NASA TT

F-9466

STATE OF THE ART O F THE THEDRY AND ANALYTICAL DESIGN METHODS FOR VORTEX PUMPS

N. N. Kupryashin and V. G.K.ovalenko

Translation of "Sovremennoye sostoyaniye t e o r i i i metodov r a s c h e t a vikhrevykh nasosov." Vestnik Mashinostroyeniya, No. 4, pp. 20-28, 1 9 5 7 .

Microfiche
ff

(MF)

653 July 65

NATIONAL A.ERONAUI'ICS AlTD SPACE ADMINISTRAnON WASHINGTQN JULY 1965 *

~

NASA

TI'F-9466
_.

STATE OF TRE ART O F THE THEORY AND A N A L Y T I C A L DESIGN METHODS FOR VORTEX PUMPS
ABSTRACT Recent a n a l y t i c a l treatment, primarily by t h e authors, of vortex (turbine, regenerative, periphery) pumps and c e n t r i fugal-vortex pumps i s reviewed, accounting i n greater d e t a i l

than formerly f o r such factors a s head, capacity, and efficiency requirements, impeller blading, outer channel cross section configuration, leakage a s d other losses, recirculation, shaft
r.p.m.,

etc.

Analytical and experimental c h a r a c t e r i s t i c curves

a r e given f o r specific Soviet models, with f a i r consistency between t h e two. High efficiencies are obtained using an

open impeller with

-

24 back-curved blades and a rectangular

channel sage and

(43$), or a closed impeller with-semicircular pas-

24 t o 32 forward-curved blades

(50%).

The theory

presented herein i s adequate f o r an a n a l y t i c a l determination

of t h e mean vortex velocity and ultimately f o r derivation of

a fundamental equation of the vortex pump. Vortex a d centrifugal-vortex pumps currently enjoy widespread applicatio& i n several branches of industry and agriculture.
pumps w e r e designed i n

The f i r s t prototypes,qf vorte

j .

4 -

1930. For 27 years, however, t h e t h e o r e t i c a l anhlysis o

the operation of vortex pumps has lagged considerably behind t h e i r u t i l s z a t i o n , which has i n t u r n held back the design of newer, more improved pump constructions.

- .-?Numbers i n t h e margin indicate pagination i n t h e o r i g i n a l foreign t&t.

I ___

In t h e SovieL l i t e r a t u r e (refs. 1 t o 7), these pumps axe referred t o a s

--

_____

__

vaned o r vortex pumps, in t h e German l i t e r a t u r e ( r e f s . 8 and

9) as self-priming

centrifugal pumps with peripheral channels, o r S i b i - pumps, i n t h e American lit+

erature (refs. 1 0and 11) as volute type vortex pumps, tangential pumps, vaned
turbine pumps, regenerative, turbulent, o r f r i c t i o n pumps, and i n t h e Japanese : l i t e r a t u r e (refs. I 2 and 13) as rotary, f r i c t i o n , o r periphery pumps. The di-

c(

1i

versity of nomenclature simplybears witness t o t h e d e s i r e t o distinguish these

1 ,

pumps from among other types, and the complexity of operation of vortex pumps.

The term "vortex pump" proposed by Prof. S. S. Rudnev i s t h e most appropriate, since it most completely embraces t h e principle of operation. a r e used predominantly i n t h e realm of low specific speeds (n =
S

Vortex pumps
-. 50), where-

4to

-_

ever t h e application of centrifugal pumps i s rendered unsuitable due t o low Given t h e i:, -r - - _-. - ___.._ . ~- -ul -- -,same impeller diameter and s h a f t r.p.m. t h e v o r t e x pump generates a head three :*7
I--

'efficiency, f a b r i c a t i o n complexity, and t h e lack of self-priming.

t o f i v e times a s great as the centrifugal counterpast, hence it i s normally designed t o operate a t high heads (25 t o 250 m) and r e l a t i v e l y l o w capacity (2 t o

60 m3/h).

A t t h e present, three types of vortex pump a r e i n use ( f i g . 1). The f i r s t

includes pumps with a b l i n d peripheral passage, where t h e suction and discharge

~

openings are located a t a lesser radius than t h e peripheral channel.

The
I ,

:second and t h i r d types are pumps w i t h a one- o r two-directional open petitipheral"

7 :

,passage, where t h e suction and discharge openings are s i t u a t e d immediat

t h e beginning and end of t h e channel o r t h e suction opening i s located l e s s e r raaius. Pumps of t h e f i r s t type have a self-priming feature, while pump

of t h e second and t h i r d types are deprived of t h i s f e a t u r e (without aux5liary jequipment) I---. --

.
~

---.----_I_-

__

._

.
c

.. .

Open t y p e impeller

Figure 1 . Diagrams of Vortex Pumps: 1) Suction; 2) Discharge; 3) Peripheral Channel; 4) Impeller; a = Channel Width; b = Impeller Width; c = Channel Depth; Ar = Distance from Impeller Hub t o Channel; s = B l a d e Thickness; I = Blade Pitch; 6 = End Clearance; 8 = Radial Clearance. P e The present a r t i c l e gives a concise c r i t i c a l analysis of t h e b a s i c hypotheses underlying t h e operating principle of vortex pumps, p r a c t i c a l recomendat i o n s f o r t h e engineering calculation and design of three types of such pumps, and t h e scheme adopted by t h e authors f o r describing t h e i r operation. The most detailed investigations of t h e working process of vortex pumps during t h e period
L, j
, I

1
, L.
c -

L kJ

* :

- *

__

1939 t o 1950

were undertaken by G. T. Berezyuk (ref. I), 0.

i

IV. Baybakov (ref. 2), and B. I. Nakhodkin (ref. 3). I

' 1
1
I

In t h e opinion of G. T. Berezyuk, an increase i n head i n t h e peripheral

L_

J -

_ _ channel of a vortex pump ( f i r s t type) i s realized as a r e s u l t of t h e onset of

secondary currents, as well as t h e entraining action of t h e impeller a d strong turbulence. Limiting himself t o these general considerations, G. T. Berezyuk

gave no s p e c i f i c rationale f o r t h e working process t h a t would be amenable t o

~

__
-1
1

_ _ - _
r

LIL.

1'

3-

-___ __
^ I
I

-.-A

--t

mthematical treatment.

H e w a s therefore unable t o give the functional rela--

t i o n s h i p between t h e pump head and t h o s e f a c t o r s contributing t o it.

0. V. Baykov investigated several vortex pumps of t h e f i r s t and second

types and attempted t o derive a fundamental equation f o r t h e vortex pump, i.e., t h e dependence of t h e head on t h e pump capacity and geometry of t h e flow-througb
1

!<I

{portion. Relying entirely on t h e hypothesis of Schmiedchen (ref. 9 )

: .

, 0.

V. Baykov &

wrote t h e momentum imparted t o t h e f l u i d by t h e i m p e l l e r per second over t h e working length of a channel with a central angle
M3.p j?f?***Udlk*
R;.

8:

- -0

where p i s t h e density of t h e f l u i d , Ri and R of t h e peripheral channel, vu and v nents of t h e velocity.

a r e t h e inside and outside r a d i i -.

are t h e peripheral and meridional compo*

_ _

Substituting i n t o t h i s equation the values of M and ment i n head over t h e angular section

0,

we obtain the incre-

bo:

where g i s t h e gravitational acceleration, F i s t h e cross section of tk p e r i 1 -

:,
r

li]
L, 2
t-

pheral channel.

*.

<

_
I

This equation, which was integrated by 0. V. Baybakov with a number of simplifications, cannot be used f o r practical purposes, since the f l u i d rotational velocity v

4-

..
- ?

i n t h e plane of the channel cross section F remains unknown.

V. I. Nakhodkin systematized and generalized a wealth of experimental d a t

-,*

1. II--on vortex
E : h
I _ _

pumps of t h e f i r s t type, experimentally investigated t h e balance of

- -_

4.

__-A

-1

___

__
e n e r g y i n a vortex pump, and conducted a series of t e s t s t o f i n d out how t h e shape of t h e peripheral channel affected t h e performance curve of t h e vortex
pump.

B. I. Nakhodkin based h i s work on t h e assumption t h a t t h e t o t a l head of

t h e vortex pump i s made up of t h e head H developed by t h e impeller i n t h e suct i o n and discharge section (as occurs i n t h e conventional centrifugal pump) and,
I

i ' ' t h e vortex head

1:

obtained as t h e r e s u l t of energy t r a n s f e r f r o m t h e rapidly

moving f l u i d p a r t i c l e s i n t h e i m p e l l e r vanes t o t h e slowly moving l i q u i d i n t h e channel i n connection with t h e intense f o m t i o n and breakdown of v o r t i c e s i n t h e working section of t h e channel. Forsaking any attempt a t deriving t h e fundamental vortex pump equation analytically,
-

B . I. Nakhodkin undertook a

calculation of the performance curves

'

f o r m d e l pumps, applying similarity principles.

-

In t h e event t h a t no model
-

' w a s available, he recommended analyzing t h e vortex pump according t o t h e follok-''

I

ing procedure.

1 .

The pump head i s calculated as H = kou /2g, where ko i s t h e pressure

'

coefficient a s determined from an e m p i r i c a l - s t a t i s t i c a l graph of ko = 4 (n ).

2.

The pump capacity i s determined f r o m t h e equation

where (c/u),
I _ !

i s the coefficient of discharge as determined from an empbical-

1
~

1 s t a t i s t i c a l graph of
3.

(c/u),

a f'

/ '

(n )
s

,u

*(

i s t h e mean peripheral v e l o c i t f o f

the ,

impeller blade, c i s t h e velocity of the f l u i d i n t h e channel. The optimum values of the various coefficients governingthe cross

sectional dimensions of the impeller

ana

channel ( f i g . 1) and t h e number and

pitch of t h e blades a r e chosen f r o m graphs of those coefficients as functions

:
-

!of n3. b--. -

For example, according t o t h e date of B. I. Nakhodnik, t h e optifnum

--

........

.--.

'-3

. . . .

^__.II

values of t h e coefficients b/c and r,/c for which t h e pump will have t h e highest; pressure coefficients and efficiency a r e near unity, while t h e coefficient a/c amounts t o 2 o r 3. The r e l a t i o n s that he advises f o r calculating t h e head and
I / C ,

t h e graphs for determining t h e optimum v a l u e s of t h e coefficients b/c, and L l i

' !

a/cp
,

as a function of t h e specific speed of t h e pump a r e i n good agreemnt

I

,with t h e experimental data t h a t he obtazned f o r vortex pumps of t h e f i r s t type, b u t cannot be used f o r calculations on pumps of t h e second and t h i r d types, which are the ones most widely used i n a c t u a l practice. For model TsVS-53 vortex pumps (n = 2980 r.p.m.,

D2 = 82 m) of t h e second

sad t h i r d types with a nonsenicircular channel p r o f i l e , V. G. Kovalenko ( r e f .

4)'

came up with t h e following values f o r t h e coefficients ( f i g . 2).

The lower

values of t h e indicated coefficients are given by t h e more sloping Q-H and Q c-s, t h e upper va

__ -

- ._

__.

The most noteworthy investigations of vortex pumps outside the Soviet

Jnion i n t h e l a s t three or four y e w s have been conducted by the American spe-

/22
,

i a l i s t s Iversen (ref. 1 0 )and Wilson, e t a l ,

(ref, 1 1 ) .
#

Vortex generator

i 1 -

. .

Suction

Discharge

, *
, -

Figure

3.

Iversen, i n explaining the operEtion o f t h e vortex pump, advanced the hypothesis of the f l u i d mass becoming entrained i n t h e channel due to shear s t r e s s e s , which a r i s e i n t h e flow as the impeller i n t e r a c t s with the f l u i d , as- - .
',T
__

__

. /

,suming t h a t the impeller represents a brind of abstract rough surface.

Accord_ .

__

ing to the-diw-iuu sEoiK-in lrigure- 3;-S the-rough s r n c e - f s s e t - i n -on,

I-- the balance of the forces acting on t h e f l u i d i n the horizontal peripheral chann e l i s expressed by the equation
P I A- PIA - ~
& c 4- Wi

=0.

where P

and P

are the pressures o n t h e suction and discharge sides, respec-

t i v e l y , A i s t h e c r o s s sectional mea of flow channel wall of area a a . i c'

7

rCi s t h e shear s t r e s s on t h e

i s t h e shear s t r e s s on the impeller surface#of are2

i Ls
1 '

. _ .'
:

-.

,.

, -

The power delivered t o the pumped f l u i d from the impeller surface moving with a v e l o c i t y u i s expressed by t h e equation N=Tau. i i

, . I

_ _

For a f l u i d density p and mean f l u z d velocity v, t h e s h e w s t r e s s

is-%rxi-

t r a r i l y determined by Iversen according t o the equation

_I ?

'where c i s an unknown coefficient.

<
*

'+
f Q

1 I
j

Assuming then t h a t , i n determining,t h e quantities

Ti

and rc, t h e velocity

;io
i
I

' .

v i s t h e difference between the velocity of m t i o n of the corresponding rough surface and t h e velocity of the f l u i d i n t h e channel Q/A, Iversen obtains a

fundamental vortex pump equation w i t h two unlmown coefficient ci and cc:

Inasmuch as c and c are unknowns, t h e solution proposed by Iversen, l i k e i C

J

_ ,

-:
_

t h a t of 0.

V . Baybakov, i s not suitable f o r p r a c t i c a l calculations, although it 2s

I
I _

. -i s based on a correct approach t o ana1yf;is of the tangential- s t r e s s e s i n the

pump flow.

Specifically, Iversen's error l i e s i n h i s ignoring the causes giving'(

r i s e t o the tangential s t r e s s e s ( f o r example, the difference i n t h e centrifugal forces a t t h e impeller and i n the channel), which was expressed i n t h e representation of' t h e impeller as an abstract rough surface. Wilson investigated a number of i n d u s t r i a l pump types and suggested t h a t

:,I'
'*

t h e f l u i d p a r t i c l e s i n the peripheral channel move i n a h e l i c a l trajecctiory.

t

The
I

42

j
1

h e l i c a l motion of t h e f l u i d occurs a t each point, of the channel with a?%angenti'S;li

- ' v e l o c i t y vt and e r i d i o n a l velocity vC directed perpendicular t o vt o Proceed4L

41 *

._
'

-*

.
c .

ing on t h i s basis, Wilson introduces two concepts:

/-

1) the t a n g e n t i a l flow, de-

fined as Q

=) vtdA, where A i s the

cross section of t h e peripheral channel;

..(A) ?L' _t h i s flow i s assumed equal t o the pump capacity; 2) t h e c i r c u l a t o r y o r meridionI

52

Qc associated with the velocit2 vc..-

__

.-.

I.

. . - .. ..

..__ .

e -

a)

V
Figure 4

A t h e r o e t i c a l investigation of the working process of the vortex pump was

conducted by Wilson f o r a simplified-moael and f o r a l i n e a r law of f l u i d motion. Considering t h e elementmy volume formed. by a section of the peripheral channel

(fig. ha) and the working impeller (fig. 4b) of width de, and drawing on t h e
I

.
1 ,

equations of hydrodynamics, Wilson derived an equation f o r t h e head:

-,-

where a and c are coefficients denoting t h e r a t i o vt/u a t t h e points 1 and 2,

' respectively, Q
I

: :

i s the circulatory capacity, which f o r the elementary volume

i s determined by t h e equation dQ = v r de, u and u are the peripheral velocc c2 2 1 2

a :

) ,

i t i e s a t t h e points 1 and 2, located a t the r a d i i r1 and r2, kt i s t h e tangentiae__loss f a c t o r , D i s t h e impeller diameter, Q i s t h e pump capacity; Q =f. " 0 , x ,
ro
*

:<I
J -

-r
12

'+3 ,

' /

42,
!,?, 1 k f L' a r e a

(here Qs = r Ab; r i s t h e radius of t h e centrofa of t h e ' ' g G

7

, .

'

A, w i s t h e angular velocity of t h e impeller;

5 and k2 are dime

1s

coefficients depending on the depth and width of t h e peripheral ch

/
L+

A -

..

The following equation i s given for determining t h e power:

"he efficiency i s ' calculated f r o m +he equation

__

The fundamental equation expressing t h e r e l a t i o n between t h e head and dis-

ichasge of f l u i d i n t h e peripheral channel of t h e vortex pump i s very complex

I
1

I 3
i
1

"

and i s not suitable f o r p r a c t i c a l application, because, l i k e t h e solutions of

?
i

0. V. Baybakov and Iversen, it contains a n a l y t i c a l l y indeterminate coefficientsp

The theory and method of calculation proposed by Wilson have been subjectea
+

t o r a t h e r severe c r i t i c i s m on the p a r t of American s p e c i a l i s t s and have not beep

given a positive rating.

The foregoing b r i e f analysis of t h e principal works i n t h e theory and

I

a n a l y t i c a l rnethods f o r vortex pumps indicates t h a t they a r e based on models of .I

"
L--

--i
~

t h e v 0 ~ x - p t . q oEerat-ion t h a t axe unsound a r i_e_t__ y of reasons and do not ' --__ 2 ~ - - 1 - f-o r a v - -----.._. r e f l e c t completelythe physical nature

of t h e process involved.

This renders

them l e s s valuable and makes t h e i r application f o r engineering calculations impossible. The experimental material acquired by various researchers i s very

meager and does not embrace a l l types of vortex pumps.

A t the present time, there are
110

r e l i a b l e data i n the practice of vortex

pump construction on the o p t i m a l r e l a t i o n s of the impeller and charnel aimenI _

sions f o r d i f f e r e n t types of vortex pumps.

~

Data are lacking on the recgncilia;

I ' .

t i o n of t h e aimensions and shape f o r which high e f f i c i e n c i e s w i l l be ebbured a t the highest heads and small size and weight of the pump. Finally, w i t

i s t i n g d i v e r s i t y of opinion r e g m d i n g t h e nature of t h e vortex pump op there i s no unl'fied theory and method for t h e i r engineering calculations.

+for engineering-purposes1
d

What theory and analytical method f o r the design of vortex pumps can be
R ~ p P r i m m ..designmrk k l and

themeI -

I-

t.

~I

--

L Q- --.A

----I

"--"..."--.L

._ .

__

_--.

research carried o u t a t the Krasnyy Fake1 Factory on the working process o f the vortex pump have fostered the following scheme t o describe the operation of the
I

vortex pump.
i
I

A s the impeller mves, it creates trortices whose axes axe p a r a l l e l and per" ,

3 ,
I

i
l

3
i

p e d i c u l a r to the r o t a t i o n a x i s of the h p e l l e r i n t h e other t w o coordinate

I

, ?

idirections.

Depending on t h e configuration and dimensions of t h e flow-through

! 13 '

! >

portion, pa.rticulas vortices are prevelent, governing the dynamics of the work-

ing process and, consequently the analytical schem.

The principal vortex formation that can be t r e a t e d by any p a r t i c u l a r anal y t i c a l scheme i s accompanied by an additional vortex formation due t o the im-

pingement of f l u i d on the blade t i p s , roughness, etc., f o r analytically.

which cannot be accounted-

I n a properly designed manufactured pump, t h i s additional

. I

vortex formation i s a second-order quantity.' A n increase i n the mass of f l u i d ' - _ . . . -- __ -- - __ . . . _ _____ _ . pushed by the impeller i s accompaniedby p a r t i a l o r complete breakdown, defo
.-.I
L .
I

c -

t i o n , o r "detachment" of the vortices.

If deformation of the vortices occurs,

the consequence i s t h a t tangential stresses occur i n the f l o w , causing the entrainment o f f l u i d i n the channel a s the impeller moves. Since there are pre-

dominant vortices determing the qualitative and quantitative aspects of the

~

working process, a s w e l l a s secondasy vortices which axe unavoidable adjunct t o t h e operation of any r e a l pump, the tangential s t r e s s e s must be arbitr
/

; r,

!/ I
7 -

+I

h2 jbmken down i n t o principal and secondary components, t h e l a t t e r being

'-? !,L

: r
!+I-.

( , t o t h e additional tangential s t r e s s e s t h a t a r i s e i n t h e turbulent motion of a

1fluid
1
t

in pipes.

'Such
~,ip&,
_ i

a pump must have a good streamline configuration i n t h e f l o w - t h o u

._
. I _ _ _

minimum blade thicknesses and wall roughness, etc. --_-_ -_ __ -I-11 ___ i
_ I -

1-

_L--

I_ I _ _ _ _

_--:

-- -

Figure 5.

In vortex pumps having a nearly semicircular channel cross section (fig.

5), especially favorable conditions are s e t

up f o r v o r t i c e s whose axis i s

directed perpendicular t o t h e p l a n e of t h i s cross section.

These vortices are

: caused by t h e action of varying centrifugal forces on t h e f l u i d mass i n t h e

~

vanes of t h e impeller and channel.

T h e cross vortices with t h e i r axes i n t h e

I

-,
-,&I1

_ I

other two perpendicular dimensions, given a s u f f ic i e-n t l y large number of t h i n --

--, - T --_

ps

blades, are negligible and will not be considered i n t h e present a r t i c l e . Pumps with an almost semicircular channel cross section have high heads and e f f i c i e n c i e s , hence they are t h e m s t widely used i n industry. The f l u i d

pumped by them i s of l o w viscosity, and the surfaces of t h e flow-through portiozl.. have l i t t l e roughness. This j u s t i f i e s considering devoting some a t t e n t i o n t o

t h e working process and a n a l y t i c a l method f o r vortex pumps with a semipircular

, channel and f l a t r a d i a l blading working with an i d e a l f l u i d (ref. I
I I

6).

rl

!i"
I i .

W e w i l l adopt t h e following notations f o r t h e elements of the peripheral

and impeller ( f i g . 5) :

.:'

' channel

l i s t h e area of t h e impeller blade, s i s t h e area of the blade t i p i n

contact with t h e f l u i d , R i s t h e radius of t h e center of g r a d t y of t h e area

-.

F, rC i s the radius of t h e center of gravity of t h e area f ,

i
. . ~

'

i s the projectiop,

L.. .

. .... .. -.I

of t h e f l u i d along l i , a r i s t h e r a d i a l clearance between t h e casing and impell e r , ae i s t h e end clearance between the casing and i m p e l l e r , p i s t h e density of
,

t h e f l u i d , 1 i s t h e point a t which the f l u i d p a r t i c l e s h i t t h e blade, 2 i s t h e point a t which they depart from the

: :
1 >

'>

bide.

'3
ii
1 .

W e w i l l assume f o r simplication t h a t t h e f l u i d i n t h e vanes of t h e i m p e l l e r ; ip

I "

11 \
I

moves with a mean angular v e l o c i t y w, aqd t h a t t h e mean angular v e l o c i t y of t h e ? ' I f l u i d i n t h e peripheral channel i s equal t o (ref.

6)

I? = - Q .

In t h e absence of losses, R
motion, i.e.,

---)

w ,

and i n t h e presence of resistance t o the&

W .

under any r e a l working conditions,R <

A s already remarked, i n t h e assumea scheme, t h e vortex motion i n t h e pump

~

t:
,
I

channel i s e l i c i t e d by a difference i n t h e centrifugal forces acting on t h e ' f l u i d messes i n %e 1 and- invanes of

_ -

,
c *

the impeller.

_TdFth

these masses.

'-' <~ moving a t d i f f e r e n t angular velocities w a n d Q .

i

Suruming f o r t h e p r o f i l e shown

_ "

i n figure 5 t h e elementary centrifugal forces dp s i n

I

acting along 1

over t h e

e n t i r e cross section (F

+ f),

we obtain
.P .P - - p / l t ( 3 - O I ) .

2'.

. -

For p r o f i l e s other than semicircular, i n place of the f a c t o r 2/7r we obtain,

as t h e r e s u l t of integration, t h e profile coefficient m, which can be w e d i n
, A

equation (2) f o r convenience. Hence, t h e centrifugal force P acting i n t h e plane of t h e cross s

F + f gives r i s e t o v o r t i c a l motion with a c e r t a i n mean v e l o c i t y cm* The mass

per u n i t time M of the f l u i d taking p a r t i n t h i s motion i s defined a s
C3 - I

5:

52

i -

M = Pac,,
_-__-_
~ _ _ _

(3)
__
L.

__-

._-:-

l--------------~ P _._ _ I _ _ I

__

f , . -

"

where a i s a length approximately equal t o t h e p r o f i l e radius a

- . . _

0'

I f t h e t h r u s t P were not balanced during motion of t h e liquid by some re-

sistive force

n, t h e motion would be expressed by the

P=hc,

Newtonian equation

'where Ac i s t h e v e l o c i t y increment t h a t t h e mass M would acquire i f it passed ,I j u s t once through t h e impeller.

I

. i

:r,
3 1

This cannot occur, however, since t h e periodic acceleration of t h e liquid,

which it can undergo a s it repeatedly passes through t h e i m p e l l e r i n a vortex
pump, would indicate a nonsteady s t a t e process, i.e.,

'

. .

<.

'

Consequently, t h e vortex motion must be described by t h e equation

-I

' i
_

&

A@

= :

0,

=IT.

(7)

Formal application of t h e i d e n t i t y
b

(6) will be of no a v a i l i n solving t h e

problem.

For t h i s purpose, it i s necessary t o examine the r e l a t i o n of t h e

vortex motion, as represented by t h e velocity c

m'

and t h e peripheral motion,

'represented by t h e v e l o c i t y vu, which v a r i e s f r o m t h e value v a t point 2 t o u2 5 /vlU a t point 1 (fig. 6). .

\ *

i f

i t

I ?

1,

,
I

5 CJ
- 2

--I'

-7
I

____
-

I _

H o w does t h e energy of the vortex motion acquired by the l i q u i d

- -_

__
I

from t h e impeller become t r a n s f o m d i n t o t h e energy of peripheral motion, i.e., i n t o t h e energy, p a r t of which i n the f i n a l analysis i s imparted t o t h e q u a n t i t y

< ;

< , o f l i q u i d Q i n t h e form of head H?

If t h e peripheral v e l o c i t y of each f l u i d
-2

19;

1:'
' .

p a r t i c l e varies from v t o vul with a constant normal component of t h e v e l o c i t y u2 t I3 1 t it then follows from t h e equation of motion t h a t t h e flow has tangential c

m'

forces

Integration of the tangential forces over the t o t a l area on which they act y i e l d s t h e mean tangential t h r u s t

(9)

-_

26 ( p e r u n i t channel length L = 2 TT R) - - _ _ _ __ -- . __ __ - - -_ - -_ . _ I . ____ Since t h e impeller i s s u b j e c t e d t o a t h r u s t T with average v e l o c i t y uc, t h e,' 72 !

I

_ I

7 -

;power spent t o accomplish t h i s i s

N = Tuc.
According t o t h e proposed scheme, t h e energy of peripheral motion i s acquired through conversion of t h e corresponding amunt of vortex energy.
> -

What

- - amount of energy i s sapped from t h e vortex motion? I f we suppose (ignoring 41 : ,losses) t h a t not a l l of t h e vortex energy i s converted i n t o peripheral Binetic 4:i
!.c' ,
I

! i

: :energy, t h e s u r p l u s could be spent i n accelerating the f l u i d mass M,

establishing Ac f 0.
, -- ,
I

I n t h i s case t h e f l u i d v e l o c i t y a f t e r passing t

impeller would be

i .

_..
IJ

f'

c :Cwhere Ac > 0.

. .

--

___

Considering t h a t the -&tion would occur i n t h i s case under t h e action o f a , nonsteady s t a t e force P (since equilibrium i s impossible according t o the equat i o n of motion

( 6 ) ) , t h e power of vortex motion would

f f v r Pe, = P (e

be

+Ae)v

ic
1 I ''
1

I n reality, we have i n t h e pump

11;
I

Ac* = O
and, hence,

Nvn = Pc me
Consequently, i n order t o ensure t h e conditions power difference

_ _ _ I .

,1

....... .............. . . . . . . . -- __ must be completely converted i n t o the power of peripheral motion (10) :
. i -

....

....

Nv -

AC % = P.

Tuc = P A C
or

Tuc =-

P2 M *
i s the r e s u l t o f applying t h e law of

._
I

The additional equation

( 1 6 ) , which

'c
*!
I 1
4

conservation of energy and t h e second l a w of Newton, makes it possible f o dei

I n 1 -

Lt2 termine t h e v e l o c i t y of vortex motion c

ii?% +
I

m '

I -~

I: i

From equations (2) and ( 9 ) w e find

51 i 52 I

5c i

..

....

. . .

. l _ l l . . l

.......... -_--

. . . .

. . . . 1. . . . . . . .
..
.......

_ I

. .

Then, analyzing t h e equilibrium o f t h e f l u i d volume i n t h e channel under

C

- - ..- .-

/es

t h e a c t i o n of t h e tangential moment M = r T and t h e drag moment against t h e

..

p r e s s u r e of t h e pump M

= RYHF, and neglecting the mment due t o f r i c t i o n of t h e

, .

f l u i d against t h e channel w a l l
13

rcT -.RyRF,
1

I1
-

we obtain t h e fundamental vortex pump equation

where

($is t h e flow containment coefficient, L i s t h e working length of t h e channel,

m i s t h e p r o f i l e coefficient, which i s calculated from the geometry);

W e obtain an expression f o r t h e expended power by multiplying t h e mment

M by t h e angular v e l o c i t y of t h e impeller:
C

1 7

- jA ~ trimThe u s e f u l power of any pump, of course, i s equal t o

(21)

N8=WQ*.
1 -

.lf
L;

Hence t h e efficiency of t h e working process of the vortex pump i s

LL
-5

lP, = Q/FRw.
The o v e r a l l efficiency of the vortex pump should take i n t o account t h e leakage, mechanical and hydraulic losses:

(23)

__

l_l

" I

Equations ( 2 0 ) , ( 2 1 ) , and (23) give a unified solution t o t h e vortex pump problem, i.e.,
1

_-

they uniquely determine i t s head, power, and e f f i c i e n c y as func-

t i o n s of t h e capacity f o r given dimensibns and pump r.p.m.

A n analysis of t h e curves calculated according t o equations (20) and (21)
i -

z
d
1:

'show t h a t t h e i r behavior corresponds quite w e l l t o t h e behavior of t h e experiI

c
!:
* )

i
I

z
'

* 1

,mental curves shown i n figures 2 and 7 Fo 10.

t If:
k

Figure 7. Characteristic Curves of the 2.533-3 Pump f o r Water a t N = 1450 r.p.m. Without Aspiration.
>

Figure 8. Experimental and Calculate Characteristics of t h e 1.5V-l.JM Pump a t n = 1450 r.p.m.

1) H without leakage (calc.)

2) N without mechanical loss o r leakage (calc. )

1) H without leakage ( c a l c
2)

42 3 : . ?
1i f 14

'
j

3)

; I

(exp.) Onset of cavitation, experimental pump

!,.

4) N (exp.) 5) N without mechanical l o s s or

leagage ( c a l c )

7s'

fr,

;I

-.
.

--

_____ _ _

--

Figure 9.

I) VS-65 Pump;

1 1 ) vs-65m Pwnp.

H, mH20'

. -

c.3
- .
.-

If the channel p r o f i l e i s not markedly d i f f e r e n t from semicircular, t h e

checking and design calculations carried out f o r a nuniber of Soviet vortex pump models a t n = 1500 r.p.m.
r;

nearly coincide with t h e experimental data.

In t h i s

' case, t h e fundamental equation (20) can be used without any correction factors.
i

'4,

If t h e channel p r o f i l e i s appreciably nonsemicircular (see fig. 1), it i s necest

b

10

kary t o u s e t h e empirical coefficients given above.

II

' I !

' ;

, "

The above solution of t h e v o r t e x pump problem i m p l i e s t h e following orderof-magnitude choice and calculation of a projected pump. One m u s t f i r s t decide on t h e type of pump, whether vortex, centrifugal or displacement. This choice i s stipulated by many factors: dimensions, weight,

a t t a i n a b l e efficiency, self-priming requirements, u s e f u l l i f e , v i s c o s i t y of t h e f l u i d t o be pumped, i t s temperature, contamination, length of operation, absoLute head, capacity, etc.
r -

----

i L

_I_.Ic

The approximate Limits of a p p l i c a b i l i t y of vortex pumps can be delineated here: f l u i d v i s c o s i t i e s up t o 5OE; coarseness of mechanical impurities up t o
0.1 am; head on a single impeller from 1 t o 200 m; capacity from 0 . 1 t o

17
.

liter/sec;

peak e f f i c i e n c i e s t o

45$; minimum outlay of metal per k W power (minS

imum s i z e and weight) down t o 2 kg/kW; specific speed n

rep.m.

from 4 t o 50; shaft

/2s

t o 6000.

In t h e event t h a t vortex and centrifugal-vortex pumps are equal apqlicable

!.to a s p e c i f i c situation, t h e choice i s determined by t w o consideration
I

/suction and a t t a i n a b l e efficiency.

As i s known, t h e centrifugal stage 'builds

.
%he d i s p a r i t y between t h e experimental and calculated curves i n figure 7
4 s a t t r i b u t a b l e t o leakage i n a r e a l pump as Q

5 -m3/b in figure 7 a . n ~ for ~ Q

!-

- l k . g3L + i n - f i . g m e ~ ~
-____I

0 and t o cavitation
-<

---l-.l ..""

I = -

--2Q_

1
i t
L
--1__--

_"_

--"

.
I --f
___
'
.

-- _-_I_j

- -

---

__

-~

_
a

up t h e necessary head f o r t h e vortex stage and improves t h e o v e r a l l efficiency

of t h e system (ref. 5 ) . secondary importance.

The vortex stage i n t h i s case may be of prime or

Space limitations _xevent u s f r o m examining a l l possible versions of pump construction.

? ; c

W e w i l l therefore confine our analysis t o j u s t those cases when

' ' 'the parameters of t h e vortex stage have already been chosen on t h e b a s i s of
;

ad-'

'

equate considerations and t h e problem i s t o determine the shape and dimensions of t h e flow-through section, then t o derive t h e pump performance curve analytically.

The decisive f a c t o r i n s e l e c t i n g t h e dimensions of t h e vortex stage i s t h e attainment of high head.

It i s known t h a t t h e head developed by a centrifugal

-'

pump i s defined i n t h e f i r s t approximation as

-_
C .

Lt? . . _ -_ -

-IT

_ .

- - -- -

- -

.ii 1 I )
, r

,'
' 8

,where

A calculation according t o equation (20) with Q = 0,

-+

1,

E :

F, rc = R,

e= 1, s = 2a yields

f o r t h e vortex pump.

i> 1

..............

~. . -

..

' . * . l l .

...___._.. &

It i s known t h a t t h e capacity a t r)
to

--.

__

--

max

f o r a vortex pump corresponds largel$

Assuming f o r t h e f i r s t approximation t h a t t h e Q-H curve i s a s t r a i g h t line,

1 4
2

9 $we define

1 I(:

Hence, by knowing H and the r.p.m. n, w e f i n d r from equation (29). 0 C Then, f r o m t h e prescribed capacity Q knowing uc, we determine F from equation 0'
(30).

Since we have assumed [It= F (semicircular p r o f i l e ) , we f i n d

2f
: o u

( 32)
. -

C -

from which we ascertain the p r o f i l e radius a

~

__

__

- .-

- __ ,~

9'

It m u s t be taken in-

This p r o f i l e i s basic t o t h e subsequent pump design.

t o account t h a t t h e i n t e r n a l leakage via t h e clearances a and a , , t h e containe ment of t h e flow by t h e blades, and the reduction i n working length of t h e channel L due t o t h e necessary linkage (connector) between t h e suction and discharge openings lower t h e head of the pump. Consequently, t h e impeller diameter

m u s t be increased accordingly.
;
I

For t h e corrected profile, t h e Q-H and Q-N curves can be constructed from
and (21) without regard for leakage or hydraulic and me

Lit

$ 2 !equations (20)
L + i
6

losses, t h e determination of which w e w 3 1 1 not consider i n t h e present a r t i c l e .

c y <

I f now some kind of special requirements a r e imposed on t h e pump character-

' i s t i c curve i n addition t o t h e generation of high head and efficiency, i t s pro,

f i l e m u s t depart considerably from semicircular.

I n t h i s case, t h e expression

provides a sensible approach t o t h e variation i n dimensions and shape of t h e flow-through p a r t of t h e pump i n order t o r e f i n e i t s characteristic. This v a r i a t i o n generally proceeds in two d i r e c t i o n s i n vortex stages which
!
r

ifunction as t h e secondary stage ( e i t h e r ' i t i s required t o obtain a maximally sloping Q-H c h a r a c t e r i s t i c or t o minimize t h e power u t i l i z e d by t h e pump).

I,;

ti?

Such

requirement a r e often imposed on self-priming vortex stages, where t h e efficiencyand head are of secondary importance. Equation

(33) shows t h a t by varying t h e r a t i o of t h e blade area f: t o t h e

'

channel cross section F, t h e pump c h a r a c t e r i s t i c can be sharply altered.

In t h e
i
-j

limit F

OD

,H -

0 ,Q -

= Fu

00,

i.e.,

the

Q-H c h a r a c t e r i s t i c tends t o -

1 ,

ward t h e horizontal axis. , It I s also-poss3bf-e t o affect tbe-pump &ara&erisbic--by e

g the n

e
,

of blades, t h e i r angle of inclination, t h e length of t h e channel, and t h e t i p s

of t h e blades s, t h e dimensions and location of t h e suction and discharge points,

t h e r a t i o r /R,
C

and shape of t h e profile.

The more t h e p r o f i l e d i f f e r s from Inasmuch a s t h e v e l o c i t y c

semicircular, t h e greater w i l l be the vortex drag. increases abruptly as a r e again minimized.

i

m
-

0 according t o equation

(17), t h e

head and power input

In practice, t h e impellers of vortex pumps a r e normally found i n

f i c a t i o n s , open and closed impellers.

i30 modi-

1 .

The open impeller represents a hub with

'long r a d i a l blades, which a r e bounded a t t h e sides and periphery by t h e pheral channel and f l a t walls of t h e pump casing. The closed impeller has a

f l a t disk with short blades, situated a t t h e periphery on both sides a 4 sepI

' m a t e d a t t h e middle by a connector.

The angle a between t h e blade and-

--_I

--

__

E_ _ ___ f
60 t o 9 0 ' .

I _ _ _ _ -

connector varies within t h e limits

These impellers a r e made with

blades of various cross sectional shape6 shaped.

rectangular, trapezoidal, crescent-

The most popular are blades with a rectangular o r trapezosdal cross The number of blades on an open i m p e l l e r i s u s u a l l y L
I

'section and curved baclrward.

-' from ' 2j
I

1 2 t o 24, on a closed impeller it i s from 18 t o 38.

From t h e q u a l i t a t i v e aspect, the number of blades a f f e c t s t h e performance

110
j '

, ^ !
I C

I'
,

--,

curve a s follows.

Increasing the number of blades within t h e indicated limits

and preserving t h e geometry of t h e peripheral channel increases t h e head and capacity of t h e pump (see f i g .

9), w h i l e t h e optimum value of t h e efficiency

i s displaced toward higher capacities; a decrease i n t h e number of blades has

. _ t h e reverse effect.
The highest e f f i c i e n c i e s ( t o impeller a t z =
-

4%)

&re given by vortex pumps with an open

--_

24 and with a rectangular channel cross section, i.e., with an

_ .

- -

- .

_-

_ .

__

-__- -

optimum value of t h e coefficient l / c = I. curved back.

The blades i n t h i s case a r e usually

'

However, with the proper combination of shape and geometric dimen-

sions of t h e peripheral channel f o r the number of blades z = 12, vortex pump c h a r a c t e r i s t i c s can be obtained i d e n t i c a l t o t h e c h a r a c t e r i s t i c s of pumps with
z =

24.
The difference between t h e characteristic curves of t h e indicated pumps i s

11 1.

contained i n t h e f a c t t h a t a pump w i t h z =
E

24 develops considerably 1
I

t)
< ? '

i ?; a t low capacities than t h e puny? with z

12, due t o t h e "looser" conneption be-

tween t h e suction and discharge sides. The closed type of impeller combines w e l l i n t h e hydraulic sense pheral channels having a semicircular cross section o r nearly
SO.

A similar

increase i n t h e number of blades increases t h e head, capacity, and e f f i c i e n c y

: 2 lof t h e pump i n t h i s case, t h e highest efficiency ( ____ to I _ . . _
-I

5%) an _ _ r e s u l t i n.g fro? - - --

. _ . I

i m p e l l e r with z = 24 t o 32 a t an angle cy
t h e impeller.

--

65 t o 70" and "forward" r o t a t i o n o

1 1 1

" "

With "backward" rotation, a l l other conditions being equal, f o r

c e r t a i n pump models t h e head and power r i s e sharply, while t h e e f f i c i e n c y i s 'reduced 3 t o 5% and t h e optimum value s h i f t s toward lower capacities.
-f

I n the
I

first instance, t h e curvature of t h e

I

Q-8and Q-N
1

c h a r a c t e r i s t i c s i s up, i n t h e 1
8 -

ic;
3 -

I 'I ;second instance it i s down, while i n t h e i n t e r v a l of low capacities it asympto-1 i

bc

t
I

t i c a l l y approaches t h e ordinate axis (see f i g . 10). This a t t r i b u t e of t h e impeller i s very widelytaken advantage of i n indust because a simple turning of t h e impeller through 1-80" on t h e pump s h a f t can re-

s u l t i n high heads a t l o w capacities, wlthout any change whatsoever i n t h e con-

s t r u c t i o n of t h e pump. Consequently, t h e nwiber of blades and t h e i r angle of inclination, l i k e

-'

,the geometry of t h e peripheral channel, exert a s i g n i f i c a n t influence on t h e __ --_ __ - - -. . . -generation a f good vortex pump characteristic curves.
~

I _

-?--

i_l

1 *

It should be pointed out i n conclusion t h a t t h e theory presented herein t o

describe t h e operation of vortex pumps permits t h e v e l o c i t y of t h e vortex motio c

m of t h e f l u i d i n t h e pump t o be determined analytically.

The value found f o r

t h e v e l o c i t y cm enables one t o derive t h e fundamental equation of t h e vortex

pump f o r construction of a t h e o r e t i c a l c h a r a c t e r i s t i c curve H
= i l ( Q )

on t h e

:c
I

isi b a s i s of t h e prescribed dimensions

I
'

.
7 "

4'1:
I -* --L
-

A compasison of t h e calculated data based on equations (20) t o (29s with

i i

--> , t h e experimental data disclose s a t i s f a c t o r y agreement f o r a series of tresious

La f *

i n d u s t r i a l models of vortex pumps.

A q p a l i t a t i v e estimate of the e f f e c t of t h e

pump geometry on i t s c h a r a c t e r i s t i c c m e according t o equation (20) permits a

r '. -

'

, r a t i o n a l approach t o t h e design of new vortex pump constructions of low weight

.and s i z e with enhanced hydraulic and o p y a t i o n a l characteristics.
--__-

I
. . --.Le-

----.-.I-_-

. .

- ."

1 . Berezyuk, G .T .
,
i

Self-Priming Rotary Pumps (Samovsasyvayushchiye vrashUkrmashgiz (Ukrainian Machinery Press),

chatel'nyye nasosy).
2 . Baybakov, 0 . V.

1 9 3 9 .

Vortex Pumps (Vikhrevyye nasosy)

Dissertaion, Moscow
i

Higher Technical School (MVTU), 1 9 5 0 .

5

1% 1

11s
;
1

1! I

'>
1 -

. Investigation o f , the Vortex Water Pump (Issledovaniye 3 . Nakhodkin, B. I

vikhrevogo nasosa na vode)

I 3

1 -

Dissertaion, Moscow Power Institute (MEI),

1 9 5 1 .
4.
Kovalenko, V. G. Investigation of Vortex Pumps for the Gasoline Motopump Dissertation,
--

/28

(Issledovaniye vikhrevykh nasosov dlya benzomotoponp) Moscow Petroleum Institute (MNI), 1953.

5 . Kupryashin, N . N . Centrifugal-Vortex Pumps (Tsentrobezhno-vikhrevyye

I

- __
L

'
i i

'

,'

6.

Vestnik Mashinostroyeniya, No. 3, 1 9 5 2 . - - - .-. - . Kupryashin, N . N . Hydraulics of the-VGrtexPump-@idi;avlika nasosy)

- --

-J

ir

vikhrevog6

---

nasosa).

Dissertation, MEI, 1955. Vestnik Mashinstroyeniya,

7 . Ustinov, A. S . Vortex Pumps (Vikhrevyye nasosy).

No. 8,

1 9 4 5 .

8 . Ritter, E. Self-Priming Centrifugal Pumps and Tests on a New Pump of this

-,
1

Type Selbstansaugende Kreiselpumpen una versuche in einer neuen Pumpe

1

&z
!J]

\>

dieser A r t ) .

Second Edition. Leipzig, 1 9 3 1 .

i -

-*

L 2 I L'> I

i 9 .
I

Schmiedchen. Investigations of Centrifugal Pumps with a Periphed Annult4~ !

1 :

LL I
.:
L7

J1

Channel (Untersuchungen uber Kreiselpumpen mit Seitlichen Ri Dissertaion, Dresden, 1 9 3 2 .

1 0 . Iversen, H . W.

* ,

Performance of the Periphery Pump. Transactions of the

- n

American Society of Mechanical Engineers (Trans. ASME), Vol. 77, No. 1

--'

-__

- - -.

. -

1 1 . Wilson, W . A.,

Santalo, M. A. and J. A. Oelrich.

A Theory of t h e Fluid-

'

Dynamic Mechanism of Regenerative Pumps.

Trans. ASME, Vol. 7'7, No. 8,

1955.
, I

x2.
5

Vasutoshi, Senoo.

Theoretical Research on F r i c t i o n Pump. Reports of t h e

I

Resemch I n s t i t u t e f o r Fluid Engineering, Japan, Vol. 1,

t

1948; Research9

I )
/ I
/ '

on Periphery Pump; Influences of t h e Suction Nozzle on t h e Characteris-i ' t i c s of a Peripheral Pump and an Effective Method of Their Removal. Reports of Research I n s t i t u t e f o r Applied Mechanics, Kyuschu University,
J a p a n , Vol.

3, No. 1 1 , 1954.
Transactions of t h e

13

Mlyadzu, A.

Theory of Westco-Type Rotary Pump.

Society of Mechanical Engineers of Japan, Vol.

5, No. 18, 1939.

Translated f o r NASA by Stemar Engineering Inc. z5

b w E n g Beach BTvti., Loag-Beach, Miftmnia-- 4 ---