. the other men could slide the heavy rafter up to

lllus. 7 - 7 . The rafter is slid along the beams and put into position.

lllus. 7-8. Half of the rafters are n o w in place. I waited until all were in place
before fastening them.

me. Eighteen-foot g r e e n hemlock 4 x 8s a r e f o u r o r f i v e d a y s a f t e r t h e t u m b l e , I c o u l d n ' t d o any

heavy, probably about 180 pounds each. p h y s i c a l w o r k , a n d t h e p r o j e c t c a m e to a halt for a
I waited until all the rafters were in place (lllus. w e e k . 1 u s e d the t i m e to t r y to s o l v e the next p r o b l e m ,
7-8) before fastening them. With rough-cut and w h i c h w a s t h a t w e w e r e a b o u t 1 5 % s h o r t o f the r e
g u i r e d roof planking. There w a s no w a y I w a s going
barn timbers, the tops of the rafters are unlikely to
b a c k t o m y o r i g i n a l s u p p l i e r I c o u l d n ' t wait a n o t h e r
be in the same plane for planking. Again, addi
six w e e k s f o r the w o o d . 1 w a s f o r c e d t o b u y f i v e p l a n t a -
tional cedar shims on the "short" rafters can bring tion p i n e s arid h a u l t h e m t o t h e s a w m i l l . I d e c i d e d that
all the tops of the rafters into the same plane for a s l o n g a s I m i x e d the p i n e p l a n k s c a r e f u l l y w i t h the
planking. ITien toenail or, better, "toescrew" the h e m l o c k . I w o u l d n ' t b e s a c r i f i c i n g t o o m u c h i n the way
rafters in place. of s t r e n g t h I t h i n k t h i s t u r n e d o u t to be a c o r r e c t
a s s u m p t i o n , though the pine w a s very g r e e n indeed
a n d b e g a n to go a little m o l d y in the i n t e r i o r until I
VVe Interrupt This Story to . ..
c o n n e c t e d the s t o v e a n d d r i e d it o u t . Today, the e f f e c t is
that e v e r y fifth or s i x t h b o a r d h a s a k i n d of a bluish
I f this w e r e p u r e l y a n i n s t r u c t i o n a l m a n u a l o n h o w t o g r e y color, r a t h e r nice, really, t h o u g h n o t exactly w h a t
build a p a r t i c u l a r h o u s e , it w o u l d be u n f o r g i v a b l e to we'd p l a n n e d . O v e r the s a u n a we used hemlock exclu-
s t o p h a l f w a y t h r o u g h a disc u s s i o n of p l a n k - a n d b e a m sively, f e a r i n g t h a t t h e 2 0 0 ° F t e m p e r a t u r e m i g h t c a u s e
r o o f i n g for a n y r e a s o n w h a t s o e v e r . But it's n o t s i m p l y the p i n e pitch to b l e e d f r o m the ceiling. E a s t e r n hem-
a n i n s t r u c t i o n a l m a n u a l ; it's a l s o t h e s t o r y o f a n lock i s n ' t a p i t c h y w o o d .
o w n e r - b u i l d e r w r e s t l i n g w i t h a l i t t l e - k n o w n style of T h e c o n c r e t e b l o c k s I laid w e r e the first c o u r s e of 4"
building b l o c k s t o g o o n t o p o f t h e 12" block wall T h e r e a d e r
H i e c o m p l e t i o n o f the r a f t e r s o n A u g u s t 2 6 m a r k e d will r e c a l l t h a t o u r e a s t a n d w e s t p l a t e s w e r e only 8 "
the e n d of a p r o d u c t i v e p e r i o d that b e g a n on July 5 wide in o r d e r to leave r o o m for these blocks. I didn't
w h e n the d i g g e r t o o k its first b i g b i t e o u t of the knoll. use s u r f a c e b o n d i n g b e c a u s e it could only have been
I-xcept for the d e l a y in g e t t i n g the h e m l o c k to the a p p l i e d t o t h e wall's e x t e r i o r . T h i s p o r t i o n o f t h e wall
sawmill, p r o g r e s s h a d b e e n s w i f t , a t least f r o m a n w a s n ' t l o a d b e a n n g , so I felt c o n f i d e n t in laying up the
o w n e r b u i l d e r ' s p o i n t o f view. J o n a t h a n C r o s s m i g h t b l o c k s w i t h m o r t a r I d e c i d e d t o lay t h e f i r s t c o u r s e
have c o n s i d e r e d o u r p a c e t o b e t h a t o f a p a r t i c u l a r l y b e f o r e nailing the roof planking, so that the planking
s l u g g i s h snail Still, f i f t y t h r e e d a y s a f t e r t h e s t a r t o f w o u l d n ' t b e i n t h e w a y o f m y w o r k , b u t 1 held o f f o n t h e
c o n s t r u c t i o n , w e w e r e r e a d y t o put the r o o f o n . . . i f s e c o n d c o u r s e s o t h a t t h e b l o c k s w o u l d n ' t b e i n the
the r o o f h a d b e e n r e a d y for us. w a y o f the p l a n k i n g .
I fell off a r a f t e r o n t o t h e w i n d o w h e a d e r below, Here I should mention a problem the reader can
b a d l y b r u i s i n g , if not c r a c k i n g , a r i b . F i f t e e n m i n u t e s e a s i l y a v o i d . W e ' d s p r e a d a b o u t half o f the p l a n k s o n
later, I w a s b a c k on t h e b e a m s , a w i s e r m a n . We fin the r o o f , loose, t o g i v e t h e m a g o o d p l a c e t o dry. B e f o r e
ished the r a f t e r s , a n d I laid b l o c k s t h e next day. but a n y r a i n , w e laid o u t r o l l s o f 1 5 - p o u n d felt p a p e r .

lapped to try to k e e p as m u c h w a t e r out of the h o u s e lock. N o luck, s o I w o r k a d e a l w i t h D e n n i s f o r p i n e .

as possible We couldn't start nailing the roof, as we W h a t we think is N o r w a y pine t u r n s out to be planta
w e r e still w a i t i n g f o r t h e p i n e l o g s t o b e milled i n t o lion p i n e . G a r y a n d D i a n e a r r i v e o n 1-abor D a y w e e k
p l a n k i n g O n e n i g h t , t h e first s t r i p o f felt p a p e r b l e w end and G a r y helps Dennis a n d me clear a road to the
up and backfokled o n t o the roof. W a t e r p o u r e d d o w n pine. I c a n ' t d o m u c h b e c a u s e o f m y r i b . S u n d a y . G a r y
the roof a n d o n t o t h e t o p o f t h e 4 " b l o c k s , f i n d i n g its b a b y - s i t s w h i l e Jaki a n d I help D e n n i s . F o l l o w i n g
w a y f i n a l l y i n t o t h e u n f i l l e d c o r e s o f t h e 12" b l o c k s , w e e k , w e g e t w o o d t o s a w mill f o r m i l l i n g D e n n i s a n d I
f h e r e ' s n o w a y t o g e t that w a t e r o u t s h o r t o f d r i l l i n g start r o o f with hemlock d u r i n g this t i m e a n d cut t h r e e
holes into the block cavities ( f r o m the inside, so as not o r f o u r t r e e s i n f r o n t o f C a v e . G e t five h e m l o c k a n d six
to break the w a t e r p r o o f i n g quality of the surface s p r u c e 2 x 6s f r o m o u r o w n trees, which had to be
b o n d i n g I this we did I he moral: Maintain vigilance c l e a r e d t o let t h e w i n t e r s u n g e t t o t h e s o u t h wall.
m k e e p i n g w a t e r o u t of t h e wall cavities. D e p r e s s i n g t i m e , d u r i n g w h i c h little g e t s d o n e . "
S t e v e a n d B r u c e left o n A u g u s t 28. T h e i r h e l p h a d
been invaluable I think my diary best c a p t u r e s the
m o o d o f t h e next t w e l v e d a y s : S o i t g o e s . O n S e p t e m b e r 10. w e h e l p e d t h e s a w y e r
p l a n e the pine, s p r u c e , a n d h e m l o r k 2 x fts to u n i f o r m
" M o n d a y , A u g u s t 29. H a u l e d 2 x 6 p l a n k s f r o m t h e thickness a n d width. This is i m p o r t a n t in g e t t i n g a
sawmill." P r o g r e s s w a s so slow that the next eleven good tight fit w h e n roofing. In addition, the planed
d a y s a r e all l u m p e d t o g e t h e r : s u r f a c e i s a lot less p r o n e t o c o l l e c t i n g c o b w e b s . A
"Tuesday, A u g u s t 30 Friday, S e p t e m b e r 9. Dis- little w o o d ( a n d s t r e n g t h ) i s lost, o f c o u r s e , b u t n o t
c o v e r t h a t w e will r u n s h o r t o f p l a n k i n g ( t h o u g h t w e e n o u g h to be significant. O u r planking actually mea-
had m o r e than enough), so I c h a s e a r o u n d for hem s u r e s 1W x 5 W.
Roof Deck
In planking, start at the bottom and work up. Got planks which will cover lour s p a n s and o n e which
the first c o u r s e of planking d e a d straight and will cover three, for example. Each span is 32".
make s u r e each end of the first course is equi- I lowever, the planks at the e n d s of each course
distant from the peak. S n a p a chalkline to mark the must be long e n o u g h to a c c o m m o d a t e an over-
location of the bottom e d g e of the first plank. We hang. I figure an extra 12" for this, 10" for the
had several long planks that were cut from the overhang and 2" to get to the outside e d g e of the
same logs as the rafters, so we were able to make o u t e r m o s t r a f t e r s . (The 32" figure is m e a s u r e d
the first course on each side of the roof from two center-to-center.) lo minimize wastage, have the
planks each—a good start. saw logs cut to useful lengths (lllus. 8 1). Take a
A word about lengths of planks and wastage: copy of these lengths with you into the woods, or
There a r e twelve rafters or eleven "spans" on each supply a list to the p e r s o n cutting for you. Make
side of our house. A course might consist of two sure that the logs a r e cut 2 or 3" longer than these

lllus. 8-1. To save wood, cut the logs to useful lengths.


Illustrations 8-2 and 8 3. IT you use green wood, nail the planks as
shown in these two photos, using a spike to keep them tight against
each other, and nailing the lower nails first.

measurements, to allow for squaring the ends of 8 - 3 . There's no r e a s o n why you can't draw 2
the planks. boards at a time. First, start the nails as shown.
If you're stuck with green wood, as we were, Then, take a star chisel or a 12" spike and whack it
nail the planks as shown in Illustrations 8 - 2 and about into the rafter tight against the plank.

Using the spike as a lever, d r a w the boards tight over the ends. 'Ihe overhang can be trimmed later
against each other and nail the lower nails first, by snapping a chalkline and cutting the whole
then the upper ones. The reason for angling the edge at once with a circular saw or a chainsaw
nails is to make it easier for the wood to shrink (lllus. 8-(>). The chainsaw made cutting the edge
without splitting I lllus. 8-4). easier. Green two-bys will choke all but the tough
1 like to use 16-penny resin-coated nails, though est circular saws. Hie only problem with the chain-
uncoated IBs would probably do. Resin-coated saw is that you've got to be very careful to keep it
nails are hard to take out once they're in, but thafs perpendicular to ensure a straight edge.
their purpose. The narrower shaft makes splitting Three things can happen at the peak (lllus. 8-7).
at the ends of the plank less likely and makes the One, the bottom sides of the opposing plank roofs
plank more likely to give with shrinkage. may meet perfectly. In this unlikely case, sit back
Try to avoid placing too many joins one above and relax. Two. they'll just miss meeting each other.
the other (lllus. 8 5). Leave the planks sticking out This is what happened to us. In this case, make a

Illus. 3-5. Avoid placing too many joins one above the other.

lllus. 8-6. L e a v e the plank ends sticking out. I he overhang can be

trimmed later by snapping a chalkline and cutting the whole edge at once.

lllus. 8 7. Af the peak, the planks might lit perfectly or just miss each other, or the last
course on each side may be too wide.

wedge from leftover pieces and fit it into the space. the last courses must be trimmed to the correct
You have to use a variable-angle table saw or good width, It's up to you if you want to incorporate the
skill saw for this Three, the last course on each proper angle into the trim cut. The angle isn't
side will be just too wide to fit. In this case, each of necessary for strength, it's hidden from view by

the girder, and it'll be covered by the waterproof- It's a good idea to use a chainsaw to cut a groove
ing membrane. 1" wide by VA" deep down the middle of the bottom
Hie fascia or retaining boards serve to keep the side of the fascia board before nailing it in place
sod roof from washing out over the gable ends of (lllus. 8-9). This groove serves as a "drip edge"
the house. The fascias frame the roof's high edges, and prevents rainwater f r o m running along the
directing all drainage down the slope. We used underside of the roof planking and into the house.
2 x 12s. w h i c h worked out well. The 2 x 12s
should be installed al ter the overhang is trimmed.
Our method of fastening the fascia board involved
lirst nailing a 2 x 6 plank right along the edge of
the roof, for its full length, then nailing scraps of
economy-grade 2 x 4s on top of the plank (lllus.
8 8). Together, the planks and scraps were now

lllus. 8-9. Use a chainsaw to cut a groove down

the middle of the fascia board before nailing the
board into place. This groove will serve as a "drip
edge" to keep rainwater from running along the
underside of the roof plunking and into the house

lllus. fi-8. fascia boards were fastened by first

nailing a 2 < fi plank along the edge of the roof
then nailing a scrap piece of economy-grade
2 < 4s on top of the plank. These t w o pieces
provide a plaie for fastening the 2 * 12 fascia.

3W higher than the roofing planks, exactly the To stop earth from falling off earth roofs, I no
thickness of our rigid-foam insulation. At the orig- longer use any form of retaining timbers, either
inal C ave, we placed particleboard over the insula- fastened or laid loose. Taking a leaf from Wells's
tion, then the waterproofing. It's much better for book. I now cut 6"-wirIe by 12"-long by 5"- to
the insulation to be installed on top of the water- 6"-deep moss sods f r o m the sand pit in front of
proofing. as we've done since 1079. The preferred Earthwood and lay them along the edge of the
method is described in chapter 9. f h e planks and roof, kicking them tight against each other. After
scraps provided a high edge for nailing the fascia the earth or sod is installed, I water the moss
board Use pressure-treated lumber for the fas thoroughly, and endeavor to keep it moist for a few-
cias. A f t e r the fascia b o a r d w a s well nailed, weeks until the moss knits together. The edge of
"ii," x 4" lag screws were used to make sure it was the roof is exposed earth for a while, but after a
on permanently -a lag screw every 3' or so along year or two, the moss completely spreads over the
the fascia edge of the socls, creating an attractive border to

lllus. 8-10. These moss sods drain well, retain the earth, and look natural.

lllus. 8-11. To accommodate skylights in the root

build the box frame on ;i flat surface and then nail it
around the opening. I recommend two courses of
2 * 8 material to build the box.

the earth roof. lllus. 8 10 shows the moss sods on you need to make must be high enough to accom-
the edge of my workshop building at E a r t h w o o d . modate the 4" (recommended) of extruded poly-
This detail is beautiful and natural, promotes g o o d styrene insulation, the 2" crushed-stone drainage
drainage, and stops the earth from falling off the layer, and the desired soil thickness. A 7" soil layer
roof. seems to maintain the green cover in northern
New York. These layers add up to 13", so I recom-
mend a 15" or 16"-inch high box, composed, per
haps, of two courses of 2" x 8" w o o d e n planks
nailed together, as shown in lllus. 8 11.1 found that
SKYLIGHTS the easiest way to install the box f r a m e is to cut the
I.og End Cave had three skylights, o n e each in the opening in the roof with a chainsaw or circular
office, living room, and bath. Now's the time to box saw, build the box f r a m e on a Hat surface, and then
in for the skylights. Follow the manufacturer's rec- nail or screw the box f r a m e a r o u n d the opening.
ommendations for sizing and have the actual sky- Stovepipe holes can be cut in later, after the
light on hand to check against. The box f r a m e that waterproofing m e m b r a n e is applied.

WATER PROOFING BEFORE h o m e m a d e m e m b r a n e , b e g i n n i n g with a

trowelled-on layer of black-plastic roofing ce-
ment, available in 5 - g a l l o n pails. Next, we
In the correct order of events, novv's the right time stretched on a layer of 6-mil black polyethylene (or
to apply the waterproofing membrane. I'he water- wsquene). pressing it into the incredibly sticky
proofing should always be on the warm side (un- roofing cement. Then, to m a k e doubly sure; and to
der! the rigid-foam insulation. This isn't what we establish beyond a doubt the masochistic side to
did at the Cave, but it is what we've done on every my persona. I repeated the exercise: another layer
earth roof since (six in number), and is the detail of black cement followed by another layer of plas-
included in the 40' x 40' Cave plans. The potential tic We used the same method on the walls, but
problem with placing the membrane on the top limited the process to o n e layer each of roofing
(cold) side of the insulation is that warm moist cement and plastic. All in all, we trowelled on 250
internal air might travel through the planking and gallons of plastic roofing cement. O n e set ol
the insulation and condense on the cold mem- i lollies and several pairs of gloves were destroyed
brane, causing a moisture problem. This is a com during this process. All this happened during a
mon occurrence in older mobile homes. The cold three-week period when n seemed to rain almost
metal roof is above the insulation. Internal moist every day, and Juki's parents were visiting from
air travels through the ceiling and, in the winter, Scotland.
reaches dew point on the underside of the roof,
causing severe condensation This is also why
vapor barriers a r e always placed on the interior
Iwarml sides of insulation in standard stick frame
construction At the Cave, we didn't actually en-
counter this condensation problem, but there's no
doubt that we didn't get the order right. That insu BITUTHENE*
laiion should be on the cold side of the vapor
barrier is now accepted as fact by all leading an On every earth shelter since, I've used an exc ellent
thorities in the underground-housing field. waterproofing membrane. Bituthene' It's com-
posed of two layers of 2-mil black polyethylene
cross-laminated over a uniform (60-mil or VW)
thickness of rubberized asphalt. This bituminous
bedding is very sticky, hut the 36" x 60' roll comes
DON'T DO THIS! protected by a layer of nonstick backing paper,
Another change I've adopted since the late 1970s is which is removed while the m e m b r a n e is applied.
my choice of waterproofing membrane. In 1977. At Hart h wood, we used Bituthene 1 - 3(XX> mem
based upon an old magazine article, I selected a brane for both the roof and earth bermed walls.
method of inexpensive waterproofing. I wouldn't There's also Bituthene" 3100 m e m b r a n e for appli
inflict this job on anyone: We built up our own cations at temperatures between 25°F and 40 T

WATER PROOFING THE WALLS smoothed with a knife. All freshly c ut edges of the
membrane are sealed in this way, but it isn't neces-
On a building like Log End Cave, where the berm sary to apply mastic to the factory edge, which
comes right over the east and west walls and comes already protected by a factory-applied
melds with the earth roof, waterproof the walls mastic along it.
first, simply because the roof membrane should Sheets are applied vertically, as shown in lllus.
overlap the wall, and not the other way around. 9-2. Cut the sheet to the right length, and have
After the surface-bonding cement is cured, the someone hold the top edge from above. On the
wall is primed with Bituthene* 3000 primer. The east and west walls of the Cave-type design, this
primer goes on very quickly with a roller and dries membrane would begin 6" onto the roof, or about
in an hour. at the center of the first course of planking. I wo
Before applying the membrane, we attend to other people work from below, one removing the
the critical waterproofing detail where the wall backing paper, the other pressing the Bituthene"'
meets the footing. A strip of Biluthene 81 about 7" into the wall. Lap the previously laid strips at the
wide and 36" long is cut from the end of a roll. The footing detail with a lull 3" overlap and s<>al the*
backing paper is removed and the strip is carefully bottom with mastic. To prevent bubbles forming
fitted into the crease where the wall and the foot- behind the membrane, adhesion is best begun at
ing meet. About 3" of the strip is now stuck fast to the middle of the sheet and spread out to the right
the footing, and the remaining 4" of the strip is and left with your hands. Continue pressing the
stuck to the rest of the wall. I'he edge on the footing Bitulhene,M into place right up onto the primed
is sealed for protection against "fish-mouthing" planking, and discard the backing paper. The sec-
(raising of the membrane) by the application of a ond 36"-wide piece laps the firv;t by ?V". A yellow
bead of Bituthene* EM3(X)0 mastic. This bead is lap line is provided on each edge of the sheet to aid
applied with a caulking gun I lllus. 9-1) and then in the placement of the next sheet. If it becomes

lllus. 9-1. The important detail where the footing lllus. 9-2. The Bituthene' waterproofing is applied
meets the wall is sealed with a strip of HitutheneK. vertically to the walls.

apparent that the second piece is wandering off Happily, the B i t u t h e n e ' m e m b r a n e sticks ex-
line, stop pressing the Bituthene*. The sheets can't tremely well to aluminum flashing. If th«»'?" x 12"
be pulled or stretched back to alignment. Cut the retaining timber is used, the detail would look like
sheet short at this point, and begin again with an thai shown in lllus. 9-3.
overlap of at least 3" at the bottom. Use heavy hand With the moss-sod method, just flash the edge
pressure at the lap seam and seal the newly cut of the planking with either an aluminum drip edge
edge with the compatible mastic. Never use an made for the p u r p o s e or a piece of 10"-wide alumi
incompatible mastic such as tar, asphalt, or pitch num flashing bent at a right angle along a straight
based materials, or mastics containing polysulfide line, so that 7" is nailed to the deck and 3" hangs
polymer. over the planking to form a protective drip edge.
The advantages of Biluthene® are: speed, ease The Bituthene® sticks to the nailed part of the
of application, and quality control. Any shape flashing, covering the nails. Keep the Bituthene*
piece can be cut to fit. but always allow for the ZVf membrane about 3" in from the edge so it's not
or 3" lap and seal all cut edges. exposed to the sun's ultraviolet rays. As long as it's

buried under the earth, the black-plastic surface

WATERPROOFING THE ROOF will last forever or 1(X) years (whichever comes
Flashing is one of the most important jobs in wa- first).
terproofing and one of the most difficult to do well. The Bituthene® goes down quickly on a simple

roof plane like that of Log End Cave. The applica- cut from the same material. The patch should ex-
tion is horizontal, beginning at the bottom of the tend 3" in every direction from the damage, the
roof, and lapping a g o o d 3" o n t o the m e m b r a n e patch pressed h a r d to the surface with the heel of
already applied w h i l e w a t e r p r o o f i n g the walls. your hand. The edges of the patch a r e sealed with
Roll out the a p p r o x i m a t e l y 30' of m e m b r a n e mastic. To diminish the chance of a c o r n e r lifting. I
across the roof with the backing paper still on. Cut make octagon-shaped patches with a razor-blade
the m e m b r a n e to the right length and establish the knife.
3" lap onto the wall m e m b r a n e which comes up
onto the roof. Now, pull up about 3" of the backing
paper along the width of the m e m b r a n e at the
starting end and stick the m e m b r a n e down onto
the roof-edge flashing, making sure you've still got
that 3" lap at the low side of the sheet. Roll up the lllus. 9 - 4 shows the aluminum flashing of a sky-
sheet again. Now unroll it again, this time while light at Log End Cave. Note that the flashing is
someone is pulling the backing paper out from tucked under the black plastic on the uphill side
under the sticky bitumastic. The m e m b r a n e should (upper right of the picture) and over the mem-
fall right into place. But if it doesn't, if it looks as if brane on the downhill side. With the Bituthene*
you're wandering off the yellow guideline, stop. membrane. I install the Hashing a little differently. I
Cut the sheet short, lap back 3", and try again. If install it over the m e m b r a n e all around, compress-
care is taken, and the technique given in this para- ing a bead of Bituthene* mastic with the nails.
graph is adhered to. you'll never have to cut short Then I cut a 6"-wide strip of Bituthene* and lap it 3"
and start again with the 3" lap, but you should onto the flashing (covering the nails), and 3" onto
know what to do, just in case. the s u r r o u n d i n g m e m b r a n e . A g a i n , caulk cut
Punctures, cuts, or other suspicious marks on e d g e s with mastic. The corners close to the sky-
the m e m b r a n e are easily remedied with a patch light box are the critical details. Here, use plenty of

lllus. 9-4. Flashing at the original Log End C a v e was under the plastic
membrane on the uphill side and over the membrane on the other sides.
I his Hashing detail is much easier now when using the Bituthene'

overlapping Bituthene*. small pieces covered by

larger pieces. The last (topmost) layer must have
the edges caulked. When the earth goes on, sur-
round all projections like chimneys and skylights
with a 6"-wide crushed-stone drainage buffer, the
crushed stone meeting the u n d e r g r o u n d d r a i n a g e
layer described in chapter 11.
I'm a great fan of the Metalbestos® chimney
system for use with wood stoves, and such a sys-
tem w o r k s particularly well with earth roofs. The
stovepipe itself is stainless steel inside and out, and
sandwiches a 1" noncombustible layer of low con-
ductivity. The pipe is rated safe within 2" of com-
bustible material There's no asbestos in Metal-
bestos ". The beauty of the system with earth roofs
is that the pipe manufacturer makes a tall alumi-
num flashing cone suitable for shallow-pitch roof
applications. The cone comes with a nice wide
flange, which makes it easy to use with Bituthene"'.
Cut out a hole in the finished roof with a diame-
ter 4" g r e a t e r than the outside diameter of the pipe
you'll be using. A chainsaw works well for cutting lllus. 9-7. A galvanized Hashing cone was
this hole. Hang the whole chimney through the specially made for this cylindrical chimney.
hole with a piece called a "Roof Support Package"
(RSP). which screws to the deck, and add at least
one piece of chimney above the roof. Lower the
flashing t o n e over the chimney and use 6" strips of
Bituthene* to seal the edges of the flashing-cone mastic applied a r o u n d the e d g e to ooze out. The
flange, the same as with the skylight (Illustrations nails c o m p r e s s a n e o p r e n e w a s h e r to seal otf wa-
9 - 5 and 9-6). ter penetration. The top of the flashing c o n e is
At Earthwood, a 3'-diameter m a s o n r y chimney sealed with silicone caulk (lllus. 9-7).
rises up out of the center of the house. I had a While there a r e o t h e r waterproofing options, I
special flashing cone made at a sheet-metal shop, don't h e s i t a t e r e c o m m e n d i n g the Bituthene 1 1
and nailed the cone through the Bituthene* 1 to the m e m b r a n e to owner-builders because of its ease
deck, using ribbed nails every 2", causing extra of application, quality, and moderate cost.
Insulation, Drainage &

EXTERIOR-WALL INSULATION We used a fairly low-density b e a d b o a r d , but even

today, 17 years later, it's still in g o o d condition. I
It's convenient to apply the wall insulation before credit the g o o d d r a i n a g e provided by the sand
installing the footing drains. In northern climates, backfill as being the r e a s o n for its preservation.
I now use 3" of extruded polystyrene d o w n to frost I've seen b e a d b o a r d completely saturated, and in
level (48" in our area) and 2" the rest of the way that condition it has no value as insulation.
down to the footings. This wall insulation should I don't r e c o m m e n d polyurethane foams, which
cover the top of the footing as well as the insulation are generally yellow in color a n d often have one or
whic h was placed on the outside of the footing two layers of aluminum foil on the surfaces. Al-
tracks. Try to prevent any direct conduction f r o m though the starting insulative value is very high at
the interior to the exterior sides of the concrete R-8 p e r inch, the foam is p r o n e to water absorp-
fabric of the building. Malcolm Wells calls any- tion, which can bring that value right d o w n to R-3
such c onduction an energy "nosebleed," and, like or less.
a drippy nose, the manifestation can be condensa- On nice straight walls, like t h o s e in the Cave, it's
tion on the inside, the result of d e w point occur- easy to spot glue the rigid foam to the walls. Six
ring on the inner surface, dollops of compatible mastic should hold a 4' x «'
At Log End Cave, we used Z of headboard for sheet in place until the wall is backfilled. There are
the first 48" of depth, and 1" from that point d o w n actually glues m a d e for r i g i d - f o a m insulation,
to the footing. We didn't w r a p the footings with available for convenient use with a caulking gun.
insulation, and suffered the worst manifestation of Don't glue the insulation too early, as we did. After
energy nosebleed: condensation at the base of the a few days of wind and rain, o u r b e a d b o a r d was
wall in late spring and early summer. Beadboard falling off the wall. Glue the insulation just before
(expanded polystyrene) may be okay to use with backfilling. At Earth wood, with its round walls,
good drainage. I )se a dense form of beadboard, we had to manually hold the S t y r o f o a m * against
given a choice, as o p p o s e d to the very light and the wall d u r i n g backfilling, b e c a u s e no glue will
easily compressible light white stuff which is com hold the foam against the wall if the wall is even
monly sold. Research the R value and the price. slightly curved. Remember, it's called rigid foam.
The denser h e a d b o a r d s will have higher R values,
some of them equivalent to extruded polystyrene's
R 5 per inch. You should be c o n c e r n e d with
R value per dollar. If there's little difference be-
tween the cost of e x p a n d e d and extruded, go with
the extruded, which is close-celled and will stay Waterproofing w a s discussed b e f o r e drainage be-
dry and crisp u n d e r g r o u n d until the end of time. cause the former's installation o c c u r s first. Drain-

age is your main line of defense against water d o w n to the footing drain. As insulation and
penetration. Give water an easier place to go than d r a i n a g e are combined in o n e product, the cost of
into your house, and it will cooperate. The keys to $1.60 per sq. ft. (1991) is quite favorable.
good drainage are the use of backfill with excellent
percolation characteristics, and a footing drain
(French drum), which carries the water away from
the house. At the Log End Cottage basement, we
installed footing drains correctly, but backfilled
with the poor claylike earth which c a m e out of the
hole. The water couldn't get to the footing drains.
Hydrostatic and frost pressures caused a move-
ment in the block wall, negating our cement-based
waterproofing. Cement waterproofings aren't
r e c o m m e n d e d , b e c a u s e they lack the ability to
b r i d g e even the tiniest s h r i n k a g e or settling
A g o o d drain system for an earth-sheltered
house like the Log End Cave is shown in lllus. 10-1.
With a freestanding roof, such as at Earthwood, a
plastic or aluminum gutter is advisable, partic-
ularly in wet climates. Using gutters eliminates the
need for the topmost drain, the one just below the
surface. The intermediate drain is optional, and is
only necessary in soils with moderately good per-
colation characteristics. If good percolating back-
fill (such as coarse sand or gravel) isn't available at
the site, it should be brought in. Backfill should be
compacted in 12" layers. Using 12" layers mini-
mizes the tamping pressure necessary, decreasing 1 Earrh
the likeliho(xl of d a m a g e to the walls. 2 Filtration mat. can be hay or straw
If no good backfill is available within hauling 3 Crushed stone
distance, several horizontal drains may be neces- 4 Waterproofing m e m b r a n e
sary. Another option is Enkadrain®, a fibrous Vz~ 5 4" perforated drain wrapped In a filtranon fabric.
thick mat (manufactured by AKZO Industrial Sys- 6 6-mll polyethylene io form underground "gutter" The top
tems Co.) which is placed against the wall. Hydro- gutter should include a fold in the plastic as shown, to
static pressure is eliminated, because water enter- allow for earth settling.
ing the tough nylon mesh is rapidly carried d o w n 7 Rigid-foam insulation
to the footing drain. Enkadrain* isn't a substitute 8. Vertically placed 4" nonperforated drains connect the hor
for w a t e r p r o o f i n g ; it's a c o m p l e m e n t . The 0.4" Izontal perforated drains. T-junctions are available to
thickness is suitable for depths of up to 10' below m a k e these connections.
grade and was about S0.50 per sq. ft. in 1991. In 9 Footing
some cases, using Enkadrain® may be cheaper 10 Concrete-slab floor
than hauling in good backfill. 11 Compacted sand, gravel, or crushed stone
A n o t h e r g o o d p r o d u c t , Dow's Styrofoam® 12 Undisturbed subsoil or heavily c o m p a a c d pad. as prev:
Therma-Dry* combines insulation and drainage ously described
in one application. An a b u n d a n c e of vertical
grooves (W deep x W wide) are cut into the out-
side of a 1 Vz"- or 2W-thick sheet of Styrofoam*
Blueboard®, and covered with a nylon filtration lllus. 10-1. A good drain system for an earth-
mat. Water seeps into the grooves and is carried sheftered house. (Not drawn to scalc)

lllus. 10 2. The f o o t i n g drain should he

perforated plastic flexible tubing, laid adjacent to
the footing on a bed of washed 1" stone. lllus. 10-3. The underfloor drain, kitchen drain,
and footing drain all take separate paths to the
same soakaway.

lllus. 10 4. From the southeast corner (the lowest lllus. 10-5. Hay spread on top of the stones
point of the footing drain's grade) the pipe will be covering the footing drains will decompose and
run at a similar slope behind the retaining wall form a mat to stop infiltration into the drain by the
and into a soakaway. subsequent sand backfill.

FOOTING DRAIN bending stresses increase as the s q u a r e of the

span of a beam (the distance b e t w e e n corners in
The other key to g o o d drainage, besides good
the case of the wall), shortening the span greatly
percolation, is a g o o d footing drain.
increases the resistance of the wall to lateral pres
The drain is constructed < >f 4" perforated flexible
sure. Cutting the span in half increases the bend-
plastic tubing, preferably covered with a nylon or
ing strength by lour times. Internal braces are one
fiberglass "sock" to prevent infiltration of silts,
way of reducing the span. Built in block pilasters
which could clog the system. The tubing is laid out
are another m e a n s of resisting the pressure, but
on a 3" to 4 -deep b e d of washed 1" (#2) crushed
they're c u m b e r s o m e and difficult to work in with
stone adjacent to the e d g e of the footing (lllus.
the floor plan. I prefer to use 12" blocks and elimi
10-2). The drain should slope slightly as it travels
nate the pilasters. And. of course, there are bond-
around the perimeter of the house and into the
beam courses, as already described.
soakaway, or (better, if it's possible at your site) out
At Log End Cave, bracing against lateral pres-
above g r a d e (lllus. 10 3). Our method of establish-
sure w a s easy, since we'd already framed most of
ing this slope was to make marks with a crayon 4"
the internal walls with 2 x 4 framing, and all we
below the top of the footing at the northwest cor-
had to do was nail a diagonal b r a c e to the framing
ner, 5" below the northeast and southwest corners,
for support. We braced at every point around the
and t>" below the southeast corner. We joined the
perimeter of the house w h e r e an internal wall met
marks by s n a p p i n g a chalkline and brought in
a block wall (lllus. 10 (5).
washed #2 stone, bucket by bucket, to the g r a d e
established. The slope, then, ran in both directions
around the house f r o m the high point in the north-
west corner to the low point in the southeast cor-
ner. about an inch of d r o p for every 30'. From the
southeast corner, shown in lllus. 10 4, the perfo-
rated pipe ran at a similar slope behind the loca
tion of the large retaining wall and on to a soaka-
way. (There's no reason why the retaining walls
couldn't be built at this time, but we were in a
hurry to get the sod roof on before winter, so the
retaining walls w e r e delayed.) The footing drain is
well covered with additional crushed stone, about
3" above the top, a n d the stone is covered with 2" of
hay or Straw (lllus. 10-5). The hay partially decom-
poses and forms a filtration mat which keeps the
crushed stone clean. Although manufactured fil-
tration mats m a d e for this purpose a r e available,
this "organic" filtration technique works well.


The footing drains completed, the only prepara-
tion before backfilling was to brace the walls from
the inside as a safety measure against the weight
of the sand that would be d u m p e d against the
Think of a below g r a d e block wall as a beam lllus. 10-6. Brace the frame from the inside to
resisting a load. In this case, the load is lateral. support it against the weight of the sand that will
Earth p r e s s u r e c o m e s f r o m the side. Because he poured against the sidewalls.

Never backfill a basement foundation or under- Backfilling can be done with a bulldozer, back-
ground house before the tops of the block walls hoe, or front-end loader. Using the f r o n t end
are tied together, either by the floor structure, or. in loader attachment for a farm tractor would be a
our case, by the roof-rafter system. These internal good choice. Maybe a neighbor can save you some
f r a m e w o r k s p r o v i d e v a l u a b l e a d d i t i o n a l re- heavy equ ipment costs.
sistance against lateral loading. During backfilling, be alert to the rigid foam
moving or coming away from the wall, and watch
for big stones which might come crashing down
BACKFILLING against the wall. Since the equipment operator
might not be able to see such happenings from his
At Log End Cave, we brought in 25 dump-truck perch high above, another pair of eyes (yours) is
loads of sand (about 125 cubic yards) to backfill the imperative. Note that the placement of the rigid
walls, because we'd learned to o u r cost at the Cot- foam over the m e m b r a n e protects the m e m b r a n e
tage basement that the subsoil at Log End doesn't from damage, within reason. The foam won't nec-
have good percolation qualities. essarily save the wall from a 500 pound boulder.
Earth Roof
loamy, a e r a t e d by the r o o t s y s t e m s ol the
THEORY grasses, and not greatly compacted. The grass
When I teach underground housing at F.arthwood itself, especially if allowed to g r o w long, supplies
Building School, there's always the studious type additional insulation. An earth roof holds snow
who listens very carefully, takes good notes, and better than any other kind of roof. A house that
arrives at some very logical questions. hus snow on the roof where there's heat inside is
"If earth is a poor insulator, then why bother a warm house. Light Huffy snow adds insulation
putting any on at all? I mean, you'd need three feet at R-l per inch of thickness. A two-foot snow-
of the stulf to be worth anything. Why engineer storm of light dry snow adds R-24 to your roof.
and build for all that load? And why risk the We certainly found, while living at Log End Cave-
leaks?" land at Earthwood), that houses are easier to
I usually reply, "Good question, and it deserves heat when they have a good layer of snow on top.
a good answer." It's true that a house built in our 9000-degree-
As I begin my reply. I write a list of seven single day climate with nothing more than 7" of earth
words. Yes. earth is poor insulation, and, yes, 3' of on the roof wouldn't work. You still need the rigid
earth is nutsy. Yes, such a heavy roof can be built, foam, at least R-20's worth. If building-code offi-
and some have been built, but what did houses cers insist on R-38, those first few inches of grass
with such roofs cost? What kind of technology and roots and loose earth and snow will make up
was used? Pre-stressed concrete planks installed for the shortfall. Remember, too, that the R-38
with a crane? Not post-and-beam. that's for sure. derived from light fluffy insulation contributes
We shouldn't cut the kinds of trees that would be nothing to the thermal capacitance of the build-
needed t<> support a load of 5(X) pounds per square ing. An earth roof is g r a s s a n d mass.
foot. The right amount of earth to have on the roof • Drainage. The conventional roof doesn't slow
is just the amount needed to maintain a green runoff during heavy rains. A lot of water has to
cover, no more, no less. In a temperate climate be dealt with quickly. In a bermed house, water
with moderate rainfall, 6" to 8" of earth seems to percolates slowly to the footing drains. Much of
work just line. If the roof dries out, as the Earth- the water just r u n s naturally away from the
wood roof did o n c e d u r i n g a long s u m m e r building, as water runs down any hill Hundreds
drought, it'll come back, maybe with some sur- of gallons of water are absorbed by the earth,
prises. It's amazing the new plants you'll discover which keeps it in reserve to replenish the atmo-
on your r< »of. I {ere are the advantages of havi ng an sphere on a dry day, which brings us to . .
earth cover: • Cooling. Water evaporating off the earth roof in
the s u m m e r promotes cooling. Even a few inches
of earth protects the substrate from the high
surface temperatures found on an asphalt or
• Insulation. Earth isn't completely without insu- metal roof. You can cook an egg on an asphalt
lative value, but by far the greatest such value is roof. Stick your finger in a wet lawn, however
found in the first 3" or 4" The soil is rich and and it comes out cool.

• Aesthetics. One of the main reasons for building of petrochemicals than burning the extra fuel oil
uii earth shelter is its minimal visual impact. An required without it. And now, with the advent of
earth sheltered home harmonizes with its sur- replacements for the chlorofluorocarbons (CFCs)
roundings If planning and zoning officials really heavily used as expanding agents in the past, the
cared about the visual amenity within their juris- ozone layer needn't suffer further damage from
diction, they'd insist upon underground houses rigid foam insulation.
.. with earth roofs. N o n b i o d e g r a d a b l e m e m b r a n e s and insula-
• Ecology, fhe earth roof supports life and oxy- tions? For a long-lived house, this is exat tly what
genation. Save 1500 sq ft of the planet's surface we want. We don't want something which is going
from being converted to lifeless desert. Wild- to break down and pollute the earth and water
llowers and leafy vegetables can be grown on a table. We don't want something that has to be
shallow earth roof. Forget the shrubs and trees. replaced someday, because it's difficult to replace
• Protection. Yes. a massive earth roof is better insulation or waterproofing in an underground
protection against sound and radiation than a 7" house.
earth roof, but again, as with insulation, the first In summary, the main gain from putting 3' of
few inches have the major impact. earth on the roof instead of 1' is that the entire
• Longevity. Do it right the first time. Read this structure is set 2' deeper into the ground. At soil
chapter carefully and follow its advice; and you'll depths of 6' to 12", there's an average difference of
never have to do it again. There are two things TF for each foot of depth, so setting the home 2'
that break down most roofing materials: ultra- deeper amounts to a 2°F advantage, both for sum-
violet radiation and freeze-thaw cycles. With an mer cooling and winter heating, plus the slightly
earth roof, the sun's rays never get to the water- greater insulution derived from the extra 2' of
proofing membrane, so there's no UV deteriora- earth. It's my contention that an extra inch of rigid
tion. I've dug up black plastic buried for 25 years foam (which weighs practically nothing) is a much
and it was just as pliable as the day it was buried. more reasonable means of achieving the same
Since the 1950s. Americans have been preserv- advantage than is building the behemoth required
ing their garbage for future posterity wrapped in to support a 3' earth load. All we should be con-
nonbiodegradable plastic bags. cerned about is maintaining the green cover. For
us, 6" to 7" of earth has worked out well. Hie roofs
have dried out in d r o u g h t s , but they always
bounce back. Don't water them. Don't mow them.
Before I can continue, I'm interrupted by an- In areas of low rainfall, 10" to 12" of earth might be
other q u e s t i o n , this time f r o m an e n v i r o n - better to promote a deeper root system Find a
mentalist. g r a s s with good ability to bounce back from ex-
tended dry periods.
"I'm very concerned about using all these pet-
rochemical products and burying them where
they'll never break down. Could you speak to
Gladly. The properly built underground house
will save fossil fuels in the long run. More efficient
use of energy is still America's best potential for On the roof, use either extruded or high-density
reducing dependence on fossil fuels. There are expanded polystyrene. You're going to install a
European countries with standards of living equal drainage layer over the rigid foam anyway, so your
to ours, but with only half the energy consumption main concern should be R value per dollar. Get
per capita. And some of these, like Sweden, have prices locally and figure out the best value.
long, cold winters. At Earthwood, we use less than In our climate (9000 degree days), I like a mini-
half the energy for heating and cooling as other mum of R 20 on the roof (equivalent to 4" of ex-
homes of comparable size in our area use. Poly- truded), but R-25 (5") would be nice if you can
styrene insulation, which saves fuel, is a better use afford it. The earth, grass, roots, snow, etc.. will

lllus. 11-1. R o o f i n g detail for a freestanding earth

roof using moss sods to retain the earth

bring the roof up to code values, and you still have earth roof using moss sods to retain the soil is
the advantage of building in an ambient (the earth) shown in lllus. 11-1.
which is 20 to 60 degrees warmer in winter than On Earthwood's 4" roof insulation, which was
the outside air enveloping a surface-built house. made in two 2" layers, we tacked the second course
Installing foam is easy. Just lay out the foam in to the first by pushing in hundreds of 16-penny
large sheets. Four-by-eight-foot sheets are easier (3W) nails. These nails were for wind protection
to work with than twice as many 2' x 8' sheets. until all the insulation was applied. As you lay out
Use 2"-thick pieces, as opposed to the very light T' the plastic described in the next paragraph, you
sheets, which are more prone to being carried can pull out the nails. I don't like the idea of hun-
away by the wind during installation. "longue-and- dreds of nails aiming towards my waterproofing
groove foam sheets are nice, if they don't cost membrane, although the nails are still W away
more money. Otherwise, lapping the first course from doing any damage. These good and valuable
with the second will do the same job as T&G. In nails can be reused.
case the sheets become separated, lap the first
course with the second, even if T&G is used, to
guard against energy nosebleed. Lots of duct tape DRAINAGE LAYER
helps. Make sure you get good overlap onto the
We create a drainage layer under the soil in three
sidewall insulation.
It isn't necessary to insulate the overhang of the
building, but I like to put an inch of foam there as a • Spread a sheet of 6-mil black polyethylene over
protection layer over the membrane. This foam the entire roof. Look again at lllus. 11-1. This
stops stones from putung a concentrated loud on layer is going to shed most of the water to the
the membrane. The edge detail of a freestanding edge of the building, taking the pressure oil the

waterproofing membrane. I consider this inex- flourishing and in need of a second mowing. (At
pensive insurance layer to be the best-spent Log End. the earth roof was an extension of the
money in the whole project. lawn area around the home, so we kept it mowed.
• Using 5-gallon buckets, spread 2" of #2 crushed At Earthwood, the roof is "freestanding," and we
stone on the plastic. Now you can relax, because rarely mow.)
a hurricane won't blow your work away. Sodding was hard work, but three of us, one
• Over the crushed stone, spread 2" of loose hay or cutting, two hauling and laying, applied the sod in
straw, creating a filtration mat to keep the drain- two days (lllus. 11-2). With the east and west
age layer clean, or buy a product made lor the berms, this amounted to about 1600 sq. ft. We
purpose. Environmentally, I like hay or straw found that lO'-square sods w e r e the best size;
larger pieces would break up in transport. S<xJ
should be cut damp, but not soaking, so that it will
hold t o g e t h e r a n d establish itsell in its new
SOD ROOF location.
We arridpntallv discovered what may he the
At Log End Cave we actually cut sods from the easiest way of all to start an earth roof. The hay
corner of a field which we had tilled, de-stoned, we'd used as insulating mulch was full of seeds,
and planted with timothy and rye back in June. We and the roof was becoming quite green on its own
found that 2" or 3" of topsoil came up with each cut in the spring. Unfortunately, we still needed an-
sod. As this was insufficient depth of cover to other 2" or 3" of earth to maintain a green cover
maintain the grass, we spread 4" of good topsoil all Had we known, we might have applied the whole
over the roof prior to installing the sods. Remem- H" of soil in the fall, and let the hay take over natu
ber that we were putting earth directly onto the rally in the spring.
membrane at Log End Cave, a practice I no longer
employ, so we screened out stones from the first
inch of soil. Such screening is unnecessary when
putting soil straight on the hay or straw filtration
mat Remove any stone larger than a tennis ball, as
such stones conduct heat and don't contribute to We found out at Earthwood that it's easier just to
the soil from the point of view of retaining mois- spread all the topsoil required (lllus. 11 -3), and
ture or supplying nutrients for the green cover. then plant it. Hie "topsoil" was a silty material
We put on the 4" of topsoil in late October. A which came f r o m a s t r e a m ' s floodplain on a
friend experienced in golf-course maintenance nearby farm. The soil is free of stones and has
advised me that it was too late in the year to put the relatively poor percolation, good for this purpose,
sod on. I was inclined to agree, but wanted to do as less earth is needed to retain water for use by
something to approximate the effect of a 6" earth the plant cover We used 8" of earth, which eventu-
roof in order to test the heating characteristics of ally compacted to 7". The drainage layer takes care
the home during the winter. 1 also wanted to know of over-saturation. We planted rye in early Septem-
for sure what would have happened if we'd put the ber and, thanks to a beautiful autumn, grass was
sod on at the end of October. On most of the roof coming through the mulch layer within weeks. By
then, we spread a 2" layer of hay to approximate late October the g r a s s w a s thick, green, and lush.
the insulative value of the missing sod and to pre- An alternative to grass is a roof of wildllowers.
vent erosion of the 4" of topsoil during the spring Vetch, purple and while, works well, as do many
thaw. We spread pine boughs over the hay as pro- other flowers. Seek local advice and choose
tection against the wind, and weighted the whole flowers suitable for the soil, climate, and sun/
mess down with long, heavy sticks. shade ratio of the roof.
As a test. 1 cut and laid 20 sq. ft. of sod near the One final point: If you decide to use heavy tim-
peak of the roof. My friend was right. The test bers such as railroad ties or pressure-treated
patch was still in bad shape late the following Jyne. landscaping timbers to retain the earth roof, be
The rest of the roof, sodded in early June, was sure to include three details shown in lllus. 11-4.

///u.s. 112. Laying sod is easy: With the side of your foot,
simply kick each new piece against those already laid.

lllus. 11-4. Retaining timbers are truss-plated to each other.


(11 Use galvanized truss plates to fasten the tim- an inch off the roof substrate. Shimming prevents
bers to each other. (2) Provide good drainage next ice-dam buildup, a problem we encountered at
to the timbers by the use of crushed stone right Earthwood and cured by retrofitting said shims
down to the main drainage layer. This drainage later. It's easier to do this shimming at installation.
will stop lateral pressure on the timbers. (3) Use 1" Thy the moss sods. That's what I'll be doing from
or %" pressure-treated shims to keep the timbers now on.
Closing In
Closing in Log End Cave involved four different The safest course is to buy them a n d store them, as
tasks: installing windows, building and hanging shown in lllus. 12-1.
the door, installing vents, and infilling with cord-
wood masonry.
There w e r e seven w i n d o w s at the original Cave,
not counting three skylights or the small Thermo-
pane* in the door. Each of the seven is a T'-thick
insulated-glass unit c o m p o s e d of two pieces each
of W plate glass enclosing Yi' of an inert gas (such
as nitrogen). All the w i n d o w s were custom-made
for the job by a local firm specializing in insulated
At the design stage of your house, go to the
nearest manufacturer of insulated glass and ask
them if they have any leftover units. All these firms
seem to store perfectly good clear units which for
some reason were never picked up by the cus-
tomer. M a y b e they w e r e cut the w r o n g size.
Maybe the customer went bankrupt. The upshot is
that many m a n u f a c t u r e r s are happy to sell you
these units at a fraction of their normal price. Ifs
actually cheaper for them to sell the units at 50% to
75% off than to take them apart, clean the glass,
and remake a smaller unit from them. Hard to Illus. 12-1. Keep the window units on planks in a
believe, but true. There are a surprising number of safe place while marking the sills and molding f o r
these small companies in business; find them in a installation.
classified telephone directory.
The three large south-facing windows at the
Cave were found this way, and they were such a Make your openings Yz' larger in each direc-
bargain that I designed the house around them. tion, so that you can leave a W dead-air space
Imagine my chagrin, then, when I went to pick around the entire unit. A 48"-square unit needs an
them up only to find that the owner of the plant had opening of 48V2" x 4HY?". If you're going to order
sold them to someone else after he'd agreed to sell the units to fit, it's probably best to o r d e r them
them to me. Ib his credit, he m a d e it up to me by after the rough opening is actually framed. This
making me b r a n d - n e w units to fit, at a very good way the actual, not the theoretical, dimensions can
price. All seven w i n d o w s (plus the little window in be quoted. Don't forget to subtract the VtH from
the door) cost a total of $322.50, in 1977 dollars. each measurement.


I wouldn't advise single-pane windows at all, al installed straight. The caulking compresses, giv-
least not in northern climates. The condensation ing a permanent seal against draft. A word about
problem would be horrific in the winter, and the caulking: Silicone is good, maybe the best, but it's
heat loss entirely out of keeping with a home expensive. A more economical (and still high-
which is supposed to be energy-efficient. Even quality) alternative is to use acrylic-and-silicone
acrylic skylights should be double-pane, as ours hybrid caulks.
were. You can even buy double-pane skylights that Using gloves and the aid of two helpers, I put the
crank open in the summer, with a screen inside to window in place against the top molding, make the
stop insects. In the winter, the screen can be re- window plumb, and mark the sill with a pencil run
placed with an internal storm pane, to further de- along the outside edge of the glass; then we put the
crease heat loss. unit back in a safe place for a few minutes.
A double-glazed window properly installed on The bottom molding can now be installed, fol-
the south wall of a house actually admits more heat lowed by the two vertical side-molding pieces. I
than it loses, even if shutters or insulated drapes don't drive the nails all the way home, just in case
aren't used. Unfortunately, even insulated-glass an adjustment is needed. You can check again with
windows are big heat losers when installed on the the window unit itself. I can't think of any other
north side of the house. Here, triple-panes might way to check that the four pieces of molding are all
be cost effective. At Log End. we shuttered off the in the same plane. If the window fits llush against
north-facing windows during the winter, with lit- all the molding pieces, we remove the unit and
tle loss of light, as the snow usually covered them drive all the nails home. A bead of caulk is run
anyway, and the sun was low enough in the sky to around the inner perimeter of the molding so that
penetrate deep into the back of the home. The the glass will compress the bead when the unit is
skylights gave a lot of light, if you went up on the permanently placed.
roof once in a while to brush the snow away.
We installed the three large south-facing win-
dows first. The procedure follows the three steps
described below SETTING THE W I N D O W IN PLACE
Set the window on VA" wooden shims, leaving a
dead-air space between the unit and the sur-
INSTALLING EXTERNAL MOLDING rounding frame. Two or three shims will suffice for
We used rough-cut 1" x 1" stock for window mold- a 7'-long window. Rubber shims are also made for
the purpose; perhaps you can get some where you
ing, inside and out. Aesthetically, such stock is in
got the glass. The shims are 1" or 2" long and just a
keeping with the rough-hewn style of the house,
whisker less than the thickness of the unit, so that
and it's cheap. Our stock was a product of our
the plate glass itself will rest on the shim without
hemlock roof, although I prefer pine, which splin-
the shim getting in the way of installing the inter-
ters less. We installed the topmost piece first, to-
nal molding. Now carefully place the unit on the
wards the exterior of the 4" x 10" lintel so that we'd shims and press firmly to compress the bead of
have useful wide shelves on the inside. caulk. The unit is held in place with a small scrap of
I cut the molding to length and set 8-penny cup- molding tacked at the middle of each end of the
headed nails so that they were just starting to window. Now do the internal molding.
break through the surface on the other side. Now-
adays, 1 use my reversible drill and screw the
molding in place. Using screws not only permits
easy withdrawal, but there's less chance of smash- INSTALLING THE INTERNAL
ing a window I run a bead of sillcone-based caulk- MOLDING
ing along the edge where the nails (or screws) are
starting to break through and fasten the piece in This is pretty much the same as the procedure just
place. 1 like to lay the molding up against a known described, except that the window unit doesn't
straight-edge guide to be sure that the molding is have to come up and down anymore. Start with

lllus. 12 2. Double-caulk the windows by running lllus. 12-3. Installing the trapezoidal window in
a bead of silicone along the glass itself and the north wall. Since it was convenient to work
another bead along the molding surface to be from the outside, I fastened the inside
placed against the sill or lintel. molding First.

the top and bottom strips ol'molding. Again, caulk on it, there's an alternative. Install the windows
against the glass and between the molding and the without the caulk. Then, a f t e r all the w i n d o w s are
frame. One way to do this double-caulking is to in place, go a r o u n d with your caulking gun and
run a bead along the glass itself and, after the nails place a small bead of caulk at both edges of the
have been started, run another bead along the molding: against the w i n d o w and against the
molding surface which is to be placed against the frame. Do this on both sides of the window. It's
sill or lintel (lllus. I2-2). Nail the molding, taking very important to use top-quality caulk with this
care not to smash the window as you drive the method.
nails home. I usually hold a piece Of cardboard or
sheet vinyl against the glass as protection while
nailing the inside molding. Another trick is to an-
gle the nails slightly t o w a r d s the glass. This an- It's a fault of the original Log End Cave design that
gling gives more room for nailing and helps com- there aren't any w i n d o w s that o p e n I trust the
press the caulk against the glass. Using pilot holes reader will remedy that situation with his own
and screws would be a n o t h e r alternative. When plans. Pre h u n g window units, be they double-
the top and bottom moldings have been fastened, hung, casement, awning, whatever, are easy to
the temporary scraps which were used as stops install if you've left the right rough o p e n i n g . Don't
can be removed and thi* side molding installed. trust the figures given in the w i n d o w catalogs
The trapezoidal w i n d o w s on the n o r t h side They give you all kinds of c o n f u s i n g dimensions:
were much smaller than the full-size windows, and glazing area, unit size, rough o p e n i n g .
thus easier to handle. It was convenient to work Have the w i n d o w unit on hand a n d make sure it
from the outside, so the inside molding w a s fas- fits the actual rough opening The installation of
tened first (lllus. 12 -3). If you don't like handling the manufactured units can be accomplished in
and nailing the molding with all that sticky caulk very much the s a m e way as for the fixed units

already described. Follow the manufacturer's in- vents that I figured would be large e n o u g h to
stallation instructions. h a n d l e a n y s i t u a t i o n s h o r t of a skunk in the
stovepipe. The floor plan in lllus. 12 5 shows the
location of the six exterior vents (the seventh is the
DOORS doorl: one in each b e d r o o m , one in the sauna or
utility room, o n e in the living room, and two in the
You can buy pre-hung, insulated, and weather- kitchen dining area. By opening and closing vents
stripped doors, or you can make and hang the or the door, we w e r e able to create any cross-draft
door yourself. The first option is easier, and, if care we needed for the removal of kitchen odors, for
is taken i n the selection of a quality unit, you'll get a heating and cooling, or simply to e n c o u r a g e a flow
weather-tight door I was sorely tempted to use of fresh air.
factory m a d e doors, and we even spent an after- Our exterior vents w e r e homemade. They were
noon looking at various doors. Jaki convinced me designed to fit u n o b t r u s i v e l y into our 10"-thick
that I could make a door more in keeping with the c o r d w o o d - m a s o n r y walls. C o n s t r u c t i o n w a s of
rough-hewn atmosphere that we were after. 1 took aluminum mosquito screen and leftover 2 x 6s
a day and built the door. {Make sure you have at and 2 x 4s (lllus. 12 6). The vents w e r e built on a
least two doors!) It's 4" thick, but w h o knows what flat surface such as t h e concrete floor, stained, and
it weighs? nailed into place. Jaki m a d e the vent covers by
I won't go into detail about our own d o o r con- gluing a piece of 2" Styrofoam® to a piece of ply-
struction, as people w h o make their own d o o r s wood c o r r e s p o n d i n g to the shape of the vent sur-
like to do it their own way, but I do include a rounds, s o m e of which w e r e trapezoidal The
cutaway view of the door (lllus. 12-4). We stayed foam is carefully s h a p e d to fit the actual vent space
with this basic d o o r design throughout all the T he cover can be held against the vent surround
buildings at Earthwood. In fact. I still refer to the with a screw (or thumbscrew) and a receiving
original diagram to build new doors. ITie diagram socket. The cover can be decorated with material,
has stood the test of time Visitors are impressed paint, varnished w i n e labels, whatever.
with the thickness and mass of the door. Nowa- During the winter of 1978, we experienced no
days, I'm inclined to hang the d o o r s with three shortage of oxygen and didn't need to use the
barn hinges instead of two, so two doors can be vents at all, so we kept them closed to conserve
hung with three pairs of hinges fuel. The stoves g o t their required air from the
My n e p h e w Steve and his friend Bruce c a m e under-floor vents. In spring and summer, we used
back up one weekend in November to see how- the vents primarily to regulate humidity, opening
things w e r e going. We put them to work, of them on d r y days, closing them on humid days.
course, moving an old kitchen stove we'd bought
at an auction and installing the h o m e m a d e door
with heavy barn door hinges.
The defining design feature at Log End Cave is the
use of log ends, the individual building blocks
whit h make up that style of building known vari-
ously as c o r d w o o d masonry, stackwall construc-
Ventilation was one of the unknowns we had to tion, stovewood building, etc. (I w r o t e about this
deal with at Log End Cave. We knew we had to style of building in Sterling's Complete Hook of
have 'cm, as there weren't any w i n d o w s that Cordwood Masonry Housebuilding.) Hie wall
opened. Where should they be positioned? What derives exceptional thermal characteristics by vir-
size? How do you build them? The subject wasn't tue of the special d o u b l e - m o r t a r matrix, the inner
covered in the few articles about u n d e r g r o u n d m o r t a r joint being s e p a r a t e d f r o m the outer by an
housing I was able to track down in 1977, so I set insulated cavity. It's amazing how many low-cost
out to design the most flexible system I could, with u n d e r g r o u n d h o u s e s m a k e u s e of this weird

lllus. 12-4. This basic door design was used at Log End Give and at Earthwood.

and wonderful building technique, as a glance contribute to the warm, rustic a t m o s p h e r e we

through the case studies in this book will show tried to create. And ifs about the least expensive
See the color section for some lovely examples of way I can think of to fill in a post-and-beam
this technique. We feel that the cordwood walls framework.

lllus. 12-5. Floor plan (left) showing the

location of six exterior vents: one in
each bedroom, one in the sauna (or
utility room), one in the living room,
and two in the kitchen dining room.
The door functions as the seventh vent.

lllus. 12- 6. Vents (below) can be

constructed from aluminum screen,
plywood, weatherstripping, rigid foam,
and scrap 2 x 4s and 2 x 6s.
Retaining Walls &

lO'-diameter by 4' deep hole filled with fieldstones

and covered with 15-pound roofing felt and earth.
It's better if the footing drains finally come out
Because we were in a h urry to get the roof on, we above grade, if your site allows, as there's always
backfilled the home before building the retaining the possibility of the soakaway being full. As with
walls on the south elevation. Actually, backfilling septic-system drain fields, it's always a good idea
and building the retaining walls are very closely to slightly crown the soil above a soakaway so that
related: part of the function of the wall is to keep it doesn't become a depository for ordinary sur
the backfill from spilling around to the front of the face runoff.
house. We should have built ihe retaining walls The retaining walls are built of massive stones
first. As it was, some of the roof soil had to be unearthed d u r i n g the original excavation. We
retained with 2 x 6 planks until the walls were ended up with an incredibly strong and beautiful
complete. wall, while getting rid of a lot of pesky boulders
The main cause of retaining-wall failure in that would have made landscaping difficult Really
northern climates is frost heaving, so good drain- nice stones, of course, can be used to help create a
age behind the wall is imperative. For that reason, beautiful natural landscape.
we continued the home's footing drains behind the A typical stone in the wall weighs several hun-
large retaining wall and backfilled the wall with dred pounds, so there's little alternative to building
sand. The two small retaining walls would stand the wall with a backhoe. There must be good coor-
on their own without extra precautions, sincc they dination between the builder and the equipment
were little more than a single course of massive operator on this sort of work. Before the operator
boulders. Even if frost was to shift them a little, arrives, spend an hour or so looking over the
there'd be nothing to come toppling down. stones in the heaps. Then have the operator push
We completed the drains before starting the all the good stones to a central depot near the
actual stonework. The footing drains meet at the work. This centralization saves running around
southeast corner of the house (lllus. 10-4) and con- for the right stone when you need it.
tinue as retaining-wall drains until this large wall The procedure is much the same as building any
becomes a single course of boulders. At this point dry stone wall. Have the backhoe clear a fiat base
the drain becomes nonperforated 4" flexible tub- for the wall, depressed a couple of inches to give
ing, crosses the wide path that leads to the south- the large stones a bit of a bed in which to rest,
side entrance, and continues on to the soakaway 'fhen, using a heavy chain with chain hooks at each
(lllus. 10-3). The u n d e r f l o o r d r a i n s a n d the end, the backhoe can lift the required stone and set
kitchen-sink drain (greywater system) take a sepa- it into position. First use the stones with two paral-
rate path to the same soakaway. The soakaway is a lel faces for the bottom courses, saving stones with

lllus. 13-1. The large triangular spaces left between the boulders can be
tilled with smaller stones.

lllus. 13-2. Railroad ties or 6 x landscaping timbers can be used to build

retaining walls. The "dead men" are timbers set perpendicularly into the bank, and
they help resist lateral pressure on the walls.

only one face for the lop course. Il will be neces- LANDSCAPING
sary to shim with smaller Hat stones during build-
ing, so keep a pile of these handy. After the back- The area in front of the exposed cordwood wall on
hoe has finished, holes can be filled with smaller the south side must have good positive drainage
stones (lllus. 13-1). away from the building. Snow can collect there
and snow always melts, even where I live. Hie
OTHER KINDS OF RETAINING frost wall footing should be exposed 6" above
grade, and protected f r o m rapid heat transfer
WALLS with extruded polystyrene on the exterior. The
Not every excavation will yield megalithic stones. ground in front of the frost wall/doorstep thus
Retaining walls can be built with smaller stones, created should slope away from the building with
blocks, bricks, old railway ties, and landscaping a pitch of at least 1" every three feet. Use a cement
timbers. I've even seen old tires used, although 1 coating to protect any exposed polystyrene insula-
wasn't impressed with their appearance. What- tion f r o m sunlight. S u r f a c e - b o n d i n g cement
ever the material chosen, the principles are the works great, if you have a little left over, or you can
same. Build on undisturbed or compacted earth, buy pre-mixed trowel on products made for the
or, better, a shallow (3"-deep) trench filled with purpose.
crushed stone. Backfill with porous material. Use The area in front of the door was wide enough
perforated drain tile behind a high wall to carry so that we could back a car or truck right up to the
away water. door. Alter bringing in groceries or firewood,
When using railroad ties or landscaping tim- we'd park the vehicles in a turnaround area away
bers, include a few d e a d men in the wall. Dead from the house. A good pathway up to the house
men are 4'-long pieces of the retaining tim would have 6-mil plastic on the exposed earth,
bers placed at right angles to the face of the wall, large Hat stones or 2"-thick concrete paving tiles
and laid back into the earth (lllus. 13 2) during on the plastic to create a walkway, and 2" to 3" of
backfilling crushed stone between the paving stones. The

lllus. 13-3. I he intent of the Log End Cave design was to have a minimal irnpai t upon the landform

plastic stops weed growth in this area. If you pre- Cave, and others have commented favorably on
fer green grass (and tight mowing) to bare stone, the way that the home seems to fit unobtrusively
the area could be planted with grass. A brick or into its surroundings (lllus. 13-3).
stone pathway could run down the center. This Because of the lateness of the season, we post
kind of landscaping decision is largely a matter of poned spreading the sods until spring. Then, after
personal taste. the sod roof w a s in place (lllus. 13-4), we spread
All too often, people don't give sufficient seven loads of topsoil to finish the landscaping. We
thought to the final appearance of the property. planted timothy and rye, then rolled and mulched
Remember that one of the best reasons to build the entire area. The grass came in well.
underground in the first place is to create a habitat To keep down mosquitoes and blackflies in the
which harmonizes with its surroundings. An extra spring and summer, we kept the area around the
measure of grace is added to life when you can house mowed as far as the old stone wall that
step back from your work and feel good about it, bordered the meadow. We built our first mega-
so don't put off the exterior-finish considerations lithic stone circle on the lawn just to the east of the
any longer than necessary. home, using the last eight stones left over from the
At Log End Cave there were two huge piles of original excavation. Years later, at Earthwood. we
earth to be spread out over the site, the larger on refined o u r stone-circle t e c h n i q u e to include
the east side. A bulldozer is the best machine to use standing stones of up to 7' high and weighing two
to spread out the earth. Because we'd taken our tons. The experience of building the retaining
east side retaining wall all the way to the old stone walls at Log End served us well at Earthwood,
wall which surrounds the meadow, we had an which also featured megalithic retaining walls.
ideal depository for excess earth. On the west side, We left the forest wild beyond the old stone wall,
there was just the right amount of earth to achieve although we did remove a few small coniferous
natural-looking contours away from the roof. The trees which blocked sunlight from the home in
desired effect is that the roofline looks as little like winter, when we really needed the solar gain. Re-
an unnatural protuberance as possible. We felt moving these trees gave us the added benefit of
that this effect was successfully achieved at the being able to look further into the forest.

lllus. 13-4. The sod roof is in glace. The north gnhlc windows admit light to the hack rooms.
The Interior
discussed previously, o u r interior w a s designed so
NOTHING SPECIAL that we only had to fit one wall a r o u n d rafters. For
There's nothing special about building internal the most part, we used M>" dryw-ill (gypsum b o a r d
walls in an u n d e r g r o u n d house. Simple st ick fram or plasterboard) for the interior walls, painted
ing with.? x 4 studs actually works very well. Such with white-textured paint. We make our own tex-
a f r a m e will receive panelling, g y p s u m b o a r d , tured paint at great savings by mixing one part
rough or finished boarding, almost any surface. cheap white latex paint with five of pre mixed joint
We did most of the framing before backfilling the c o m p o u n d . Mix the ingredients thoroughly with a
home so that we could help buttress the walls. It's stick and apply with a roller. Vou can buy textured
easier to build the f r a m e for a panel on the floor rollers or make y o u r own by scorching an ordi-
and then stand it up (lllus. 14-1) than it is to toenail nary roller with a p r o p a n e torch. Such a roller
each stud to the floor and ceiling plates. gives a nicely textured 3-D look to the wall, al-
Because finished 2 x 4 s actually m e a s u r e though you'll use quite a hit of the mixture.
1V/' x 3Vz", we made the floor plate fast to the floor
with 2" m a s o n r y nails. Twelve-penny (12d) com
mons are fine for all other framing purposes. As


Our r a f t e r s are exposed and carried by three large
barn beams, and the internal walls are planned to
meet the underside of the beams. In order to di-
minish sound transfer f r o m the livinci to sleeping
or b a t h r o o m areas, we had to till the space be-
tween rafters (from the top of the beam to the roof
planking). But how? A simple and practical solu
tion, well in keeping with the motif of Log End
Cave, w a s to use c o r d w o o d masonry. For these
internal areas, we used 5" log e n d s and laid them
with a full width m o r t a r bed. Unlike external cord
wood walls, there's no need to include an insulated
space between the inner and outer mortar joints.
The mortar mix for c o r d w o o d m a s o n r y is 9
parts sand, 3 p a r t s sawdust, 2 parts Portland ce-
e a s i e r tn build the frame for a panel on the floor ment. and 3 parts builder's (or Type S) lime, equal
and then stand it up, rather than toenail each stud parts by volume. Use sand that's s u g a r y in texture
to the floor and ceiling plates. rather than coarse, as the mix will be less crumbly.

both the kitchen and parlor stoves had their backs

to the stonework, the r o o m divider acts as a heat
sink or storage heater. A m a s s of stone 5' high. 8'
long, and V thick, it has forty cubic feet of stone
m a s o n r y . At an a v e r a g e w e i g h t of a b o u t 145
p o u n d s per cubic foot, the m a s s w e i g h s 5800
pounds, almost 3 tons. That's a lot of BTUs of heat -
storage capacity, and it'll store "coolth" in the sum-
mer. Because the stone m a s s isn't in direct contact
with the earth (remember the under-floor insula-
tion), all the stored heat f r o m the stoves is given
back into the room as the interior temperature
tries to fall below the m a s s t e m p e r a t u r e . The heat
stored in the m a s s exerts a m o d e r a t i n g influence
lllus. 14-2. Log ends fill the space between rafters on the room temperature. The block walls and the
from the top of the beam to the roof planking. floor also act as heat sinks, but they don't get
They go well with the rough-hewn style of the "charged up" quite as well as the stone heat sink
house. located right next to the radiant stoves.
Building the room divider was fun, although I
The sawdust should be softwood, such as pine or w a s hesitant about starting b e c a u s e I'd never built
fir, passed through a V" screen, and soaked over- a freestanding stone wall before. Building the wall
night to absorb water before it's added to the mix. and a large stone hearth on the lloor took 5Vz days,
The use of dry sawdust would be a big mistake. It with Jaki doing all the m o r t a r pointing with the
would rob moisture from the mortar, accelerating homemade pointing knife. We took the time to
the mortar set (curing time), instead of retaining include several design features, like stone shelves
moisture and retarding the set. The mortar is left and a massive stone table (lllus. 14-3), stones with
recessed so that the wood is "proud" of the ma- fossils, and a raised scat in the hearth in the shape
sonry background by VI to W. Finally, the mortar of a butterfly (lllus. 14 4). My m o r t a r mix was 5
is smoothed with a pointing knife made from an parts sand. 1 part Portland cement, and 1 part
ordinary stainless-steel butter knife which has had m a s o n r y cement. It's a s t r o n g mix, light in color,
its last inch bent to a 30" angle. recommended to me by a m a s o n . The stones were
Each panel between rafters has some different taken from the old stone wall (in front of the house)
design feature, such as symmetry, a specific pat- which had to be cut through with the bulldozer to
tern. or a particular kind of rubble effect. Cord- provide vehicular access to the front door.
wood masonry (lllus. 14-2) ties the interior archi- At Earthwood. we took the heat-sink idea to
tecturally to the exterior, a n d p r o v i d e s an another level. There, we included a 23-ton ma-
interesting and pleasing design feature to the sonry stove at the center of the home. The ma-
open-plan living-kitchen-dining area. Also, it's fun s o n r y stove has a firebrick-lined firebox at its
to do. heart. We actually charge this great m a s s from
Alternatives to c o r d w o o d masonry in these lo- within. Wood is deliberately b u r n e d hot and f a s t
cations would be: m o r t a r e d bricks, sections of so that there a r e no u n b u r n e d w o o d g a s e s to be
4 x 8 timbers, o r drywall sections nailed t o fur- wasted up the chimney, or, worse, to condense in
ring strips. the chimney in the form of dangerously flammable
creosote. The heat from the hot burn is transferred
to the stone (or brick) by way of a series of horizon-
tal flues. When the c h a r g e of w o o d is reduced to a
bed of red-hot coals, with no m o r e colored flame,
the stove is closed d o w n almost airtight. A blast
The room divider between the living and dining gate (damper) is closed so that the coals last a long
areas is of solid stone masonry a foot thick and, as time, continually charging the mass. Such a hot

Illustrations 14-3 and 14-4. Left: The room

divider includes stone shelves and a massive stone
table. Above: The hearth features a raised seat in
the shape of a butterfly.

burn is the most efficient way to burn wood, about

35% more efficient than using most wood stoves,
and the house (even a surface house) stays at a
stable temperature thanks to the BTUs stored in
the 23-ton mass (lllus. 14-5).


Because the floor level of the u n d e r g r o u n d home is
normally several feet below that of a surface dwell-
ing, special care must be taken to assure that the
septic system is d o w n g r a d e from the home and
that the house drain lias a positive slope. Outside
of that, plumbing and wiring are really no differ
ent from the systems used in any other house.
At the Cave, and again at Earthwood, we made
use of alternate systems, solar (photovoltaic cells)
and wind energy. Neither home is connected to
commercial power lines. At Earthwood, we p u m p
all our water with a piston p u m p married to an old
bicycle. Six or eight minutes of pedalling a day
provides all the w a t e r we need for our family of
four. lllus. 14-5. The masonry stove at Earthwood

Electrical wiring i an be included in the internal FLOOR COVERING

frame walls, in the same way as in conventional
houses. Because the floor is concrete, the National The floors at the Cave w e r e covered with sheet
Electrical C o d e (NEC) r e q u i r e s that all plugs, vinyl in the kitchen and bathroom, and carpeting
called duplex receptacles (DRs), must be ground- in all other r o o m s except the m u d r o o m a n d the
fault protected, either by the use of ground-fault- utility room, w h e r e concrete floor paint w a s used.
interrupted duplex receptacles or by the use of Because of o u r homesteading life-style and two
ground-fault circ uit b r e a k e r s in the electrical- German shepherds with perpetually m u d d y paws,
supply panel box. With g r o u n d - f a u l t - p r o t e c t e d we chose a used industrial-grade (short n a p ! car-
devices, the d a n g e r of electrical shock is elimi pet in almost-new condition. We carpeted half the
nated, particularly in highly conductive locations, house for $120 plus a little for a foam pad, impor-
such as near a sink or w h e r e a concrete floor tant for extended life of the carpet. Friends in the
might get wet. The g r o u n d - f a u l t i n t e r r u p t e r s carpet-laying trade laid both the carpet a n d the
sense a conduc tion of electricity to a location other sheet vinyl in e x c h a n g e for seven weeks' f r e e rent
than where it's intended to go (the ground fault), at the Cave while we visited Scotland, a g o o d deal
and the GFIs trip instantly to an open position. all around.
This happens so fast that the person w h o might It's possible to construct a w o o d e n lloor on
provide a g r o u n d won't even feel a shock. Years furring strips laid over the concrete floor, but this
ago in Scotland. Jaki w a s filling an electric kettle reduces the value of the concrete floor as thermal
from the kitchen sink while the kettle w a s still mass. If a w o o d e n floor is desired, you could leave-
plugged in. She experienced 240 volts (standard out the concrete lloor altogether. Cover the earth
house current in the United Kingdom), which she with 6-mil plastic a n d run floor joists f r o m footing
described as like being hit h a r d in the back with a to footing, keeping the joists a few inches clear of
plank. Had the circuit been g r o u n d - f a u l t pro- the plastic. Vent the underfloor space with floor
tected, she w o u l d h a v e m i s s e d this u n i q u e registers. You might even consider r u n n i n g duct-
experience. work u n d e r there if a f o r c e d - h o t - a i r s y s t e m is
It will probably be necessary to have at least a planned. A w a r m floor translates into a warm
few duplex receptacles a r o u n d the perimeter of house. A grade beam (a fancy term for an internal
the floor plan, in o r d e r to satisfy the code, which footing) or t w o might be n e c e s s a r y to r e d u c e
requires a duplex receptacle at least every 12' floor-joist spans. Another benefit to this type of
along a wall. With c o n c r e t e or concrete-block floor is that the underfloor space can be u s e d for
walls, and. incidentally, with c o r d w o o d masonry, running plumbing and electrical wiring.
the easiest way to install DRs is by using either Another flooring alternative is to apply hard-
metal conduit or code-approved surface-mounted wood parquet right to the concrete. This type of
wire mold. Wire-mold systems are m a d e by sev- floor is applied in tiles or in lengths of h a r d w o o d
eral companies, and come in different styles and flooring. Both systems involve the use of a mastic
colors. Wire mold is often used now in commer- made lor the purpose.
cial buildings, even restaurants, which might have In the past few years, I've b e c o m e e n a m o r e d of
stone or brick walls. It's attractive, safe, and easy to lloors made of old recycled roofing slates bedded
alter should a change be desired. Another advan- in a thin m o r t a r joint right over the concrete floor.
tage is that use of conduit or wire mold allows you With a b r u s h or roller we apply a coat of b o n d i n g
to build the basic structure first and wire it later. agent to both the concrete floor and to the under-
One manufacturer I've used is the Wiremold Com- side of the slates to be installed. The "underside,"
pany. Ask them for their catalog and wiring guide. in this case, is actually the side that used to be
Similarly, plumbing can be located within inter- exposed on the slate roof.
nal framed walls. And. personally. I'm not dis Using w o o d e n lath as a screeding guide, we lay
turbed by an occasional pipe which has to be left down a %" bed of strong m o r t a r (5 parts sand. I
exposed, although the reader may feel differently. part Portland cement, 1 part m a s o n r y cement).
Exposed plumbing can be repaired or added onto I'hen, with a r u b b e r mallet, we set the slates into
easily. the mortar, trying to keep them all in the same

plane. Finally, any excess mortar is scraped away coats of a mix of 3 parts u r e t h a n c floor sealer and 1
with a trowel, and the 1" mortar joints between part wood stain, using a long-handled roller
slates are pointed with o u r old friend, the home-
made pointing knife (lllus. 14- 6). When the m o r t a r fhe liquid mixture is, of course, deepest in the little
is fully cured, in about four days, a slate sealer can valleys between die tiny ridges, so it remains darkest
be applied to seal the lloor and b r i n g out the inter- there, fhe overall color is a dark antique brown 'I'he
esting colors of the slate. We initially apply t w o best color to use . . . is Minwax's Istainl called
coats of sealer, and we reseal the tloor every year
or two (lllus. 14-7).
Malcolm Wells likes a particular concrete-floor Concrete floors are hard on the skeletal system,
technique that he calls the least expensive, the as two hours' shopping in a mall will quickly con-
toughest, the best for solar-energy a b s o r p t i o n , firm. Slate, sheet vinyl, h a r d w o o d tile—even Wells'
and the least slippery of all finishes. Wells says: faux leather—don't help m u c h . C o n s i d e r using
throw rugs, rush mats, or padded carpet where
The final product looks for all the world like antique you'll be doing a lot of standing; or use a wooden
Italian leather, and we've fooled many people into be floor.
licvincj that that's exactly what it is when they rave Furnishing and decorating are matters of inch
about it. vidual taste, of course, and I've really no intelligent
comment to make except to advise keeping the
In short. Wells s c r e e d s the concrete floor, b u t walls and floors as bright as possible. The excep-
doesn't trowel it smooth. After the construction tion would be a floor which receives direct solar
work is finished and the house is ready to be occu- gain. A dark floor absorbs heat better than a light
pied, he applies a "cheap sealer" and then t w o reflecting tloor.
We lived in Log End Gave, summer and winter, for
three years. We've seen the outside temperature LEAKS IN SKYLIGHTS
vary from 40° to 90°F, while the interior tempera- Proper flashing of skylights is tricky business, and
ture ranged from 58* to 77°E The drainage and we had flashing leaks early on. Back in chapter 11,1
waterproofing systems met the severest possible described the flashing process as it should be
test one spring when it rained steadily for three done. In 1978, we weren't as careful in our detail-
days on top of 36" of compressed snow. The only ing, and didn't know about the Bituthene* mem-
problem encountered during that time was one or brane, which would have made waterproofing
two minor Hashing leaks at the base of the front easier and better than our efforts with the black
windows, easily repaired. plastic. But our main error was in not providing
Other problems have occurred, and I'll relate positive drainage around each skylight. Leakage
them now with comments about correcting them. in a roof isn't likely to occur on the wide o p e n
spaces, where the membrane can be easily ap-
plied. leakages are far more likely to occur where
ENERGY ' NOSEBLEED'' there's any projection (such as a chimney or sky-
light) through the roof.
On the south side of the home, at the corners, the We cured our flashing leaks by digging up all
concrete-block walls transfer heat from the inte- the earth within 12" of the skylights and the main
rior to the exterior. Or, think of "coolth" being
transferred from the exterior to the interior. It's all
the same to me: energy "nosebleed." We didn't
notice it until December of our first year in the
Cave. The bedroom wall, and, later, the mudroom
wall, were getting wet at their southern ends. Our
first thought was "Leak!" After all, the waterproof-
ing detail at those corners was particularly tricky.
Luckily, before tearing up all that earth, I remem-
bered si unething I'd read by Wells about conduc-
tion of heat at parapet walls and the need for a
thermal break.
I quickly covered the exterior of the block wall
with 2" of headboard. Within 48 hours, the wall
had dried up, its temperature now able to rise
above dew point. This sort of detail should be
considered at the design stage. One solution is to
install the rigid foam during construction, held in
place by the retaining wall or mechanical fas- lllus. 15-1. SkvligMs, chimneys, and other root
teners. Then cover the foam with a protective coat projections should be surrounded by good
ing to prevent UV damage. drainage material, such as #2 crushed stone.


stovepipe. We then cleaned and dried the area (lllus. 15-2). Thus the space next to the retaining
thoroughly and applied plenty of roofing cement boards w a s now well drained, removing the hy-
and 6-mil black polyethylene, well lapped. Finally, drostatic pressure.
we used sand to backfill the skylights and spread
wood chips over the sand for the sake of appear-
ance (lllus. 15-1) The sandy areas around projec- CONDENSATION
tions were connected to each other with 4" non-
perforated flexible tubing. The bathroom and In late spring we began to notice condensation
office skylight areas were drained by the same around the base of the wall Here's what happens:
tubing to the sandy backfill at the edge of the Because of the so-called "flywheel effect," the
building The 4" tubing was sod-covered and invis- earth's temperature is still quite cool at (•»' of depth
ible. Nowadays I use crushed stone instead of sand in May and June, This "coolth" is conducted up
through the footing to the area where the poured
us drainage around any projections through the
floor meets the first course of blocks. Warm moist
roof Stone drains much faster than sand, prevent
air hits the area, followed by: dew point, conden-
ing any i hance of a pool of water collecting next to
sation, damp. Because we'd failed to insulate right
Hashing details.
around the footings and under the floor, we didn't
have a lot of choices by way of cure. Someone
connected to commercial electricity could have
PRESSURE ON RETAINING BOARD run a dehumidifier. We didn't have that option. We
vented in the corners where internal walls met the
The earth roof exerted much pressure on the external walls, as per the dotted line shown in lllus.
2 x 12 retaining boards, probably because of ex- 12-5. We rolled the carpet back from the wall, so
pansion of the wet or frozen soil during the winter. that the rug wouldn't get mildewed We used some
By May the b o a r d s were almost 1" out of plumb. moisture absorbent material to dry things out We
The cure wasn't difficult. We removed the topsoil toughed it out, remembering architect lohn Bar-
within 5" of the retaining boards, and replaced it nard's warning that u n d e r g r o u n d houses have a
with a thin layer of sand over the membrane and a high humidity until all the concrete fully c ures.
l»" depth of #2 crushed stone up to the surface

Don't get nervous, Whenever possible, leave the build-

ing open on dry days. Drying out may take two
years. 10

In our case, we also had a good ton of water

trapped in the 4 x 8 hemlock rafters, which had
been trees just days before they were installed.
Hemlock has what is known as 110% moisture
content, which means that the green wood weighs
110% more than it d o e s when the wood is fully dry.
Because of the waterproofing, all that green-wood
moisture can only c o m e into the house. Wood
takes about a year per inch to dry through side
grain. Using woodstoves helped dry things out
quite well, and we used the stoves every two or
three days during that first June.
lllus. 15-2. Removing the sod within 5" of the
The second spring was better. Humidity was
retaining hoard and replacing it with a layer of lower, but there was still some condensation. The
sand covered by ti of crushed stone improved the third year was even better, but not completely free
drainage and relieved the lateral pressure from of condensation in the corner rooms. An ounce of
the retaining board. prevention is worth a pound of cure, as we found

at Earthwood. where we w r a p p e d the footings were phenomenal; the view to the south w a s excel-
with insulation as described in this book. We've lent. The home w a s rustic and cozy. There were
had no wall or floor condensation at Earthwood. unexpected joys, too. like the sun's rays on sum-
Humidity is comfortable, upstairs and down. mer evenings giving natural illumination to the
d a r t b o a r d , and being able to lie back in the bath-
LIVABILITY tub bathed bv sunshine f r o m above. Of course, it's
possible for folks to walk o n t o the roof and peer
ITie hvabih(\ ithe quality of life in the home) of Log into the skylights, but they're unlikely to do this
End Cave exceeded our expectations. There was more than once.
plenty of light; heating and cooling characteristics Our goals f r o m the outset were: to create a

home with superior heating characteristics; to little doubt that a masonry stove at Log End Cave
own it ourselves instead of the bank owning it for would have reduced wood consumption to 2Vz
us; and to maintain the beauty and natural har cords or less. As I write in 1994, we can still buy
mony of the land. good hardwood for less than $100 per four-foot
cord, so we're talking about an annual heating cost
HEATING of $300 to $350 per year, even if we have to buy our
Over three years, our average fuel consumption at While living at the Cave we found that we
the Cave was between 3 and 3V<\ cords of medium- burned 95% of our fuel in our airtight kitchen
grade hardwood per year. This converts to about cookstove. We fired the parlor stove fewer than ten
3B0 gallons of fuel oil or about 9500 kilowatt hours times a year, occasionally to charge the thermal
of electricity (see Table 3). F.arthwood uses 3V* to mass of the north wall during particularly cold
:iVi turds of wood per year, but has over twice the times, and sometimes just to enjoy the stove's open
usable floor area as the Cave, so it's twice as effl- fire.
uent in terms of heating units per square foot of We found that the home required less fuel when
living space. I attribute this efficiency to Earth- there w a s a g o o d snow load on the roof. During
wood's round shape (the least skin to enclose a the w i n t e r of 1979-80, a l t h o u g h considerably
given area), insulation better than what we'd used warmer than the record-breaking cold winters of
previously, 16" cordwood walls above grade, and 1977-78 and 1978-79, we actually required a little
the use of a highly efficient masonry stove. There's more fuel because of the total lack of snow until

mid-February and the resulting frozen roof. On month. In an u n d e r g r o u n d home, temperature

February 11.1980. we left lor a month's vacation. A curves have shallower slopes than those of surface
friend living next door at I.og End Cottage came in homes.
to the Cave every day to look after the dogs and to We've experienced these cooling advantages at
record the home's inside temperatures. There was both the Cave and again at Earthwood. Others
no heat in the Cave except for solar gain and the re|x>rt similar observations. Underground houses
heat produced by two German shepherds. Win- don't require air conditioning, although they may
dows and skylights weren't shuttered. The tem- require some form of dehumiditication at certain
perature in the house bottomed out at 39°F on times of the year, particularly in the southern U.S.
March 1 (with the temperature - 15°F outside) and
then began to r ise (lllus. 15-3). It should be noted
that the temperature was always taken at 7:30 A.M., ECONOMY OF CONSTRUCTION
the very coldest time of day in the house. On sunny-
days. the internal temperature would rise six or The m e t h o d s of construction described in this
eight degrees by 2 P.M. book have been c h o s e n b e c a u s e they can be
I wo worthwhile points are shown by lllus. 15-3. readily learned by the owner-builder, and they're
One is that earth-sheltered homes, set in the moderate in cost compared with poured-concrete
above-freezing earth temperature, can be left for walls, pre-stressed concrete planks, and the like.
extended periods in the winter without employing The materials and contracting (excavation and
special efforts (such as threshold heating) to keep landscaping) for Log End Cave cost $6750.57. We
pipes and food from freezing. lllus. 15—3 also re- also spent $660 on labor, bringing the total cost of
flects the primary reason that an earth-sheltered the basic house to $7410.57 Covering the floors
house is easy to heat: The home is set into an with g o o d used carpets and new vinyl (concrete
ambient temperature that's warmer than the out- paint in the mudroom and utility room) added
side air. This theory was presented way back in $309.89, and fixtures and appliances added an-
chapter 2 (lllus. 2-8). other $507, bringing the total amount spent on our
house to $8227.46. I'm sure that this figure is
within 3% of the actual spending, as we kept re-
ceipts for practically every item. An itemized ai
counting appears in fable 4. The final total doesn't
In addition to the benefits of evaporative cooling include the cost of our septit system, well, wind
off the earth roof, there are two more reasons for plant, storage batteries, or furniture, all of which
ease of cooling in underground houses. One is the were already in use at the Cottage. The cost of the
"thermal flywheel cffcct." The earth's coolest tem- septic tank and the drain field in 1975 was about
perature lags about two months behind the cli- S600. In 1994, a similar system with a thousand-
mate's lowest average temperature readings at a gallon concrete tank and a medium size drain field
depth of 5.3 feet. I"he earth just outside the walls is would (if contracted out) run about $2200 in our
likely to be coolest then, in early March, gradually area. A lot of money can be saved if you can oper
warming up to its warmest temperature in early ate equipment yourself. The wind plant, tower, and
September (Another great thermal mass, Lake batteries cost about $1300. We later upgraded to a
Champlain, performs similarly: The highest water better wind plant. The spring-fed well was already
temperature is generally recorded about Septem- there when we purchased the property.
ber 1.) This thermal lag also helps with heating 'ITiese figures don't place a dollar value on the
during the first months of the heating season. owner-builder's labor, but I believe this value
The second cooling advantage is that the under- should be measured in terms of time, not money.
ground house, with its characteristic slow tem Money saved is a lot more valuable than money
perature change (thanks to the moderating effect earned, because we have to earn so darned much
of its own great mass as well as the surrounding of it to save so precious little.
cool earth), can effectively average out variations Granted, the cost figures are old. People ask me
in temperature over a day, a week, or even a what they could expect to spend for such a home in

Roughly 1800 hours of work w e r e required to

build l.og End Cave. 'ITiis includes about 450 hours
of outside help, s o m e of it paid, s o m e of it volun-
teered. The bulk of the work was d o n e during a
five-month period ending on December 17, 1977.
During .January and F e b r u a r y 1978. we completed
most of the interior work. Final landscaping and
placement of the sod w a s d o n e in June 1978. For
about half of the days. 1 noted actual h o u r s worked
in my diary, especially when this was required to
pay help, but for the other half I've had to read my
notes to see what w a s d o n e and g u e s s at the labor
time. Therefore, the 1800 hour figure is probably
not as accurate as the cost analysis, but I feel sure
that it's within 200 hours of reality. Each builder
will work at a different rate anyway, so the actual
time it takes s o m e o n e else will be different.


More than a third of the average American's after-
tax income is devoted to shelter, usually rent or
m o r t g a g e payments. If a man works f r o m age 20
to age 65, it can be fairly a r g u e d that he's put in 15
years (20 in California!) just to keep a roof over his
head. With a piece of land, six months' work. and.
say $16,000, he and his family could have built their
own home.
To save 14V£ years of work, you can't afford not
to build, even if it means losing a job. Granted, the
land and the $16,000 has to c o m e f r o m some-
where, but this a m o u n t is no m o r e (and probably
less) than the d o w n payment on a contractor-built
home, and about half the cost of a n e w mobile
home (figuring either to be the same s q u a r e foot-
age as an u n d e r g r o u n d home). The s u m is about
what many people pay for their car, which depre-
ciates rapidly. And you can't c o m p a r e what you
get for your bucks.
What do you get for your time and money? You
get a comfortable, long-lasting, energy-efficient,
1994 Since 1977. concrete, concrete blocks, wood environmentally compatible, low-maintenance
products, and heavy equipment have all just about home. You get the design features that suit you, so
doubled in price in our area. Improved techniques that the house fits like an old slipper. You get built
and building materials described in this book will in fire, earthquake, and t o r n a d o insurance. You get
add another 20% or so to the cost. My best guess is intimate k n o w l e d g e of the h o m e so that when
that a Log End Cave type of home can still be built maintenance or repairs are required you're the
in rural areas for a materials cost of about $16 per one who's best placed to make them. You get tre-
square foot, or S16,000 for a 1000 sq. ft. home, not m e n d o u s personal satisfaction; and you get free-
counting lot. well, septic, and gold doorknobs. dom from a lifetime of economic servitude.
Case Studies in the North
later to be the c e n t r a l divider. T h e result: an
EARTHWOOD energy-efficient, d e b t - f r e e h o m e , f h e trade-off:
West Chazy, N Y. t w o or t h r e e y e a r s of l e s s - t h a n - d e s i r e d living
We were comfortable at the Cave, but the family The o p p o r t u n i t y t o build E a r t h w o o d c a m e
was growing and the building wasn't designed for when, as the result of a contract g o n e a w r y Jaki
expansion. U n d e r g r o u n d houses are difficult to and I acquired s o m e of the m o s t important mate-
add on to unless expansion is addressed specifi- rials we needed to build a 39-' -diameter round
cally at the design stage. A few years ago, a young cordwood home, a n d a piece of land which we
couple with several small children visited us to accepted back f r o m a neighbor as satisfaction of a
discuss earth-sheltered housing. They insisted mortgage on a larger parcel w e ' d sold s o m e years
that they absolutely had to have 1600 sq. ft. of living earlier.
space, minimum. The trouble was that they could The house site w a s a mess, t w o acres of badly
only afford 800 sq. ft. of living space. I advised scarred gravel pit. Taking a leaf f r o m Malcolm
them to build the 800 sq. ft. that they could afford, Wells's extensive notebook, we decided to build on
leaving the east wall of surface-bonded blocks in- this marginal p r o p e r t y and r e n d e r it living and
sulated. but not backfilled. As they were both mak g r o w i n g and productive again. We need to re-
ing good money, they could afford to complete the verse the 20th-century attitude of building in all of
other 800 sq. ft. two or three years down the road. nature's most beautiful places. Instead, let's re-
This expansion could be accomplished by reusing claim blighted lands and make them beautiful.
the rigid foam (as long as it had been protected), We figured that it wouldn't a d d a lot to the cost
and utilizing the internal wall as a thermal flywheel of the h o m e to build a second storey, since the
and effective noise buffer between one side of the price of the roof and foundation, two of the most
house and the other (lllus. 16-1). An alternative expensive items, would be constant. S a m e with
detail would be a m a s o n r y stove in that east wall, the well, septic system, and driveway. This time
we'd build a house we could g r o w into, and that's
indeed how things turned out.
Although Earthwood looks completely differ
ent from Log End Cave, it is, in many ways, an
evolution of the same design, and makes use of
many of the s a m e t e c h n i q u e s : p o s t - a n d - b e a m ,
plank-and-beam, surface b o n d i n g , c o r d w o o d ma-
sonry. We corrected the glitches we'd encountered
at the Cave, and retained the positive features,
such as the large open-plan living-kitchen-dining
area, and the positioning of the b e d r o o m s away
f r o m the centrally located h e a t i n g system, this
time a m a s o n r y stove. E a r t h w o o d has twice as

lllus. 16-3. View of Earthwood and the Stone

Circle. Photo was taken from halfway up the
wind plant.

que. So the house w a s finished in stages, internally,

between 1980 and 1986, always on a pay-as-you-go
basis. Like most of the homes in this chapter, there
were no bank loans, no mortgage
When wood w a s the primary source of heat (he
used a small amount of electric backup) Richard
burned about 15 face cords a year, which trans-
lates to about 3% full cords, or 15% more wood
than we used at the original Log End Cave. As his
useful square footage w a s about 58% greater than
the original Log End Cave, 1440 sq. ft. versus 910
much space as the Cave, but the new house I Illus- sq. ft., the Guay home w a s very efficient, even with
trations 16-2,16-3) uses about the same amount of electric-backup heat figured in
fuel as the Cave used, while keeping a very con- There have been t w o recent changes to the
stant temperature. home, precipitated by Lisa's becoming a five-time
c h a m p i o n on "Jeopardy." The couple spent a
chunk of the winnings on a 9' x 40' solar room
attached to the south side of the home. I he addi-
tion, a manufactured unit, was professionally in
Richard was one of the first students at o u r school, stalled, with all low-E insulated glass. It supplies a
and the first. I think, to build from the 40' x 40' wonderful break to o u r long North-Country win
Cave plans. The home is located in Champlain, ters and a super play area for Monique.
N.Y., just 3 or 4 miles from the Quebec border. Hie The other major change was the installation of
basic plan (Illustrations 2 - 5 and 2-7), was mod- oil -fired hot water baseboard heat, which greatly
ified only slightly to a c c o m m o d a t e individual simplifies heating the home. Their fireplace insert
needs. Richard was single when he first built the is still available for emergency backup, however,
home llllus. 16 41. He married Lisa a while later, 'fhe small efficient oil furnace is located in the far
and the couple now have a daughter of six, Moni- corner of the new solar room llllus 16-5).

lllus. 16-4. Richard Guay's 441' x 40' Log E n d Cave

during construction in 1980, The home is now fully lllus. 16-5. The new oil furnace occupies a corner
earth-sheltered: berms on the north, east, and of the solar room.
west sides: earth on the roof.

We sometimes visit the Guays with a class of afford them, thus avoiding debt. Elaine's father,
underground-housing students. During my last who had a background in construction, helped the
visit. I asked Richard what he liked, what he'd couple design a roof f r a m e which would support
change, etc. He cited the low maintenance, no the 6" earth roof, and then he helped to build it
freezing pipes, good noise reduction, and har One night a couple of years after the earth was
mony with the environment as the positive fea- placed on the roof, Elaine was startled to find that
tures. The house is much brighter than he ex- their 3 horses and 4 donkeys were all grazing
pected it to be Changes? He'd install in-floor happily on the lush roof grass.
heating right from the start and use 16" thick cord- Walls are 8" poured concrete, and the home is
wood walls instead of 12". wrapped with Styrofbam' insulation, including 2"
of the foam under the floor. Like so many builders
of low-cost underground homes, the Cosgroves
choose wood heat, and they incorporated at the
UNDERGROUND STUDIO center of the house a magnificent stone masonry-
chimney, measuring 5' square by 6' high, and ta-
Elaine Cosgrove RieUy
pering to 3' square for the next 6' before exiting
Elaine has lived in and around Jay N.Y., for most of the roof. Two wood stoves are vented into this
her life. She was brought up just down the hill mass, which weighs some 16V^ tons. This beautiful
from her underground home in the farmhouse stonework exerts a moderating influence over the
where her mother still lives. In f a c t the property temperature of the home, in both summer and
where she built her home, land with a wonderful winter. Eight face cords (2% full cords) are used
view of the Adirondacks, was given to her by her each winter to heat, and they have about the same
parents. climate as we do, close to 9000 degree days. As the
Elaine and her husband started the home in July useful living area is about 4<)% greater than the
1981 with a $5000 grubstake, and while they built, original Cave (1271 sq. ft., not counting a small
they lived in an old trailer. They moved into the sleeping loft, versus 910 sq. ft.) and it uses only
32'4" x 42'4" underground home in October of 85% of the wood which we used, 1 think it's fair to
the same year, but it was "pretty rough," accord- say that the home is extremely energy-efficient.
ing to Elaine. They made things less rough over the The main differences between the two homes are
years by paying for improvements as they could the 2" of Styrofoam® under the floor and the inter-

nal chimney's ability to store a lot more heat than humidity level not commonly found in most stove-
the Cave's relatively small 5-ton mass. heated log or stackwall homes. In truth, the humid-
Elaine is on her own now and the home has ity level is a little too generous at times, and I mop
become the Underground Studio, where she pur up condensation on the windowsills frequently in
sues her various artistic talents, including the morning. But 1 look at this as a labor of love; it
underground-design work using a CAD drawing keeps the sills from rotting and keeps them fresh-
program on her computer. Her 27' long mural of looking as well.
the Adirondack countryside g r a c e s the living The rough plaster w a s smoothed out with a
room and studio. Creative works and books on art stucco paint mix made from four parts plaster-
are everywhere. The house is alive, inside and out. board taping compound and one part inexpensive
See the color section for photos of this house. white latex paint. Fhis coating gives the walls the
appearance of the lovely adobe buildings in the
desert Southwest. It creates a pleasing contrast
SIEGFRIED BLUM with the posts, rafters, girders, and exposed plank
ceiling, and greatly brightens the interior, lb add
Priceville, Ontario, C a n a d a visual interest, we left the colored bottle ends ex-
posed in these white walls.
"We built o u r e a r t h - r o o f e d r o u n d c o r d w o o d
home in 1987, slaving nine long m o n t h s from The exterior of the house (unplastcred cord-
morning till night, using one of Rob's earlier books wood walls) hasn't caused us any problems. Log
as a manual. I lowever, we decided on a single- shrinkage was minimal (we used white cedar ex-
storey version, adding our own touches as we clusively) and there were very few hairline cracks
went along. Would we do it again? There's a in the mortar joints, firmly pointed smooth by my
loaded question, not easily answered. Yes, we like wife.
our unique home (Illustrations 16 6 and 16-7) and On the roof, the thatch is getting richer with
its special qualities, like the small amount of fuel each year, and I still wonder how onion plants
required to keep it snug in winter. Our wood stove found their way up there, in place of the wildflower
is normally lit around 4 P.M. and left to go out by seed. I guess it gets too dry up there during the hot
midnight. The heat stored by the house keeps the summer. 1 don't water the roof. We built the roof as
temperature cozy until the next afternoon. outlined in Rob's book on c o r d w o o d - m a s o n r y
Part of our goal was a low-maintenance home, house building, only substituting 6" of beadboard
since retiring on a low: fixed income doesn't allow (two 3" layers) for the more expensive Styrofoam*.
for lavish expenses such as painting the exterior, and we've had practically no trouble in the seven
y e a r s w e ' v e lived u n d e r i t
or reroofing. A lot of chores connected with the
o w n e r s h i p of a conventional h o m e are non- The minor exception occurred during the sec-
existent here. ond winter, when a few d r o p s showed up near the
l he primary method of construction (curd- outside of the ceiling. The cause turned out to be a
wood m a s o n r y ) lends itself well to o w n e r - tiny air bubble in the overlap of the membrane. It
building. Hut beware, the patience required for was easy to locate the spot, press down the area of
seemingly endless, repetitive work, such as mix- Bituthene* concerned, and apply some Bituthene5*
ing 'mud' and laying up the log ends, becomes mastic. So, extreme caution is advised when laying
apparent only after you've started, and then there's the membrane. It's probably best to choose a
no turning back! It took us seven weeks to com- cloudy day to lay the membrane and to give the
plete the perimeter wall of our round house. overlap extra attention. We didn't use the recom-
We decided to rough-plaster the inside of the mended primer with the membrane, but it still
cordwood walls, applying Lhe same mix we used seemed to stick down to the wooden roof deck
for the laying up of the log ends, a mix which may really well.
have been too rich just for plastering. The benefit Our lloor is covered with local llagstone, an
of this plaster layer is that it acts as a quite effective other tedious job best forgotten. We sealed it with
vapor barrier, which, in combination with the roof an acrylic sealer. I wish now I'd used 2" of insula
membrane (Bituthene® 3000), keeps a comfortable tion under the slab instead of 1". I think that would

keep the cold stone floor somewhat warmer in and John (an English teacher soon to take early
winter, although this poses no great problem for retirement) wanted to deal with.
us. It's just something that could be improved Our planning, devising, and designing led us
upon with very little extra expense. In the summer, across the road, to a wild forty acres we owned.
the lloor puts its best foot forward. The delicious Planning was influenced by our environmental
coolness to walk on. together with the 'air- instincts and knowledge, and the design finally
conditioned' feeling in the home, make the hot coalesced into a small earth-sheltered, post-and-
summer months a pleasure to experience. beam, cordwood home. The wing-shaped floor
By now it should be apparent how we'll answer plan has but one bedroom, a galley kitchen, dining
the question at the outset of this story. Yes, we'd do area, and living r o o m . A small b a t h r o o m and
it again, but we're glad we don't have to. It was, at walk-in closet are hidden behind the kitchen wall.
times, a gruelling effort. However, it was a revela- We called it 'Earthward,' from Frost's sonnet To
tion of one's physical and mental makeup. We're Earthward.'
happy now to putter around the gardens and feel As the floor plan shows (lllus. 16-8), 'Earthward'
content with our achievement. The realization that is long and narrow, almost banana-shaped, l i t e
we did all this with our own hands is extremely end angles a r e a b o u t 15° to the southeast and
gratifying!" southwest. The result is that we get not only mid-
day sun, but morning and afternoon sun, too.

by John & Edith Rylander Digging In
See the color section for photos of this remarkable There's a utility room in the rear with ample stor-
home. age shelving, and access to the garden by a pene-
"From 1973 to 1987, we lived in a house we built trational entrance through the earth berm. A front
on five acres near a lake in central Minnesota. Our entry way porch (lllus. 16-9) is screened in summer,
three children eventually moved out on their own, enclosed in winter. Altogether, we have 875 sq. ft
and by 1985 we were alone in a 2000-sq.-ft. home, of interior space, and figure the home cost about
more room than Edith (poet, novelist, columnist) $20/sq. ft., exclusive of our labor.

lllus. Hi 9. "Earthward " (John Hylander photo)


The excavation was made in the autumn of 1986 the color and grain. Exterior ends were soaked in
and we poured the footings for the 55' x 15' struc- a wood preservative and old motor oil. so they're
ture on June 1,1987 When our house across the dark and contrast well with the paint.
road was suddenly sold, we moved into the Rafters and beams came mainly from a farm
weather-tight but very unfinished ' E a r t h w a r d ' several miles away. The farmer took two crops of
that same October. hay from our land for fifteen trees from his, a good
After six years, we find the advantages of earth- deal for all. Barter circumvents the need for cash.
sheltered living fully meet our rosiest anticipa- For structural and ornamental use, we used .i
tions. The southern orientation allows lots of sun- lot of field stone, inside and out Almost anyone
light and solar gain. The roof overhang and Vene- can learn to mix 'mud.' handle a trowel, and las
tian blinds keep out the hottest s u m m e r rays. stone in satisfying and structurally strong ways.
During the record breaking heat of 1988, we kept Recycled materials (lumber, doors, windows,
interior temperatures between 73° and 80"F, using plumbing, etc.) are much less expensive than new
only a ceiling fan and a box fan. commercial stuff, and they come in greater vari
Electric baseboard heat is available, but it hasn't eties. We used plenty, including a quantity of
been turned on except during vacations. The neatly stacked lumber which cost about 10% of
house won't freeze without backup, but we feared new material. In short, the use of cordwood ma-
that houseplants might succumb during our ab- sonry, indigenous materials, and recycled goods
sence Our heat comes from us. appliances and all helped save us trips to the bank.
lights, and a wood stove. Our firewood is dead or
downed wood from our property, yet there's still
plenty of tangles and stumps for wildlife shelter.
We've never exceeded 2Vz cords of wood per year,
and seldom need to stoke the fire at night to main- S o m e Relevant S p e c s
tain our 68°-72°F comfort range. Humidity is a • Footing: 3' wide and 4' deep on the front, with I
comfortable 30 40%. Except for the most humid of foam insulation on the outside. 2' wide and 3'
summer days, we've had almost no condensation d e e p on the back wall.
problems, and these minor ones were easily cured • F l o o r : 5" concrete slab, with b a t h r o o m and
with a small fan to stir the air around. kitchen drain lines embedded in the floor.
• Back wall a n d utility-room wall: 12" block with
conventional mortar bond, every fourth block
cavity slushed with concrete and rebar Asphalt
coating applied to exterior. One-inch foam insu-
We built ' E a r t h w a r d ' from materials we could lation top to bottom, 2" down to the 4' frost level
provide and work with. The south-facing wall, for Four-inch flexible drainpipe around entire back
example, is cordwood masonry. We'd never seen a wall at bottom, middle, and 6" below top.
cordwood wall or talked to anyone who'd built • Posts a n d b e a m s : 4 x 8 red oak rafters, 24" o.(
one, so we practised our techniques on the back ranging up to 16'6" long, and supported by the
wall of a garage, where mistakes wouldn't be too front and back walls, and a n 8 x 8 center beam
serious. We saw possibilities for both solidity and on 8 x 8 posts.
esthetic satisfaction, possibilities which became • Roof: 2"-thick white oak rough planking is nailed
realities, f h e cordwood walls in the home are 20" to the rafters '?" of rigid-foam insulation (H ID
thick and composed of aspen, oak, red elm, birch, a b o v e the p l a n k i n g . Two l a y e r s o f c r o s s -
and basswood logs which John had Cut from our laminated Bituthene' come next, and V liber-
woodlot. board protects the Bituthene". Drainage layer is
The large unsplit log ends shrank quite a bit, 2" of gravel covered with a filtration mat Roof is
and we caulked, inside and out. that first year. We topped with 9" of black dirt. (The Bituthene'
used wide mortar joints (painted off-white) with hasn't leaked, but we'd recommend placing it
the cordwood The inside cordwood ends were over the roof deck, below the rigid foam A few
planed, sanded, and varnished, which brings out mice have found their way into the insulation.)

• Front wall: 20" c o r d w o o d infill between posts on written for the standard, not the unconventional.
7' centers, with about 7" of cellulose insulation Better to cross every T and dot every 'I' than be
laid in between the inner and outer m o r t a r joints. forced into expensive changes.
No split wood. 14 different species of wood w e r e • Can you tolerate frustration? Things don't al-
used in the 'library wall.' a design feature of the ways go smoothly. A j o b you thought would take
home. five hours may take five days. 1<x)ls break, mate-
• Ceiling: 3V1/ fiberglass insulation was stapled rials run out or aren't available, or the weather
between the rafters. Molding strips of 3/V oak doesn't cooperate. Have a Plan B. even a Plan C.
were nailed to the rafters, up 2" from the bottom. Ulcers will greatly reduce your satisfaction in
Oak boards were cut to length and placed loose building your own home.
on the molding strips. Estimated insulation value • If you have a domestic partner, are you sure he/
of ceiling and roof is R-30. she really w a n t s to get into this? Not every pair
that lives together can work together. Mothers-
in law s o m e t i m e s r e g a r d house-building as a
Before You Build form of spouse abuse. House-building isn't a task
with which all w o m e n will be happy, nor will all
Building your own h o m e should start with self- • If you're working with a p a r t n e r w h o has fewer
appraisal. A few questions now may save anguish building skills than you do. can you transmit in-
later. Here's what you need to know before you struction and criticism in a way which doesn't
start: leave your p a r t n e r loathing both you and the
• Can you motivate yourself? Building a h o m e project? Can you take instruction and criticism
takes a long, steady effort. Will you keep at it, without too much ego-bruising?
good weather or bad, 'in the mood' or not? • Can you put up with each other's work habits?
• Have you the physical stamina to see the j o b Domestic tensions don't improve with the addi-
t h r o u g h ? Weight-litter s t r e n g t h isn't n e c e s s a r y , tion of h a m m e r s and saws
but some strength, balance, and general g o o d S a g e Advice
health is.
• Can you get professional help when needed? We In 1973 we built our first home, starting in May and
hired pros for s o m e of the block laying, plumb - moving into a very unfinished house in July. At
ing. wiring, and carpet installation, doing the about the same time, folks*we knew were building
rest ourselves. We wanted to finish in one sum- a house about the s a m e size as ours. During con-
mer and were willing to trade money for time. struction, h u s b a n d and wife and t w o small chil-
• Can you organize other people? We hired two dren lived in a small drafty cabin without indoor
recent high-school g r a d u a t e s for much of the plumbing. Five years later they w e r e still in that
summer, and had to plan each day's activities, cabin, while they worked to finish every small de-
making sure the right tools and materials w e r e tail of their new home, before moving in.
on hand . . . and do our o w n work at the s a m e Our willingness to move into an unfinished
time. home was to s o m e extent shaped by Ralph, who'd
• Are your finances adequate? Be sure you'll wind once helped John retrofit a b a t h r o o m in an old
up with a building fit to live in, if not entirely out-of-square f a r m h o u s e w h e r e we w e r e living.
finished, before you begin. Discuss your cost The last sheet of panelling caused major problems.
estimates with people w h o have experience in John and Ralph would trim it. bring it into the
building and building costs. Don't start unless bathroom, mark the sheet, take it out, and trim it
you're sure that you have at least the bare mini- again. After three tries it w a s still not exactly right
mum. Lenders a r e skeptical a b o u t o w n e r - Ralph eyeballed the situation and uttered a
builders using unconventional techniques. m e m o r a b l e word of advice. 'John, nail her down.
• 1 lave you checked local building codes and regu- We ain't buildin' no piano!'
lations? Which permits are needed? Which in- If you're building a piano, or a space shuttle,
spections are required, and w h e n ? Codes a r e then everything must be perfect. But for amateur

builders, doing your best, but settling for the solid than it had to. Building an energy-efficient, low-
and safe, if slightly imperfect, is a valuable lesson. cost h o m e with mostly local materials, and as
We needed to move into 'Earthward,' and did. much as possible with our own hands, is our state-
ment of how we think h u m a n beings ought to live.
Underground house, for many, c o n j u r e s up im-
Joining the U n d e r g r o u n d
ages of Grandma's fruit cellar or lorn Sawyer's
We made a conscious choice to go u n d e r g r o u n d . Cave. It's gratifying to see visitors c o m e in, blink,
We didn't want the most important investment of and say in startled tones, ' W h y this, this is. . nice!"
our lifetime to encourage clear-cutting of ancient In its comfort and beauty, ' E a r t h w a r d ' a r g u e s that
forest, global warming, the production of toxic ecologically sound housing need not mean shiver-
wastes, or mindless assembly-line labor, a n y m o r e ing in the dark."
Case Studies in the
Southern U.S.
After m o n t h s of collecting unwanted c e d a r s
ROGER DANLEY & BECKY from neighbors'" lawns, we decided against cord-
GILLETTE wood masonry. Too many of the cedars we cut had
These two first case studies describe Log End deteriorated sapwood and some heartwood dam-
Cave type houses built in the Deep South. Roger age. I guess everyone says this about their own
starts the story: region, but the South has a harsh environment.
"We'd just finished college in 1981, gotten jobs, Western red cedar won't last five years outside,
and dreamed of sailing around the world. Having even off the ground. In Canada. I've seen 500-year-
no desire for a long-term mortgage, we bought old logs of the stulf still in good shape. Instead of
land in rural hilly southern Mississippi, built a cordwood, we ended up with cypress shiplap sid-
pole-barn cabin, and moved in while we decided ing milled on the planer, with a 2 x 6 stud wall for
which house best suited our site. extra insulation.
Hie day we discovered Rob Hoy's book on un We got a deal on a large sheet of EPDM, a
derground housing was the day o u r course was synthetic-rubber membrane, g u a r a n t e e d for 40
set An earth-sheltered home seemed perfect for years in an exposed industrial setting. Our only
us and good for the hilly site. And, by building problem has been fire ants, who'll cat a hole right
ourselves, we could escape that d r e a d e d through a 120-mil (W) sheet. So we have to be
mortgage. vigilant in treating any m o u n d s we see on the roof.
We look back on the years of construction with I don't think it would have m a d e any difference
pleasure. We met many of our friends as a result of which m e m b r a n e we'd have used, they would
our alternative-housing choice. They, too. built have eaten through it. We m o w about twice a year
their own homes. so we can see any m o u n d s that appear. We have a
Building parties were the order of the day. I'll beautiful spring-to-fall wildflower meadow on
never forget the roofing party. We must have had lop.
20 people installing those oak planks with 5-pound
mini-sledges and 20d cement-coated nails. Our
only shortage was crowbars to remove the bent
nails. At the end of the day 1 swept up 20 pounds of Hie first winter after we closed in the house w a s
nails from the floor. very severe, - 2DF for days, and below freezing for
But the bulk of the work was done by Becky and a month. 'ITiat may not sound bad, but with the
me I lllus. 17-1). We used very few contractors, humidity we have . . .
mostly for the digging and cover-up. And did we We had a box stove set up in the house shell so
learn some things! For one, a certain brand of we could stay warm, and we started sleeping on
planer has a tough time with 5" x 10" southern an old hide-a-bed in the corner, That was a mis-
yellow pine and an even tougher time with 2" x 6" take. Once you move in. work becomes much
red oak. Hour after hour was spent feeding that more difficult. As a result, it took us several extra
machine. years to finish.

income from the earth house to further our sailing

Becky adds some bard facts:
"In a nutshell, here's what we've learned about
earth homes in the South:

• Since the ground temperature averages 70 to

72°F, the earth h o m e didn't stay cool or dry
enough without air conditioning. We live in a
hot, humid environment, and things like shoes in
the closet tended to mold, so we added a small
window air conditioner in the back room. It
didn't use much electricity, but made the house
comfortable. Our total electric bill tno gas ap-
lllus. 17-1 An early look at Bccky and Roger's pliances) averaged less than $50 a month year-
underground house in the "Deep South" (Becky round. ranging f r o m $90 in the summer to $30 in
Gillette photoI the winter. If you're building only for energy sav-
ings, an earth h o m e may not be the right choice.
But o n e day we were finally finished. Cedar Our stilt house on the coast has 2 x 6 insulated
planking In the dining room and cedar cabinets In walls and low-E gJass and has summer electric
the kitchen. Yellow poplar in the living room and bills as low as those for the earth home. But
kids' rooms, pine in the pantry, cedar in the mud- energy efficiency isn't the only consideration
room. Blackjack oak and cherry molding in the 'tornadoes have struck within a couple of miles of
bathroom. Ceramic tile throughout. A very warm us quite frequently, including a huge one that
and earthy place, all-electric with an average bill of wiped out an entire community. Then there was
$45 a month. the aftermath of Hurricane Andrew; In the rnid
With a west-facing exposure, we were forced to die Of the night, after high winds had knocked
put a porch in front to shade the face of the house. the power out, my brother-in-law James, staying
One sunny afternoon, while sitting with friends on in the pole cabin, came banging on the door
the porch, a man dropped by with an aerial photo- When the cabin started swaying, James headed
graph of our place. He had mistakenly centered for safety like a rabbit into a burrow.
the cabin, as the house was nearly invisible from • Roger mentioned the fire ants. They caused leaks
the air! I low's that for blending into the landscape? in our FPDM membrane, and it's a pain to fix
Al ter 12 years, our cattle pasture had turned into a leaks on a sod roof. A leak doesn't always drip
beautiful forest The most glaring object in the exactly where the hole is, so leaks are hard to
photo was our 22' sailboat, sitting unmoved these find. You can dig up a large area to find a small
last five years. Oh, yes, the dream. Sitting there hole. With 60" of rain a year, and as much as 8" in
that day, we realized we still had it. a day, even small holes can be big problems. You
So here we are a year later, moved 90 miles to can kill all the ants, and new colonies will spring
the coast, where we built a house on 8 x 8 piers. up overnight. I don't know the answer. A nearby
We have a long-term lease on the earth-house, to home was earth-sheltered with a conventional
people who really appreciate and take care of the roof, and it looks funny. Our earth home looks
place. Someday it will be our retirement home. In much nicer with the sod roof, especially when the
the meantime, the boat is in the water, we sail daffodils and coreopsis are in bloom. The house
regularly, and we're looking for 'the* boat. blends into the environment so well that people
In conclusion, I'd like to thank Rob Roy for all he drive up and ask, 'Where's the house?'
did for us. giving us the confidence to do what we • With our warm g r o u n d temperatures, we can
did. I don't think we would have our favorable have windows that open. In the Southern earth
financial situation had we bought into a mortgage, home we don't have the problem of condensa-
so the hard work was worth it. As it is, we'll use the tion on the walls.

• While most owner-builders probably don't plan 72"F. December, January, a n d F e b r u a r y are the
to sell, if you know in advance that you will, you'd coldest months, with t e m p e r a t u r e s in the teens
be safer with a conventional home. Most banks not u n c o m m o n . W h e n it's nippy, we fire the wood-
sell their m o r t g a g e s to Fannie Mac (Federal Na- stove, p e r h a p s 8 or 10 times during the winter.
tional M o r t g a g e Association), which requires f h e heat in the s u m m e r is another matter. Actu-
t h r e e c o m p a r a b l e sales a s r e f e r e n c e . T h e r e ally, Ray a n d 1 w e r e fairly comfortable, f h e house
weren't three comparable earth-home sales in would heat up to 84"E while outside temperatures
our area, hence it may have been difficult to get were in the high 90s. We have ceiling fans in every
bank financing. We could have owner-financed, room; so as long as the air w a s moving we w e r e
of course, as we didn't have a mortgage. okay. I lowever, o u r 'city folk' family wouldn't c o m e
• When potential buyers came to look at the house, to see us in t he s u m m e r b e c a u s e of the heat, so we
the c o m m o n response was that they loved how it gave in and installed air conditioning in 1985. I
looked, but were scared about it being different, keep the setting on 80°F Just that little difference,
scared they might not be able to get insurance plus the dehumidification, makes it much more
(we finally bought insurance ourselves just to p l e a s a n t inside. I must admit that even I enjoy the
show it could be done), scared they couldn't re- a / c now. f h e unit r u n s i n f r e q u e n t l y , a n d o u r
sell if necessary. Now that the home is being monthly p o w e r bill goes f r o m $50 to $80. Humid-
ity is a problem in the South. We use a dehu
leased, we've had several opportunities to sell it,
midifier in the winter. But a great many 'conven-
but we now want to keep it for our retirement
tional' homes here have more moisture problems
when the coast gets too crowded.
than we do. We've never had a mildew problem
• Our total building cost (excluding the value of
and many of my friends do."
our labor) w a s about $50,000. That included 4.2
acres of land, a pole-frame cabin we lived in for But they did have s o m e minor roof leaks, ac
two years, several sheds and a shop, a well and cording to Linda's letter of 1986, and termite prob-
septic system. By building as we could afford it, lems. T h e / d used a combination of waterproofing
and only taking out small short-term loans, the techniques, including black plastic and bentonite
house was paid for when it w a s finished." clay, but they had to dig up the earth at the e d g e of
the house, patch it and, importantly, fill in with
gravel so that the w a t e r could find its way to the
footing drains. Ray's view was, "I built the thing. I
can fix it!"
LINDA & RAY HURST Like Becky and Roger, Linda reported that she
I indn and Ray's home in Alabama is less than 100 and Ray moved into the house the minute the roof
miles away from Becky and Roger's in a virtually was on, covered temporarily with felt. They lived
identical climate that's not far from the Gulf of in "total confusion and disarray for what seemed
Mexico. In a 1986 letter. Linda wrote: like years."
"It was such a pleasant surprise to open the They r e m o v e d the c r o s s t i e r e t a i n i n g walls,
mailbox and find a letter from our mentor. You which had b e c o m e infested with termites. "This
may be sorry you asked about the house. I, too, w a s too close to the h o u s e for c o m f o r t . " said
could write a book! Bottom line: We love the Linda. "We replaced the ties with surface-bonded
house. But we've had our pitfalls. block walls."
We used your Log End Cave plan, but with Linda and Ray built their h o m e during the early
d i m e n s i o n s of 35' x 5 4 ' . Later, we a d d e d a 1980s, when u n d e r g r o u n d housing had captured
25' x 15' greenhouse on the south side. We prob- the public's imagination. " O n e drawback." Linda
ably shouldn't have faced the house west, but the wrote, "is the endless onslaught of the curious
view was too g o o d to pass up. In the s u m m e r They c o m e f r o m nowhere, people we've never
months, we put up solar screens to cut out the heard of even people from other states. It w a s
direct heat f r o m the evening sun. In the winter, the kind of neat at first, but we have long since tired of
sun helps w a r m the house, but staying w a r m is no having o u r solitude invaded. Every soul w h o has
problem. The coolest the house has ever been is ever even thought about building o n e of these

comes, picks our brains, and calls for advice, al- AUTHOR'S COMMENTARY
though they seldom follow it."
In 1993. Linda wrote: "Building this place seems
like a vague memory. We don't remember the hard
parts or the fact that it took years. It was quite an Ifs enlightening to hear from people like Becky
undertaking for two city dwellers with little and Roger, and Linda and Ray. It's so easy to forget
knowledge of building. I was a secretary and Ray- that climate and local conditions can make such a
worked on a river barge. But I'm glad we did it. We huge difference in the way a home should be de-
have great financial freedom now in our retire- signed and detailed. While the basic construction
ment as a result of no mortgage and very low- m e t h o d s d e s c r i b e d in this b o o k should hold up
upkeep and utility bills. We've also planted a seed throughout the U.S., 1 admit to having little experi-
in the minds of the boys, who both plan to build ence with destructive ants and termites. Once,
their own homes just as soon as they get their about ten years ago, a band of carpenter ants
degrees and get 'real' jobs. made a tiny hole in our Bituthene®'. We repaired
We used cedar for the cordwood masonry It's the hole with a patch and eradicated the ants, and
still beautiful inside, but we should have sealed the we've had no problem since. The Earthwood roof
outside. We have so much rain and moisture in the has the advantage of being freestanding. The
South that the logs have mildewed and turned earth doesn't berm up onto the roof, as in the
dark on the outside. I think we could use bleach Cave-type design, so i f s hard for undesirable
wildlife to gain access. Maybe a freestanding
and a brush to get it off. and then seal it. but we've
earth roof with a termite-shield drip edge would
got too many irons in the Fire to worry about it."
be a good idea in the "Deep South." Seek local
advice f r o m y o u r c o u n t y o r t o w n building
Linda reported further problems with termites
("insidious little beasts"):
"We chose not to treat the ground under and all If you live in the southern U.S.. I hope that you,
around us (like most people do here) because we too, will gain from the experiences of these coura-
didn't want to surround ourselves with poison. geous people.
But we had a problem with termites on the roof.
Fortunately, we knew what to watch for and got to
them before any significant damage was done.
However, we chose at that time to remove the d i r t
and replaced it with a more conventional roof. We
built the walls two courses higher, and then added OUR UNDERGROUND HOUSE
2 x l()s on 24" centers to make the roof even with by Geoff H u g g i n s
the blocks We added insulation, decking, and
shingles. So the surrounding earth is 10" below the "Here in Virginia an underground house faces a
roof. Once a year, now, Ray goes around the pe- mixed climatic situation. Its design must contend
rimeter ami sprays the g r o u n d and under the with both cold winters and hot, humid summers
eaves litis spraying has worked fine. Thus, one's motivation for going u n d e r g r o u n d
I guess it's a trade-off of sorts. I wanted the could be justified by hoping to save either heating
beauty of the wood in the ceiling, but I w o n d e r if it energy in the winter or cooling energy in the sum-
wouldn't be better to choose something termite- mer. Since we (Louisa Poulin. my construction
proof for the roof and ceiling. Can't use treated partner, and I) happen to have plenty of firewood
wood: more poison. I think sometimes we wind up on our land, the scales were tipped towards sum-
harming o u r s e l v e s m o r e than we h a r m the mer considerations. Design choices thus were of-
termites! ten driven by a concern to keep our house cool in
ITiere's another fellow here in town who had an hot weather.
underground home professionally built. I've no- We chose to use earth-bermed walls and an
ticed that he, too. has taken the dirt off his roof. I'm earth roof for three reasons. (1) The earth layer
inclined to think he had a similar experience." provides a buffer that greatly reduces seasonal

lllus 17 2. Geoff and Louisa 's underground home

in 1993 (Geoff Hug gins photo)

temperature fluctuations at the roof. A roof ex-

posed to air sees seasonal temperature fluctua-
tions of at least 110°F in o u r area, while sub-
m e r g e n c e u n d e r only 4" of soil r e d u c e s that
seasonal fluctuation to about 40°E (2) Because
earth doesn't insulate (it stores heat), solar energy
absorbed by soil during hot weather can be re-
turned for use during cold weather. (3) Vegetation
growing on the roof provides a natural cooling
process during hot summer days, due to evap- lllus. 17-3. Geoff with his power compactor. He
otranspiration. (Plants pull moisture from the soil supplied the power. (Louisa Poulln photo)
and evaporate it into the air.)
Our house (lllus. 17-2) is 25' x 40' (1000 sq. ft.),
inside dimensions. There are no interior walls, and Two retaining walls hold back the earth (lllus.
ceilings are 10' high, which makes the living space 1/-b): a short m o r t a r e d - s t o n e wall at the south-
seem larger than it is. lllus. 17-3 shows me at work west corner and a longer concrete block (surface-
preparing the sand pad prior to the floor pour. For bonded) wall m i d w a y on the east wall. The longer
good thermal contact with the earth, the floor is a wall is supported by six tiebacks Qllus. 17-6), which
cement slab. Above-ground walls are constructed w e r e made by attaching o n e end or steel cables
of cordwood, while u n d e r g r o u n d walls are dry- (through small holes in the back of the block) to
stacked c o n c r e t e block, c o a t e d with s u r f a c e - reinforcing b a r inside the wall, a n d the other end
bonding cement. Liking the w a r m t h of wood, we to large stones buried in the backfill.
chose not to have a concrete ceiling, and instead Large windows in the south and east walls pro-
designed 6" x 12" rafters (at somewhat less than vide a b u n d a n t natural light to most of the house A
4 foot spacing) supported on a 12" x 12" beam clerestory over the buried northwest corner
(lllus. 17-4) down the center of the house. Both brings much-needed light to an otherwise dark
beam a n d rafters were assembled by bolting to- location. Roof penetrations are held to a minimum
gether 3" x 12" rough-sawn oak timbers. Using by having water a n d cooking-gas lines and the
small c o m p o n e n t s made the weight of any one septic ventilation stack share the clerestory strut
m e m b e r m a n a g e a b l e e n o u g h that two people ture. There's also an air-ventilation d o o r in the
could lift it for assembly. clerestory to allow w a r m air in the house to rise

and flow outdoors. Fresh inlet air is provided by humidity throughout the house remains high all
three earth tubes entering the back walls of the summer. On those sultry days when the outdoor
house, down near the floor level (as well as a sepa- temperature is 95°F and the relative humidity is
rate outside air-feed supply for the wood stove). high, when some of that air moves indoors it's
A major challenge here in Virginia is condensa- cooled to about 75°F as its heat is absorbed into the
tion in the summer. Although we rarely get mois- mass of earth behind the walls. A lot of water
ture actually condensing on the lower block walls. vapor is dropped in the process. We must run a

dehumidilier 4 to 5 hours each day on hot and with the most important): (1) surface water is en-
humid s u m m e r days. S o m e w h a t in compensation, couraged to flow away f r o m the house by con-
our interior humidity in the winter rarely falls be- structing a p p r o p r i a t e s l o p e s a n d swales; (2) the
low 65%. umbrella sheds water that s e e p s into the soil, keep-
A major feature of our earth-cover design is the ing the earth d r y below it; (3) air spaces are pro-
insulation/watershed umbrella (lllus. 17-7) we con- vided by gravel layers arid plastic fibre mats im-
structed over the house. This concept w a s devel- mediately against the roof a n d walls, such that any
oped by John N. Hait, based on research done by water that may someday s o m e h o w penetrate the
the University of Minnesota and described in his umbrella will freely fall to the footing drains and be
book Passive Annual Heat Storage. The umbrella carried away; and (4) the outside of the roof and
is a multilayered treatment of 3 sheets of 6-mil walls are surfaced with asphalt roof coating and
polyethylene sandwiching 2 layers-of 2"-thick ex- 6-mil polyethylene sheets (primarily as a v a p o r
panded polystyrene (headboard) insulation. This barrier, not as a bulk-moisture defense).
'umbrella' a p p r o a c h appealed to us because it si- Note that the roof and walls of the house aren't
multaneously tackles ' w a t e r p r o o f i n g / insulation, insulated f r o m the s u r r o u n d i n g earth, but are in
and heat storage. direct thermal contact with it. It's the house and its
I use the w o r d ' w a t e r p r o o f i n g / although 1 be- s u r r o u n d i n g earth ball that are insulated by the
lieve that there's no such thing as a permanently umbrella f r o m the wide s e a s o n a l t e m p e r a t u r e
waterproofed u n d e r g r o u n d structure; sooner or variations of the air. If the walls are in thermal
later, moisture can be expected to breach the de- contact with this a d j a c e n t e a r t h ball, heat can
fenses. We implemented a 'water-resistant" ap- readily flow into the e a r t h during the s u m m e r
proach that consists of a 4-step process (starting (keeping the house cool), g e t stored in all that m a s s

of dirl, and bo released hack into the house during department very early in the process, and find out
the winter, sort tjt an annua! heatexchanger, John everything you can about the taw. The department
Halt points out that its critical thai ihe earth stay will have pamphlets and books to inform you
dr\ WfeI earth, rather than storing heat for later about regulations, By thoroughly planning your
use, will simply conduct it away. A main function of underground house beforehand, you can impress
the umbrella is to keep the soil dry iin addition to your code officials with your sincerity Have the
its insulating role) building well planned, but not to unreasonable
A few words on how we did it: The early first detail. Use Ihe expertise of the building officials to
step was kits of research We read every book and help design final details. Gel them involved in the
periodical on underground construction we could process. For example, the plumbing inspector
line I rhcre isn't an extensive literature, so inform helped me with drain-line details When he subse
ing oneself is not a gargantuan task. We took Rob quently came to the site to perform the inspection,
Roy's hands-on building courses, and 1 took an he hardly looked at the system, because he was
applied building science course. The value of a already familiar with how it would look. Give your
sound education cannot be overstated. We began code official credit for being cooperative. An
us neophytes, boot-strapping ourselves up from ounce of friendly persuasion in his office can avoid
ground zero. We'd never buill a structure before. a pound of grief when he a p p e a r s on site to
We tried a small cabin as proof we could pull it off, inspect.
and art1 now living in our second structure: our Louisa and I were fortunate to have good jobs in
underground house. the city, which allowed us to build some savings
Another phase whose importance is hard to We were able to leave those jobs, nearly 10 years
overstate is the design/planning sLage. Don't ab- ago, move out to the country and build full-time,
breviate this stage The complexity of the whole while residing in our cahin. The savings freed us
process seems overwhelming, especially to those from the need to borrow construction money. We
who've never built before. It's a complex pro- did all Ihe work ourselves, except digging the hole
t edure of many, many steps. Keep in mind, though, into the hillside and finishing the concrete slab.
that through careful planning the process can be (The t h o u g h t of 14 y a r d s of c o n c r e t e being
reduced to a sequence of simple steps: line them dumped on the ground and wanting to be evenly
up and take the first one. spread out in a couple hours' time w a s in timid at
If you have a building code to contend with, deal ing to me,) We've been building the house for over
with your local building officials with care, consid- 9 years now, and living in it for 7 years. It is about
eration, and honesty. They hold the power; don't BO 90"i, complete and very livable. Does it seem as
buck Ihem, but go with the flow. Compliance with though we should be further along, al ter nearly 10
the code can be a particular challenge when build- years? Maybe so. It used to concern us, ai times
ing an underground house, since your local code- But then we chanced upon an old Chinese proverb
enforcement official may be unfamiliar with the that evaporated this concern, and which we have
techniques you plan to use, Consult the permit since adopted: 'Man who finish house, die.'"
Appendix 1: Radon
Several years ago my good friends. Peter and in Ro's upstairs room. We were particularly con
Eileen Allen built a wonderful earth-sheltered cerned because we'd included into the walls of the
home. The home features two "double domes," home earth tubes for summer cooling. We wanted
built on 4' e a r t h - b e r m e d knee walls. A 26'- to be sure that we weren't going to subject a
diameter dome is constructed within a 29' dome, 9-year-old to the equivalent of smoking a pack of
the geometries matching exactly. The space be- cigarettes every night while he slept. We sent for a
tween the dome frameworks is filled with sheet- test kit, and were relieved to receive an analysis of
rock. a plastic vapor barrier, 12" of fiberglass insu just 0.4 pCi/l, or Ym of the EPA's action-level figure.
lauon, and a vented air space. The vapor barrier is Lucky for us. but what about our friends? They
quite tight, as are the concrete knee wall and floor, tested again in the summer and found that radon
except as noted below: Heat is delivered to the levels were only 1 or 2 pCi/l, well within safety
c o n c r e t e floor by a w o o d - f i r e d in-slab limits, but in winter, levels were high again. Re-
conduc tion-and-convection system, carefully laid member that their dug well was right in the house,
out by Eileen's father, dome designer George Bar- and quite close to the location of the wood stove
ber. It was decided to build the home over the well, which powered the in-slab heating system. Peter
so that there would never be any concern about observed that cobwebs near the top of the well
frozen pipes during severe winters. were rising up substantially whenever the wood
A couple of years after building the home, Pe- was burning. The wood stove was creating a nega-
ter's brother a physician, made a gift of a radon- tive pressure in the soil gases as the stove tried to
testing device to the y o u n g couple, c o n c e r n e d , find air to replace the air that was being exhausted
perhaps, about some of the construction tech- by the chimney. Given the seasonal nature of the
niques used. It may have been the best present radon problem, it was obvious that the wood stove
they ever received. When the test results came pulling soil gases from the well was the culprit.
back, they learned that the level of radon in the The last upper ten feet of the well was a 3<r-
winter in the home was about 40 picocuries per diameter metal culvert. Peter was able to seal off
litre (pCi/l), or ten times the "action level" recom- the well by pouring concrete in the bottom of the
mended by the U.S. Environmental Protection culvert. In addition, he sealed up other direct ex-
Agency. At 4 pCi/l. the bPA recommends that re- posures to soil gases, such as a house-drain clean-
medial work be done to lower the levels of radon. out. And, to help decrease the negative pressure
Radon's greatest health threat is increased risk while the stove was in use, he opened a second
of lung cancer. For example, breathing 20 pico- direct-air inlet to the outside. The efforts paid off,
curies per litre is considered to increase the risk of and the home's winter radon levels are now at or
lung cancer about as much as smoking two packs just under the F.PA's recommended action level
of cigarettes a day. Pete and Eileen, organic gar- The Aliens figure that the average level for the year
dener's and clean-air freaks, didn't m u c h like the is about 2 pCi/l. Eileen asked me to emphasize that,
after construction, any airtight house should be
idea of having SO times more risk of lung cancer
tested for radon gas.
than a nonsmoker living in a low-radon house."
About the time that Peter and Eileen were bat-
tling radon, Jaki and I were thinking of moving our Is radon a serious problem? The EPA estimates
oldest son. Rohan, to a bedroom in the fully earth- that eight million American homes may contain
sheltered part of Earthwood, down on the first unsafe levels of radon. The National Cancer Insti
lloor, in order to make room for little brother Darin tute considers radon exposure to be the greatest

cause of lung cancer after smoking, accounting between the lloor a n d footings, as this will provide
for 30,000 deaths a year in the U.S.* a place for infiltration of contaminated soil gases.
What are the ramifications for earth-sheltered Strictly speaking, the 40' x 40' Log End Cave plan
housing? Michael Lafavore. writing in his book is the largest h o m e which can be built without
Radon: The Invisible Threat, says: expansion joints in the concrete. If you must in-
clude expansion jointing, be sure to tool and caulk
Earth-sheltered and u n d e r g r o u n d homes have a the joint thoroughly.
larger surface area in contact with the soil, which
If you're planning on wood heat. \>e sure to
means m o r e places for radon to enter. "We've yet
provide a positive s o u r c e of outside combustion
to see an underground house that doesn't have
air to the stove. This pipe should be a 6" solid-
some radon in it," says B.V Alvarez,president of
walled pipe, fully glued, and sealed as it enters the
Airchek, a radon-testing company. 13
You should find out if the soils on the site where There's a chapter on radon in Wood ITame
you intend to build hold a lot of radon gas. Gener- House Construction (Sterling Publishing Co., Inc.,
alized m a p s of radon "areas" are only somewhat 1992). The chapter includes several pages of detail
helpful, however. Sometimes, one building lot may on both prevention in new construction and mit-
have high levels of radon and another one 100 igation in existing buildings. l"his is the best and
yards away may have no radon at all. The only way most u p - t o - d a t e i n f o r m a t i o n I've seen on the
to be s u r e is to conduct a soil test. One company subject.
which supplies a site-testing kit is Airchek. They'll Finally, there's a p r o d u c t called Enkavent* (a
mail you the kit and provide an analysis after you thicker version of AKZO's Enkadrain* drainage
mail the charcoal packet back to them. Set up the product discussed in chapter 10), which is installed
little c a r d b o a r d tent cubicle they supply on the under the floor pour. Enkavent* provides an air
subsoil itself, so if you're planning on a deep-hole space that intercepts r a d o n before it seeps into the
test for septic-system purposes, this would be a home. If concentrations of the g a s are discovered
good time to check for radon. after the house is completed, pipes can be added to
Your state health department is a good source route the g a s o u t d o o r s .
for getting names of reputable testing agencies in In conclusion, k n o w the enemy. Find out if
your area. Also, the EPA has two useful (and free) you're in a high-risk urea, f h e local or state health
pamphlets on radon. A Citizen's Guide to Radon department or cooperative-extension agent may
and Radon Reduction in New Construction. be able to help. C o n d u c t a site test, particularly if
What if the site is "hot"? 'fhere are techniques you're in an a r e a with granite, shale, or phosphates
that allow you to build on a site w h e r e there's near the surface. II nec essary, incorporate radon-
radon in the soil. You want to be sure that the lighting details at t h e design stage. It's hard to
house is completely sealed against infiltration of retrofit an u n d e r g r o u n d h o m e for radon protec-
soil gases. The Bituthene® m e m b r a n e will help a tion. Build with i n f o r m e d confidence, and don't
great deal, as will a 4" rein forced-concrete slab allow smoking in the home. Radon actually g r a b s
with 10-mil black polyethylene beneath it (in addi- onto smoke particles in the air, exacerbating the
tion to any rigid foam). Avoid expansion joints problem.
Appendix 2: Sources

Underground Space C e n t e r 790 Civil and Engi- E a r t h w o o d Building School, 366 M u r t a g h Hill
neering Building, 500 Pillsbury Dr. SE, Min- Road. West Chazy, NY 12992. Rob Roy, Director.
neapolis, MN 55455. Director: Raymond Sterling. Earthwood has offered workshops in earth-
Associate Director: John Carmody. Sterling and sheltered housing and cord w o o d - m a s o n r y con
(larmody have collaborated on a number of books struction since 1980, a n d the school acts as a
on earth sheltered housing, most of which are out books-and-plans clearinghouse in these areas.
of print. The center does photocopy s o m e of these
American U n d e r g r o u n d Space Association, 511 Surface-Bonding Cements
11th Avenue South, Box 320, Minneapolis, MN W. R. Bonsai Co., PO. Box 38, UlesvilJe, NC 28091.
55415. ' T h e AUA is an organization of profes- Surewall* Surface Bonding Cement.
sionals involved in e v e r y aspect of the under- Conproco, 24 Industrial Park Dr., Hooksett, NH
ground construction industry. More than 40 pro 03106. Foundation Coat*
f e s s i o n a l d i s c i p l i n e s a r e r e p r e s e n t e d i n its Quikrete, 1790 Century Cir., Atlanta, <JA 30345.
membership, including e n g i n e e r s , contractors, Quikwalf Cement.
suppliers, researchers, and architects." While the
AUA has been dealing mostly with commercial Stone Mountain Manufacturing Co.. Lafayette Ctr.
u n d e r g r o u n d s p a c e a n d tunnelling d u r i n g the Suite 304, PO. Box 7320, Chesapeake Blvd.. Nor-
past ten years, it reports a recent increase of inter- folk. VA 23509. Fiberbond*.
est in u n d e r g r o u n d homes. Waterproofing M emb r an e
British Earth Sheltering Association, Caer Llan W R. Grace and Co.. 62 Whittcmore Ave.. Cam
Berm House, Lvdart, M o n m o u t h , Gwent, Wales, bridge, MA 02140 Bituthene*
U.K. NP5 4JJ. Director: Peter Carpenter. "BESA is a
D r a i n a g e , Radon-Venting, & R a d o n - D e t e c t i o n
non-profit organization aiming to encourage the Materials
design and construction of earth sheltered build
ings in the U.K." It publishes at least three journals Airchek, Box 2000, Arden, NC 28704
a year Membership d u e s include the journals. AKZO Industrial S y s t e m s Co., P.O. Box 7249,
Asheville, NC 28802. EnkadrainEnkavent*
AUTHORS, ARCHITECTS Dow Chemical Co., 2020 Willard H. Dow Ctr., Mid-
land, MI 48674. Styrofoam* Therma-Dry«
Malcolm Wells, author, architect. U n d e r g r o u n d
Art Gallery, 673 Satucket Road, Brewster, MA Bonding A g e n t
02631. Ask Wells for his current book list. He's Thoro System Products. 7899 NW 38th St.. Miami.
also a v a i l a b l e f o r a r c h i t e c t u r a l s e r v i c e s a n d FL 33166. Acryi-60*.
Wire Mold
Don Met/, Metz & Thornton, Architects, P.O. Box
52. Lyme, NH 03768. Architect Metz has designed Wire mold Co., 60 Woodlawn St.. West I lartford.
a number of earth-sheltered homes. CT 06110


Most of the books listed below are available from U.S. Dept. of Health & Human Services. A Citizen's
the Earthwood Building School. Guide lo Radon. W a s h i n g t o n , DC: U.S. G.P.O.,
Alth. Charlotte and Max. Wells and Septic Sys- 1992.
tems. Blue Ridge Summit, PA: TAB Books. 1992. U.S. Environmental Protection Agency. Radon Re-
Alth, Max. Do-It-Yourself Plumbing. New York: duction in New Construction Washington, DC:
Sterling Publishing Co., Inc., 1987. U S. G.PO.

Armpriester, K. E. Do Your Own Wiring. New Wells, Malcolm. An Architect's Sketchbook of Un-
York: Sterling Publishing Co., Inc., 1991. derground Buildings. Brewster. MA: self-pub
lished, 1991.
Hait. John N. Passive Annual Heat Storage. Mis-
soula. MT: Rocky M o u n t a i n Research Center, . How to Build an Underground House.
1983. Brewster, MA: self-published. 1991.

Heldmann, Carl. Be Your Own House Contractor. OUT-OF PRINT TITLES

Pownal, VT: Garden Way Publishing, 1986.
Thy to find these books in your public library or in a
McClintock, Mike. Alternative Housebuilding. used-book store. S o m e U n d e r g r o u n d Space Cen
New York: Sterling Publishing Co., Inc., 1984. tcr b<x>ks a r e available from the Center in photo-
copied form
McRaven Charles. Building with Stone. Pownal.
VI": Garden Way Publishing. 1989. Campbell, Stu. The Underground House Rook.
Charlotte, VT: G a r d e n Way Publishing. 1980
Nearing, Helen and Scott. Living the Good Lite.
New York: Schocken Books. 1987. Lafavore, Michael. Radon: The Invisible threat.
Emmaus. PA: Rodale Press. 1987.
Oehler, Mike. The $50 and Up Underground
House Book. Bonners Ferry, ID: Mole Publishing Underground Space Center. University of Minne-
Co., 1992. sota. Earth Sheltered Homes. Plans and Designs.
New York: Van Nostrand Reinhold Co.. 1981.
Roy, Rob. Complete Bonk of Cordwood Masonry
Housebuilding: The Earthwood Method. New -. Earth Sheltered Housing Design. Second
York: Sterling Publishing Co., Inc., 1992. Edition. New York: Van Nostrand Reinhold Co.,
Sherwood, Gerald E., a n d Robert C. Stroh. Wood
Frame House Construction: A Do-It-Yourself —-. Earth Sheltered Residential Design Man-
Guide. New York: Sterling Publishing Co., Inc., ual. New York: Van Nostrand Reinhold Co., 1982.
Wells, Malcolm. Notes from the Energy Under-
Sobon, Jack Build a Classic Timber-Framed ground. New York. Van Nostrand Reinhold Co.,
House. Pownal, VT: G a r d e n Way Publishing, 1994. 1980.

A Closing In, 97-102
Door, 97, 100
Afford ability, 11
Infilling, 97, 100-102
Air Conditioning, 129
Vents. 97,100,102
Atrium. Covered, 10
Windows, 97-100
Bull-Float. 53
Backfilling. 90 Calculations, 52
Beadboard, 86 Dry Portland Cement. 54
Before You Build. Questions, 125 Engine Oil. Used, 37
Bending Failure, 12 Iron Silo Hoops, 37
Bending Strength. 11-12, 64 Mid-Floor Fence, 51-52
Bern ted I louse, 10. 23 Pouring Floor, 52 54
Bituthene®, 66, 80, 81. 82, 83, 85,112 Pouring Footing, 38-39
Applying, 81 85 Power Trowel, 54
Lap Seam. 82 Reinforcing, 51
Mastic, 81 Reinforcing Rods/Bars, 37
Blocks Construction Economy, Table, 117
Dry-Stacked, 6 Cooling, 116
Keying. 35 36 Cooling/Heating Efficiency, 11
Laying, 44-46 Cordwood Masonry, 55, 63,100-102,107, 108,
Mark Corners, 42 121,124,130
Surface Bonding, 40-41. 55, 56, 70 Alternatives, 108
Building Parties, 127 Mortar Mix, 108
Courtyard, Below Grade, 10
Case Studies, North, 118-126 D
Minnesota, 122-126 Dehumidifier. 129, 133
New York, 110 119, 119 120, 120-121 Doors
Ontario, Canada, 121-122 Make Your Own, 100
Case Studies, South, 127 134 Pre-Hung, 1CX)
Alabama, 129-130 Drainage, 17. 26, 86-88
Mississippi, 127-129 Materials, 86 88
lips, 128 129 Gutter, 87
Virginia. 131-134 Insulation. 87
Caulk. 99 Driveway Access, 17
Ceiling Spans, 11
Chambered I louse, 10. 23
Flashing, 83-85 Earth
Metalbestos® System, 84 85 As Capacitor, 14-16

As Conductor. 16 Framework, 6 3 - 7 2
As Insulator, 14-16 Barn Timbers, 63
Earth Sheltered Housing Design, 8,16 Log Ends, 63
Earthwood. 23, 26. 31, 37, 4<), 46. 51. 53, 55, 70. Post-and-Beam. 6 7 - 6 8
77. 85. 93. 94. 106. 108, 110.114. 116. 130 Recycled Timbers, 63 -64
Elevational Approach. 19 Rough-Cui Timbers, 63 64
Energy Shrinkage, 64
Crisis, 9 Frost Heaving, 3 2 - 3 3
Efficiency, 92 Fuel Equivalents, Table, 114
"Nosebleed." 46. 112
Solar. 131
Sources, 8
Enkadrain' 8 ', 87 Grayboard*, 37
Enkavent* 136
I-PPM. 127. 128 H
Escape. Means of, 11
Excavation Heating, 115-116
Backfilling, 26 Heating/Cooling Efficiency, 11,16, 92
Backhoe. 26 In-Slab, 30
Bulldozer, 27 Humidity, 113-114,124
Educated Guesswork. 2 8 - 3 0
Flat Surface Site, 28 1
Footing Trench, 30, 31 Impact. Planetary, 7
Front-End Loader. 26 Insect Pests, 128, 130
Insulation, 3 0 - 3 2
Exterior. 16
Floors Nonbiodegradable, 2
Blocks, 40-46 Overhang, 93
Carpeting, 110 Overlap, 93
Concrete vs. Wood, 110-111 Insulation/Watershed Umbrella, 133
Covering, 110-111 Integration, Floor/Support Structure. 64 65
Flagstone, 121 Internal Heal. 11
Grade Beam, 110
Monolithic Slab, 40
Hoofing Slate. 110 111
Sand Compacting. 40 Joints
Thickened Edge, 40 Keying, 31
Vinyl. Sheet. 110 Keyway Pieces. 38
Drains, 87. 89, 103 L
Frost Wall, 105
Measurements. 3 3 - 3 5 l a n d s c a p i n g , 105-106
Pillar. 46-47 Paving files. 105
Pour, 37- 39 Life Quality/Living Standard. 9
Ratios. 28 Lintels, 67
Reinforced Concrete, 28 Living the Good Life, 55
Forms. Forming. 33-39 Log End Cave, 11-20, 25, 31, 32, 37, 41, 56, 69, 79.
Levelling, 36-37 86. 94, 97. 106, 108, 114, 116
Preparing. 33-39 Log End Cottage, 87,116
Stakes, 35-36 Log Ends. 107-108

Materials Retaining Walls, 103-105,131
Availability, 11 Backfilling. 103
Indigenous, 124 Backhoe, 103-105
Recycled, 124 Drains, 103
Footing Drains, 103
Materials, 105
Megalithic Stone, 105
Nails Roof
Resin-Coated, 75 B i r d ' s - M o u t h , 69
Scaffolding/Duplex, 35 Crushed Stone, 94
Drainage, 9 3 - 9 4
Earth, 91-96
Earth. Advantages, 91-92
Paint, lextured, 10/, 121 Earth/Sod, 16, 77, 93, 94
Passive Annual Heat Storage, 133 Flashing, 82
Penetration Hay/Straw, 94
Entranceways, 10 Leaks, 128,129
Door, 16 Moss Sods, 77-79, 82
Sidewall, 10 Plank-and-Beam, 67, 6 8 - 7 2
Peripheral Rooms, Temperature, 13 Planted-Earth, 9 4 - 9 6
Planning, 134 Rafters. Dimensions, 68- 69
Backfill Materials, 23, 24- 25 Rafters, Filling Between, 107-108
Contour Map, 18-20 Rafters, Installing, 69-71
Contoured Site, 18 20 Rafters, Thickness, Width, 72
Deep-Hole Test, 23 Retaining Board, 112
Fit H o m e to Site, 18-27 Termite-Shield Drip Edge, 130
Flat Site, 2 3 - 2 5 Timbers, I Ieavy, 94
Location for Excavation. 29-32 Tbpsoil, 94
Marking Excavation, 2 0 - 2 2 TTuss Plates, 96
Radon Test, 23 Vents, 79
Septic System, 2 2 - 2 3 Wildllowers, 94,127
Soil Percolation Qualities, 23 Roof Deck, 7 3 - 7 9
Subsoil Quality, 22 Chainsaw, 77
Survey Land, 18-20 Fascia/Retaining Boards. 77
Water Table, 26 Joins, 75
Work Space Around Walls. 20 Peak, 75-77
Plumbing Planking, 7 3 - 7 7
Drains, 47 48 Planks, Nailing, 75
Kitchen Sink Outflow, 49 Roof Support Pac kage, 85
Supply, 49-51 Room Divider. 108
Toilet Cleanout, 49 As Heat Sink, 108
Vent Stack, 49
Waste, 49-51
Polyethylene. 6-Mil, 47, 80, 93, 105
Polystyrene, 92 Safety, Structural. 11
Expanded, 86, 9 2 - 9 3 Screws, I-ag, 77
Extruded, 47, 48, 86, 9 2 - 9 3 Shear Failure, 12, 70
Poured Concrete, 55 Shear Strength, 12, 64, 69

Site Considerations. 11 Underbill 10

Skylights, 79. 97,112
Double-Pane, 98 V
Drainage. 112
Flashing, 8 3 - 8 5 Vents, 100,102, 131, 133
Sodding. 94 Crossdraft, 100
Soil Stoves. 100
Compaction, 32
Percolation. 87. 90
Specifications. 124-125 Walls
Stove, Airtight, 121 Anchor Bolts, 60
Blast Gate/Damper, 108 Backfilling, 90
Masonry 'type, 108-10^) Bending Stresses, 89
Vent Line, 4 8 - 4 9 Bond Beam, 59 60
Wood, 129 Choices, 5 5 - 6 0
Wood-Burning Efficiency, 108-109 Condensation, 113-114
Stress-Load Engineering, 14 Conventionally Mortared, 56
Styrofoam® 77 Gypsum B o a r d / P l a s t e r b o a r d , 107
Styrofbam* Blueboard®, 31, 37, 46, 51,121 Insulation, Exterior, 86
Styrofoam® Therma-Dry®, 87 Internal 106
Support Structure, 66 lateral L.oad, 89- 90, 90
Surface Bonding, 56 Plate Dimensions, 61-62
As Waterproofing, 60 Riser Piece, 62
Surface-Bonding Cement, 56 Siding, Cypress Shiplap. 127
Sill Plates, 60
Sill Seal, 62
Surface Bonded, 56
Thermal Waterproofing, 81
Advantages. 7 Wastewater Systems, 2 5 - 2 6
Break, 16 Water, Pumping, 110
Contact, 133, 134 Waterproofing, 8 0 - 8 5
Mass, 16 Inside Corner, 81
Thermopane*, 97, 98 Membrane, 80
Tools Membrane, Home-Made, 80
Drill. 65 Under Insulation, 80
Level. Contractor's, 35 Wind Plant. 7,110
Windows, 131-133
Pre-1 lung Units. 99-100
Rescue/Egress, 16
Underground Construction 'techniques, 11 Setting, 98
Underground House Wire Mold, 110
Livability, 16 Wiring, 110
Strength, 16 Ground-Fault Protection, 110
Waterproof, 16 Solar Energy (Photovoltaic Cells). 110
Underground House Book, 17 Wind Energy, 110