Beruflich Dokumente
Kultur Dokumente
Deep
pwaterr Reverrse‐Circculatio
on Prim
mary Ceementiing
10121‐4
4502‐01.FFINAL
Deepwater Reeverse‐Cirrculation Primary Cementiing
1012
21‐4502‐001
Septem
mber 21, 2014
Jeff Watters, P
Principal Innvestigatorr
General Managger
CSI Technologiees
2202 Oiil Center C
Court
Houston TX, 770073
www.csi‐tech.nnet
LEGAL NOTICE
This report was prepared by CSI Technologies, LLC as an account of work sponsored by the
Research Partnership to Secure Energy for America, RPSEA. Neither RPSEA members of
RPSEA, the National Energy Technology Laboratory, the U.S. Department of Energy, nor any
person acting on behalf of any of the entities:
a. MAKES ANY WARRANTY OR REPRESENTATION, EXPRESS OR IMPLIED WITH
RESPECT TO ACCURACY, COMPLETENESS, OR USEFULNESS OF THE INFORMATION
CONTAINED IN THIS DOCUMENT, OR THAT THE USE OF ANY INFORMATION,
APPARATUS, METHOD, OR PROCESS DISCLOSED IN THIS DOCUMENT MAY NOT
INFRINGE PRIVATELY OWNED RIGHTS, OR
b. ASSUMES ANY LIABILITY WITH RESPECT TO THE USE OF, OR FOR ANY AND ALL
DAMAGES RESULTING FROM THE USE OF, ANY INFORMATION, APPARATUS,
METHOD, OR PROCESS DISCLOSED IN THIS DOCUMENT.
THIS IS A FINAL REPORT. THE DATA, CALCULATIONS, INFORMATION, CONCLUSIONS, AND/OR
RECOMMENDATIONS REPORTED HEREIN ARE THE PROPERTY OF THE U.S. DEPARTMENT OF
ENERGY.
REFERENCE TO TRADE NAMES OR SPECIFIC COMMERCIAL PRODUCTS, COMMODITIES, OR
SERVICES IN THIS REPORT DOES NOT REPRESENT OR CONSTIITUTE AND ENDORSEMENT,
RECOMMENDATION, OR FAVORING BY RPSEA OR ITS CONTRACTORS OF THE SPECIFIC
COMMERCIAL PRODUCT, COMMODITY, OR SERVICE.
i
Abstract
In Reverse‐Circulation Primary Cementing (RCPC) fluids are pumped downhole via the annulus
and then up into the casing, in contrast to a conventional cement job where fluids are pumped
down the casing then up into the annulus. The objective of the RPSEA Deepwater Reverse‐
Circulation Primary Cementing (RCPC) project is to assess the viability of performing RCPC to
reduce circulation pressure requirements for deepwater wells, to determine required
technology to apply RCPC for deepwater wells, and to present development strategies for
required technologies. This report is submitted in fulfillment of Task 19 of the Deepwater
Reverse‐Circulation Primary Cementing project and contains a summary of Phase I and II
activities.
ii
Signature and Date Stamp
X
Jeff Watters Date
Principal Investigator
iii
THIS PAGE INTENTIONALLY LEFT BLANK
iv
Acknowledgement Statement
CSI Technologies wishes to acknowledge the support of RPSEA, Weatherford and the University
of Houston, as well as the Working Project Group for their invaluable expertise. Working
Project Group members include Anadarko, BP, Chevron, Conoco Phillips, Hess, Schlumberger
and Shell.
Funding for the project is provided through the “Ultra‐Deepwater and Unconventional Natural
Gas and Other Petroleum Resources Research and Development Program” authorized by the
Energy Policy Act of 2005. This program—funded from lease bonuses and royalties paid by
industry to produce oil and gas on federal lands—is designed to assess and mitigate risk
enhancing the environmental sustainability of oil and gas exploration and production activities.
RPSEA is under contract with the U.S. Department of Energy’s National Energy Technology
Laboratory to administer three areas of research. RPSEA is a 501(c)(3) nonprofit consortium
with more than 180 members, including 24 of the nation's premier research universities, five
national laboratories, other major research institutions, large and small energy producers and
energy consumers. The mission of RPSEA, headquartered in Sugar Land, Texas, is to provide a
stewardship role in ensuring the focused research, development and deployment of safe and
environmentally responsible technology that can effectively deliver hydrocarbons from
domestic resources to the citizens of the United States. Additional information can be found at
www.rpsea.org.
v
Nomenclature:
API: American Petroleum Institute
BBL: Barrel
BDF: Backward Difference Formula
BH: Bottom Hole
BHCT: Bottom Hole Circulating Temperature
BHP: Bottom Hole Pressure
BHST: Bottom Hole Static Temperature
BOP: Blow Out Preventer
BPM: Barrels per minute
BWOC: By weight of cement
BWOW: By weight of water
DST: Drill Stem Test
ECD: Equivalent Circulating Density
FEA: Finite Element Analysis
FEM: Finite Element Method
HEC: Hydroxyethyl Cellulose
HS&E: Health, Safety and Environment
HTHP: High Temperature High Pressure
ID: Inner Diameter
RKB: Rig Kelly Bushing
MD: Measured Depth
MWD: Measurements while Drilling
vi
OBM: Oil Based Mud
OD: Outer Diameter
PBR: Polished Bore Receptacle
PPG: Pounds per gallon
PSI: Pounds per Square Inch
Pv: Plastic Viscosity
RCPC: Reverse‐Circulation Primary Cementing
RFID: Radio Frequency Identification
RP/GS: Rotating Paddle Gel Strength
SBM: Synthetic Based Mud
TD: Total Depth
TOC: Top of Cement
TOL: Top of Liner
TVD: Total Vertical Depth
UCA: Ultrasonic Cement Analyzer
UDW: Ultra‐Deep Water
WBM: Water Based Mud
WOC: Wait on Cement
Yp: Yield Point
vii
List of Figures
Figure 1: (a) Reverse‐Circulation Primary Cementing (b) Conventional Primary Cementing ........... 3
Figure 2: (a) Generic Deepwater Well Schematic #1 (b) Generic Deepwater Well Schematic #2 .............. 5
Figure 3: (a) Conventional Flow Path (b) RCPC Flow Path ......................................................................... 13
Figure 4: Mechanical Fixed Crossover Tool ................................................................................................. 13
Figure 5: Crossover tool system that is switchable on demand ................................................................. 16
Figure 6: Full Open Reamer Shoe example ................................................................................................. 18
Figure 7: Example of a pH Chemical Trigger Operation .............................................................................. 22
Figure 8: (a.) Conventional Placement Simulation (b.) Reverse Placement Simulation with commercial
software ...................................................................................................................................................... 25
Figure 9: (a) Circulating Pressure at Previous Shoe (b) Circulating Pressures at Casing Shoe ................... 29
Figure 10: Coupling of Equations for Multiple Fluids .................................................................................. 31
Figure 11: (a) Comparison of BHCT between commercial and COMSOL (b) comparison of first/last sack
temperatures .............................................................................................................................................. 32
Figure 12: Well geometry representation in COMSOL ............................................................................... 33
Figure 13: Temperature surface plots after mud circulation for conventional and reverse circulation .... 33
Figure 14: Bottom‐Hole Circulating Temperature (BHCT) in conventional and reverse circulation .......... 34
Figure 15: Temperature in the cement at the end of placement ............................................................... 34
Figure 16: Pump, lift and back pressure(well 1, region 2) (a) Conventional placement (b) reverse
placement ................................................................................................................................................... 36
Figure 17: Conventional and reverse temperature schedules (Well Schematic #2, 16” liner) ................... 38
Figure 18: Thickening Time comparison (Well Schematic #2, 16” liner) .................................................... 39
Figure 19: Conventional and reverse temperature schedules (Well Schematic #1, 7 5/8” liner) .............. 39
Figure 20: Comparison of compressive strength development during WOC time after placement (Well
Schematic #) ................................................................................................................................................ 40
Figure 21: Setup of tube flow experiments ................................................................................................ 43
Figure 22: Light weight viscous pill density of discharged fluid and photos of test ................................... 44
Figure 23: Heavy weight viscous pill test .................................................................................................... 45
Figure 24: Density and corresponding samples of discharged fluid for high density viscous pill ............... 45
Figure 25: Rheology of discharged sample for heavy density and high viscosity viscous pill ..................... 46
Figure 26: Fresh water flow ........................................................................................................................ 46
Figure 27: Interface between fresh water (blue) and viscous pill (white) .................................................. 46
Figure 28: Viscous pill flow displaced fresh water ...................................................................................... 47
Figure 29: Interface between viscous pill (white) and brine (red) .............................................................. 47
Figure 30: Brine flow displaced viscous pill ................................................................................................ 47
Figure 31: Interface between brine (red) and viscous pill (white) .............................................................. 47
Figure 32: Viscous pill flow displaced brine ............................................................................................... 47
Figure 33: Interface between viscous pill (white) and brine (red) .............................................................. 48
Figure 34: Brine flow displaced viscous pill ................................................................................................ 48
Figure 35: Interface between brine (red) and fresh water (blue) ............................................................... 48
Figure 36: Fresh water flow displaced brine ............................................................................................... 48
viii
Figure Apx A‐1: Comparison of temperature profiles between drilling mode and workaround production
mode ......................................................................................................................................................... A‐1
Figure Apx A‐2: Comparison of First Sack and Last Sack Temperatures between drilling mode and
workaround production mode .................................................................................................................. A‐3
Figure Apx A‐3: (a.) Conventional Placement Simulation (b.) Reverse Placement Simulation ................. A‐4
Figure Apx A‐4: Conventional (Drilling mode) and Reverse Circulation (Production mode) First and Last
Sack Temperatures ................................................................................................................................... A‐5
Figure Apx A‐5: Simulated Reverse‐Circulation Temperatures using Workaround .................................. A‐5
Figure Apx A‐6: General Fluid Temperature vs Depth Profile ................................................................... A‐6
Figure Apx A‐7: Simulated Reverse and Conventional WOC temperature profile ................................... A‐7
Figure Apx B‐1: Rescaling of the Wells. ...................................................................................................... B‐2
Figure Apx B‐2: Papanastasiou's constitutive equation compared to Bingham behavior . All units are
dimensionless. ........................................................................................................................................... B‐6
Figure Apx B‐3: Summary of temperature boundary conditions. .............................................................. B‐9
Figure Apx B‐4: Boundary conditions for the Navier‐Stokes equations. ................................................. B‐11
Figure Apx B‐5: Coupling of the heat equation and the Navier‐Stokes equations. ................................. B‐12
Figure Apx B‐6: The interface between the fluids is assumed to be perfectly sharp, and moves with the
average velocity of the fluids. .................................................................................................................. B‐13
Figure Apx B‐7: Spatially and Temporally Dependent Thermal Properties. Represents an arbitrary
property. The graph on the right is the overall property plotted along the dashed line in the schematic
on the left. ................................................................................................................................................ B‐14
Figure Apx B‐8: Calculation of the combined velocity field. The “ ” is not a true addition operator, but
rather a combinatorial operator. ............................................................................................................. B‐14
Figure Apx B‐9: Coupling of equations for multiple fluids ....................................................................... B‐15
Figure Apx B‐10: COMSOL predicted bottom‐hole circulating temperature compared to commercial
simulator results. ..................................................................................................................................... B‐17
Figure Apx B‐11: Cement temperatures predicted by the COMSOL model compared to the first‐sack/last‐
sack temperatures in the commercial software. Time is at the end of cement placement. ................... B‐18
Figure Apx C‐1: Geometry for Well 1, Region 1. ........................................................................................ C‐1
Figure Apx C‐2: Surface plot of the temperature at the end of mud circulation for well 1, region 1 in
reverse circulation. .................................................................................................................................... C‐2
Figure Apx C‐3: Surface plot of the temperature at the end of mud circulation for well 1, region 1 in
conventional circulation. ........................................................................................................................... C‐2
Figure Apx C‐4: Maximum Bottom‐Hole Circulating Temperature (BHCT) in conventional and reverse
circulation. ................................................................................................................................................. C‐3
Figure Apx C‐5: Temperature in the cement at the end of placement. Reverse is higher than
conventional, but less so than the BHCT. .................................................................................................. C‐3
Figure Apx C‐6: Pump, lift and back pressure for well 1, region 2 conventional placement ..................... C‐5
Figure Apx C‐7: Pump, lift and back pressure for Well 1, Region 2 reverse placement. ........................... C‐6
Figure Apx D‐1: Generic well diagram used for hydraulic analysis. .......................................................... D‐1
ix
Figure Apx D‐2: Flow‐paths for offshore reverse, conventional, and traditional reverse circulation. ..... D‐2
Figure Apx D‐3: Pressure diagram for conventional circulation. .............................................................. D‐7
Figure Apx D‐4: Pressure diagram for (offshore) reverse cementing. ...................................................... D‐8
Figure Apx D‐5: Pressure schematic for (traditional) reverse cementing. ................................................ D‐9
Figure Apx D‐6: Total pressure for both conventional and offshore reverse cement placement. ......... D‐10
Figure Apx D‐7: Total pressure for both conventional and (traditional) reverse cementing. ................ D‐11
Figure Apx E‐1: Rheological Compatibility Results for SBM, Spacer and Cement ..................................... E‐3
Figure Apx E‐2: Rheological Compatibility Results for SBM and Spacer .................................................... E‐4
Figure Apx E‐3: Rheological Compatibility Results for SBM and a Viscous Pill .......................................... E‐4
Figure Apx E‐4: Rheological Compatibility Results for SBM, Viscous Pill and Cement Slurry .................... E‐4
Figure Apx E‐5: Rheological Compatibility Results for Cement Slurry and a Viscous Pill .......................... E‐4
Figure Apx E‐6: Compressive strength comparison between slurry and 25% intermixing with viscous pill E‐
5
Figure Apx E‐7: Rotor Test ......................................................................................................................... E‐6
Figure Apx E‐8: Rotor test with SBM and chemical tag ............................................................................. E‐6
Figure Apx E‐9: Contact Angle Test Results ............................................................................................... E‐7
Figure Apx E‐10: Generic well schematics used for cement slurry design ................................................ E‐8
Figure Apx E‐11: Well #1, 16” Liner thickening time results, TOC @ 16,500’ MD ..................................... E‐9
Figure Apx E‐12: Well #1, 16” liner compressive strength results at the shoe ........................................ E‐10
Figure Apx E‐13: Well #1, 16” liner compressive strength comparison at TOC ....................................... E‐10
Figure Apx E‐14: Well #1, 9 7/8” Liner thickening time results, TOC @ 26,500’ MD .............................. E‐13
Figure Apx E‐15: Well #1, 9 7/8” Liner thickening time results, TOC @ 21,000’ MD .............................. E‐13
Figure Apx E‐16: Well #1, 9 7/8” liner compressive strength results at the shoe ................................... E‐14
Figure Apx E‐17: Well #1, 9 7/8” liner compressive strength results at the TOC. Note: Conventional slurry
TOC is 26,500’ .......................................................................................................................................... E‐14
Figure Apx E‐18: Well #1, 7 5/8” liner thickening time results, TOC @ 28,500’ MD ............................... E‐16
Figure Apx E‐19: Well #1, 7 5/8” liner compressive strength results at shoe ......................................... E‐16
Figure Apx E‐20: Well #1, 7 5/8” liner compressive strength results at TOC........................................... E‐17
Figure Apx E‐21: Well #2, 16” Liner Thickening Time Comparison, TOC @ 12,500’ MD ......................... E‐19
Figure Apx E‐22: Well #2, 16” Liner Thickening Time Comparison, TOC @ 5,500’ MD. Note: Conventional
result is for the 12,500’ MD system and is shown for comparison. ........................................................ E‐19
Figure Apx E‐23: Well #2, 16” liner compressive strength results at shoe .............................................. E‐20
Figure Apx E‐24: Well #2, 16” liner compressive strength results at TOC ............................................... E‐20
Figure Apx E‐25: Well #2, 13 5/8” Casing Thickening Time Comparison ................................................. E‐22
Figure Apx E‐26: Well #2, 13 5/8” Casing Thickening Time Comparison, TOC Slurry .............................. E‐22
Figure Apx E‐27: Well #2, 13 5/8” Casing Compressive Strength Development Comparison at Shoe .... E‐23
Figure Apx E‐28: Well #2, 13 5/8” Casing Compressive Strength Development Comparison at TOC ..... E‐23
Figure Apx E‐29: Well #2, 11 7/8” Liner Thickening Time Comparison, TOC 21,500’ MD ....................... E‐25
Figure Apx E‐30: Well #2, 11 7/8” Liner Thickening Time Comparison, TOC = 20,500’ MD .................... E‐25
Figure Apx E‐31: Well #2, 11 7/8” Liner Compressive Strength Development Comparison at Shoe ...... E‐26
Figure Apx E‐32: Well #2, 11 7/8” Liner Compressive Strength Development Comparison at TOC ........ E‐26
x
List of Tables
Table 1: Example operations ....................................................................................................................... 27
Table 2: Example Temperature and Pressure Data for Deepwater RCPC Workaround ............................. 28
Table 3: Example Data for Deepwater RCPC Workaround ......................................................................... 29
Table 4: Maximum ECDs in well 1 ............................................................................................................... 35
Table 5: Maximum ECDs in well 2 ............................................................................................................... 35
Table 6: Low Density Viscous Pill Test Fluids .............................................................................................. 43
Table 7: High Density Viscous Pill Test Fluids ............................................................................................. 44
Table Apx B‐1: Entrance Lengths of Mud and Cement. .......................................................................... B‐16
Table Apx B‐2: Pressure drops for mud though well regions calculated with the COMSOL model and with
analytical methods ................................................................................................................................... B‐18
Table Apx B‐3: Pressure drops for cement though well regions calculated with the COMSOL model and
with analytical methods. .......................................................................................................................... B‐19
Table Apx C‐1: Maximum ECDs in well 1. ECDs reduced at bottom‐hole, but increased at the previous
shoe. ........................................................................................................................................................... C‐4
Table Apx C‐2: Maximum ECDs in well 2. ECDs reduced at both bottom‐hole and the previous shoe. ... C‐5
Table Apx E‐1: Well #1, 16" Slurry Design Comparison ............................................................................ E‐8
Table Apx E‐2: Well #1, 16” Liner Slurry Test Results Comparison ........................................................... E‐9
Table Apx E‐3: Well #1, 9 7/8” Slurry Design Comparison ...................................................................... E‐11
Table Apx E‐4: Well #1, 9 7/8” Liner Slurry Test Results Comparison .................................................... E‐12
Table Apx E‐5: Well #1, 7 5/8” Slurry Design Comparison ...................................................................... E‐15
Table Apx E‐6: Well #1, 7 5/8” Slurry Test Results Comparison ............................................................. E‐15
Table Apx E‐7: Well #2, 16” Liner Slurry Design Comparison ................................................................. E‐17
Table Apx E‐8: Well #2, 16” Slurry Test Results Comparison .................................................................. E‐18
Table Apx E‐9: Well #2, 13 5/8” Casing Slurry Design Comparison ........................................................ E‐21
Table Apx E‐10: Well #2, 13 5/8” Casing Slurry Test Results Comparison .............................................. E‐21
Table Apx E‐11: Well #2, 11 7/8” Liner Slurry Design Comparison ......................................................... E‐24
Table Apx E‐12: Well #2, 11 7/8” Liner Slurry Test Results Comparison ................................................ E‐24
xi
Table o
of Conten
nts
Abstract .................................................................................................................................................ii
Acknowle
edgement Statement .............................................................................................................. v
Nomencllature: ..................................................................................................................................... vi
List of Figgures ..................................................................................................................................... viii
List of Taables ........................................................................................................................................ xi
Table of Contents ................................................................................................................................ xii
1 Execcutive Summ
mary ...................................................................................................................... 1
2 Proje
ect Introducction ..................................................................................................................... 2
2.1 RCPC Background Informaation ........................................................ .................................................. 2
2.2 Project Challenges ........................................................................... .................................................. 3
2.3 Project Apprroach ............................................................................ .................................................. 4
3 Phasse 1 Summary and Conclusions .............................................................................................. 7
4 Phasse 2 Summary and Conclusions .............................................................................................. 9
4.1 mmendationss .............................................................. ................................................ 11
Future Recom
5 Performance Capabilities an
nd Future Re
equirementss ............................................................... 12
5.1 Equipment aand tools ....................................................................... ................................................ 12
1
5.1.1 Analysiss of switchablle crossover .......................
. ..................... ................................................ 12
5.1.2
2 er for Liner Haanger and Cassing Hangers ................................................ 14
Switchaable Crossove
5.1.3
3 Floatingg equipment ................................................................. ................................................ 17
5.1.4
4 Considerations .............................................................. ................................................ 18
Other C
5.1.5
5 Tool operations mecchanisms .................................................. ................................................ 19
5.2 Modeling and Numerical Simulations .......................
. ..................... ................................................ 24
5.2.1
1 Comme ors ........................................................... ................................................ 24
ercial Simulato
5.2.2
2 Finite Element Mode pwater RCPC ...................... ................................................ 30
eling for Deep
5.2.3
3 OL Validation ................................................................. ................................................ 32
COMSO
5.2.4
4 COMSO Results for Temperature ......................... ................................................ 32
OL Modeling R
5.2.5
5 COMSO
OL Modeling R essure ................................ ................................................ 35
Results for Pre
5.2.6
6 Hydraullic Analysis ................................................................... ................................................ 36
5.3 Materials ....................................................................... ................................................ 37
Cementing M
1
5.3.1 RCPC Slurry Design .................................................................. ................................................ 37
xii
5.3.2
2 Spacerss .................................................................................... ................................................ 41
5.3.3
3 Viscous Pills for Fluid
d Separation ............................................ ................................................ 41
6 Deepwater RCPC
C Operational Performance ............................................................................. 49
6.1 ency Planningg ......................................... ................................................ 49
Operational and Continge
6.2 derations ...................................................................... ................................................ 49
Other Consid
7 Tech
hnology Tran
nsfer Efforts ....................................................................................................... 50
Referencces ......................................................................................................................................... 52
Appendix A ‐ Alternativve method fo RCPC simulatiion using WELLCAT .............. A‐1
or performingg deepwater R
Appendix B ‐ COMSOL Governing Eq quations, resccaling metho ds, boundaryy conditions aand other
modeling parameters ...................................................................................... ............................................... B‐1
e and Pressurre Results ........................... ............................................... C‐1
Appendix C ‐ COMSOL Temperature
draulic Analyssis ............................................................ .............................................. D‐1
Appendix D ‐ RCPC Hyd
ng Materials TTesting ..................................................... ............................................... E‐1
Appendix E ‐ Cementin
Appendix F ‐ Deepwate
er RCPC Operrational Guide
elines ................................. ............................................... F‐1
xiii
1 Executive Summary
The objective of the RPSEA Deepwater Reverse‐Circulation Primary Cementing (RCPC) project is
to assess the viability of performing RCPC to reduce circulation pressure requirements for
deepwater wells, to determine required technology to apply RCPC for deepwater wells, and to
present development strategies for required technologies. In RCPC fluids are pumped
downhole via the annulus and then up into the casing, in contrast to a conventional cement job
where fluids are pumped down the casing then up into the annulus. The application of RCPC to
deepwater wells is expected to reduce bottomhole circulating pressures and prevent lost
circulation during cementing as well as increase safety, enhance environmental sustainability,
provide zonal isolation, and improve cement seals. Other benefits that have been seen
through the application of RCPC include reduced placement time, reduced excess pumped and
a reduction in the amount of retarding additives used in the cement.
Challenges in applying RCPC to deepwater include the necessity of a downhole crossover tool to
divert fluids into the annulus downhole below the BOP stack, the availability of specialized float
equipment, the availability of simulators that are able to model the complex flow path of
deepwater RCPC circulation, and cementing fluid design considerations. The approach to this
project was divided into two phases and the scope of work comprised analyzing the RCPC
placement method, preparing a development path for technology required to apply RCPC to
deepwater wells, and creating preliminary operational procedures with associated contingency
plans. The objectives of Phase I were to confirm, evaluate and address expected challenges and
benefits of RCPC in deepwater. Phase I results were analyzed to determine technical issues and
to identify technologies to be developed for implementation of deepwater RCPC. Phase II of the
Deepwater RCPC project focused on a more detailed analysis of deepwater RCPC technical
requirements and the development of finite‐element simulations.
During the course of this project, finite‐element software package has been used to develop a
robust model capable of handling deepwater RCPC. Also, a method was developed to be able
to use workarounds within commercial cementing simulation software to perform deepwater
RCPC simulations. The resulting temperature and pressure profiles provided basic estimates of
placement. Laboratory intermixing studies have shown that rheology is a key parameter in fluid
design and placement while use of a conventional fluid hierarchy can result in interface
instability and fluid swapping. Key mechanical components required to perform deepwater
RCPC were analyzed to determine the current state‐of‐the‐art, as well as future, performance
requirements.
Overall, the applicability and benefits of RCPC to deepwater should be evaluated on a case‐by‐
case basis. Existing gravel pack and sting‐in float technology can be modified for use in the near
1
future. HHowever, tecchnology needed for futture developpment includ des the mod dification of float
equipme ent and a swwitchable cro ossover thatt will divert fluids on deemand. The next step in n tool
developm ment should d add capabiilities that allow for nonnmechanicall operation o of tools fromm the
surface by incorporrating techn nologies succh as RFID,, chemical‐aactivated trriggers, or mud‐
pressure pulses. Mud d removal and fluid sepaaration will remain a maajor challengge for deepw water
RCPC sinnce physical separation will need to be maintaained throu ugh the use of viscous plugs
instead of
o traditionaal plugs, daarts, or ballss. The desiggn methodo ology of cemmenting fluids is
affected by this chan nge in placem ment metho od since the leading edge of cementt will become the
critical sh
hoe slurry. A
A close review of simulattions of variious wells an nd casing strrings revealss that
cement slurry is offten expose ed to a higgher downhhole circulatting temperrature, and that
placement time can n be shortened significantly in som me cases. Hydraulic aanalysis of tthese
deepwatter strings has confirme ed the criticaal depths att which placcement by RRCPC resultss in a
lower equivalent circculating density (ECD).
2 Projject Introd
duction
2.1 RCPC Backgr
R round Inforrmation
In Reversse‐Circulatio on Primary C Cementing (RCPC) fluidss are pumpeed downholee via the annulus
and then n up into the e casing, in ccontrast to aa conventionnal cement job where fluids are pum mped
down the casing the en up into the
t annulus.. RCPC has been in use as a placeement techn nique
since the e 1960’s. Ap pplications haave included d casing leakk repair, tiebbacks, and production caasing.
Reverse cementing has also bee en used in geothermal
g wells with foamed cemment due to o low
formation fracture pressure an nd a high risk
r of lost circulation.. Most preevious published
applicatio ons have be een on land w wells, but RC CPC has alsoo been impleemented offfshore in shallow
water wells.
w The primary
p bennefits that have
h been seen with cementing through revverse
circulatioon is that fricction pressu ures and equ uivalent circuulating densities (ECD) aare reduced since
it is no longer nece essary to lifft fluids all the way uup the annuulus as duriing conventtional
cementin ng placemen nt. Gravity aalso helps to o place the ffluids down the annuluss. Other ben nefits
that have e been seen through the e application n of RCPC innclude reducced placemeent time, red duced
excess pumped
p and d a reductio
on in the am mount of reetarding add ditives used
d in the cemment.
Keeping placement p pressures low is critical in formations with low fracture graadients or w where
lost circuulation is a concern. Deepwater
D wells
w continnue to be d
drilled deepeer in challen
nging
condition ns, with increasingly co omplex arch hitecture. TThese deepwwater wells will need tto be
effectively cemente ed and sealed to ensu ure zonal issolation forr production n efficiencyy and
environm mental prote ection. A placement te echnique thaat can poten ntially reducce lost circulation
risk, redu uce ECDs an nd lower fricction pressures in formaations with a narrow po ore‐frac gradient
has a cle ear appeal to o the deepw water industry. Howeveer, since a reeverse‐circulation placement
2
technique has not yet
y been implemented in deepwater, the poteential beneffits and risks will
need to b
be thorough hly evaluated
d prior to implementatioon.
2.2 Project Chal
P llenges
An example of a dee epwater RCP PC configuration for a linner is shown n in Figure 1 (a), compared to
a conven
ntional floww path show RCPC operattion, fluids were
wn in Figure 1 (b). In laand‐based R
injected directly into
o the annulu us. For deep pwater RCPC C placement to occur, ffluid will need to
be pumpped down a a workstringg so that no o cementingg fluids are pumped in n the riser, then
through a crossover tool to be d diverted intoo the annulu s. Fluids will continue tto circulate d down
the annu
ulus, back up
u into the casing, thro ough the croossover tool and then into the annular
space aro
ound the wo ork string.
(a) (b)
Figgure 1: (a) Reverse‐Circulation P ng
Primary Cementting (b) CConventional Prrimary Cementin
A review
w of past literrature on the current state‐of‐the‐aart of RCPC h has shown that the follo
owing
challenge
es apply:
Accurately
A id
dentifying when
w cemen nt reaches the casing shoe to avoid incomplete
ce
ementation around the shoe or a large amount of cement tto drill out frrom the casiing.
o When n cement has filled the shoe track aarea, pumpiing can be sstopped. If more
cemen nt than neceessary is pumped the c ement left iin the casingg will need tto be
drilledd out, which
h increases cost and rigg time. Ho
owever, stop
pping beforee the
3
cemen nt has reach hed the shoe e can resultt in incompleete cementaation around the
shoe.
The availabiliity of speciaalized float equipment
e tto circulate the well an
nd perform RCPC
operations.
o A challenge with R RCPC is that convention al float equiipment cann not be used. Past
RCPC jobs have ussed either sp pecialized flooat equipmeent, or no flo oat equipmeent at
all. A
A technique e that has be een used on land wells is to reversee cement thrrough
an inn ner string without float equipment.. A disadvantage of thiis method iss that
pressu ed to be held on the c asing until tthe cementt sets, which
ure will nee h can
increaase the risk of micro‐an nnulus formmation. Thiss technique is unlikely tto be
used in deepwater. Vario ous patents for speciaalized float equipment and
placem ment metho ods for revverse cemennting have been estab blished, how
wever
viabiliity of the application to deepwater nneed to be eevaluated.
Cement desiggn modificattions due to temperaturres variations during placement affeecting
WOC time an
W nd thickeningg time since e cement neaar the botto om of the an nnulus is exp
posed
to
o the Bottom m Hole Circulating Tem mperature (B BHCT) while cement neaar the top o of the
annulus is exp posed to a lo ower temperature.
o Cement designs for f RCPC arre measuredd by the sam me test parrameters as with
conve entional cem menting, in ncluding thhickening time, comprressive streength
develo opment, ge el strength developmennt, free fluiid, fluid losss and rheo ology.
Howe ever, there aare some de esign consideerations and d modifications that maay be
made to slurries for RCPC. Considerati ons includee rheological hierarchy, mud
removval efficiencyy, and conce entration of cementing aadditives.
Computer sim mulation and modeling are necess ary for paraameters succh as ECD, p pump
raates, frictionn pressures, temperature e gradient and mud rem moval efficien ncy.
o A cha allenge in ap
pplying RCPC cement ddesigns to d deepwater aapplication is the
develo opment of software to o model annd monitor RCPC jobs. Consideraations
includ de placeme ent simulattions, mudd removal, ECD mod deling and the
tempe erature grad dient as the ccement is puumped.
2.3 Project App
P roach
Phase I of the Deepwater RCP PC project has
h focused on investiggating the ccurrent status of
technoloogies and documentingg those thatt need to be developed to RCPC C for deepwwater
applicatioons. The obbjectives of tthe project ttasks in Pha se I have beeen to confirm, evaluatee and
address expected ch hallenges and benefits o of RCPC in ddeepwater. The first task of the prroject
was to fo
orm a Working Project G Group consisting of operrating compaanies and seervice compaanies.
The Working Projecct Group provides technical guidaance on co onditions an nd challengees of
deepwatter operations and how w they wouldd relate to R
RCPC. Next,, an industrry‐wide survvey of
4
experts in the field wwas conducted. Industryy experts in deepwater and expertss experiencee with
reverse‐ccirculation placement
p techniques were surveeyed througgh an onlin ne survey. The
purpose of this surve ey was to gaather opinion
ns, expectattions and concerns abou ut the appliccation
of reversse‐circulation cementingg technique
es to deepwwater. The information collected in
n this
survey was
w combine ed with Working Projecct Group feeedback and informationn produced from
the Technology Statu us Assessmeent to guide subsequent tasks.
Figure 2: (a) Generic Deepw
water Well Sche
ematic #1 (b) G
Generic Deepwater Well Schematic #2
5
There were many anticipated challenges to the application of RCPC to deepwater including
modeling and simulations, mechanical placement controls, and cementing materials.
Subsequent tasks of this project included the development of a series of simulations and
numerical models applicable to RCPC operations in deepwater including finite element
modeling, temperature simulations, and job placement simulations. The evaluation and
analysis of this task was performed by the University of Houston and by CSI Technologies.
Mechanical components required to implement RCPC in deepwater wells were also
investigated and analyzed. This evaluation was performed by Weatherford, and included
analysis of cement flow and placement controls required to direct fluid down an annulus of a
deepwater well casing or liner, and a means to separate fluids during placement. Cementing
material considerations were also studied to identify potential design and performance benefits
and issues based on cement performance under deepwater conditions when placed by RCPC.
Considerations included fluid intermixing, key aspects of spacer design, additive sensitivity, and
wait‐on‐cement (WOC) time. This analysis and evaluation was conducted by CSI Technologies.
Operational performance of RCPC in deepwater was another key aspect investigated in Phase I.
While some operational procedures will remain the same as with conventional placement, the
application of a novel technique will require special consideration and planning. This included
the identification and addressing of expected issues that come from the application of the RCPC
process on a deepwater rig. Anticipated issues included liquid additive proportioning for
accurate cement formulation, mixing rates and cement deliverability, measurement of
displacement volumes and measurement of downhole and return flow rates. Overall, it was
expected that there would be numerous technical issues that must be addressed before routine
RCPC operations can occur in deepwater wells. Phase I results have been analyzed to
determine technical issues and to identify technologies to be developed for implementation of
deepwater RCPC.
Phase II of the Deepwater RCPC project focused on a more detailed analysis of deepwater RCPC
technical requirements and the development of the COMSOL simulations. Preliminary models
developed in Phase I were refined and developed in more detail, with a focus on ECDs and
pressures during placement. Also, workarounds for RCPC in commercial software were further
investigated. Next were comparisons of results from both the COMSOL model and the
commercial simulators on temperatures and pressures. Phase II work on the mechanical
placement controls also built on Phase I results. Since the development of a switchable
crossover tool was identified as a development area in Phase I, Phase II analysis included tool
requirements for long strings and tie‐back strings. Further analysis was conducted specifically
for crossover tool applications in conjunction with a liner hanger. Areas of interest were ways
to combine the crossover tool with the liner hanger assembly, compatibility with existing liner
hangers and tools, and operations with both set and unset liner hangers. Since the selection
6
and use of float equipment would need to be determined on a case by case basis, further study
was done on methods to control fill‐up and flow back and minimizing flow restrictions during
placement.
Phase II work also included the design optimization of cement and spacer design based on the
fluid intermixing studies conducted in Phase I. Other considerations were that slurries for RCPC
applications will typically not be exposed to the same circulating temperature as those during
conventional placement. The performance of optimized RCPC cement designs were compared
with conventional designs, with thickening time, compressive strength development, gel
strength development, stability and fluid loss as areas of focus. Another area of fluid
optimization looked at the design and use of viscous pills for downhole fluid separation as an
alternative to mechanical separation; balls and wiper darts could be used in the work string
section only. Phase II tasks also included the preparation of an operational plan to successfully
implement RCPC in deepwater and a contingency plan for all major expected contingency
situations.
In Phase 1, the following development areas were identified:
Mechanical Placement Controls
One of the biggest project challenges from the mechanical side will be the
development of a switchable crossover. The crossover needs three positions: 1) going
in the hole with flow directed in the conventional direction through the ID of the
casing; 2) switched to direct flow to the annulus side of the casing while allowing
returns to come through the ID of the casing then exit to the annulus above a pack off
above the reverse flow port; 3) switch back to allow flow to go in the conventional
7
direction to set the liner hanger and possible activate a shut off valve to keep cement
in place while pulling the work string out of the hole.
Further study is needed for float equipment. Float equipment to keep cement in
place on the back side of the casing needs to be decided upon on a case by case basis
at this juncture of reverse cementing.
Modeling and Simulations
Due to the specific nature of the wells under consideration, standard commercially
available software packages are unable to model the flow path through the complex
configuration of a deepwater reverse‐circulation cementing process. Further
refinement and analysis is needed to evaluate downhole pressures and temperatures in
a full‐scale well.
A COMSOL Multiphysics finite‐element software package has been used to develop a
robust model capable of handling the reverse‐circulation cementing process. The model
has been successfully applied to a scale well to predict temperatures and pressures, and
work is currently being performed to scale the model to the full‐size wells. Once the
full‐size simulations are complete, comparisons with commercial software simulation
results are needed to evaluate the applicability of the commercially‐available software
to deepater RCPC operations.
Cementing Materials
Fluid intermixing is dependent on fluid rheology, fluid density, well geometry and flow
rates. These properties affect the design of both slurries and spacers. In Phase I,
rheological and density hierarchy effects were investigated. Further investigation is
needed on the effects of well geometry, eccentricity and deviation, as well as the
quantification of these effects in order to provide best practices for RCPC fluid design.
Operational Considerations
Many operational considerations are expected to remain the same with RCPC as with a
conventional deepwater primary cementing. However, there are many anticipated
additional considerations introduced by RCPC.
Considerations that need further study include hole cleaning, circulation direction,
mixing rates, measurement of displacement volumes and measurement of downhole
and return flow rates.
Additional considerations specifically for deviated wells need to be investigated.
8
4 Phase 2 Summary and Conclusions
Overall, the applicability and benefits of RCPC to deepwater wells should be evaluated on an
individual basis. In some cases, RCPC can significantly shorten operational time for cement
placement. Simulations and determination of the critical depth can determine if RCPC is viable
for that string since in RCPC can result in an ECD reduction during placement. Weak formation
or zone needs to be below the critical depth for RCPC to have an advantage
Mechanical Placement Controls
A switchable crossover is preferred to give flexibility in desired flow direction during
various operations ranging from running casing, cement placement and setting liners.
A mechanically operated crossover, which could be a modification of a gravel pack
crossover, may be the first to market for reverse cementing.
A crossover tool that is strictly operated with ball(s) and/or dart(s) would be a major
challenge.
A crossover tool can be located above the liner/casing hanger, or below the hanger if
port collars are used.
From an operator’s perspective the selection and placement of floating equipment for
reverse cementing should be done on a well by well basis.
The float equipment can be run near the top of the liner and held open by a stinger, or
near the bottom of the casing if needed to aid flow into the casing ID.
Various methods could be used to detect the location of fluids downhole or trigger the
operation of downhole crossover tools or valves. These methods include balls, darts,
RFID tags, chemical tags, mud pressure pulse, wire drill pipe, hydraulic control lines from
the surface, mechanical or gas powered springs, or any combination.
Modeling and Simulations
A method was developed to be able to use workarounds within commercial cementing
simulation software to perform deepwater RCPC simulations. The resulting temperature
and pressure profiles are somewhat simplistic.
A robust model has been developed which predicts temperatures and pressures/ECDs
during a RCPC job. This model is built on the COMSOL Multiphysics software package,
and uses Finite Element Analysis to simultaneously solve the heat equation and Navier‐
Stokes equations.
9
In the developed model, secondary calculations such as the load on the pump, the
presence of U‐Tubing and the required back‐pressure to prevent U‐Tubing are available.
This model was applied to several example cementing jobs, results indicate that overall
the bottom‐hole circulating temperature is increased significantly in reverse cementing.
ECDs at bottom‐hole are reduced significantly, and ECDs at the previous shoe can either
increase or decrease depending on the configuration of the well. RCPC reduces ECDs at
bottom‐hole by changing the flow path and reducing the frictional component of the
pressure at bottom‐hole. The static pressure contribution does not change from
conventional to reverse.
The fact that ECDs can be increased at the previous shoe is a significant issue. If the ECD
is reduced at bottom‐hole and increased at the previous shoe, then there is a point
between those two where the pressures in conventional and reverse circulation are
equal. The location of this point with respect to a weak zone or zone of interest in the
annulus will be a factor in determining if RCPC is advantageous in that particular
application.
If the previous shoe is at the same depth as the crossover tool, then the pressure in
RCPC will be higher than for conventional placement. However, if the overlap between
the casing being cemented and the previous casing is large enough, such as with a tight
liner lap annulus, then the previous shoe could be lower than the critical depth. In such
a configuration, the pressure exerted against the formation could be reduced.
Cementing Materials
Many slurry considerations for RCPC are expected to be similar to conventional
deepwater primary cementing slurry considerations.
Slurry systems can be tailored to optimize compressive strength development. If there
is a significant difference between the static temperatures of the top and bottom of the
cement column, additive staging with a decrease in retarders in near the top of cement
is possible to optimize compressive strength development throughout that interval.
Alternative density hierarchies are possible with RCPC placement. Fluid placement
downhole is assisted by gravity down the annulus. A conventional density hierarchy,
where each fluid pumped is progressively heavier than the preceding fluid, can result in
fluid swapping and intermixing in the annulus due to the difference in density.
10
Rheology is a key param
meter in flu
uid design aand placem
ment. As w
with conventtional
placement, an increasiing rheologgical hierarrchy is neeeded to assist with fluid
displacementt.
Operatio
onal Conside
erations
Equipment tyype and con
nfiguration can
c be seleccted based on operato or and regulatory
needs, and w
what is most appropriatee for the exp ected well cconditions.
Preparing perrsonnel on location for the reversee operation w will be speciific to the tyype of
equipment that is used d and/or de
eveloped foor the operration. Possible continggency
siituations ne
eed to be reviewed
r nd plans sh ould be covvered as paart of perso
an onnel
trraining/pre‐jjob procedure.
After the cro
A ssover tool, the next crritical area oof developmment will be to use emeerging
te
echnology in
n deepwaterr such as RFID and fiber‐‐optics to reliably detectt the presen nce of
ce
ement insid
de the casin
ng shoe. This
T technology shouldd be used iin addition to a
vo
olumetric m
method of mo onitoring fluids in and reeturns.
Modelingg and Simulations
Placement simulators shhould be exp
panded andd made avaiilable within
n the industtry to
model deepw
m water RCPC fflow paths and cement pplacement.
11
Simple equattions to estiimate the difference in pressure frrom conventional to revverse
have been de eveloped. Th hese can be used as a prreliminary evvaluation to ool for deepw water
RCPC to determine if RCP PC will be ad
dvantageouss, and if the resulting EC CDs downholle are
acceptable.
Cementing Materials
Fluid designss will need to be tailo
ored for eacch applicatioon dependeent on mod deling
reesults and de
esired param meters.
Operatio
onal Conside erations
The personne el on location may or may not haave previouss experiencee with a revverse
ceementing op peration. Th he job timin
ng and inherrent procedu ures will havve to be clarified
prior to the jo
ob for optim mal performaance.
5 Perrformance
e Capabilities and Future Requireme
ents
This secttion details tthe current “State of deevelopment”” of deepwaater RCPC. Included are the
functionaal requireme ents and cap each major eequipment, software an
pabilities of e nd techniquee.
5.1 Equipment
E t and toolss
Key mechanical components forr reverse‐circculation opeerations werre analyzed for the follo
owing
string typ
pes:
Lo
ong string – casing typiccally hung offf near the w
well head
Liiners – casin
ng typically hung off near the end oof the previously run caasing with a liner
hanger
Tie Back Strinngs – casing typically hu
ung off near the well heead tied backk to the PBR
R of a
previously run liner
5.1.1 Analysis of sw
A witchable crrossover
For deep pwater RCPC C, a downho ole crossoverr tool is neeeded to diveert flow from m the work sstring
into the annulus to b be cemented (Figure 3). Phase 1 woork on crosssover tool deesign looked d at a
fixed croossover and a switchable e crossover design optioon. Fixed crrossover typ pes are similar to
those used on port collars for ggravel packin ng, and mayy require a ssecond run in hole afteer the
casing haas been land ded. The fixe ed crossoverr tool is thenn run in on aa work stringg and stung into a
reverse‐ccementing p port collar fo or pumping. After placem ment the to ool and workkstring are ppulled
to close tthe port and d then reverrsed out. Th he operationn of a fixed‐ttype crossovver tool operated
by balls is shown in
n Figure 4. In
I comparison, a switchhable crosso over tool may not need
d the
casing too be landed prior to pum mping and th he flow direcction can bee changed do ownhole witthout
additionaal tool runs.
12
Figure 3 nal Flow Path ( b) RCPC Flow Paath
3: (a) Convention
anical Fixed Crosssover Tool
Figure 4: Mecha
F
13
The switchable crosssover tool lo ocated abovve the top oof the liner n needs to be compatible with
existing lliner hangerrs and casingg hangers. Itt has to alloow for the reeverse cemeenting proceess to
take placce, before thhe liner hangger is set orr the casing is hung off. The switchaable crossovver of
this desiggn will be disscussed further in the neext section.
For a swiitchable crosssover type system thatt uses a portt collar located below th he liner hangger or
casing hanger, then s or the casing is hu
n the liner hanger is set ung off beffore the revverse
cementin ng process takes
t place. This type of
o placemen t or reversee cementingg operationss will
require aa method to o divert floww into the an nnulus below
w the liner hhanger, such h as use of aa port
1
collar. Port collarss have alsoo been use menting opeerations in relatively weak
ed for cem
2
formations.
5.1.2 Sw
witchable Crossover for Liner Hanger and Cassing Hangerss
The crosssover tool iss located aboove the top of the liner hanger or caasing hanger and is inteended
to be ussed to place
e cement on
n the annulu
us side of tthe casing b
by cementin
ng in the revverse
directionn before the casing is hung off.
Minimum
m requireme
ents for the sswitchable ccrossover aree:
To be univer
T rsal, the crossover need
ds to work w
with existingg liner hangeers and/or ccasing
hangers.
h
Allow flow in
A n the conven nto the hole.
ntional directtion, while tthe casing is being run in
When the ca
W asing is at th
he desired lo ocation, the crossover wwill switch ddirections so
o that
all the fluids
a pumped do own the work string will be diverted d to the annu ulus between the
casing OD an
c nd open hole e ID with retturns being ttaken up thee casing ID aand then diveerted
at
a the crosso over tool to the annular area betw
ween the wo
ork string OD and previously
run casing ID
r D.
To switch flo
T ow directionss the crossover must paack‐off the annular area between th he OD
of the tool a
o nd the ID off the previouusly run casi ng between
n the exit po
ort from the work
sting ID to th
s he exit port ffrom the cassing ID.
After
A the ce
ement is plaaced the cro
ossover will then switcch back to tthe conventtional
direction to a
d allow the lin
ner hanger to o be set or tthe casing hung off. A siimple ball an nd/or
dart
d drop opperated cro
ossover can be designe d to meet the minimu um requirem ments
liisted for the
e crossover tool.
1
US Patent 7,857,052 “Stage Cementing Method ds used in Casing w
while Drilling” by G
Giroux, et al, startiing with Figure 12A through 12F
2
Jim McNicol, Archer Oiltools. SPE‐168048‐MS.
14
Functional requirements for the crossover tool when located at the top of the casing are:
Ability to wash/ream the casing to bottom with standard or reverse flow.
o Reverse flow can be done by pumping down the riser with returns being taken
up the liner ID and through the work string ID.
o Reverse flow can also be done with concentric drill pipe where the fluid is
pumped down the outer annulus and returns are taken through the liner ID back
up the inner ID of the drill pipe. Note this option pre‐supposes a traveling pack‐
off to keep the fluids separated.
o Another option is by pumping down the work string with flow crossed over at
the crossover tool and returns taken back up the liner ID until the flow is crossed
over to the annular area above the crossover tool. Note this option pre‐supposes
a traveling pack‐off to keep the fluids separated.
Ability of maintaining rotation during washing/reaming in – once the hanger pack‐off is
set, rotation must stop – the crossover will stay attached to the casing until it’s hung off.
Maintain a minimum ID to allow the dropping of balls and/or darts through the tool(s)
when in the conventional circulation mode.
When used on a casing system that requires a liner hanger, the crossover as a minimum
must allow conventional circulation while running in the hole, crossover to reverse
cement, and switch back to the conventional flow direction to set the liner hanger.
Desired capabilities for a switchable crossover are to:
Have a four position crossover tool system that can be switched on demand (Figure 5):
1. Conventional flow direction where all fluid pumped down the work string goes
through the liner ID and back up the annular area between the liner OD and the
previously run casing ID or open hole ID. This flow direction also allows fluid flow down
the external annulus and back up the liner ID and the work string.
2. Reverse flow direction where all fluid pumped down the work string is directed
to the annular area between the liner casing OD and hole/or previously run casing ID. All
returns are taken up the ID of the liner and diverted to the annular area between the
work string and previously run casing and riser ID.
3. Circulation mode where all fluid pumped down the work string is diverted at the
tool(s) to the annular area between the working string OD and previously run casing and
15
riser ID. No fluid
f enters the liner ID
D or the annnular area between the liner casingg and
open hole ID.. (Needed to o place a ball or dart to sset a liner haanger, witho out disturbin ng the
previously plaaced cementt.)
4. Bullhe
ead or Squeeze mode where
w all fluuid pumped into the wo
ork string caan be
used to presssure up the work stringg to set the liner hanger and operaate other too ols in
th
he system.
Figure 5: C
Crossover tool syystem that is sw mand
witchable on dem
In review
wing the reequirements to the operation modde of the ccrossover to ool, is has been
determin ned that a crrossover too ol that is striictly operateed with ball((s) and/or daart(s) would d be a
major chhallenge. To add the deesired capab bilities to thee operating mode of thhe crossoverr tool
system means
m that the tool syystem operaating methood(s) will neeed to be eevaluated fro om a
multiple discipline sttand point. This includees the use oof traditionaal operating systems such as
balls andd/or darts syystems, the u use of battery powered systems, su urface powered systemss (e.g.
16
hydraulicc and/or electrically powered),
p stored
s enerrgy systems such as springs – both
mechanical and gas powered, orr vacuum ch hamber pow ered system ms that’s activated by vaarious
methodss such as dro opped balls, pump down darts, RFID D tags, chem mical tags, w
wired drill piipe, a
wire line
e plug in runn from the surface, fiber optics, m mud pulse, m mud pressure pulse, tim mers,
hydraulicc lines run from
f the surrface, conceentric drill sttring with th
he outer annulus of thee drill
pipe useed as a hydrraulic fluid carrier, etc. have all beeen consideered. The feeasibility of each
system o or combination(s) of each system will need to bee carefully evvaluated.
In summary, a mechaanically operated crosso over, which ccould be a m
modification of a gravel p
pack
crossover, may be th
he first to maarket for revverse cementing.
Functional requirements determ
mined in Phasse 2 for RCPC float equip
pment are:
The ffloating equipment musst have the minimum EC
CD possible,, e.g. as largge a flow paath as
possible
The floating equipment must remain open whi le the cem ment is placced. If it ccloses
premmaturely then the reverse cement job j could sqqueeze off leaving cem ment in the work
stringg, which wo
ould be diffiicult to dum
mp or reversse out whilee still attach
hed to the liner,
beforre the liner h
hanger is sett.
Floating equipmeent must nott limit the opptions of whhat can be ru
un on the bo ottom of thee liner
casing for reamin
ng in the hole – particulaarly while reeaming in wh
hile circulating in the revverse
mode e.
An in
nter string ru
un inside the e liner casing can be useed to keep tthe floating equipment open
and take
t returnss back to th
he top of th
he liner durring reversee cementingg. Once the liner
hanger is set the inter string can be movved up and leet the float vvalve(s) close. Excess cement
pumpped into the
e ID of the inter string can be dum
mped on top p of the floaat equipment by
pumpping in the cconventional manner, orr it can be reeversed outt by pumpingg down the riser.
17
(Note
e: The frictio
on of pump
ping up the inner stringg needs to b
be taken intto account w
when
considering ECDss.)
The floating
f equ
uipment can be run neaar the top oof the liner and held open by a sttinger
(inne
er string).
If the
e floating equipment iss needed to o aid in flow
w into the casing ID, tthen the flo oating
equip pment shoulld be run ne ear the botto om of the caasing. (Note:: Floating eq quipment is not a
primaary well conntrol valve, it
i is not dessigned to seeal gas and iit has a limiited capabiliity to
hold differential pressure fro om below.)
All floating equipme ent run on reverse cem menting appplications sh hould probaably seal in both
directionns once it’s closed,
c e.g. opposed flaapper valve floating equ uipment – TThis preventts the
circulatioon of fluids inside the ID
I of the linner or abovee the top o
of the liner, e.g. dumpin
ng or
reversingg out excesssive cement from a inte er string from adverselyy affecting tthe newly placed
cement.
Figure 6: Full Op
F pen Reamer Shooe example
18
5.1.5 Tool operatio
T ons mechan
nisms
Phase I results
r have shown thatt a challengge of deepw
water RCPC w
will be to acccurately ideentify
when cement reache es the casing shoe to avvoid incompplete cemen ntation around the shoee or a
mount of cem
large am ment to drill out from the
t casing. When cemeent has filled the shoe track
area, pum mping can b be stopped. If more cement than nnecessary iss pumped th he cement leeft in
the casinng will need to be drille ed out, whicch increases cost and rig time. Ho owever, stop pping
before the cement has reached d the shoe can result in incompleete cementaation around d the
shoe, wo ould result inn reduced zo onal isolatio on at the shooe and increeased cost to o the rig thrrough
remediall costs. Pre
eviously on some land d‐based RCPPC jobs radioactive tracers were used,
howeverr this is conssidered unddesirable offfshore so otther methodds of fluid identificationn will
need to be investigaated and tessted. Variou us methods could be ussed to detecct the locatio on of
fluids do
ownhole or trigger the e operation of downhoole crossoveer tools or valves. TThese
methodss include baalls, darts, RFID
R tags, ch
hemical taggs, mud pressure pulse, wire drill pipe,
hydraulicc control liines from the surface e, mechaniccal or gas powered springs, orr any
combination of these e.
RFID Ap pplications Downhole: RFID techn nology helpps to replacce or comp plement cuurrent
downhole applications. RFID tecchnology hellps to improove commun nication withh downhole tools
and can be used for a wide varie ety of operations which include opeening and clo osing valvess. For
downhole purposes, RFID comp ponents conssist of a tagg and a read der. The tag is dropped from
the surfaace into the pipe while the reader is attached to the tool going downhole. The ttag is
programmed with a command through the use of softw ware in a com mputer. The reader is ussually
accompaanied by an aantenna, batttery pack, aand a motor..
Ultimately, RFID application is only favoraable if the TTag/Reader system can
n function u
under
normal ddownhole op perating con nditions. P.M Sinder annd Tom Doigg showed th he use of RFFID in
3
closing and opening a downhole circulating sub at 10,1000 feet. Durring circulatiion, multiplee tags
were droopped with ccommands instructing th he reader too open the sub. Once the first tag paassed
the reader downhole e, pump pre essure was ggreatly reduuced indicatiing commun nication betwween
the tag and reader. MMultiple tagss were dropped to ensu re that there was no case of tags geetting
3
Phillip and D
Doig, SPE‐113842‐‐MS
19
stuck to drill pipe and not reaching the reader. The reader was programmed to only decipher
commands from one tag thereby rendering the other tags redundant once the command was
received. This eliminated the possibility of multiple commands being sent to the reader. Testing
conducted at 302oF on the batteries indicated a max operating time of twenty (20) days. Also,
velocities in excess of 10m/s were achievable for tag detection for the antenna.
Iain Adan, presented a case study of a successful RFID enabled completion in the Middle East.4
It studied the use of a single RFID module to control a completion system that includes a
packer, circulation sleeve and downhole barrier valve. RFID was used to close and open a
barrier valve for tubing testing, to set production packer and open/close sliding sleeve for
annulus circulation. The system was run at 275oF and 5,000 psi. Mads Grinrod et al. studied the
application of RFID for drill pipe, downhole applications, and the temperatures and
shock/vibrations needed to be overcome.5 RFID tags have been shown to have good
performance in drilling fluids, pipe dope, and water. In order to show that RFID technology can
withstand downhole conditions, several tests have been conducted including temperature
cycles between 86oF and 410oF and pressures up to 10,000 psi. Shock, vibration and direct
impact tests were also conducted.
Passive radio identification devices can be used as means of detecting drill pipe depth
downhole. In the United States patent application, Joseph Zierolf described an innovative
approach for determining casing and drill pipe position with the aid of radio identification
devices combined with a downhole tool that has an inbuilt transmitter and receiver.6 Precise
depth measurements are vital in the drilling and completions industry since activities such as
cementing/fluid placement volumes are largely based on well depth. Inaccurate depth data
leads to inaccurate volumes and subsequently bad cement jobs. The modulus operandi involves
the placement of several passive radio identification devices on each casing or drill pipe joint. In
order to determine casing or drill pipe depth, the downhole tool, which has a radio frequency
transmitter and receiver attached, is lowered down hole and sends a constant signal. Upon
receiving this signal, the passive radio identification device resonates and transmits a response.
The response is received by the downhole tool and sent to a surface computer. The downhole
tool can also be modified with a battery so that all the information can be stored downhole and
deciphered when the tool is retrieved from the well.
The need to properly identify the interface between drilling and completion fluids during a
cementing operation cannot be overemphasized. In conventional cementing operations,
4
Iain Adan, SPE. SPE‐166182‐MS
5
Grinrod, et al. SPE‐163449‐MS
6
US 6,759,968 B2
20
inaccurate depiction of these interfaces could lead to drilling mud contaminating the cement in
the annulus. Reverse circulating aims to reduce this problem by pumping through the annulus
rather than through the drill pipe. As a result, there is a need to properly identify the
cement/mud or cement/spacer interface to determine when the cement reaches the casing
shoe. Robert Dillenbeck and Bradley Carlson, in a patent application described an apparatus for
detecting the cement/mud or cement/spacer interface7. The apparatus consists of a sensor that
is attached to the lower part of the casing shoe, and a detectable device that emits radio
frequency identification signals (RFID). The detectable device or transponder emits signals that
are detected by the sensor attached to the base of the casing shoe. Once the signal is detected,
more fluid is prevented from being pumped through the activation of a valve and the
subsequent pressure increase noted by the operator at the surface.
RFID technology has been shown to withstand downhole conditions and the technology is
generally expected to be successful in the industry. The downhole conditions/obstacles include
tag size, vibrations downhole, loss of communication, and reliability. The technology has been
successfully used to open/close valves and set packers. As a result, application in well
completions/cementing is expected to be positive. Potential applications of RFID technology to
deepwater RCPC is its use in the detection of cement location downhole or the activation of
downhole tools. Limitations for RFID applications to deepwater RCPC are that if the receiver is
located on the casing in the open hole section, there is a possibility for tags to get lost or stuck
in the formation before reaching the receiver.
Chemical Tags: Currently, there is limited literature on practical applications of downhole tools
controlled via chemical changes such as pH. The more common methods of tool activation
widely available with field tests include tools activated via specific flow and pressure changes.
While electronic pH meters are available, their actual application in a down hole trigger system
is not well documented even for drilling fluids and as such is even more remote for cement.
Available literature focuses on actual pH measurement using pH meters and pH sensitive dyes
that provide in‐situ measurements. For example, a pH downhole sensor has been used for in‐
situ pH measurement of formation water at reservoir conditions.8 It involved using pH‐sensitive
dyes which tend to change color depending on the pH of the accompanying fluid.
A preliminary laboratory investigation into possible chemical triggers or markers for downhole
tools was conducted. The objective was to evaluate a chemical system that acts as a chemical
trigger for potential application in deepwater RCPC. Assumptions are that the method of
activation would be a sequential change in pH from pumping a low pH chemical tracer. The
7
US 6802373 B2
8
Raghuraman, et al. SPE‐93057‐MS
21
detection n point for tthe strongesst signal is located insidde the workk string befo ore the crosssover
point. Other potential placemen nt for the deetector woulld be inside the casing sshoe or as paart of
the crosssover tool fo or detection n of cement location or for activatio on of floats or the crosssover
tool. A d
detector placed lower in n the well m may not be rreliable on itts own sincee the risk off fluid
contamin nation and intermixing increases in i the openn hole annu ulus, causingg an uncleaar pH
A differential pH of 5 waas targeted,, and the tri gger processs would be activated by the
signal. A
detection n of two changes in pHH. The prim
mary fluid beeing pumped
d (spacer orr mud) would be
assumed d to have a re elatively high pH. This ffluid would bbe followed by a low pH H chemical trracer,
then ano other high pH H fluid. Thiss drop in pH and subseqquential incrrease is inten nded to reduce a
‘false possitive’ effectt. This scenaario for a cheemical triggeer is shown iin Figure 7.
Figure on
e 7: Example of aa pH Chemical TTrigger Operatio
22
activated using a wiper plug. For shallow applications, the packer is set using an acoustic wave
generator on the rig floor. The signal is sent downhole via the casing string to a receiving sensor
in the packer’s electronic module. This signal is decoded and the setting sequence is initiated.
For deep well applications, sensors are located downhole that sense the magnetic field
produced by the wiper plug. Once the magnetic field is detected by the sensors, the setting
position of the packer is initiated. The electronic module of both systems are powered by
battery packs with operating temperatures ranging from 100 ‐ 392⁰F and run times of 336
hours.9
Future development of deepwater RCPC mechanical controls requires implementation of a way
to reliably detect cement location downhole during placement, particularly in the casing shoe
track to confirm job completion. Technology investigated by RPSEA project # 10121‐4504‐01 on
Intelligent Casing‐Intelligent Formation Telemetry (ICIFT) could have applications to address this
particular challenge of RCPC. Stalford et al, describe the implementation of intelligent casing
sensors to combat the age old problem of downhole information gathering.10 Obtaining real
time downhole information during or before completion operations is vital. To accurately
obtain distributed downhole data in producing wells, an intervention is usually required. The
need to eliminate the time, money, and effort required for well interventions justifies the
design of an intelligent casing sensor capable of sending downhole data. The sensors can be
placed in cement, the formation, and on the casing. The system can be equipped with fiber
optic distributed sensing in conjunction with radio frequency intelligent devices to enhance
data transmission. The system is beneficial as it reduces the need for well intervention, is not
affected by weather conditions, and ensures constant well monitoring. On the other hand, the
need for specific steel grades of casing for sensor attachment, hole attachments in the casing,
and unreliable wireless communications may hinder this technology. Real‐time monitoring
capabilities could allow for the positive identification of cement position downhole in the
annulus during RCPC. Also, pressure and temperature sensing capabilities would allow for
validation of RCPC placement simulations, and data for refinement if needed.
Also, RPSEA project # 10121‐4501‐01 may also provide a future avenue for cement detection
downhole. This project focused on the creation of a smart cement system with better sensing
capabilities to aid wellbore monitoring and small‐scale laboratory tests to evaluate
performance.11 The electrical resistivity of the cement is measured to determine the level of
contamination with oil based mud (OBM), and in small‐scale experiments, resistivity was used
9
Pleasants et al., SPE 160217
10
Stalford et al., OTC 25161‐MS
11
Vipulanandan and Heidari, OTC‐25200‐MS
23
to determ
mine the possitions of various fluids iin a modeledd annulus. This type off instrumenttation
and methodology caapable of diffferentiatingg OBM and ccement dow wnhole has cclear applicaability
to RCPC. While this method of using resistivity measurrements to differentiatee fluids has been
d in the laboratory, scaling this techn
validated nology to th e field is a challenge.
5.2 Modeling a
M and Numerrical Simu
ulations
Commercially available software e packages are
a unable tto model th he flow‐pathh of a deepw
water
reverse‐ccirculation cementing
c process
p due to the commplex well aarchitecturee and changges in
downhole flow path h. The COMSSOL Multiph hysics finite‐‐element sofftware packkage was useed to
develop a robust model
m capab
ble of handling the deeepwater reeverse‐circulation cemeenting
process. In Phase I tthis COMSOL model had d been succcessfully app plied to a smmall‐scale well to
predict temperatures and pressu ures along thhe well. Furrther develoopment and expansion o of the
FEA mod deling capabilities were extended in Phase 2. EEvaluations using comm mercial simulators
currentlyy available to the industry and the developmeent of workaarounds within the softtware
were connducted.
24
Figu
ure 8: (a.) Conventional Placeme
ent Simulation ((b.) Reverse Placcement Simulattion with commeercial software
In an efffort to obttain as reasonable comparison aas possible between simulations using
conventional and deepwater RCPC R techniqques, the bbaseline con
nventional ssimulations from
Phase I wwere re‐ran and their re m those pressented in the Phase I intterim
esults differ slightly from
report. For this po
ortion of the project, all
a simulatioons were ru un using thee same softtware
package to minimize e discrepancies. The sam me fluid propperties weree used for both conventtional
and deep pwater RCPC C simulation.
Conventio
onal Simulatiion
Deepwate
er RCPC
25
Above the Crossover: Configure a cementing simulation as for a conventional job (tubing‐side
downward flow). This simulation is meant to model the cementing operations above the
crossover point.
Define all well parameters as for a conventional cementing job. Define pumping
operations as for a reverse cementing job as seen from surface, but performed as a
conventional job.
Operations should be defined such that the simulator produces results for the fluid
fronts at key depths (e.g. spacer at the mudline or cement at the crossover point). This
may lead to operations with the same fluid and rate being split into smaller volume
steps for later convenience.
Rates and pressures for using a reverse cementing job in a conventional configuration
may lead to the simulation to show losses (fracturing) or influx (blowout) in the open‐
hole section. Simulator “safeties” should be disabled, bypassed, or ignored for the
region below the crossover. Burst and collapse pressures above the crossover should
still be considered.
Outputs from the “conventional” simulation should include the fluid temperature and pressure
as each fluid reaches the mudline and the crossover. Additional points of interest can be
analyzed if desired. The fluid temperature and pressure at the crossover will be used as inputs
for the “reverse” simulation.
Below the Crossover: Configure a cementing simulation for a reverse job (annulus‐side
downward flow). This simulation is meant to model the cementing operation below the
crossover point.
Define the surface of the well as the depth of the crossover. Pore pressure, fracture
gradient, geothermal temperature, “surface” temperature (geothermal temperature at
the crossover), and other well parameters will need to be translated to the new depth
reference.
Define the pumping operations for reverse cementing. Use the same fluids, volumes,
and rates as for the conventional simulation. The simulation should include original
fluid displaced from the drill pipe above the crossover after any pre‐job mud circulation.
The reverse operations should mirror the conventional operations as seen from the
crossover point.
Again, operations may be split into smaller volumes for convenience. An example table
with pumping operations is shown below; the parenthetical comments for the
“conventional” simulation refer to fluid front locations at the end of the operation.
26
Table 1: Example operations
The fluid temperatures at the crossover from the “conventional” simulation are used as the
injection temperatures for the reverse simulation. Tubing‐side fluid pressures can similarly be
used for injection pressure. Annular‐side fluid pressures just above the crossover should be
input as back pressure. Alternatively, hydrostatic pressure in the annulus above the crossover
can be used as an approximation if friction pressure is not significant. If required, additional
operations for shut‐ins simulating wait‐on‐cement can be added. Outputs from the reverse
simulation should include fluid temperature and pressure profiles in the annulus.
Interpretation: Using the results from both simulations, a data table can be constructed. An
example is given below in Table 2:
27
Table 2: Example Temperature and Pressure Data for Deepwater RCPC Workaround
First Sack Last Sack
Time Temperature Pressure Time Temperature Pressure
Description
[min.] [°F] [psi] [min.] [°F] [psi]
As defined above, the results at the end of each operation yielded data points with the fluid
fronts at relevant locations. The “Injection”, “@ Mud Line”, and “@ Crossover” data points
were taken from the “conventional” simulation for above the crossover. The “@ End of Job”
and “WOC” data points were taken from the “reverse” simulation for below the crossover.
Similarly, a data table can be constructed using annular fluid pressures from the “reverse”
simulation to analyze circulating pressures. In the example below (Table 3), only the pressures
at the previous shoe and the casing shoe were analyzed. However, any depth of interest can be
chosen.
28
Table 3:: Example Data ffor Deepwater R
RCPC Workarou
und
Fluid Pressuree Fluid Preessure
“Co
onventional” Sim
mulation Reverse
e Simulation Time
@ Previous Sho oe @ Shoe
(Above Crosso
over) (Below
w Crossover) [min.]
[psi] [psii]
Pre‐job Circulaation Pre‐job
b Circulation 0.00 13163 175008
Cem
ment (Spacer @ Mudline) Mud 31.60 11092 152007
ment (Cement @ Mudline)
Cem Mud 59.42 11091 152006
Cem
ment (Spacer @ C
Crossover) Mud 167.47 11090 152005
Ceme
ent (Cement @ Crossover) Spacer
S 407.47 11076 152003
Cement (Remainder) Ce
ement 647.47 10698 151998
Displacem
ment (Displacement @ Mudline) Ce
ement 887.47 10483 151997
Displaceme
ent (Displaceme
ent @ Crossover) Ce
ement 1127.47 10187 151995
The presssures can be compared d to the fraccture and poore pressurees at the speecified deptths to
analyze d
dynamic welll security (Figure 9).
Figure 9: (a) Circulating P
Pressure at Prevvious Shoe (b) CCirculating Presssures at Casing SShoe
29
Breaking the pumpin ns into smaller steps inc reases the rresolution, b
ng operation but also increeases
the time to perform the simulation.
Bittleston wo‐dimensional temperaature simulaator (1990).112 He began with
n published the first tw
an analysis of one‐dimensional simulators aand concludded that theey are unablle to capture the
complicaated heat traansfer duringg cementingg and provideed a conven nient simplifiications and non‐
dimensio onalization o
of the governing equatio ons. He solvved this equaation with aa finite differrence
method, but any meethod can be used. For laminar floow, he assum
mes that a vvelocity proffile is
known in n advance. W
With minor m modifications, his methood is used to model the ttemperaturee.
The following assumptions were
e made for th
his model:
12
S. H. Bittlesston, SPE‐20448‐M
MS
13
Chabra and d Richardson, 2008
8
14
Nelson and d Guillot, 2006
15
Papanastassiou, 1987
30
The wells of interest are perfectly vertical, with nno eccentric annuli.
n regions without
In w flow
w, such as the casing wall and tthe surroun nding formaation,
coonvective he eat is negleccted.
The governing equation ffor pressure and velocityy is the Naviier‐Stokes eq quation.
The velocity during cem menting wells is assum ed to be u
unidirectionaal in the veertical
direction.
The Navier‐Sttokes equation and the equation off continuity assume fluiids with con nstant
density.
Multi‐phase f
M flow modelin ng is flow‐rate constrainned, rather tthan pressurre constraineed.
Fluid‐fluid intterfaces are
e perfectly sharp
s and m
moves with the averagee velocity oof the
fluids
The existence e of an interfface has a negligible effeect on the reest of the flo
ow.
In COMSOL, the simulation was set up as a multi‐phasee flow probleem to solve for temperature
and presssures simulttaneously. Flow rates were speciffied in the siimulation, w which alloweed for
multiple velocity fie
elds to be calculated th n the locatiion of each fluid
hen combineed based on
interface
e. This iterattive approacch is shown in
Figurre 10: Coupling o
of Equations forr Multiple Fluidss
A more detailed discussion off the govern
ning equati ons, the reescaling metthods, boun
ndary
conditionns and otherr modeling p
parameters ccan be foundd in Appendix B
31
5.2.3 COMSOL Valiidation
To validaate outputs of the CO
OMSOL model, the moddel was alteered to run
n a conventtional
primary cement sim mulation an nd the resuults were ccompared too a simulation made with
commerccial software e. The same e fluids, rheo
ology, forma tion temperratures and pump rates were
used for both simulaations. Bottom‐hole circculating tem mperature (BBHCT) and th he first‐sack/last‐
sack tem
mperatures were comp pared and are
a shown in Figure 111 (a) and (b). Commeercial
simulator results shoow only the ffirst‐sack/lasst‐sack tempperature and
d the COMSOL output shows
the temperatures th hroughout the
t cement depth. Teemperaturess shown are at the en nd of
placement.
Figure 11: (a) Comparisson of BHCT bettween commerccial and COMSO
OL (b) compariso
on of first/last ssack temperaturres
32
Figure
e 12: Well geom
metry representaation in COMSOL
Temperature surface plotts for reversse and conveentional circulation for tthe Figure 12
2
well geom
metry are sh
hown in Figu
ure 13. The arrows in thhe plots indiccate the flow
w path.
Figure 13: Tem
mperature surfaace plots after m
mud circulation ffor conventionaal and reverse circulation
33
fluid in the annulus. In conventional circulattion, less heeat is transfeerred to the fluid on thee way
down than it does in reverse circulation, producing a lower BH HCT. A comparison of the
conventional and revverse BHCT o over time is shown in Figgure 14.
3
Figure 14: B
Bottom‐Hole Circculating Temperrature (BHCT) inn conventional aand reverse circulation
3
Figure 15: T
Temperature in the cement at tthe end of placeement
34
etailed descriiption of this analysis can b
A more de be found in A
Appendix C
Table 4: Maaximum ECDs inn well 1
C
Conventiona
al Reverse
Botttom Hole 14.5 12.6
Casing 1
C
Prrev. Shoe 12.7 13.5
Botttom Hole 14.6 14.2
Casing 2
C
Prrev. Shoe 13.4 14.6
Botttom Hole 14.3 13.2
Casing 3
C
Prrev. Shoe 12.8 14.4
Table 5: Maaximum ECDs inn well 2
Conventional Reverse
e
B
Bottom Holee 38.2 16.1
Casing 1
Prev. Shoe 41.8 16.6
B
Bottom Holee 26.1 15.8
Casing 2
Prev. Shoe 30.2 17.1
B
Bottom Holee 18.1 14.0
Casing 3
Prev. Shoe 18.1 14.3
ECD incre
ease at the previous shoe is an imp
portant phennomena, an nd the reasoons why invoolve a
more th
horough hyddraulic analysis, which
h is summaarized in Section 5.2.6
6. From tthese
35
comparative simulattions, deepw water RCPC C is most e ffective in wells with tight liner gaps,
reduces ECDs at botttom‐hole. A Additional deetails of presssure resultss can be found in Appen ndix A
The pummp pressure,, lift pressure, and back pressure for one con nventional jo
ob are show wn in
Figure 166 (a). The longest portio on of the jobb is the mudd conditionin ng phase, wh here drillingg mud
is circulated. Becausse there is o one fluid thrroughout thee well, the llift pressuree is zero, and d the
pump pressure is rellatively consstant. Once cement is b eing pumpeed, the lift pressure beco omes
negative and drives the pump p pressure belo ow zero, inddicating U‐Tubing if bacck pressure iis not
applied. Once the ce ement reach hes bottom‐hole, the liftt pressure in ncreases draamatically, w which
drives the pump presssure up.
a) b)
Figgure 16: Pump, lift and back pre
essure(well 1, re
egion 2) (a) Connventional placeement (b) reversse placement
5.2.6 Hydraulic An
H nalysis
If the ECD is reduced d at bottom‐hole and in ncreased at the previous shoe, then n there is a point
between those two w where the p pressures in conventionaal and reverse circulatio on are equal. This
16,17
point is kknown as thhe critical de
epth or the ccritical zone . When designing aa well for revverse
circulatio
on it is important to enssure that the
e weakest paart of the fo ormation is b below this crritical
depth in order for RCCPC to be efffective. A m
more detailedd discussion can be foun nd in Append dix A
16
E. Kuru, S. SSeatter, 2005
17
R. Moore e et al., 2005
36
5.3 Cementing
C g Materialss
Cementin ng material consideratioons were studied to ideentify potential design aand perform mance
benefits and issues b based on cem ment performance undeer deepwateer conditionss when placeed by
RCPC. Consideratio ons included fluid inte
ermixing, keey aspects of spacer design, add
ditive
sensitivitty, and wait‐‐on‐cement (WOC) time.
In conveentional plaacement, most of the slurry pum mped is exp posed to thhe bottom hole
circulatin
ng temperatture (BHCT) near the sh hoe. The sluurry is pump
ped down thhe casing annd up
around in nto the annu ulus until the
e desired annular heightt is reached,, and retardiing additivess are
added soo that the sllurry remain
ns pumpable e during plaacement. The leading ed dge of slurryy will
have eno ough retarding additive tto remain pu umpable forr the duratio on of placemment plus a ssafety
factor at BHCT or att the maximum circulatiing temperaature in the annulus. A As a result o of this
necessarry retarder d dosage and ccooler formaation temperratures high her up in the annulus, the top
of cemen nt (TOC) will be slower to o set and gaain compresssive strength h after placeement.
Dependin ng on string size and depth, placement by RCPC will often result in a diffeerent
temperatture profile from conventional place ement due tto the modiffied flow path. With revverse
placement, slurry ne ear the shoe e and bottom m of the annnulus is exposed to the BHCT, while the
slurry neear the TOC is exposed tto a lower ttemperaturee. Deeper sttrings with rrelatively sm maller
37
volumes of cement pumped will have less of a tempeerature diffeerence betw ween reversee and
conventional placem ment. Slurrie es used in sh
hallow to m mid‐depth weell strings w with large annular
volumes or longer columns
c of cement
c are exposed too higher tem mperatures in general during
placement, aas well as a larger formaation tempe rature differrence betweeen the shoee and
reverse p
the TOC. These diifferences affects
a both required tthickening ttime and W WOC/compreessive
strength developme ent. Based on literaturre review off past land‐‐based RCPC C jobs, adeq quate
compressive strengtth developm ment at cooller temperaatures near the top of the annuluss is a
concern.
An exam
mple of a reeverse temp perature schhedule com pared to th he conventional schedu ule is
shown in
n Figure 17. These temp peratures were
w used too design thee slurries forr the 16” lin
ner in
generic w
well #2, with h the TOC 50 00’ below the e previous caasing in this example.
Figure 17
7: Conventional and reverse tem
mperature scheddules (Well Scheematic #2, 16” liiner)
In this example, RCPPC placement has resulted in a higgher maximmum temperrature (120⁰⁰F for
reverse compared to 100⁰F for conven ntional), a significantlyy shorter placement time
(Approximately 2:45 h:min shortter for reverrse placeme nt), and com mparable WO OC temperatures
for the top
t and bottom of the cement ann nulus. The pplacement ttimes for these temperature
simulatio ons were based on fluid volumes an nd rates, andd do not incclude tool op peration tim
mes or
safety factors. Due to the chan
nge in BHCTT and placemment time, a shorter th
hickening tim
me is
required for the reve erse cement design (Figuure 18).
38
omparison (Welll Schematic #2, 16” liner)
Figure 18: Thicckening Time co
Simulatio
ons with dee eper strings with relative
ely smaller vvolumes of ccement pum mped had less of a
temperatture difference betwee en reverse and
a convenntional placeement. Figure 19 showss the
temperatture estimattes of a 7 5/8” at 27,500 0’ MD. The first sack an
nd last sack ttemperaturees for
reverse p
placement aare exposed to a compaarable tempeerature as cconventional placementt, and
WOC temmperatures aat the TOC and shoe are e similar. Durring placemeent the maxximum down nhole
temperatture for the
e conventional slurry was
w 204⁰F, ccompared tto a maximu um of 210⁰⁰F for
reverse p
placement w with a simulaated placeme ent timed diifference of about 10 miinutes.
Figure 19: Conventional an
nd reverse temp
perature scheduules (Well Schem
matic #1, 7 5/8” liner)
Cite figure 20 The temperature
t e of the shoe after revverse placement is alm
most identiccal to
conventional, and compressive strength de evelopment is comparaable. The teemperature after
placement varies at the TOC and d in this casse the reversse cement d design devellops compreessive
strength before the e conventio
onal cementt design, w which can bbe attributed to the h
higher
39
temperatture at the end of placeement and more retardder in the co
onventional cement to meet
the longe
er thickeningg time requirement.
If there iss a significan
nt differencee between th he static tem
mperatures o of the top an nd bottom o of the
cement column, add ditive stagin
ng with a de
ecrease in rretarders in
n near the ttop of cemeent is
possible to optimize compressive e strength developmentt.
Figure
e 20: Comparison of compressivve strength deve
elopment duringg WOC time afteer placement (W
Well Schematic ##)
Overall typical
t RCPC C slurry design modificaations are tthat higher concentrations of retarding
additive will be requ uired for the slurry located aroundd the shoe and lower aannular secttions.
Retarderr can be redu uced when sslurry will be e placed neaar the top o of the annulu us retarder w when
the place ement time and temperratures will b be reduced ccompared to o a conventiional job. Staging
the retarrder in this way can re esult in an environmenntal and eco onomic ben nefit through h the
reduction n in the am
mount of add ditives usedd. Also, by vvarying and staging the additives in
n the
slurry, th
he entire co olumn of ce ement can be b designedd to set and d gain comppressive streength
simultaneously.
Another consideratio on for RCPC C design, thaat is also siggnificant in cconventionaal slurry design is
the gel strength development of slurries aftter placeme nt. The tran nsition timee is considered to
be the tim me that the gel strength h develops frrom 100 lbf//100 ft to 5000 lbf/100 ftt2. At 500 lbf/100
2
ft2 gases are conside ered to be no longer able to migraate through h cement slu
urry. Duringg this
time the e slurry is neither a liquid nor a solid
s and hyydrostatic pressure tran nsmitted byy that
column of cement will decline e as the cement deveelops gel strength and d becomes self‐
40
supportin ng. If the hyydrostatic prressure of th he cement ccolumn drop ps below thee formation pore
pressure, gas from the formatio on may ente er the slurry and create pathways fo or gas migraation.
To reducce the risk of
o gas migration during RCPC, any sslurry used should be d designed witth an
adequate e transition time. If multiple
m slurrries system
ms with stagged additivees are used,, it is
o verify that the slurry aat the TOC does not bec ome self‐supporting before the slurrry at
critical to
the shoe e since this w will remove hydrostatic pressure froom the shoee slurry and increase thee risk
of gas miigration at th he shoe.
Free Fluid, fluid loss and stability are expected to be sim
milar to req quirements oor recommeended
best pracctices for coonventional cementing d designs. As with critical cement job
bs such as liiners,
fluid losss control is e
expected to be an impoortant designn parameterr, especially with the leaading
edge of tthe cement slurry. This leading edgge will be thee shoe cemeent and will have the lon ngest
exposure e to the formmation during placementt.
5.3.2 Sp
pacers
Spacer m
materials and d basic design methodss for RCPC w will be similaar to those for conventtional
cementinng. One maajor difference is with design criteriia for densitty due to the change in flow
path; cem
ment and sp pacer system ms could be placed at thhe same den nsity as the mud, or even at
decreasin
ng densitiess as long as a rheologiccal hierarchyy is maintainned. In ord
der to effecttively
design sp
pacers to bee placed in re eversed circulation this rheological hierarchy sh hould be a m major
consideraation in the spacer desiggn, along witth criteria foor mud remo oval.
41
effect of rheological and density hierarchies on RCPC cement performance was investigated
through small‐scale laboratory simulations in Phase 1. In Phase 2, a similar experimental setup
was used to evaluate the effectiveness of a viscous pill to separate and displace other fluids as
an alternative to mechanical separators.
The objective of a viscous pill in RCPC is to minimize intermixing between spacer, mud, or
cement slurry during placement. The placement of a viscous pill can be used at the leading or
trailing edge of the cement or spacer to achieve plug flow. Considerations for design are
downhole stability, rheology, pumpability, and compatibility with mud, spacer and cements.
Solids such as fibers or weighting agents can be used, however impact on fluid ECDs and flow
areas of downhole tools should be taken into consideration since excessive solids increase risk
of bridging. Secondary considerations for a viscous pill are the ability to aid mud removal off of
the casing or formation wall and post‐placement stability. After placement, a viscous pill
pumped ahead of the cement slurry will end up inside the casing shoe track. An additional
benefit of a viscous pill remaining stable above cement in the shoe track is to prevent static
fluid interchange, which can affect shoe integrity.
Materials used in viscous pills for lost‐circulation, sweeps for hole cleaning, or reactive pills
used to set cement plugs include natural polymers, polysaccharide, synthetic polymers,
oligomers. Examples of materials that have been used in viscous pills are xanthan gum.
polyether amines, and hydroxyethyl cellulose (HEC). Some viscous pill systems are controlled
by pH, where the crosslinking process occurs only when polymers are exposed to a certain pH
range. In laboratory setup for this project, a viscous pill can be loaded to displace wellbore
fluids, or be displaced. The displacement efficiency and cohesiveness can be evaluated
through visual means. Materials tested in this study were combinations of crosslinked guar,
barite, bentonite, and xanthan gum. The experimental setup is shown in Figure 21. Fluids are
placed in a clear tube and are displaced by either a water or air driven floating piston.
42
p of tube flow exxperiments
Figure 21: Setup
F
Two systtems of fluid
ds were testted: a low‐d density viscoous pill and aa higher density viscous pill.
Propertie
es of the low
w‐density visscous bill arre shown in Table 6. This low‐densiity pill was m
made
from 20p
ppt crosslinkked guar andd was set up to displace water from the tube, ass well as separate
the wate
er from a sepparate fluid located beloow the pill.
Table 6: Low Den
T nsity Viscous Pil l Test Fluids
Fluid Density Yp
Water Spa
acer 8.3 ppg 2
Viscous piill 8.5 ppg 6
Tail fluid 8.9 ppg 2
43
Figu
ure 22: Light weiight viscous pill density of dischharged fluid and
d photos of test
Taable 7: High Den
nsity Viscous Pilll Test Fluids
Tail cemen
nt 15.8 ppg 26 143
44
Figure 23: Heavvy weight viscouus pill test
Figure 24: Density and corrresponding sam
mples of discharrged fluid for higgh density viscous pill
45
Figure 25
5: Rheology of d
discharged samp
ple for heavy deensity and high vviscosity viscouss pill
Fluid testting through
h large‐scale
e horizontal tube was allso run to evvaluate the ability of visscous
pills to separate andd displace flu
uids while b
being pumpeed laterally. In this testt a 10’ long clear
tube wass set up. Onne end was tied into a ppump and thhe other end discharged. The follo owing
fluids weere pumped in successio on and are shown in Figgure 26 throough Figure 336: 8.3 ppg fresh
water dyyed blue, thee 8.5 ppg low
w‐density visscous pill dessign, and a 88.9 ppg brinee dyed red.
Figure 26: FFresh water flow
w
Figure 27: In
nterface betwee
en fresh water ((blue) and viscou
us pill (white)
46
Figure 28: V
Viscous pill flow displaced fresh water
Figure 29: In
nterface betwee
en viscous pill (w
white) and brine
e (red)
Figure 30: B
Brine flow displa
aced viscous pill
Figure 31: In
nterface betwee
en brine (red) an
nd viscous pill (w
white)
Figure 32: V
Viscous pill flow
w displaced brine
e
47
Figure 33: In
nterface betwee
en viscous pill (w e (red)
white) and brine
Figure 34: B aced viscous pill
Brine flow displa
Figure 35: In
nterface betwee
en brine (red) an
nd fresh water ((blue)
Figure 36: FFresh water flow e
w displaced brine
In additioon, the use of simulatio
on software can be usedd in conjuncction with laaboratory daata to
estimate the effect o on displacem ment efficiency when a dding a visccous pill to tthe job proggram.
Future considerationns for studyy include maaterial depeendent sheaar effects on
n the viscou
us pill
downhole as it passses throughh tool openiings and ceentralizers in
n the annulus. To be used
downhole in a deepw water well, any fluid ussed needs too be tested to ensure co ompatibilityy with
the drillin
ng mud, spaacer and cem ment so thatt potential d ownhole isssues from co ontamination can
be avoided. Additional compatibility testingg conductedd with the visscous pill, SB BM, cementt, and
conventional spacerss. Results caan be found in Appendixx E
48
6 Dee
epwater R
RCPC Ope
erational P
Performa
ance
Some special considerations forr operationss will be neccessary during the desiggn, planningg and
execution of a deep
pwater RCPC C job. Overrall, many ooperational processes wwill be similar to
those forr conventionnal cementinng operation ns in deepwaater. Excepttions to thesse consideraations
and proccesses will often be dep pendent on tthe choice ddownhole to ools, specificc well condittions,
and operrator prefereences.
6.1 Operationa
O al and Con
ntingency P
Planning
Operatio
onal considerrations and guidelines fo or the desiggn, planning and executiion of deepwwater
RCPC is located in Appendix F Pre‐job planning inncludes con nsiderations when seleecting
equipmeent, running simulations, designing ffluids, laboraatory testingg, and contingency plan nning.
Equipmeent type and configuratio on should be e selected bbased on opeerator and rregulatory needs,
and what is most ap ppropriate fo or the expeccted well coonditions. TTypical down nhole equipment
for deepwater RCPC are centraliizers, a crossover tool too divert flow w into the annulus, and float
equipmeent. Other ppossibilities ffor use are port collars,, dart catcheers or a ream
mer shoe. O
Other
areas add
dressed include logisticss, maintenan nce, trainingg, HS&E, and personnel.
On‐site o
operational p planning inccludes attenttion for holee conditionin ng and contingency plan nning
for pre‐jo
ob losses or cuttings bloocking down nhole flow. Other pre‐jo ob consideraations coverred in
Appendixx F are operations to ru
un equipme ent in hole, potential w
well control ssituations, rrig‐up
and gene eral job proccedures. Coonsiderations during job execution in nclude down nhole equipment
consideraations, pipee movemen nt, mixing and
a pumpinng operations, and maajor continggency
situations that may o occur duringg cementing operations.
18
Mackay, Pe ogy Digest, March 2003
etroleum Technolo
49
under pressured reservoirs. The current limitations is that this is a specialty technique that has
only been tested on land on a small number of pilot wells, and would need to be validated for
deepwater as well.
The following reports, technical publications, and presentations have related project progress,
technology and findings to the industry.
PRESENTATIONS:
50
PUBLICATIONS:
Wreden, C., J.T. Watters, R. Giroux, Weatherford; M. Nikolaou, K. Macfarlan, D.A.
Richardson. “Deepwater Reverse‐Circulation Primary Cementing: Applicability and
Technical Path Forward for Implementation.” Paper OTC 25194‐MS presented at the
2014 Offshore Technology Conference, May 5 ‐8. Houston, TX 2014.
Wreden, Crystal, Kyle Macfarlan, Richard Giroux, and Michael LoGiudice. “Reliability is
key to developing deepwater RCPC.” Hart’s E&P March 2014.
Wreden, Crystal. “RCPC Holds Promise for Deep Water.” The American Oil & Gas
Reporter Feb. 2013.
Deepwater Reverse‐Circulation Primary Cementing Phase I Interim Report
51
References
Badalamenti, Anthony, et al. Casing Shoes and Methods of Reverse‐Circulation Cementing of Casing.
United States of America: Patent US 7,503,399 B2. 17 March 2009.
Badalamenti, Anthony, et al. Cementing Methods and Systems for Initiating Fluid Flow with Reduced
Pumping Pressure. United States of America: Patent US 7,252,147. 7 August 2007a.
Badalamenti, Anthony, et al. Methods and Systems for Reverse Circulation Cementing in Subterranean
Formations. United States of America: Patent US 7,303,008 B2. 4 December 2007b.
Badalamenti, Anthony, et al. Methods for Reverse‐Circulation Cementing in Subterranean Formations.
United States of America: Patent US 7,401,646 B2. 22 July 2008.
Bergman, Theodore L., Adrienne S. Lavine, Frank P. Incropera, and David P. DeWitt.
Fundamentals of Heat and Mass Transfer. 6th ed. John Wiley & Sons. 2006.
Beirute, Robert and John Kearns. Reverse Circulation Float Equipment Tool and Process. United States of
America: Patent 5,890,538. 6 April 1999.
Bird, R Byron, Warren E Stewart, and Edwin N Lightfoot. Transport Phenomena. 2nd ed. New York:
John Wiley & Sons. 2002
Bittleston, S.H. 1990. “A Two‐Dimensional Simulator To Predict Circulating Temperatures
During Cementing Operations.” MS 20448 Prepared for the SPE Annual Technical Conference
and Exhibition, 23 ‐ 26 September 1990.
Bour, Daniel and Rafael Hernandez. "CO2 Resistance, Improved Mechanical Durability, and Successful
Placement in a Problematic Lost‐Circulation Interval Achieved: Reverse Circulation of Foamed
Calcium Aluminate Cement in a Geothermal Well." Geothermal Resources Council Transactions
2003: 163‐167. October 12 ‐ 15
Bour, Daniel L. Reverse Circulation Cementing System and Method. United States of America: Patent US
6,920,929 B2. 26 July 2005.
Branch, Alton, Donny Winsolow and Almee Greening. Selectively Activated Float Equipment. United
States of America: Patent US 7,644,774 B2. 12 January 2010.
Buckingham, E. “On Plastic Flow Through Capillary Tubes.” ASTM Proceedings 21: 1154–56. 1921.
Chabra, R P, and J F Richardson. Non‐Newtonian Flow and Applied Rheology Engineering
Applications. 2nd ed. San Diego: Elsevier Science & Technology Books. 2008.
“COMSOL Multiphysics.” 2012. Stockholm, Sweeden. www.comsol.com.
52
Chase, Stephen and Gary Maler. Reverse‐Circulation Cementing of Surface Casing. United States of
America: Patent US 8,162,047 B2. 24 April 2012.
Chen, Zhongming and Rudolf J. Novotny. “Accurate Prediction Wellbore Transient Temperature Profile
Under Multiple Temperature Gradients: Finite Difference Approach and Case History”. In:
Society of Petroleum Engineers. 2003.
Colvard, R. Kee and Michael LoGiudice. Methods and appartus for actuating a downhole tool. United
States of America: Patent US 7252152 B2. 18 June 2003.
Davies, Jason, et al. "Reverse Circulation of Primary Cementing Jobs ‐ Evaluation and Case History."
IADC/SPE 87197 presented at the IADC/SPE Drilling Conference, Dallas, Texas, 2‐4 March 2004.
Deen, William M. Analysis of Transport Phenomena. 2nd ed. New York: Oxford University Press. 2012.
Edgar, Mike and Greg Horton. Reverse Cementing Float Shoe. United States of America: Patent US
6,802,374 B2. 12 October 2004.
Fredrickson, Arnold G, and R Byron Bird. “Non‐Newtonian Flow in Annuli.” Ind. and Eng. Chem. 50
(3), pages 347‐352. 1958.
Gaddy, D.E. "Burlington Reverse Circulates Cement, Sets World Record." Oil & Gas Journal 98.16 (2000):
pages 58‐59. <http://www.ogj.com/content/ogj/en/articles/print/volume‐98/issue‐16/in‐
thisissue/drilling/burlington‐reverse‐circulates‐cement‐sets‐world‐record.html>.
George, Matthew and Ramadan Ahmed; Rheological Properties of Fiber‐Containing Drilling Sweeps at
Ambient and Elevated Temperatues, American Association of Drilling Engineers, 2011, AADE‐14‐
NTCE‐35.
Giroux, Richard Lee and Steven M Rosenberg. "Stage Cementing Methods Used in Casing While Drilling."
United States Patent Application Publication US 2007/0261850 A1. 15 November 2007.
Griffith, J.E., D.Q. Nix and G.A. Boe. "Reverse Circulation of Cement on Primary Jobs Increases Cement
Column Height Across Weak Formations." SPE 25440 presented at the Production Operations
Symposium, Oklahoma City, Oklahoma, USA, 21 ‐ 23 March 1993.
Griffith, James and Caleb Carroll. Method and Means to Seal the Casing‐by‐Casing Annulus at the
Surface for Reverse Circulation Cement Jobs. 1 July 2008.
Griffith, James, et al. Reverse Circulation Cementing Process. United States of America: Patent US
7,013,971 B2. 21 March 2006.
Grinrod, M., et. al. " RFID. A Key Enabler in Drilling Atomation" SPE‐163449‐MS. Presented at the
SPE/IADC Drilling Conference 5 ‐ 7 March 2013. Amsterdam, The Netherlands.
53
Halliburton. Operator Used Reverse Circulation Placement Method for Foamed Variseal™ Cementing
System. n.d. 12 July 2012.
<http://www.halliburton.com/public/cem/contents/Case_Histories/web/evc/EVC_4976.asp>.
Heathman, James, Method of Setting a Balanced Cement Plug in a Borehole. United States of America:
Patent US 5,368,103.A 29 November 1994.
Hemphill, Terry, Wellington Campos, and Ali Pilehvari. “Yield‐Power Law Model More Accurately
Predicts Mud Rheology.” Oil and Gas Journal 91 (34): 45. 1993.
Hepburn, B. and B. Beekman. "Reverse Circulating Cement Technique for Subsurface Pipeline River
Crossing." Paper 2005‐265 presented at the Petroleum Society's 6th Canadian International
Petroleum Conference, Calgary, Alberta, Canada, 7 ‐ 9 June 2005.
Hernandez, Rafael. “Unconventional Cementing Method Enables Drilling in Lost‐Circulation Zones:
Case History.” MS. 154564 In SPE Asia Pacific Oil and Gas Conference. Perth, Australia: Society
of Petroleum Engineers. 2012.
Hernandez, Rafael and Daniel Bour. "Reverse‐Circulation Method and Durable Cements Provide Effective
Well Construction: A Proven Technology." Proceedings, Thirty‐Fifth Workshop on Geothermal
Reservoir Engineering. Stanford, California, 1 ‐ 3 February 2010.
Hernandez, Rafael and Hao Nguyen. Foamed Latex Slurry and Reverse‐Circulation Cementing Enable
Drilling in Lost‐Circulation Zones. n.d.
<http://www.halliburton.com/public/solutions/contents/Geothermal/related_docs/Foamed%2
0Latex%20Slurry_paper.pdf>.
Houwen, OH, and T Geehan. “Rheology of Oil‐Base Muds.” SPE Annual Technical Conference and
Exhibition. 1986. http://www.onepetro.org/mslib/servlet/onepetropreview?id=00015416.
Kuru, E. and S. Seatter. "Reverse Circulation Placement Technique vs. Conventional Placement
Technique: A comparative Study of Cement Job Hydraulics Design." Journal of Canadian
Petroleum Technology 44.7 (2005): 16‐19.
Iain A., “Radio Frequency Identification (RFID) Leads The Way In the Quest For Intervention Free Upper
Completion Installation" SPE‐166182‐MS. Presented at the SPE Annual Technical Conference
and Exhibition 30 September ‐ 2 October 2013.
Leturno, Richard, Clay Bretches and Johnny Shields. "Casing Leak is Repaired Using Reverse Cementing
Procedure." World Oil 1992: 44‐45.
MacEachern, Doug, et al. "Reverse Foam Cementing an Offshore Production Tieback." Paper AADE‐03‐
NTCE‐32 presented at the AADE 2003 National Technology Conference "Practical Solutions for
Drilling Challenges", April 1‐3. Houston, Texas, 2003a.
54
MacEachern, Douglas P., et al. "Advances in Tieback Cementing." Paper SPE/IADC 79907 presented at
the SPE/IADC Drilling Conference, Amsterdam, The Netherlands, 19‐21 February (2003b).
Mackay, P., K2 Energy Corp. “Reverse circulation drilling avoids damage to low‐pressure gas reservoirs.”
Petroleum Technology Digest, Calgary, Canada. March 2003. http://www.worldoil.com/March‐
2003‐Petroleum‐Technology‐Digest‐Reverse‐circulation‐drilling‐avoids‐damage‐to‐low‐pressure‐
gas‐reservoirs.html
Mariott, Tim, Steve Chase and Isaac Khallad. "Reverse‐Circulation Cementing to Seal a Tight Liner Lap."
Paper OTC 18839 presemnted Offshore Technology Conference, Houston, Texas, USA, 30 April ‐ 3
May 2007.
Marquairi, R. and J. Brisac. "Primary Cementing by Reverse Circulation Solves Critical Problem the North
Hassi‐Messaoud Field, Algeria." Journal of Petroleum Technology 1965: 146‐150.
McNicol, J., "New Applications of Cementing Value Technology for Gulf of Mexico Deepwater
Operations" SPE‐168048‐MS. Presented at the IADC/SPE Drilling Conference and Exhibition,
4 ‐ 6 March 2014, Fort Worth, Texas.
Mendes, Paulo R Souza, Eduardo S S Dutra, and Rio De Janeiro. “A Viscosity Function for
Viscoplastic Liquids.” Annual Transactions of the Nordic Rheology Society 12: 183–88. 2004.
Mirakyan, Audrey, Triggered Polymer Viscous Pill and Methods of Using the Same. United States
America: Patent US 8,579,030 B2
Moore, Robert D, Daniel L Bour, Shawn Reed, and Rafael Hernandez. 2005. “High‐Temperature Wells
with Lost‐Circulation Demands and Reverse Circulation Placement Technique Using Foamed
Cement Systems: Two Case Histories.” 84563 ‐ PA SPE Drilling & Completions. Society of
Petroleum Engineers.
Nelson, Erik B., and Dominique Guillot.. Well Cementing. 2nd ed. Sugar Land: Schlumberger. 2006
Papanastasiou, Tasos C. “Flow of Materials With Yield.” Journal of Rheology 31: 385–404. 1987
Phillip M. S., et al. "RFID Actuatuin of Self‐Powere downhole Tools" SPE‐113842‐MS. Presented at the
SPE/ICoTA Coiled Tubing and Well Intervention Conference and Echibition, April 1‐2, 2008, The
Woodslands, Texas
Pleasants, Charles W; Basil J. Joseph; and Amy L. Farrar; Baker Hughes. “Reliable Completions Systems
for Gas Migration Prevention.” SPE 160217. Presented at the SPE Annual Technical Conference
and Exhibition, San Antonio, Texas, USA, 8‐10 October 2012.
Prebensen, Ole Lacob et al., Fluid Pressure Transmission. Patent CA2682949C. 29 November 1994.
55
Raghuraman, B., et al. "Real‐Time Downhole pH Measurement Using Optical Spectroscopy" SPE‐93057‐
MS. Presented at the SPE International Symposium on Oilfield Chemistry, 2‐4 February, The
Woodlands, Texas.
Renpu, W. Advanced Well Completion Engineering. 2011.
Rogers, Henry and Earl Webb. Apparatus and Method for Reverse Circulation Cementing a Casing in an
Open‐Hole Wellbore. United States of America: Patent US 7,290,612 B2. 6 November 2007.
Rogers, Henry , et al. Removable Surface Pack‐Off Device for Reverse Cementing Applications. United
States Patent Application Publication US 2005/0183857 A1. 25 August 2005.
Sharqawy, Mostafa H, John H Lienhard V, and Syed M Zubair. “Thermophysical Properties of
Seawater : A Review of Existing Correlations and Data.” Desalination and Water Treatment 16:
354–80. http://web.mit.edu/seawater/. 2010.
Shaughnessy, John and John Helweg. "Optimizing HTHP Cementing Operations." Paper IADC/SPE 74483
presented at the IADC/SPE Drilling Conference, 26‐28 February 2002.
Spielman, Paul, Rafael Hernandez and Hao Nguyen. "Reverse Circulation of Foamed Cement in
Geothermal Wells." Geothermal Resources Council Transactions 10‐13 September 2006: 303‐
308.
Stalford, Harold, Ramadan Ahmed, Victor Hugo Soriano Arambulo; “Intelligent Casing‐Intelligent
Formation (ICIF) Design.” OTC 25161‐MS presented at the Offshore Technical Conference in
Houston, Texas, USA, 5‐8 May 2014.
Sullaway, Bobby and David Szarka. Reverse‐Cementing Method and Apparatus. United States of
America: Patent US 6,244,342 B1. 12 June 2001.
Vipulanandan, C ., and M. Heidari, CIGMAT‐University of Houston; Q.Qu, Baker Hughes; H. Farzam,
Cemex; PappasJ. M, RPSEA. “Behavior of Piezoresistive Smart Cement Contaminated with Oil
Based Drilling Mud.” OTC‐25200‐MS presented at the Offshore Technical Conference in
Houston, Texas, USA, 5‐8 May 2014
Vradis, George C, John Dougher, and Sunil Kumar. “Entrance Pipe Flow and Heat Transfer for a
Bingham Plastic.” International Journal of Heat and Mass Transfer 36 (3): 543–52.
doi:10.1016/0017‐9310(93)80030‐X. 1993.
Webb, Earl D. and Henry E. Rogers. Apparatus for Autofill Deactivation of Float Equipment and Method
of Reverse Cementing. United States of America: Patent US 7,357,181 B2. 15 April 2008.
56
Webb, Earl, Brett Fears and Henry Rogers. "Method and System for Cementing." United States Patent
Application Publication US 2009/0260816. 22 October 2009.
Wilkes, James O, and Stacy G Bike. Fluid Mechanics for Chemical Engineers. Prentice Hall International
Series in the Physical and Chemical Engineering Sciences.
Wreden, Crystal, Michael Nikolaou, Jeffrey Thomas Watters, Kyle Macfarlan, Richard Giroux, and Donald
Alexander Richardson. “Deepwater Reverse‐Circulation Primary Cementing: Applicability
and Technical Path Forward for Implementation.” 25194‐MS. Presented at the Offshore
Technology Conference. Houston, 1993.
Zhu, H., Y.D. Kim, and D. De Kee. 2005. “Non‐Newtonian Fluids with a Yield Stress.” Journal of Non‐
Newtonian Fluid Mechanics 129: 177–81.
57
Appen
ndix A ‐ Alternative
e method ffor perforrming deeepwater
RCP
PC simula ation using
g WELLCAAT
Landmarkk’s WellCat® temperatu ure simulato or was used to study aand model tthe temperaatures
throughout the wellbo ore for RCPC iin deepwaterr wells, and thhe Opticem® placement simulator wass used
to evaluate deepwater liner and caasing strings. The limitat ion of the W Wellcat® softw ware used is tthat a
deepwate er reverse‐cirrculation cem
menting configuration cannnot be appllied. Howevver, a workarround
reverse circulation operation wherre fluids are injected intoo the well thrrough the an
nnulus can bee run.
This workkaround includes running the temperatture simulati on in ‘producction’ mode vvs. ‘drilling’ m mode,
and defining the casin ng as a produ uction tubing.. Once thesee modification ns have been n made, a revverse‐
circulation
n temperaturre simulation can be run.
To validatte using this P
Production m
mode workaro ound, the we ll temperaturres for a convventional 16”” liner
job was ssimulated in b both Producttion and Drilling modes. A A comparison of the geneeral wellboree fluid
temperature profiles ccan be seen inn Figure Apx A A‐1. The maxximum tempeerature differrence betweeen the
annulus flluids in produ
uction and drilling mode iss 3.8⁰F. For tthe fluid in the drill pipe (D
DP) and casing, the
maximum m difference b between prod duction mode e and drilling mode tempeerature is a m maximum of 6.4⁰F.
Overall, the
t temperature profiless in Figure Apx
A A‐1 shoow that tem mperature sim mulations ussing a
workarou und setup in n production
n mode are e comparablee to tempeerature simu ulations run in a
conventioonal drilling m
mode setup.
Figu
ure Apx A‐1: Com
mparison of tem
mperature profiles between dril ling mode and w
workaround pro
oduction mode
A‐1
Interface temperatures of the first sack and last sack of cement were also compared, since these
temperature profiles are used to determine laboratory testing parameters. In drilling mode, the
temperature of these interfaces over time is easily generated. However, these interface temperatures
are not generated in the production mode due to the simulation format. The temperatures provided
are at the top of tubing, mudline, and bottom of tubing at the end of each operation. To determine a
first sack/last sack temperature profile, each general operation (pumping spacer, pumping cement,
displacement) is broken down volumetrically so that at the end of each operation the first or last sack
interface is located at the mudline or bottom of tubing. The general temperature profile vs. depth of
the well is also given at the end of each operation, so that intermediate temperatures points can be
obtained by back calculating the volume and depth. The temperatures are taken at these points are
then plotted to obtain a first sack and last sack temperature profile for the tricked simulation. A
drawback to this method is that the fluid positions need to be manually calculated and entered into
separate operations, so the number of data points collected is dependent on the number of operations
and is limited compared to data generated in drilling mode. A comparison of the first and last sack
interface temperatures between production and drilling mode is shown in Figure Apx A‐2. The first sack
temperature profiles generated in both modes correlate well over time. At the end of the job, the BHCT
in production mode is only 0.5⁰F higher than the BHCT of drill mode. The last sack temperatures follow
the same trend, however the temperature during the job in production mode is underestimated
compared to that in drill mode. This could be attributed to differences in calculation of temperatures in
each mode, or that there is difficulty in matching temperature profile curves due to the limited number
of data points can be collected by the workaround method. At the end of the job the difference in BHCT
between the modes is 1.0⁰F. Overall, the temperature profiles in Figure Apx A‐2 show that temperature
simulations using a workaround setup in production mode are comparable to temperature simulations
run in a conventional drilling mode setup.
A‐2
Figure Ap
px A‐2: Comparisson of First Sackk and Last Sack T
Temperatures bbetween drillingg mode and workaround producction
mode
The main limitation off this programm workaroun nd is that it ccan be directtly applied to
o land‐based RCPC
jobs but not
n to a deep pwater configguration. On
n land‐based jobs, cementing fluids arre injected directly
into the aannulus from the surface. With a deep pwater configguration, fluids are expected to be pumped
down the e drillpipe, through a crosssover tool, th hen into the annulus so th hat cement is injected intto the
annulus below
b the risser BOP stackk. WellCat® and Opticem m® are unablee to model tthis flow pathh and
assume th he flow is down the entire e length of thhe annulus, i ncluding the riser. This liimitation is shared
with othe er commercially available ssimulation so oftware. A coonventional fflow path for a deepwaterr liner
string sim
mulated in Optticem® is show wn in Figure A Apx A‐3a. Thhe reverse plaacement flow w path for the same
liner strin
ng is shown in n Figure Apx A‐3b. While e the softwarre can be con nfigured for rreverse‐circulation,
placemen nt of cementin ng fluids dowwn the riser is unrealistic annd operationaally undesirab ble.
A‐3
Figure Ap
px A‐3: (a.) Convventional Placem
ment Simulationn (b.) Reverse Placement Simulaation
A‐4
Figure Apx A‐4: Conventio
onal (Drilling mode) and Reversse Circulation (PProduction mod
de) First and Lastt Sack Temperattures
Additionaal workarounds within the e program are necessaryy to address this softwaree assumption n that
fluids are injected dire ectly into the annulus from m the rig floorr, even with aan offshore cconfiguration. This
second workaround fo or a commerccial deepwate er RCPC simullation dividess the full conffiguration into two
separate simulations using both drilling
d and production
p m workaround assumes thaat the
mode. This w
temperature profile off the fluids fo or deepwater RCPC are thee same as forr a conventional job from initial
injection aat the rig floo
or until the flu
uids reach the e crossover ppoint, since a w will be used in
work string w n both
cases. Initial temperaature simulatiions assumed d that the croossover pointt would be lo ocated at thee liner
hanger. To
T obtain a temperature
t profile after the crossoveer, a production mode sim
mulation is broken
down intto smaller, sequential
s operations. The data outtput from th hese simulattions and sm maller
operationns is collected and com mbined to obtain
o an eestimate of deepwater RCPC placeement
temperatures.
Figu
ure Apx A‐5: Sim
mulated Reverse‐‐Circulation Tem
mperatures usin
ng Workaround
A‐5
The first aand last sack ttemperature profile estim mates prepareed with this w workaround are shown in FFigure
Apx A‐5, ccompared to conventionall and the initiial productionn mode workkaround. In th he example sshown
the first sack
s cement temperature mpared to 94⁰F for
e at the end of placementt is 115⁰F in reverse com
conventio onal. A comparison of the general wellbore fluid circculating temp perature proffiles can be seeen in
Figure Appx A‐6. As with the first/llast sack tem
mperature proofiles the fluid temperatures during reeverse
circulation n placement are higher than conventio onal placemennt. Reverse B BHCT and the max BHT is 1 115⁰F,
compared d to a BHCT of 88⁰F and a m max BHT of 1 103⁰F during cconventional placement.
In generaal, the simuulated deepw water RCPC temperaturee estimates a higher BH HCT compareed to
conventioonal simulatio ons. This can be attrib
buted to thee modified fflow path; in
n the annulu
us the
cementing fluids are in n contact with h the formation allowing ffor direct heaat transfer du uring placemeent. A
limitation of this worrkaround is that
t the tem
mperature of the return fflow in the simulated reeverse
temperature profile is underestimaated from the e crossover baack up to the rig floor. Ho owever, cemeenting
fluids will not be in thhis section du uring placeme ent so this is not expected d to have an effect on thee first
and last sack temperatture estimate es for cement design.
px A‐6: General Fluid Temperatuure vs Depth Profile
Figure Ap
A‐6
Figu
ure Apx A‐7: Sim
mulated Reverse
e and Conventioonal WOC tempeerature profile
Circulating
C te o the mud at the x‐over (Drill string fluid temp aat x‐over MD after
emperature of
circulation)
c
First
F sack/Lasst sack tempe
erature profile. First sackk/last sack intterface temp peratures from
m the
nd
rig
r floor to the crossoverr point are assumed to bbe the same for the 2 ssimulation ass in a
conventional
c simulation.
A‐7
Circulating
C te o 1st sack at
emperature of a x‐over pooint. This w
will be your cement injeection
temperature f
t for the 2nd sim
mulation.
2. Enter fluid properrties and builld a 2nd simu on using the production mode
ulation for re verse injectio
19
workaaround. Asssumptions for this simulattion are:
TThis is a land well, with reverse injectio
on of all fluidss into the ann
nulus at the su
urface.
The ground re
T eference poin
nt is located aat the x‐over so that the xx‐over depth n
now equals 0
0’ MD.
Your MD/TVD
Y D of everythin
ng located bellow the x‐oveer will be adju usted accordingly.
19
“Building a Reverse‐Circulatio
on File Using the W
WELLCAT Software
e for Import Into t he WELLPLAN Sofftware OptiCem M
Module” WELLCAT Help
file.
A‐8
Enter
E erties. Theirr values at thhe ground in the 2nd simu
the folllowing prope ulation will b
be the
same as at the
s e x‐over point in the conve entional simuulation. The values at TD will be equal.
o Formatioon Temperatu ure
o Fracture Gradient
o Pore Pressure Gradien nt
3. Define operations. Production
n mode doess not give fluuid interface temperaturees (first/last sack).
Fluid temperatures are given att the top of tuubing, groundd, and at the bottom of tubing. Each off your
operaations will nee
ed to be set uup so that a ffluid interfacee of interest is at one of tthese points aat the
end oof the operatio
on. Suggesteed operationss are:
d ‐ enter duraation, circulation direction and injection
3.1. Circulate mud
C n temperature
3.2. Reverse Space
R er – enter vollume and inje
ection temperrature
3.3. Reverse Ceme
R ent – enter vo
olume, injectiion temperatture
3.4. Displacement
D t – (if TOC is lo
ower than x‐o
over point) ennter volume, injection tem
mperature
A‐9
mary’ (Bottom
Summ m hole MD, A
Annulus Fluid column). Laast sack final BHCT can bee found undeer the
MD of its final position.
5. Interm
mediate dataa points can be obtained during eachh operation b
by calculatingg fluid positiion in
annulus at the end d of each ope eration, then looking up thhe corresponding circulating temperatu ure at
that depth
d erature’ Table (Annulus TTemperature at fluid MD)). Cementingg and
in the ‘Fluid Tempe
displa
acement operations can b be broken down into sma ller operation to acquire more data points.
Increaasing the nummber of data p points improvves the rever se temperatu ure profile.
6. These
e collected daata points can peratures from the x‐over point
n be plotted aas first/sack laast sack temp
downn. Since First sack/last sack interface te
emperatures ffrom the rig ffloor to the ccrossover poin
nt are
assum
med to be the e same for thee 2nd simulation as in a connventional sim
mulation. Daata from these two
curves can be com mbined for a first/last sack estimate.
WOC temperattures for the first and last sack:
7. For W
7.1. First sack: De
F efine a shut‐in
n operation lasting 24 ho urs that directly follows tthe last placeement
operation. Th
o he temperatu nd of the tubi ng in the Tem
ures at the en mperature vs Time table arre the
first sack WOC
f C temperaturres. Annulus should be se lected under ‘Flow path data selection’.
hut‐in periods that
7.3. If the last sackk is located below the x‐ovver point: Creeate multiple successive sh
total 24 hours
t s (ex: 4 shut‐ins, 6 hour duration each ). The first shut in operattion will follow the
A‐10
last placement operation. WOC temperatures for each time period can be found in the
‘Wellbore Temperature Summary’ table at the last sack MD.
8. For general circulating temperature data that can be imported into Wellplan: Copy the circulating
temperatures down the drill pipe and up the riser annulus from the rig floor to the x‐over from the
conventional simulation. Copy circulating temperatures from the 2nd simulations. Combine the data
so that you have a temperature profile of reverse‐circulation flow. Import this data in to Wellplan as
detailed in the production mode workaround help file1.
An example of this workaround is below:
NOTE: Example is RPSEA, Schematic 2, 13⅝‐in. Casing, with top of tail at 20,000 ft. MD and top of lead at 14,500 ft. MD.
1. Define Fluids
a. Muds and spacers as “standard muds” or “synthetic muds”. “General” can be used as
well.
b. Cements as “standard muds” or as “general”.
c. Note temperature and pressure effects on density and rheology depend on how a fluid
is defined.
2. Define target casing string as drill pipe.
3. Build full well as if for a conventional cementing job.
4. Create operations as “Trip Pipe & Circulate” in DRILL mode using target string as lowest segment
of drill pipe.
a. Casing is not in the hole.
b. Break operations up such that the end of each operation is a relevant time (i.e. cement
at crossover).
c. Set “additional depths” at points of interest, particularly the mud line and crossover.
A‐11
Current Well Schem atic
0.0 ft 0.0 ft
f RKB
0.0 ft Mean Sea Level
5000.0 ft
f Drill Pipe
e: Od = 6 5/8", Id = 5.965"
5300.0
5000.0ft
f ft 36" Line
Mud Condductor Driven
5500.0 ft
f TOL
7000.0 ft
f TOL
8000.0 ft
f 22" Surfa
ace Cas ing
12000.0 ft
f 18" Drilli ng Liner
14000.0 ft
f 16" Interm ediate Liner
Drill Pipe
e: Od = 13 5/8", Id = 12 3/8 "
21000.0 ft
f Hole Sizze = 16 1/2"
A‐12
5. Calculate results. Copy the following infformation forr all operation ns to an Excell spreadsheett:
a. Fluid TTemperature vs. Measured d Depth in thee Tubing
b. Tempe erature vs. Timme in the Tub bing at the m ud line and at the crossovver.
c. Fluid PPressure vs. MMeasured Dep pth in the Tubbing
6. Build a secondd well with the e crossover at ground leveel.
a. Translate depths, p pore pressure e, fracture graadient, and geeothermal teemperature to o new
depth references.
b. Build well as for conventional up to the previous string. Createe target string as
producction tubing.
7. Trranslate operrations from first well to operations ffor the secon onsidering what is
nd well by co
go
oing on below w the crossovver.
a. Use chhecksums for volumes of spacer, lead, aand tail plus o overall job tim
me.
b. Displacement volum me in the reverse scenarioo should be reeduced by the drill pipe vo olume
above the crossover. (Reverse e Displacemeent Volume ++ DP Volume = “Conventiional”
Displacement Volu ume).
"Conventional" Reverse (Below
w XO)
Rate Rate Volume Time Tim
me Rate Rate V
Volume Time Time
O
Operation Operatioon
[b
bbl/min] [gal/m
min] [bbl] [min.] [hr.] [bbl/min] [gal/min] [bbl] [min.] [hr.]
Rev. ‐ Pre‐job
10.0 420.0 10200.00 1020.0 17.00
00000 Pre‐job CCirc. 10.0 420.0 100200.00 1020.0 17.000000
Circ.
Reev. ‐ Spacer 8.0 336.0 100.00 12.5 0.208
8333 Mud (1 oof 2) 8.0 336.0 1100.00 12.5 0.208333
Rev. ‐ Tail (Spc @
8.0 336.0 72.82 9.1 0.151
1715 Mud (2 oof 2) 8.0 336.0 772.82 9.1 0.151715
XO)
R
Rev. ‐ Tail Spacer (11 of
8.0 336.0 11.31 1.4 0.023
3566 8.0 336.0 11.31 1.4 0.023566
((Balance) 2)
A‐13
"Conventional" Reverse (Below XO)
Rate Rate Volume Time Time Rate Rate Volume Time Time
Operation Operation
[bbl/min] [gal/min] [bbl] [min.] [hr.] [bbl/min] [gal/min] [bbl] [min.] [hr.]
Rev. ‐ Lead (Tail Spacer (2 of
8.0 336.0 88.69 11.1 0.184767 8.0 336.0 88.69 11.1 0.184767
@ XO) 2)
Rev. ‐ Lead (Lead
8.0 336.0 84.13 10.5 0.175281 Tail 8.0 336.0 84.13 10.5 0.175281
@ XO)
Rev. ‐ Lead
8.0 336.0 289.92 36.2 0.603997 Lead (1 of 2) 8.0 336.0 289.92 36.2 0.603997
(Balance)
Rev. ‐ Disp (Disp
8.0 336.0 172.82 21.6 0.360048 Lead (2 of 2) 8.0 336.0 172.82 21.6 0.360048
@ XO)
Rev. ‐ Disp (Fast) 8.0 336.0 294.70 36.8 0.613953 Disp Fast 8.0 336.0 294.70 36.8 0.613953
Rev. ‐ Disp (Slow) 4.0 168.0 100.00 25.0 0.416667 Disp Slow 4.0 168.0 100.00 25.0 0.416667
8. Create operations as “Circulation” with “Production Tubing” in PROD mode.
a. Fluids and volumes/times based on translation of operations.
b. Injection temperatures taken from results of previous simulation. If DRILL operations
were defined correctly, then the outputs should directly yield the required
temperatures.
c. Choke pressure can be defined as the hydrostatic pressure in the annulus above the
crossover.
d. For lead and tail, or for top of cement below the crossover, define a wait‐on‐cement
shut‐in in multiple steps.
Current Well Schem atic
2000.0 ft TOL
A‐14
9. Calculate the rresults. Copy the followingg informationn for all operaations to an EExcel spreadsh
heet.
a. Fluid TTemperature vs. Measured d Depth in thee Annulus.
A‐15
b. Temperature vs. Time in the Annulus at End of Tubing.
c. Temperature vs. Time in the Annulus at Top of Tubing.
d. Fluid Pressure vs. Measured Depth in the Annulus.
10. Using the information exported to Excel and the definitions of the operations, you can
determine the following:
a. First (“Conventional”) operations:
i. Fluid time, temperature, and pressure at mudline.
ii. Fluid time, temperature, and pressure at crossover.
b. Second (“Reverse”) operations:
i. Fluid time, temperature, and pressure at end of placement.
ii. Fluid temperature at various times during wait‐on‐cement.
11. Build a table with the exported information to determine thickening time and compressive
strength temperature and pressure schedules.
“Tail” First Sack
Elapsed Time Temperature Pressure
Description
[min.] [degF] [psi]
Injection 12.50 80.00 1882.54
@ Mud Line / Crossover 34.10 72.45 6431.01
@ End of Job 214.87 146.83 13642.67
WOC (1 of 6) 454.87 158.28 14031.61
WOC (2 of 6) 694.87 163.49 14031.16
WOC (3 of 6) 934.87 167.13 14030.87
WOC (4 of 6) 1174.87 169.97 14030.66
WOC (5 of 6) 1414.87 172.28 14030.49
WOC (6 of 6) 1654.87 174.20 14030.35
“Tail” Last Sack / “Lead” First Sack
Elapsed Time Temperature Pressure
Description
[min.] [degF] [psi]
Injection 23.02 80.00 3417.76
@ Mud Line / Crossover 44.62 72.74 6431.01
@ End of Job 164.30 143.95 12995.59
WOC (1 of 6) 404.30 155.49 13382.26
WOC (2 of 6) 644.30 160.51 13381.81
WOC (3 of 6) 884.30 163.93 13381.52
WOC (4 of 6) 1124.30 166.55 13381.31
WOC (5 of 6) 1364.30 168.66 13381.14
WOC (6 of 6) 1604.30 170.40 13381.00
“Lead” Last Sack
Elapsed Time Temperature Pressure
Description
[min.] [degF] [psi]
Injection 80.86 80.00 2603.76
@ Mud Line / Crossover 102.46 73.24 4379.10
@ End of Job 164.30 117.61 9139.31
WOC (1 of 6) 404.30 134.32 9451.29
WOC (2 of 6) 644.30 138.32 9451.29
WOC (3 of 6) 884.30 140.63 9451.29
WOC (4 of 6) 1124.30 142.31 9451.29
WOC (5 of 6) 1364.30 143.61 9451.29
WOC (6 of 6) 1604.30 144.66 9451.29
A‐16
Schematic 2
13⅝‐in. Casing
Top of Lead = 14,500 ft. Top of Tail = 20,000 ft.
Placement Temperatures & Pressure vs. Time
250 20000
225 18000
200 16000
175 14000
Temperature [degF]
150 12000
Pressure [psi]
125 10000
100 8000
75 6000
50 4000
25 2000
0 0
0 30 60 90 120 150 180
Elapsed Time [min.]
"Tail" First Sack Temperature "Tail" Last Sack / "Lead" First Sack Temperature
"Lead" Last Sack Temperature "Tail" First Sack Pressure
"Tail" Last Sack / "Lead" First Sack Pressure "Lead" Last Sack Pressure
12. Build a table using the Fluid Pressure vs. Measured Depth in the Annulus information from the
second (“reverse”) simulation and the pore and fracture pressures to monitor dynamic stability.
@ Previous Shoe @ Shoe
Operation Elapsed Time
Pore Fracture Pressure Pore Fracture Pressure
A‐17
Rev. ‐ Disp (Disp @ XO) Lead (2 of 2) 102.46 8153.60 11138.40 9126.68 13104.00 17472.00 13644.85
Schematic 2 Schematic 2
13⅝‐in. Casing 13⅝‐in. Casing
Top of Lead = 14,500 ft. Top of Tail = 20,000 ft. Top of Lead = 14,500 ft. Top of Tail = 20,000 ft.
Circulating Pressure at Previous Shoe vs. Time Circulating Pressure at Shoe vs. Time
12500 20000
12000 19000
11500 18000
11000 17000
10500 16000
Pressure [psi]
Pressure [psi]
10000 15000
9500 14000
9000 13000
8500 12000
8000 11000
7500 10000
0 30 60 90 120 150 180 0 30 60 90 120 150 180
Elapsed Time [min.] Elapsed Time [min.]
Known limitations:
Injection temperature is constant for PROD circulation operations. This can affect the results of
the pre‐job circulation particularly.
Cannot see ECD in PROD mode. Properly translated pore pressure and fracture gradient
combined with fluid pressures in annulus will give good approximation of ECD in terms of
circulating pressure (but not equivalent density).
Cannot determine displacement efficiency and fluid inter‐mixing.
A‐18
Appendix B ‐ COMSOL Governing Equations, rescaling methods,
boundary conditions and other modeling parameters
The following is an excerpt from Chapter 2 of: Macfarlan, K. (2014) "Reverse Circulation Primary
Cementing of Offshore Oil wells" Unpublished Master’s Thesis. The University of Houston.
Governing Equations
The wells of interest are perfectly vertical, with no eccentric annuli. A cylindrical coordinate
system with axial symmetry is therefore perfectly suited to this problem. The governing equation for
temperature in the heat equation with a convective term,
T
C p v T k 2T , (1)
t
where is the density, C p is the heat capacity, k is the thermal conductivity, v is the velocity vector,
and T is the temperature (Bird, Stewart, and Lightfoot 2002). In regions without flow, such as the casing
wall and the surrounding formation, the convective term is neglected, producing the classic form of heat
equation
T
C p k T .
2
(2)
t
The governing equation for pressure and velocity is the Navier‐Stokes equation. The velocity
during cementing wells is assumed to be unidirectional in the vertical direction. The Navier‐Stokes
equation under this scenario is
vz p
evz (3)
t z
The law of mass conservation, expressed as the equation of continuity, is trivially expressed in a
unidirectional case as
vz
0 . (4)
z
The Navier‐Stokes equation and the equation of continuity assume fluids with constant density. This is
the usual assumption for flow of liquids, but many of the newer simulators for conventional cementing
B‐1
treat the ould include such
ese fluids ass weakly compressible. Future workk on this siimulator sho
compresssibility effectss.
Rescaling
Th
he wells of in nterest in this study have aan aspect rati o on the ordeer of 10 5 . This poses num merical
difficulties with the fin
nite element analysis. In fact,
f the defaault meshingg algorithm in
n COMSOL faails on
such geom metries. Even n if a custom algorithm iss used and a mesh is geneerated, such a mesh is off such
poor quality that the results are unacceptable
u . It is therefoore necessaryy to rescale the geometry and
equationss to a more re easonable asp pect ratio.
1:105
Resca
ale
1:1
Figure Apx B‐1
1: Rescaling of t he Wells.
Rescaling is accomplished by non‐dime ensionalizing the governiing equations with a diffferent
ertical and radial directio
length scale in the ve ons. Bittlestoon used this approach to
o rescale thee heat
20
equation in his tempe erature simulator. Follow wing this appproach and reegrouping the parameters into
the traditional dimensionless group ps produces th he following rescaled heatt equation wiith convection,
20
S.H. Bittlestton, Schlumberger Cambridge Reseaarch. SPE‐20448‐M
MS.
B‐2
1 2
v*z (r ) * r , (5)
t *
z Pe
and the heat equation without convection,
1 2
r . (6)
t * Pe
In this equation, the small terms (convection in the radial direction and conduction in the vertical
direction) have been discarded, and all parameters are dimensionless and defined as follows:
r
r* , (7)
rmax
z
z* , (8)
zmax
rmax vin
Pe , (9)
T TML
, (10)
TBH TML
vin
t* t , (11)
zmax
rmax
, (12)
zmax
where TM L is the geothermal temperature at the mudline, TBH is the geothermal temperature at the
bottom of the well, and vin is the fluid velocity in the riser.
This same method has been applied to the Navier‐Stokes equations as well, taking care to define the
rescaling in such a way that it is compatible with the rescaled heat equation. The resulting equation is
vz * dp* 1
* *
r * r vz * . (13)
t dz *
ReB
The parameters which are not shared with the rescaled heat equation are defined as
* , (14)
mud
B‐3
mud vin rmax
Re B , (15)
p gz
p* , (16)
po
The Bingham Reynolds number ReB , the viscosity at infinite shear , and the non‐dimensional
viscosity will be more fully explained in the section on viscosity.
*
Equations (5) and (13) are the equations which are inputted to COMSOL. Conveniently, they are
structurally the same as the original equations, and are almost identical to the usual non‐
dimensionalized forms.21 The only significant difference is the additional scaling term and the lack of
the vertical direction in the Laplacian operator. This allows many of the built‐in methods in COMSOL to
be used with only minor modifications.
Equivalent Circulating Densities
Equivalent Circulating Densities are a convenient way to express pressures downhole by
converting the pressure units to “equivalent” density units. Mathematically,
P
equiv , (18)
gh
The ECD is non‐dimensionalized in order to match the form of the Navier‐Stokes equations as
P*
*
equiv * * , (19)
gh
where all terms are dimensionless and defined as
equiv
equiv
*
, (20)
mud
21
R. B. Bird, W. E. Stewart, E.N. Lightfoot. Transport Phenomena. 2007.
B‐4
zmax g
g* , (21)
vin 2
h
h* (22)
zmax
and all other terms are the same as their definition in the Navier‐Stokes equations.
Viscosity
Drilling fluids are usually modeled as either Bingham fluids or Herschel‐Bulkley fluids (Hemphill,
Campos, and Pilehvari 1993). The rheological data was provided to us from CSI Technologies as Bingham
fluids. Therefore, the model was built for Bingham fluids.
The Bingham model treats fluids as a solid when the shear stress is below a yield stress, and as a
Newtonian fluid when the shear stress is greater than the yield stress. Mathematically, this is expressed
as
y
y
e , (23)
y
y 1 exp n
e . (24)
B‐5
Figure Apx B‐2: Papan
nastasiou's consttitutive equation compared to Bingham behavvior . All units arre dimensionlesss.
The
e Bingham‐Paapanastasiou expression iss used in a noon‐dimension
nal equation,, so it too mu
ust be
non‐dimeensionalized in such a wayy to be comp
patible to thee non‐dimenssional Navier‐‐Stokes equaations.
The Bingh
ham Reynoldss number is defined as (Chabra and Richhardson 20088)
vin rmax
Re B . (25)
(Tref )
Th ear, , is the
he viscosity aat infinite she e parameter which is used
d as a characcteristic viscossity in
this dime
ensionless grroup. This paarameter varried with te mperature, aand so its vvalue at a ceertain
reference
e temperature is used. Th
his implies ussing as thhe characteriistic viscosityy and defininng the
dimension
nless viscosityy as
e (T ) y T 1 exp(n )
* (T ) . (26)
(Tref
r ) (Treff ) (Tref )
where the temperature dependencies have be een shown exxplicitly. The reference teemperature in this
model is ttaken to be th
he temperatu
ure at the rise
er inlet.
If a Herschel‐Bulkley mmodel is desireed instead off the Bingham m model, thenn the same baasic procedurre can
be used, but substituting a Herschel‐Bulkley version
v of Paapanastasiou’’s constitutive equation. Many
authors have suggeste ed forms of thhis constitutivve equation ( Mendes, Duttra, and Janeiro 2004; Zhu, Kim,
and Kee 22005).
B‐6
Temperature Dependence
As alluded to in the previous section, the viscosity is strongly dependent on temperature. This
dependence is usually modeled with an Arrhenius expression (Houwen and Geehan 1986)
B
A exp
, (28)
T
B y
y A exp . (29)
y
T
The values of the parameters can be determined by doing nonlinear regression on experimental data at
different temperatures.
Boundary Conditions
Temperature
The boundary conditions for temperature have already been established by Bittleston(Bittleston
1990). At the centerline of the well, a symmetry boundary condition is imposed
T
0 . (30)
r r 0
Far away from the well, it is assumed that the geothermal temperature remains undisturbed,
Infinite distances are not possible to implement in FEA, and so infinity was approximated by a
sufficiently large distance
At the riser inlet, the temperature is assumed to be equal to the surface temperature 80 F ,
T ( z inlet , r , t ) Tsurface . (33)
At boundaries between radially adjacent sections of the well (e.g. between the formation and the casing
annulus) which have different governing equations (heat equation and heat equation with convection,
for the formation and casing annulus, respectively), continuity of temperature and flux is imposed:
B‐7
T T
kleft kright . (35)
r left r right
This is the boundary condition natively implemented in COMSOL, and so no special programming is
required.
Bittleston’s model only dealt with single string wells, so he did not have to deal with vertically
adjacent sections. Therefore, additional boundary conditions but be developed for these situations.
Because vertical conduction is negligible, the only mode of heat transfer between vertically adjacent
sections of the well is convection. This is implemented in the boundary conditions at the inlet and outlet
of regions with fluid flow. Because these regions differ in size and shape, no point‐wise boundary
condition can be applied. It is therefore necessary to define a boundary condition such that the total
amount of thermal energy is conserved from one region to the next. This is done by setting the inlet
temperature of a region equal to the mixing‐cup temperature (sometimes known as the bulk
temperature) at the outlet of the previous region. The mixing cup temperature is defined as
Tmc
vT
Ac
z
. (36)
v Ac z
This boundary condition is also applied at the shoe, where the fluid flows from casing center into the
casing annulus (in conventional circulation), or vice‐versa (in RCPC). This prescribes the inlet, but a
different boundary is required for an outlet. The usual assumption for outlet boundaries is that there is
no heat conduction across the outlet boundary (“COMSOL Multiphysics” 2012),
nT 0 (37)
where n is the normal vector of the boundary. In vertical directions without flow, a simple insulation
boundary condition is applied,
T
0 . (38)
z
The temperature boundary conditions are summarized in Figure Apx B‐3. Arrows indicate flow direction.
A is (32), B is (34) and (35), C is (30), D is (38), E is (36), and F is (37). The arrows indicate flow direction.
B‐8
E D F D
Casing Annulus
Casing Center
Formation
Casing
C B B B A
F D E D
Figure Apx B‐3: Summary o
of temperature boundary cond itions.
Th
he initial con
ndition of the temperatu
ure is the geeothermal tem
mperature o
of the surrounding
formation
n,
This does not include tthe effects off drilling histo
ory on the weell. This simp
plification is ju
ustified becauuse of
the long cconditioning time of the wwell, based on the pumpi ng scheduless provided byy CSI Technologies.
Because mmud is circulaated for two ccomplete hole volumes, thhe effect of tthe initial con ndition is forggotten
by the sysstem before tthe cementingg process beggins.
Riser
In
n the riser sysstem, there iss no surroundding rock form mation. Insteaad, heat is traansferred bettween
the riser wall and the
e surroundingg ocean. Thiss requires a different boundary conddition at the outer
radius of the riser to capture the co onvective effeects of the occean current.. Here, Newto on’s law of coooling
and an en nergy balance e is applied at the boundarry,
T
k h T Tocean ( z ) . (40)
r r wall
ensionless form,
Or in dime
Bii ocean ( z * ) , (41)
r * *
r wall
where
B‐9
hrmax
Bi . (42)
k
The heat transfer coefficient h can be calculated from the Nusselt number (Bergman et al. 2006)
4/5
0.62 Re1/2 Pr1/3 Re 5/8
Nu D 0.3 1/4 1 . (43)
1 0.4 / Pr 2/3
282, 000
This requires knowing the temperature of the ocean throughout the depth of the riser, and the
properties of the ocean water at those temperatures. The data on the thermocline was provided by CSI
Technologies, and the properties as a function of temperature were calculated from correlations
(Sharqawy, V, and Zubair 2010).
Fluid Flow
The boundary condition for the Navier‐Stokes equations at the solid wall is the classic no‐slip
condition (Bird, Stewart, and Lightfoot 2002)
vz wall
0 . (44)
At the center‐line, a symmetry condition is applied (Deen 2012),
vz
0 . (45)
r r 0
At fluid outlets, the default outflow condition in COMSOL, no viscous stress, is used
e
v v T n 0 . (46)
The fluid outlet was also where the boundary condition on the pressure was supplied
The pressure should match the inlet pressure of the next section of flow, if the current section is not the
outlet of the overall well, but specifying a non‐zero pressure caused issues the COMSOL solvers.
Therefore, each outlet was set to a zero pressure, and then properly adjusted in post‐processing.
At the inlet of each region, the natural boundary condition was used, with a average velocity
constraint. This forces the flow to enter as fully‐developed laminar flow, by converting the Navier‐Stokes
partial differential equation to an ordinary differential equation on the boundary, and then solving that
ODE (“COMSOL Multiphysics” 2012), the boundary equation is
B‐10
penntr I e v v
T
p
entr
n , (48)
where pentr
e is a fictitio
ous pressure C
COMSOL calcculates in ord er to obtain tthe desired flowrate.
Figure Apx B‐4
4: Boundary con
nditions for the N
Navier‐Stokes eequations.
Th
he initial condition for fluid flow is a stteady‐state soolution of thee Navier‐Stokkes equationss. This
ignores an
ny ramping efffects, but theese are short compared too the lifetimee of the well jo ob.
Equation Coupling
One
O of the ch hallenges in solving this problem
p is t he coupling of the heat equation and the
Navier‐Sto
okes equations. The heat equation inccludes a veloccity term from m the Navier‐Stokes equaations,
and the viscosity depends on the teemperature.
B‐11
Figu
ure Apx B‐5: Cou
upling of the heaat equation andd the Navier‐Stokes equations.
COMSSOL is able to solve the e equations simultaneoussly, but doin ng so can bee computatio onally
inefficientt in some caases. Therefo
ore, an iterattive techniqu e was used to generate the solution
n. The
algorithmm is as follows:
1. et the temperature in the well to the ge
Se eothermal teemperature
2. So
olve for the fllow field usin
ng the temperrature from sttep 1.
3. So
olve for the teemperature u using the flow
w field from sttep 2.
4. So
olve for the fllow field usin
ng the temperrature from sttep 3.
edure can be repeated as long as is neccessary, but thhose four steeps are usuallyy sufficient.
This proce
Multi‐Pha
ase Flow
Thhe problem at hand iss, fundamen ntally, a muulti‐phase flo ow problem. It involvess the
displacemment of one fluid (drilling m mud) with an nother (cemeent). The Phasse Field and Level Set methods
are the twwo different mmethods for ttwo‐phase flo ow in COMSO OL. They havee a large degrree of similariity, so
they will be discussed together. They solve the
t usual Naavier‐Stokes equations, w with an additional
equation which contro ols the volume fraction of each fluid. Thhe details of tthese schemees are not relevant
urrent discussion, but it is worth notting that theey are compu
to the cu utationally expensive, and are
designed for very com
mplicated inte us fingering. Such method
erfaces, such as interface s with viscou ds are
used in thhe industry to
o calculate dissplacement effficiency(Nelsson and Guillo ot 2006).
B‐12
phase flow equations near the intterface. To properly captuure the dynaamics of the interface, a much
higher ressolution is req
quired.
Figure Apx B‐6: The interfface between th
he fluids is assumed to be perfeectly sharp, and moves with thee average velociity of
the fluids.
B‐13
Figure Apx B‐7: Spatially and Temporallyy Dependent The ermal Propertiees. Represents an arbitrary pro
operty. The grap
ph on
the right is the overall prroperty plotted along the dasheed line in the schematic on the left.
Figure Apxx B‐8: Calculation
n of the combined velocity field
d. The “ ” is not a true addition operator, but raather a combinaatorial
operator.
B‐14
Figure A
Apx B‐9: Couplin
ng of equations for multiple fluids
Rather than calculate the combined pressure fie
eld explicitly,, the total preessure drop aacross each rregion
is calculatted accordingg to
N
PTotal i Pi , (49)
i 1
uids, i is the
where N is the total number of flu d i , and Pi is the
e fraction of tthe region filled with fluid
total pressure drop thrrough the reggion when fille
ed with fluid i .
L
0..061Re , (50)
D
where Ree is the Reyno olds number,, L is the length of the enntrance regio
on, and D is the pipe diam
meter
(Wilkes annd Bike). The calculated en ntrance lengtths in the riseer and drillpip pe for the wellls of interest using
this equation are pressented in Tab ble Apx B‐1. While the aabove equatio wtonian fluids, the
on is for New
effect of a yield stresss only decre
eases the en ntrance lengtth (Vradis, DDougher, and Kumar 1993 3), so
performin ng the analyssis as a Newttonian fluid with
w viscosityy equal to th he plastic viscosity is the most
conservattive estimate for the entraance length.
B‐15
Table Apx B‐1: Entrance Lengths of Mud and Cement.
Entrance Length (ft) Percent of Pipe Length
Riser Drillpipe Riser Drillpipe
Mud 206.9 206.9 4.1% 3.0%
Cement 30.9 36.4 0.62% 0.52%
So ultimately, ignoring the flow features near the plug will introduce an error of less than ~4%.
Therefore, the effect of the interface can be safely neglected when calculating the overall flow.
Solution Techniques
The COMSOL Multiphysics software package was used to solve the equations. COMSOL uses a Finite
Element method, and has powerful, built‐in solvers. Many of the specific details of the solution methods
are proprietary and hidden from the user, but the some important ones can be adjusted. The
adjustments used in this model are as follows:
Meshing: Triangular.
Time‐Stepping: Backward Difference Formula (BDF). In order to avoid unphysical oscillations in
the temperature, it was necessary to limit the maximum BDF order to 2.
Numerical Solver: PARDISO or MUMPS.
Secondary Calculations
In addition to the primary parameters the model produces, a number of secondary calculations
are desired; namely, the load on the pump, the presence of U‐Tubing (where the static pressure moves
the fluid faster than the pump, also known as free fall), and the required back‐pressure to prevent U‐
Tubing.
For these calculations, it is convenient to introduce another parameter known as the lift pressure PL .
This is the static pressure differential between the regions where fluid travels towards the surface and
regions where fluid travels towards bottom‐hole (i.e. in conventional circulation, the difference between
the annuli and the central regions). If there is only one fluid being pumped, such as during the mud
conditioning phase, the lift pressure is zero. The lift pressure can be positive or negative; a positive
number indicates that the pump has to provide additional work to lift the heavier fluids up against
gravity, and a negative number indicates that gravity is assisting the flow.
The load on the pump is simple the sum of the pressure drops across each region plus the lift
pressure,
N
PPump PL Pi . (51)
i 1
B‐16
By analyzzing the calcculated pum mp pressure during the jjob, U‐Tubin ng and back pressure caan be
determine ed. If U‐Tubing is present, the calculaated pressuree on the pum
mp will beco
ome negativee. The
magnitude of the pump pressure in n this situation n in the back pressure required to allevviate the U‐Tu
ubing.
Model Va
alidation
It is importantt to verify thaat the model accurately ddescribes the system. To d do this, the m
model
was used to analyze co onventional ccirculation of well 1, regio n 1, and then
n the results w
were comparred to
existing siimulators. The simulationss on commerccial simulatorrs were perfo ormed by CSI TTechnologiess.
Temperatture
Th he two main temperature es of interest are the bottoom‐hole circu
ulating tempeerature (BHCTT) and
the first‐ssack/last‐sackk temperature
es. The BCHTT results are ccompared in Figure Apx B‐‐10. After thee mud
conditioning period, w where the BHCT drops quickly, the maxximum BHCT matches up very well witth the
commercial simulator maximum. Once the ce ement reachees the botto
om the prediicted temperrature
increases,, due to the hhigher thermaal conductivity of cement.
Figure A
Apx B‐10: COMSSOL predicted bottom‐hole circulating temperaature compared
d to commercial simulator resullts.
Th
he first‐sack//last‐sack tem
mperature comparison iss shown in Figure Apx B‐11. Comm mercial
simulator results are only for the first‐sack/lasst‐sack tempperature, but COMSOL is able to showw the
temperatures through hout the cem ment plug. Th he COMSOL results will b
be more fullyy explained in the
results section, but thee first‐sack/last‐sack temperatures in CCOMSOL are vvery close to tthose predictted by
the comm mercial simulaator.
B‐17
Figure Apxx B‐11: Cement ttemperatures predicted by the COMSOL modeel compared to tthe first‐sack/lasst‐sack temperaatures
in the commercial software. T
Time is at the ennd of cement placement.
Pressure
Because equivvalent circulatting densities are dominanntly composed of static pressure effectts, it is
inadvisable to compare ECDs directtly to validatee friction loss models. Insteead, pressuree drops acrosss well
regions are
a calculated d with both the COMSO OL model annd with anallytical metho ods developeed by
Buckinghaam and Bird ffor flow in pipes and annu uli, respectiveely (Buckingh
ham 1921; Freedrickson and d Bird
1958). The results for mud are shoown in Table A Apx B‐2 andd results for ccement are sh hown in Table Apx
B‐3. No te
emperature e effects were included, and d the viscositties were equ ual to their vvalues at 80°FF. The
Papanastaasiou accuraccy parameter n 30 .
Table Apx B‐2: Pressure drops for mud th
hough well regio
ons calculated w
with the COMSO L model and witth analytical meethods
∆P (psi)
Region COMSSOL Analyticcal
Casing 42.32 36.00
0
Drill Pipe 221.6
65 263.70
0
Dril l Pipe in Riser 108.3
38 101.50
0
Casing Annulu us 62.9
92 62.10
0
Ri ser Annuluss 47.16 39.50
0
Lowerr Casing Ann nulus 101.223 164.40
0
Upperr Casing Ann nulus 104.779 169.40
0
B‐18
Table Apx B‐3: Pressure drops for cement though well regions calculated with the COMSOL model and with analytical
methods.
∆P (psi)
Region COMSOL Analytical
Casing 53.63 22.40
Drill Pipe 501.94 702.00
Drill Pipe in Riser 164.72 228.50
Lower Casing Annulus 244.78 286.80
Upper Casing Annulus 351.03 546.20
B‐19
Appen
ndix C ‐ CO
OMSOL Te
emperatu
ure and Prressure Reesults
Th
he simulator was used to
o analyze tw
wo different wwells, with thhree differen
nt casings in each.
Simulations were run in both reverse and convventional placcement in orrder to clearlyy see how reeverse
placemennt affects the
e cementing process. As far as possibble, job paraameters (e.g. flow rates) were
maintaineed between reverse and co onventional p placement.
Figgure Apx C‐1: Ge
eometry for Welll 1, Region 1.
Temperatture
C‐1
°F
Figure A
Apx C‐2: Surface plot of the temperature at the end of mud circculation for welll 1, region 1 in rreverse circulatiion.
°F
Figure Apx C‐3: Surface plo
ot of the temperrature at the en
nd of mud circulaation for well 1,, region 1 in con
nventional circulation.
C‐2
much lesss heat on the way down th han it does inn reverse circuulation, producing lower ttemperaturess. This
qualitative analysis is verified by the quantitaative BHCT ddata presenteed in Figure Apx C‐4. Reeverse
circulation
n increases th he BHCT by about 40ºF in tthis simulatioon.
200
190 Coonventional
180 Reeverse
Max BHCT (°F)
170
160
150
140
130
120
110
100
90
80
0 2 4 6 8 10 12
Dimensionless Time
Figurre Apx C‐4: Maxximum Bottom‐H
Hole Circulating Temperature (B
BHCT) in conven
ntional and reveerse circulation.
While
W the BHCCT may be im
mportant for characterizinng the overalll temperaturre of the well, the
temperatures the cem ment experien nces are equ
ually importannt. In convenntional circulation, the ceement
passes through bottom m‐hole, and th herefore experiences the BHCT. But fo or reverse circculation, as GGriffith
et al. poin
nt out, only th
he leading ed dge of the cem
ment reachess bottom‐hole. Therefore,, the BHCT is a less
importantt parameter ffor reverse cirrculation than n it is for convventional circculation.
Figure Apx C
C‐5: Temperature in the cemen
nt at the end of p
placement. Reveerse is higher th
han conventionaal, but less so th
han the
BHCT.
C‐3
circulation. This is contrary to the claims of Griffith et al., who claimed that the amount of retarders
could be reduced in RCPC because only the leading edge of the cement would experience temperatures
as high as conventional placement.
The shape of the temperature curves in Error! Reference source not found. merits additional
discussion. In reverse circulation, the cement enters the annulus at the top, and then picks up heat as it
travels down to the hotter regions of the well. Because the flow is in the same direction as the
geothermal temperature gradient, the temperature line is fairly linear. In conventional circulation, the
relatively cool fluid enters the annulus at the bottom of the well (the region with the highest formation
temperature), and quickly picks up temperature as it travels up the annulus. As the cement reaches the
cooler regions towards the top of the annulus, it no longer draws heat from the formation, and
eventually begins to lose heat to the formation and decrease in temperature.
Pressure/ECDs
The ECD results are the most important result from the simulation, because it determines
whether the cementing job is feasible or not, and whether RCPC is an improvement over conventional
cementing. ECDs at two locations, bottom‐hole and at the previous shoe, will be compared in
conventional and reverse placement. These two locations are at the top and bottom of the open holes,
so unless there is an abnormally weak region in the middle of the open hole, these locations provide a
good indication of whether the job will induce fracturing.
The maximum ECDs for well 1 are shown in Table Apx C‐1. ECDs are significantly reduced at
bottom hole, but significantly increased at the previous shoe. Table Apx C‐2 shows the maximum ECDs
for well 2. As with well 1, ECDs are reduced at bottom‐hole, but unlike well 1, are also reduced at the
previous shoe. The ECDs in conventional placement for well 2 are extremely high, because of extremely
tight liner. While such a configuration may not be practically advisable, such an extreme scenario allows
the differences between reverse and conventional placement to be seen more clearly.
Table Apx C‐1: Maximum ECDs in well 1. ECDs reduced at bottom‐hole, but increased at the previous shoe.
Conventional Reverse
Bottom Hole 14.5 12.6
Casing 1
Prev. Shoe 12.7 13.5
Bottom Hole 14.6 14.2
Casing 2
Prev. Shoe 13.4 14.6
Bottom Hole 14.3 13.2
Casing 3
Prev. Shoe 12.8 14.4
C‐4
Table Apx C‐2
2: Maximum ECD
Ds in well 2. ECD
Ds reduced at booth bottom‐holee and the previo
ous shoe.
Pump Pre
essure and Ba
ack Pressure
Thhe pump pre essure, lift pressure, and back pressuure for one cconventional job are show wn in
Figure Appx A‐6. The lo
ongest portio on of the job
b is the mud conditioningg phase, wheere drilling mmud is
circulatedd. Because thhere is one fluid
f through hout the welll, the lift pressure is zerro, and the pump
pressure is relatively cconstant. Oncce cement is being pumpped, the lift p pressure beco omes negative and
e pump presssure below zero,
drives the z indicatiing U‐Tubingg if back presssure is not applied. Oncce the
cement re eaches botto om‐hole, the lift pressure increases draamatically, w which drives the pump preessure
up.
Figure A
Apx C‐6: Pump, liift and back pressure for well 11, region 2 conveentional placem
ment
C‐5
Thhe analogouss graph for reverse placem ment is shownn in Error! Refference source not found d.. The
shape of the pressure
e curves is allmost identiccal during muud circulation
n and the firrst part of ceement
placemen nt. The samee U‐Tubing effect is seen n shortly afteer cement placement begins. But because
reverse placement
p do
oes not require pumping the heavy ccement again nst gravity, the lift pressure is
always neegative, whichh keeps the pump pressure e much lowerr.
Figure
e Apx C‐7: Pump, lift and back pressure for Welll 1, Region 2 revverse placement.
C‐6
Ap
ppendix D
D ‐ RCPC H
Hydraulicc Analysiss
pT p f ps . (52)
Because tthe total presssure can be split in this m
manner, the ffrictional porrtion and the static portio
on will
be analyzed independe ently. The tw
wo main pointts of interest downhole arre the bottomm of the annu ulus of
the casingg being ceme ented, bottomm‐hole (BH), and the top of the casingg annulus, wh here the crosssover
tool is located in offsho
ore RCPC (XO O).
DR
DRA
CA
A
Figure Apx D
D‐1: Generic well diagram used for hydraulic an
nalysis.
D‐1
further diivided into central regionns and annulaar regions. T his grouping results in fo our regions: 11. the
drillpipe/rriser (DR) 2. tthe drillpipe//riser annuluss (DRA) 3. thee casing (C) aand 4. the caasing annuluss (CA).
No assum mptions are m made about th he geometry of the well, existence of valves or floaat equipmentt, etc.
except for the existennce of a crosssover. This allows the ressults obtained
d to be as geeneral as posssible.
Three diffferent flow paths will be aanalyzed in th his section. TThese flow paaths are illustrated inFigurre Apx
D‐2.
Figure Apx D‐2: Flow‐paths ffor offshore reve
erse, conventionnal, and traditio
onal reverse circculation.
Conventio
onal Circulatiion
Thhe static presssure at botto
om‐hole is the
e sum of the sstatic pressurres across eacch region abo
ove it,
in this casse the static p
pressures in th
he annuli,
The staticc pressure at tthe crossoverr can be calcu
ulated similar ly as
p XO , s pDRA, s . (54)
onal pressure
The frictio e component at bottom‐ho m of the pressure drops acrross each region
ole is the sum
“downstre eam” of bottoom‐hole,
ossover,
At the cro
p XO , f pDRA, f . (56)
al Reverse Circulation
Traditiona
In
n traditional rreverse circulaation, where fluids are pu mped directlyy down the annulus, the
expressions for static ppressure do nnot change, soo
D‐2
pBH , s pCA, s pDRA, s , (57)
p XO , s pDRA, s . (58)
The alternative flow path does change the frictional pressure at each point,
Offshore Reverse Circulation
The same procedure can be performed for reverse circulation offshore, taking care to change
the definitions of “above” and “downstream” to include the flow path through the crossover tool
suggested by Wreden et al (2014):
p XO , s pDR , s , (62)
Conventional vs. Traditional Reverse
By defining a pressure change as the difference between traditional reverse and conventional
circulation, convenient expressions for pressure savings can be found:
pBH , s 0 , (65)
PXO , s 0 . (66)
Because the expressions for static pressure are identical in conventional and traditional reverse
cementing, the static pressure has no effect on the overall pressure change from conventional to
reverse circulation. The change in pressure is therefore solely due to the change in frictional pressures,
expressed as
Because the annular regions are typically narrower, and the casing annulus is filled with more viscous
cement at the end of cement placement, pC, f pDR , f pCA, f pDR A, f , which implies that
pBH , f 0 and ECDs are reduced at bottom hole.
D‐3
The pressure savings at bottom hole are offset by a pressure increase at the previous shoe
(making the reasonable assumption that the magnitude of pDRA, f is not greater than the sum of the
other pressures),
caused by the additional flow path length “downstream.” This is in agreement with Kuru (2005), who
found that above a critical point the pressure would be increased.
Conventional vs. Offshore Reverse
The same analysis can be performed for conventional circulation and offshore RCPC. The
expressions can be further simplified by specifically analyzing the end of cement placement, where
cement is localized to the casing annulus and mud is everywhere else:
The displacement mud is the same density as the drilling mud. As before, the static pressure does not
change the overall pressure.
Because of the hybrid nature of the reverse flow, the expressions for the frictional pressure
change are simplified in this case. The pressure at bottom‐hole is expressed as
The pressure change at bottom hole is dependent upon the relative magnitudes of the pressure drop
through the casing and the casing annulus. Because the casing is filled with mud, and the casing annulus
is more narrow and filled with more viscous cement, pCA, f pC , f . This is the cause of the pressure
savings in RCPC. The pressure change at the crossover tool is expressed as
As before, the pressure at the crossover tool is increased in RCPC.
Critical Depth
If the ECD is reduced at bottom‐hole and increased at the previous shoe, then there is a point
between those two where the pressures in conventional and reverse circulation are equal. This point is
known as the critical depth or the critical zone (Kuru and Seatter 2005; Moore et al. 2005). By taking
expressions for the pressure in the casing annulus for both conventional and reverse circulation, an
analytic expression for the critical depth can be derived, assuming a constant pressure drop per unit
D‐4
dp f
length in the casing annulus . When designing a well for reverse circulation it is important to
dz
CA
ensure that the weakest part of the formation is below this critical depth in order for RCPC to be
effective.
Offshore Reverse
The pressure at any height in the casing annulus in conventional placement is expressed as
dp f
pconv (z) pDRA, f (z PS z ) . (73)
dz CA
Similarly, for offshore reverse placement,
dp f
prev ( z ) pC , f pDRA, f ( z) . (74)
dz CA
Setting the two pressures equal and solving for the critical depth zc yields
z XO pC , f
zc . (75)
2 dp
2
dz CA
Traditional Reverse
As before, the pressure at any height in the casing annulus in conventional placement is
expressed as
dp f
pconv (z) pDRA, f (z PS z ) . (76)
dz CA
Similarly, for traditional reverse placement,
dp f
prev ( z ) pC , f pDR , f ( z) . (77)
dz CA
Setting the two pressures equal and solving for the critical depth zc yields
D‐5
z XO pDRA, f pDR , f pC, f
zc . (78)
2 dp
2
dz CA
Lift Pressure and Back Pressure
In conventional circulation cement is pumped all the way to bottom hole through the casing,
and then back up into the annulus. The pump must therefore exert additional pressure to pump the
heavier cement against gravity. This additional pressure is known as the lift pressure p L , and is
defined as the difference in hydrostatic pressure between the regions where flow is towards the surface
and the regions where the flow is towards bottom‐hole. In reverse circulation, the lift pressure is
eliminated because the cement is always flowing with gravity. However, in reverse circulation it is
usually necessary to introduce back pressure p B in order to prevent the cement plug from coming up
into the casing (Moore et al. 2005). This pressure is applied at the outlet, and is theoretically equal in
magnitude to the lift pressure.
The absence of lift pressure has led to misunderstandings and confusion about what causes the
pressure savings in reverse circulation (Davies et al. 2004; Hernandez and Nguyen 2010; Hernandez and
Bour 2010; Hernandez 2012). At the end of cement placement, the heavier cement fills the casing
annulus, and the other regions are filled with drilling mud. The cement is heavier than the mud, and so
the static pressure at the bottom of the annulus is greater than the static pressure at the bottom of the
casing. This pressure imbalance would cause flow from the annulus into the casing, unless the pump
provides additional pressure to rectify this imbalance. So the lift or back pressure simply increases the
pressure inside the casing to match the pressure in the annulus. It does not increase the pressure in the
annulus, and therefore the absence of lift pressure cannot create ECD savings.
Graphical Analysis
The proceeding analysis can be visualized by plotting the pressures through the well in both
conventional and reverse circulation. Such a plot can be easily constructed once the fluid densities and
pressure drops across each region are known. Begin at the outlet, where the pressure is known (zero for
conventional circulation, and equal to the back pressure in reverse circulation), and draw lines for the
static and frictional pressures. The slope of the static line is equal to g and the slope of the frictional
dP
line is equal to the pressure drop per unit length . Then draw the total pressure line which is the
dz
sum of the static and frictional pressure lines. The next region is drawn similarly, but shifted along the
pressure axis so that the outlet pressure of the current region is equal to the inlet pressure of the next
region, so as to maintain pressure continuity. This process should be repeated for each region until the
whole well is completed. Once diagrams for both conventional and reverse circulation are completed,
the total pressure lines from each can be transferred to a single graph for comparison. The point where
the total pressure lines in the casing annulus intersect is the critical depth. Examples of such diagrams
are shown in Figure Apx D‐3‐ Figure Apx D‐7.
D‐6
Inlet/
Outlet
Friction Pressure
Static Pressure
DRA
Total Pressure
DR
Depth
XO
CA
C
BH
Pressure
Figure Apx D‐3: Pressure diagram for conventional circulation.
D‐7
Inlet/
Outlet
DRA
Friction Pressure
Static Pressure
DR Total Pressure
XO
Depth
CA
C
BH
Pressure
Figure Apx D‐4: Pressure diagram for (offshore) reverse cementing.
D‐8
Inlet/
Outlet
Friction Pressure
Static Pressure
DR
Total Pressure
DR
XO
Depth
CA
C
BH
Pressure
Figure Apx D‐5: Pressure schematic for (traditional) reverse cementing.
D‐9
PL PB PL PB
Inlet/
Outlet
Conventional
Reverse
Depth
XO
Critical
Depth
BH
Pressure
Figure Apx D‐6: Total pressure for both conventional and offshore reverse cement placement.
D‐10
Inlet/
Outlet
Conventional
Depth Reverse
XO
Critical
Depth
BH
Pressure
Figure Apx D‐7: Total pressure for both conventional and (traditional) reverse cementing.
Alternative Placement of the Previous Shoe
If the previous shoe is at the same depth as the crossover tool, then the pressure in RCPC will
necessarily be higher than for conventional placement. However, if the overlap between the casing
being cemented and the previous casing is large enough, then the previous shoe could be lower than the
critical depth. In such a configuration, the pressure exerted against the formation could be reduced
throughout the casing annulus. In fact, this is the case in well 2, where the extremely tight liner gap
shifts the critical depth above the previous shoe.
D‐11
Appendix E ‐ Cementing Materials Testing
1) Test methods
Laboratory test methods are described below. When applicable, tests were performed according to
standard API procedure.
Rheology: Slurry surface rheology is measured at ambient temperature with a rotational viscometer.
Downhole slurry rheology is measured after conditioning in an atmospheric consistometer if the BHCT is
190°F or less. If the BHCT is greater than 190°F the slurry is conditioned under temperature and
pressure in a pressurized consistometer.
API Thickening Time: Slurry thickening time is tested using a pressurized consistometer to simulate
downhole pressure and temperature to determine how long the slurry can be pumped before setting.
API Static Fluid Loss: The slurry is conditioned to temperature in an atmospheric consistometer and
placed in a fluid loss cell. A 1000 psi differential pressure is applied across the slurry and the amount of
fluid released in 30 minutes is recorded. The fluid loss test is a representation of fluid loss from the
slurry into the formation during placement.
API Stirred Fluid Loss: The slurry is conditioned at 190⁰F or above in the fluid loss cell. After
conditioning, the cell is rotated and a 1000 psi differential pressure is applied and the amount of fluid
released in 30 minutes is recorded.
Free Fluid: A column of slurry is left static at downhole temperature and the volume of free fluid
collected at the top of the sample is measured. This is an indication of static slurry stability.
UCA: Compressive strength and time to initial set is measured non‐destructively with an Ultrasonic
Cement Analyzer (UCA) for 24 hours or 48 hours.
Dynamic and Static Settling: Density variation is measured after completion of the thickening time test
(dynamic settling) and after completion of the UCA (static settling). A density variation greater than 10%
typically indicates an unstable slurry system.
HTHP Settling: The slurry is brought up to bottom hole temperature and pressure in an atmospheric
consistometer. Two 15 minute shutdowns are simulated by stopping the consistometer motor. After
each shutdown, any spike in consistency is recorded. A high spike after the motor has been restarted is
an indication of stability issues. The density variation between the top, middle and bottom of the slurry
is measured after the second shutdown. A density variation greater than 10% typically indicates an
unstable slurry system.
Gel Strength Development: Static gel strength development is measured by a RP/GS consistometer.
Transition time is considered to be the time that the gel strength develops from 100 lbf/100 ft2 to 500
lbf/100 ft2.
E‐1
API Rheological Compatibility: Compatibility testing includes a visual and rheological comparison of 5
ratios between two intermixed fluids (95/5, 75/25, 50/50, 25/75, and 5/95). These fluids can be
cement, spacer or mud. Testing is conducted on a rheometer after fluids are conditioned at BHCT, and
the intermixed fluids are also observed visually for signs of incompatibility.
Rotor Test: In a rotor test, a specialized rotor is used on a standard atmospheric rheometer to measure
how effectively the spacer can remove mud filter cake. Mud and spacer will be conditioned on an
atmospheric consistometer at BHCT for 30 minutes. Mud is transferred to a rheometer cup and stirred
at 100 rpm for 5 minutes, then rinsed with water. Spacer is transferred into the rheometer cup and
stirred, followed by a second water rinse. After the second water rinsing, erosion of the mud film is
checked. If the water beads stay on the rotor, it is considered as fail since this indicates that the rotor
surface is oil wet. If the water streams down the rotor, then it is considered as pass since this indicates a
water wet surface.
Coupon Contact Angle: Glass coupons are used as a test surface in the coupon contact angle tests. The
contact angle of a clean glass coupon is taken at the beginning of the test as a control. The glass coupons
are then soaked in mud for 5 minutes, removed, then air‐dried for 10 minutes to leave a mud film.
Coupons are also soaked in spacer for 5 minutes, and then air‐dried for 10 minutes. Contact angle of a
water droplet is measured with a goniometer on the surface. To evaluate the mud removal
effectiveness, contact angles of spacer treated coupons will be compared with those of the neat glass
coupon and mud coated coupon. If the treated coupon has the same or lower contact angle than the
neat glass coupon, it is considered as a pass. If higher than neat glass coupon contact angle, it is
considered as a fail.
2) Fluid Descriptions
Cement Slurry: Cement systems were designed for this project using commercially available oilfield
additives and Class H Lafarge cement. Additives used were KCL, high and low temperature retarders,
dispersants, viscosifiers, and fluid loss control additives. Slurries were designed in ranges from 12.5 ppg
to 16.4ppg.
Spacers: Due to common fluid incompatibilities between cement, SBM, OBM, and in some cases WBM,
spacers are used to physically separate cement slurry from muds and displace the existing mud in the
wellbore. Spacers used in this study were mixed from commercially available generic oilfield chemicals:
bentonite and polymer based gelling agents, surfactants, and barite as a weighting agent.
Viscous Pill: Since mechanical separators cannot be used in the annulus, a viscous pill in RCPC is
intended to minimize intermixing between spacer, mud, or cement slurry during placement. Materials
tested in this study were combinations of crosslinked guar, barite, bentonite, and xanthan gum.
Chemical Tag: Chemical tags that trigger a pH sensor downhole are a potential downhole tool activator.
For design purposes, it was assumed that the chemical tag will be pumped in between the spacers. The
pH of the spacers ranges between 8 and 9 while the pH of the chemical wash designed is 2.8. The
chemical tag fluid designed for this study was made up of ~99% water, 0.1% citric acid, and 1%
E‐2
surfactantt. As with any fluid pum
mped downho
ole during ceementing jobss, the cemen
nting tag fluid
d was
evaluated
d for compatib bility.
3) Test Results
T s
Rheologiccal Compatibility
Fluid com
mpatibility wass tested betw ween a 14.5 p ppg SBM, a 144.8 ppg spaceer and a 15.0 ppg cement slurry
system. TThe test temp perature was 150⁰F. Rheo ological and viisual notes arre shown in FFigure Apx E‐1 1. For
the designns used, therre was compllete compatib bility betwee n the spacer and the mud d, and the ceement
and mud. The spacer and cement were rheologgically compaatible, however some visual indicationss such
on and separaation occurred. The limitaation of visuaal results is th
as gelatio hat they are q qualitative an
nd can
be open to
t interpretation. For fie
eld application, this type oof result sho
ould be review
wed in detail with
other testt results to de
etermine acce eptability andd a system reddesign may b be needed.
Figu
ure Apx E‐1: Rhe
eological Compaatibility Results for SBM, Spaceer and Cement
Fluid com
mpatibility waas tested betw ween a 12.0 ppg SBM an d a 12.4 ppgg spacer. Ressults are shown in
Figure Ap
px E‐2. This test
t result highlights the importance oof screening spacer perfo ormance prior to a
cement jo ob. The 50:5 50 ratio of mud to spacer is incompatiible both rheeologically and visually. SSevere
gelation aand fluid reacctions downhole as a result of intermixxing can negaatively affect fluid perform mance
during thee job, and in aa worst case sscenario, resu ult in job failuure.
E‐3
Figure Apx E‐2: Rheologicall Compatibility R
Results for SBM and Spacer
Fluid com
mpatibility waas also tested
d between a 14.7ppg visccous pill, a 114.5 ppg SBM
M, and a 16.4
4 ppg
slurry. Results betwee en the mud and the viscou us pill are shoown in Figuree Apx E‐3. TThere was exttreme
incompatibility both viisually and rh heologically be
etween the ppill and the SB BM. Due to tthis incompattibility
mmend that a pill with th
is it recom his compositio
on should bee pumped between the sp pacer and cement,
and a visccous pill used in the field sshould be com
mpatibility tessted with anyy fluids it will be in contactt with
for validattion.
Dial Re
eadings (R1-B1-F1
1.0) PV YP
Mud Pill ogical
Rheolo Visual
300 200 100 60 300 6 3 [cP] [ f/100ft2]
[lb
100 0 23 34 21 16 111 6 4 23 9
95 5 38 56 35 24 177 8 6 39 14 Yees Flocculaation
75 75 53 83 55 41 299 15 11 51 25 N
No Flocculaation
50 50 141 236 107 119 944 71 66 110 88 N
No Clabbeering
25 25 78 130 102 86 722 50 45 45 70 Yees Clabbeering
5 95 42 70 52 45 355 22 18 30 33 Yees Viscosific
cation
0 100 27 47 36 31 244 14 8 20 21
Figure Apx E‐3: Rheologicall Compatibility R
Results for SBM and a Viscous PPill
Dial Re
eadings (R1-B1-F1
1.0) PV YP
Mud/Pill/Cem
m ent Rheological Visual
300 200 100 60 300 6 3 [cP] [ f/100ft2]
[lb
25% / 50% / 25% 136 119 94 77 655 41 35 97 52 es
Ye Clabbe
ering
Figure Apx E‐4: Rheologicall Compatibility R
Results for SBM, Viscous Pill annd Cement Slurryy
Compatib bility results b
between the ccement slurryy and the visc ous pill are sh
hown in Figurre Apx E‐5. TThe
fluids are visually and rrheologically compatible.
Dial Re
eadings (R1-B1-F1
1.0) PV YP
Pill Ce
ement ogical
Rheolo Visual
300 200 100 60 300 6 3 [cP] [ f/100ft2]
[lb
100 0 54 47 36 31 244 14 8 42 18
95 5 74 62 48 38 322 22 19 54 26 Yees Homogen
neous
75 75 98 75 52 41 299 15 11 84 19 Yees Homogen
neous
50 50 98 76 52 40 277 14 10 86 18 Yees Homogen
neous
25 25 120 91 60 45 355 20 15 102 22 Yees Homogen
neous
5 95 168 123 76 53 344 13 10 157 17 Yees Homogen
neous
0 100 174 126 75 52 322 13 9 164 15
Figure Apx E‐5: Rheologicall Compatibility R
Results for Cement Slurry and aa Viscous Pill
E‐4
Compatib bility – UCA
Rheologiccal compatibility tests conffirmed that the viscous pi ll design used d in this studyy should be p placed
between the cement aand spacer. U UCA tests we ere conductedd to determin ne the effect on intermixing on
compresssive strength. Since the 16 6.4 ppg slurryy and the visccous pill weree compatible, a mixture of 25%
pill and 75%
7 cement slurry was seelected for fu
urther compaatibility testing. UCA ressults are show
wn in
Figure Apx E‐6. With tthis viscous pill design, com mpressive strrength develo opment began earlier than n with
a 100% slurry. Howeve er, overall compressive strrength was loower.
Figure Apx E‐6: Compressivve strength comparison betwee
en slurry and 25%
% intermixing w
with viscous pill
Rotor Tesst
The rotorr test was conducted with h a 14.5 ppg SBM and a 14.8 ppg spaacer to evalu
uate mud rem moval
potential of the spacerr. Fluids were tested at 15 50⁰F. Resultss are shown b below in Figu ure Apx E‐7. B Barite
from the spacer is seeen on the ro
otor surface, with water sstreaming down the rotorr after the seecond
wash indicating a pass criterion.
E‐5
Figure A
Apx E‐7: Rotor TTest
The rotorr test was conducted with BM and the chemical tagg to evaluate the mud rem
h 14.5 ppg SB moval
potential of the chemiical tag. Fluid ds were teste ed at 150⁰F. Results are sshown below w in Figure Appx E‐8.
Mud is remaining on the rotor, so tthis composittion of chemiccal tag should d be used in conjunction w with a
designed spacer and su urfactant pacckage.
Figure Apx E‐8: Rotor test w
with SBM and ch
hemical tag
Contact AAngle Test
This test w
was conducte ed with a 14.55 ppg SBM, cchemical tag aand a 14.8 pp pg spacer at 1150⁰F. Results are
shown in Figure Apx E‐9. The pe
erformace of the chemicaal tag alone versus the pperformance when
combined d with a space er was evaluaated. As showwn by Figure A Apx E‐9 (2) and (4), the ch hemical tag did not
turn the surface
s he coupon to water wet, indicated by the high con
of th ntact angle o
of the water drop.
When the e coupon wass treated with the spacer (Figure Apx EE‐9 (1), (3) an nd (5), the w
water contact angle
was low, iindicating a w water wet surrface.
E‐6
Figure Apx E‐9: C
F Contact Angle TTest Results
Cement SSlurry Tests
Cement slurry perform mance testing included rhe eology, thickeening time, fluuid loss, free fluid, compreessive
strength u using a UCA, settling and gel strength developmentt. Design crriteria for inteermediate caasings,
intermediiate liners, an nd tieback strrings was a desired API fluuid loss less tthan 100 mL and a desired d free
fluid less tthan 0.2%. Foor production n liners, the d
desired API fluuid loss was less than 50 m mL and the deesired
free fluid was zero. In all cases, the thickenin
ng times weere based onn pump timee obtained frrom a
simulation n plus a saffety factor. Test temperature and pressure raamps were determined using
commercial simulation n software.
In Phase I, conventionnal slurry dessigns were teested for commparison. In
n Phase 2, th
hese slurry deesigns
were mod dified accordiing to the sim mulated deepwater RCPC ttemperature schedules. P Pressure scheedules
were assu umed to be siimilar to thosse estimated for conventioonal placemeent. For certtain strings slurries
were designed for tw wo TOC : 500 0’ below the previous ca sing shoe an nd to the liner hanger. Some
conventio onal designs were
w not included if the simulation
s inndicated that placement w would break down
the formaation. For soome strings, slurries were e designed aas a lead/tail system or aas staged ad
dditive
systems to optimize co ompressive sttrength or plaacement timee. Systems w were then furrther optimizeed for
reverse placement through modificcation of the density and rrheological hierarchy inveestigated in Phase I
of the stu
udy. Slurry systems werre designed and a compareed for selectt casing strin ngs in two geeneric
deepwate er wells. The wells used arre shown in FFigure Apx E‐110 (not to scaale) and the sselected stringgs are
highlighteed in yellow. Simulated placement tim mes were esti mated from vvolumes and flow rates and do
not includde time for do ownhole equiipment operaation or safetyy factors.
E‐7
Figure Apx E‐10
0: Generic well sschematics usedd for cement slu
urry design
Table Apx E
E‐1: Well #1, 16"" Slurry Design C
Comparison
Scchematic #1 16" LLiner, TOC = 16,,500'
E‐8
Table Apx EE‐2: Well #1, 16”” Liner Slurry Teest Results Comp
parison
Schematic #1
1 16" Liner, TOC == 16,500'
R
Reverse First Sacck
Conventional Reveerse Last Sack (TTOC)
(Shoe)
Free fluid
d, fluid loss an
nd settling re esults are accceptable for aall slurry systtems. Rheolo ogy of the reeverse
first and last sack syste
ems could be e further optim mized so thatt the viscosityy of the last ssack slurry is equal
to or greaater than thatt of the first ssack system. Thickening tiime results arre shown in FFigure Apx E‐1 11 for
the TOC aat 16,500’ sce enario. For re everse placemment, the reqquired placem ment time is aabout 2:45 (h h:min)
shorter thhan conventio onal placeme ent. This allo
ows for both the operational placemen nt time needed on
location aand the thickeening time to be reduced ssignificantly.
Figure Apx E‐11: W
Well #1, 16” Liner thickening tim
me results, TOC @
@ 16,500’ MD
E‐9
Compresssive strength results at the
e shoe are sho
own in Figuree Apx E‐12. The temperatu ure profile ovver
the 24 hour WOC periood varies fromm convention nal to reverse,, however thee strenghts obtainined witth
these designs are comparable.
Fiigure Apx E‐12: W
Well #1, 16” line
er compressive strength resultss at the shoe
Figgure Apx E‐13: W
Well #1, 16” line
er compressive sstrength compaarison at TOC
E‐10
Slurry designs and results for the 9 7/8” liner in this well can be found in Table Apx E‐4 and Table Apx
E‐6. A conventional system is shown for comparison, and reverse designs were modified for two TOC
scenarios: at 26,500’ MD, which was 500’ below the previous shoe, and at 21,000’ MD (liner hanger).
Table Apx E‐3: Well #1, 9 7/8” Slurry Design Comparison
E‐11
Table Apx E‐4: Well #1, 9 7/8” Liner Slurry Test Results Comparison
Free fluid, fluid loss, rheology and settling results are acceptable for all slurry systems. Thickening time
results are shown in Figure Apx E‐14 for the TOC at 26,500’ scenario. For reverse placement, the
required placement time is about 20 minutes shorter than conventional placement. This allows for both
the thickening time to be reduced slightly. Thickening time results for the TOC at 21,000’ scenario are
shown in Figure Apx E‐15. Here, the required placement times for the shoe slurries are similar. The
thickening time for the reverse TOC slurry could be reduced slightly by approximately an hour to set up
closer to the shoe slurry system.
E‐12
Figurre Apx E‐14: We
ell #1, 9 7/8” Lin
ner thickening ti me results, TOC
C @ 26,500’ MD
Figurre Apx E‐15: We
ell #1, 9 7/8” Lin
ner thickening ti me results, TOC
C @ 21,000’ MD
Compresssive strength results at th he shoe are shown in Figgure Apx E‐116. At the shoe temperaatures
during thiis WOC perio od were highe er for the revverse slurries due to the h higher circulaating temperaatures
during plaacement. Fo
or the two reverse designss, compressivve strength ddeveloped aftter about 5 h
hours,
compared d to approxim
mately 12 ho ours for the conventionall slurry desiggn though the convention nal 24
hour compressive strength was high her overall.
E‐13
Figure Apx E‐16: Well #1, 9 7/8” liner comp
pressive strengtth results at the shoe
Figure Apx E‐17: Well #1, 9 7/8” liner comp
pressive strengtth results at the TOC. Note: Con
nventional slurryy TOC is 26,500’’
E‐14
Slurry designs and results for the 7 5/8” production liner in this well can be found in Table Apx E‐6 and
Table Apx E‐6. A conventional system is shown for comparison, and reverse designs were modified
placement with TOC at 27,500’
Table Apx E‐5: Well #1, 7 5/8” Slurry Design Comparison
Schematic #1 7 5/8” Liner, TOC = 27,500’
Table Apx E‐6: Well #1, 7 5/8” Slurry Test Results Comparison
Schematic #1 7 5/8" Liner, TOC = 28,500'
Reverse First Sack Reverse Last Sack
Conventional
(Shoe) (TOC)
E‐15
Free fluid, fluid loss, rh
heology and ssettling results are accepttable for all slurry systemss. Thickeningg time
results are shown in Figure Apx E‐1 18. For reverrse placemennt, the requirred placemen nt time is abo out 20
minutes shorter
s than conventionaal placement so the requuired thickenning times fo
or the slurriees are
similar.
Figure Apx E‐18: We
ell #1, 7 5/8” liner thickening tim
me results, TOC
C @ 28,500’ MD
Compresssive strength results at the shoe are sh
hown in Figuure Apx E‐19. The temperrature profilee over
the 24 ho
our WOC perio od are compaarable for revverse and connventinal placcement. For this design,o
overall
compresssive strength developmentt at the shoe is lower for tthe reverse slurry. This m
may require fu
urther
optimization dependinng on desired strength.
Figure Apx E‐19:
F Well #1, 7 5/8” liner compress ive strength ressults at shoe
E‐16
Compresssive strength results at th
he TOC are shown
s in Figuure Apx E‐200. The reversse slurry devvelops
compresssive strength e
earlier than the reverse slu urry and overrall strenth affter 24 hours is similar.
Figure Apx E‐20:
F : Well #1, 7 5/8”” liner compresssive strength ressults at TOC
Table Apx E
E‐7: Well #2, 16”” Liner Slurry De
esign Compariso
on
Schem
matic #2 16" Liner, TOC = 12,500' 16” Lineer, TOC = 5,500’
Reversee First Sack Reverse Last Sack
Conventio
onal Reverse
(SShoe) (TOC)
Cement Class H
H Class H Claass H Class H
Yield (ft3/sk) 1.07 1.07 11.07 1.08
Antifoam (gal/sk) 0.02 0.02 00.02 0.02
Dispersannt B (gal/sk) 0.03 0.03 00.03 0.03
Fluid Lo
oss Control
0.10 0.12 00.12 0.12
Additivee A (gal/sk)
Free Water Control
‐ 0.01 00.01 0.02
Agent (gal/sk)
Retarder A (gal/sk) 0.04 0.004 00.01 ‐
Deionized W
Water (gal/sk) 4.20 4.20 44.20 4.02
Potassium Chloride (KCl)
3.00 3.00 33.00 3.00
% B
BWOW
Calcium Chhloride (CaCl2)
‐ ‐ ‐ 0.25
(ga
al/sk)
E‐17
Table Apx E‐8: Well #2, 16” Slurry Test Results Comparison
Reverse First
Conventional Reverse Reverse Last Sack (TOC)
Sack (Shoe)
Free fluid, fluid loss, rheology and settling results are acceptable for all slurry systems. In conventional
placement, most of the slurry is exposed to the BHCT as it passes the shoe during placement. This
exposure to temperature, as well as the additional pump time requires a longer thickening time to
assure placement. When the same volume of slurry is placed by RCPC, the required pump time is
shortened and only the slurry near the shoe is exposed to the BHCT. This results in a shorter required
slurry thickening time, shown in Figure Apx E‐21, since the simulated placement time of the reverse
slurry is approximately 2:30 (h:min) shorter than that of conventional placement. Thickening time
results for the TOC at 5,500’ scenario are shown in Figure Apx E‐22. There would be challenges to
design a conventional system for this string to bring cement up to the liner hanger; simulation indicated
that the pressure needed to lift the cement to this height resulted in breakdown of the formation. The
conventional design from the 12,500’ TOC is shown for reference.
E‐18
Figure
e Apx E‐21: Well #2, 16” Liner Th
hickening Time CComparison, TO
OC @ 12,500’ MD
Figure Ap
px E‐22: Well #2
2, 16” Liner Thickening Time Comparison, TOC @@ 5,500’ MD. N
Note: Conventio
onal result is forr the
12,5
500’ MD system and is shown foor comparison.
E‐19
Figure Apx E‐23
3: Well #2, 16” liiner compressivve strength resu
ults at shoe
mpressive strength
Compresssive strength results at the TOC are shown in Figgure Apx E‐224. The com
developen
nt of the re
everse slurry for the 5,50
00’ TOC sce nario is sign
nificantly lowwer, and could be
optimized
d further if gre
eater strength is required..
Figure Apx E‐24
4: Well #2, 16” liner compressivve strength resu
ults at TOC
E‐20
Table Apx E‐9: Well #2, 13 5/8” Casing Slurry Design Comparison
Schematic #2 13 5/8” Casing, TOC = 5,500'
Reverse First Sack Reverse Last Sack
Conventional Lead Conventional Tail
(Shoe) (TOC)
Cement Class H Class H Class H Class H
3
Yield (ft /sk) 1.3 1.07 1.07 1.07
Antifoam (gal/sk) 0.02 0.02 0.02 0.02
Dispersant B (gal/sk) 0.05 0.03 0.03 0.03
Fluid Loss Control
0.24 0.10 0.15 0.15
Additive A (gal/sk)
Free Water Control
0.01 ‐ 0.01 0.01
Agent (gal/sk)
Retarder A (gal/sk) 0.10 0.07 0.02 0.01
Deionized Water (gal/sk) 5.68 4.18 4.15 4.16
Potassium Chloride (KCl)
3.00 3.00 3.00 3.00
% BWOW
Table Apx E‐10: Well #2, 13 5/8” Casing Slurry Test Results Comparison
Schematic #2 13 5/8" Casing, TOC = 5,000
Free fluid, fluid loss, rheology and settling results are acceptable for all slurry systems. Thickening time
results are shown in Figure Apx E‐25 and Figure Apx E‐26. For reverse placement, the required
placement time is shorter than conventional placement so the required thickening times for the slurries
are similar.
E‐21
Figure Apx E‐2
25: Well #2, 13 5
5/8” Casing Thic kening Time Co
omparison
ell #2, 13 5/8” Caasing Thickeningg Time Compari son, TOC Slurryy
Figurre Apx E‐26: We
Compresssive strength results at the
e shoe are sho
own in Figuree Apx E‐27.
E‐22
Figure Apx E‐‐27: Well #2, 13 5/8” Casing Com
mpressive Strenngth Developmeent Comparison at Shoe
Compresssive strength results at the
e TOC are sho
own in Figure Apx E‐28.
Figure Apx E
E‐28: Well #2, 13
3 5/8” Casing Co
ompressive Strenngth Developmeent Comparison
n at TOC
Slurry dessigns and resu
ults for the 11 7/8” liner in this well ca n be found in
n Table Apx EE‐11 and Tablle Apx
E‐12.
E‐23
Table Apx E‐11: Well #2, 11 7/8” Liner Slurry Design Comparison
Schematic #2 11 7/8" Liner, TOC = 21,500
Table Apx E‐12: Well #2, 11 7/8” Liner Slurry Test Results Comparison
Schematic #2 11 7/8" Liner, TOC = 21,500
Thickening time results are shown in Figure Apx E‐29 and Figure Apx E‐30.
E‐24
Figure Apx E‐29: Well #
#2, 11 7/8” Line
er Thickening Tim
me Comparison, TOC 21,500’ M
MD
Figure A
Apx E‐30: Well #
#2, 11 7/8” Linerr Thickening Tim
me Comparison, TOC = 20,500’ M
MD
Compresssive strength results at the
e shoe are sho
own in Figuree Apx E‐31.
E‐25
Figure Apx E
E‐31: Well #2, 11
1 7/8” Liner Com
mpressive Strenggth Development Comparison aat Shoe
Compresssive strength results at the
e TOC are sho
own in Figure Apx E‐32.
Figure Apx E
E‐32: Well #2, 11 7/8” Liner Com
mpressive Strenngth Developmeent Comparison at TOC
E‐26
Appendix F ‐ Deepwater RCPC Operational Guidelines
The introduction of reverse‐circulation primary cementing (RCPC) operations to deepwater is expected
to introduce many operational challenges. Operational considerations are anticipated to be similar to
those of a conventional deepwater job, however due to the specific challenges and configuration of a
RCPC job there will be some variances. Contingency plans are also included.
1) Pre‐Job Planning ................................................................................................................... F‐3
1.1) Equipment Selection .................................................................................................................. F‐3
1.2) Centralizers ................................................................................................................................ F‐3
1.3) Crossover tool ............................................................................................................................ F‐4
1.4) Cementing with a port collar run below a liner hanger: ............................................................ F‐5
1.5) Multiple port collars run in the casing ....................................................................................... F‐6
1.6) Other Considerations ................................................................................................................. F‐6
1.7) Float equipment ......................................................................................................................... F‐7
1.8) Wiper assembly for drill pipe ..................................................................................................... F‐8
1.9) Reamers ..................................................................................................................................... F‐8
1.10) Simulations ............................................................................................................................. F‐8
1.11) Fluid design ............................................................................................................................ F‐9
1.12) Lab testing .............................................................................................................................. F‐9
1.13) Procedure ............................................................................................................................... F‐9
1.14) Contingency Scenarios ......................................................................................................... F‐10
2) Logistics and Maintenance ................................................................................................. F‐10
2.1) Equipment to and on location: ................................................................................................ F‐10
2.2) Materials to location ................................................................................................................ F‐10
3) Personnel ............................................................................................................................ F‐10
3.1) Training .................................................................................................................................... F‐10
3.2) Number of Personnel ............................................................................................................... F‐10
4) Hole Conditioning ............................................................................................................... F‐11
4.1) Considerations for high deviation wells ................................................................................... F‐12
4.2) Contingency: Pre‐Job Losses .................................................................................................... F‐12
4.3) Contingency: Cuttings or Debris Blocking Equipment Flow Path ............................................ F‐12
5) Running Equipment In hole ................................................................................................ F‐13
F‐1
5.1) Centralizers .............................................................................................................................. F‐13
5.2) Liner Hanger ............................................................................................................................. F‐13
5.3) Crossover tool .......................................................................................................................... F‐13
5.4) Float equipment ....................................................................................................................... F‐13
5.5) Wiper balls and/or darts for drill pipe ..................................................................................... F‐13
5.6) Equipment triggers ................................................................................................................... F‐13
5.7) Others ...................................................................................................................................... F‐14
5.8) Contingency: Well Control Situations ...................................................................................... F‐14
6) Rig Up .................................................................................................................................. F‐14
7) Pre‐Job Procedure .............................................................................................................. F‐14
8) Job Execution ...................................................................................................................... F‐15
8.1) Downhole Equipment Operation – Operating the Crossover Tool .......................................... F‐15
8.2) Pre‐Job Well Control Considerations ....................................................................................... F‐15
8.3) Job procedure recommendations ............................................................................................ F‐16
8.4) Pipe movement ........................................................................................................................ F‐16
8.5) Additive proportioning ............................................................................................................. F‐17
8.6) Mix and Pump Rates ................................................................................................................ F‐17
8.7) Density Control ........................................................................................................................ F‐17
8.8) Downhole Equipment Operation – Operating the Crossover Tool .......................................... F‐18
8.9) Downhole Equipment Operation – Operation of the Floating Equipment .............................. F‐18
8.10) Downhole Equipment Operation – Setting the Liner Hanger .............................................. F‐19
8.11) Contingency: Equipment Issues ........................................................................................... F‐19
8.12) Contingency: Unplanned Shutdowns ................................................................................... F‐21
8.13) Contingency: Lost Returns During Placement ..................................................................... F‐21
8.14) Contingency: Circulate out Cement ..................................................................................... F‐21
8.15) Contingency: General Well Control ..................................................................................... F‐21
9) Safety and Environment ..................................................................................................... F‐21
9.1) Safety Considerations .............................................................................................................. F‐21
9.2) Environmental Considerations ................................................................................................. F‐22
9.3) Personnel and Training ............................................................................................................ F‐22
10) Post‐Job Evaluation ............................................................................................................ F‐22
F‐2
1) Pre‐Job Planning
1.2) Centralizers
As with a conventional cement job, centralization will play a key factor during the run in process as well
as during the cementing operation for optimum placement. Required standoff need to be taken into
account for centralizer selection and positioning. Low standoff can lead to poor displacement in the
open hole section and possible mud channeling. In reverse placement poor standoff and channeling
could contribute to inadequate zonal isolation at the shoe, since the leading edge of cement being
placed is the shoe cement. Because the casing is normally not 100% centered in the annulus there will
be a resulting eccentric flow in the cross sectional area of the annulus. The velocity and pressure profiles
are related to the stand‐off percentage and the differences in rheology between the various fluids. The
widest portions of the annulus will have the highest flow rate with the converse being true on the
narrowest side, and the velocity/volume differential will be most critical at the shoe. Increasing the
percent standoff between the casing and the annulus will reduce the risk of a wet shoe due to poor
displacement or reduce the required displacement volume up inside the shoe track to ensure cement
coverage over the entire annulus at the shoe.
Centralizer selection should be based on the clearance between casing strings, e.g. OD of the liner and
the ID of the previously run casing string. Also, if the liner is moved while placing cement, e.g.
reciprocation and/or rotation, then this will influence the type of centralization, if any, that can be used.
Casing or liner size, along with the annular clearance with the previous string limits the types of
centralizers available. The liner string can be made up of flush joint with no collar, or with a standard
box and pin joint. Conventional bow spring centralizers that are hinged and are designed to go over a
conventional casing collar may be too large to fit in tight clearances. This also applies to slip on ridge
collars and bow spring centralizers that are designed to go on the OD of the casing and to ride between
collars. Rigid centralizer subs and bow spring centralizer subs have been developed to meet the
requirement for tight annular clearance. The subs screw into the pipe and become part of the casing.
Rigid centralizer subs are designed to fit the drift ID and drift length (generally 6” or 12” depending on
casing size) of the previously run casing to give some standoff in the lap area between the two casing
strings. Bow spring centralizer subs have the bow spring centralizers designed go thru tight drift IDs and
centralize in open hole.
If the operator wants to move the pipe during cementing operations, e.g. rotation and/or reciprocation,
then centralizer selection is very critical. Most current centralizers are not designed for planned
rotation. Slip on bow spring and slip on ridge centralizers are better suited for rotation than the hinged
type centralizers. However, if the ID of the centralizer collar that slips on the OD of the pipe is not the
proper material or is not properly coated and/or sleeved rotation could cause the centralizer to wear
out and/or wear a hole in the casing. (This scenario assumes the pipe will rotate, while the centralizers
F‐3
stay stationary.) In cases where the centralizer is rotated with the pipe, similar to some drill pipe
protectors, then the classic bow spring centralizer would be a poor choice. In cases where the centralizer
is rotated with the pipe, ridge type centralizers should be used. Close attention to material selection
and/or coatings must be done to prevent the blades of the centralizer from wearing out.
When rotating and/or reciprocating pipe with centralizers the operator must consider how the
centralizers will endure movement along the open hole section Each operator needs to go over the
possible scenarios and determine what type of centralization is appropriate for each liner‐hole situation.
A “single” standard secondary option may not be appropriate each situation.
One of the criteria that the crossover tool must have is that it allows flow in the conventional direction,
e.g. down the work string, thru the ID of the liner, and back up the annulus area, while going in the hole.
With the liner in place a pack‐off between the work string to annulus port and the liner returns exit port
is set (typically mechanically with a ball or dart or through rotation). Setting the pack‐off also switches
flow direction from conventional to reverse flow. With the cement in place the crossover tool is
switched back to the conventional flow direction and the pack‐off is un‐set so that the liner hanger can
be set, or the casing hung off an existing shoulder.
Crossover tool control
A more advanced version of the crossover tool will allow pipe movement throughout the job, e.g.
reaming the liner in the hole, switching from reverse to conventional circulation on demand, e.g. allows
fluid pumped down the annulus side while reaming in with returns being taken up the ID of the liner.
This system uses batteries to power a hydraulic pump(s) to shift the sleeves that control the direction of
flow in the tool. An on board computer will interpret commands and control the start and stop of the
pump(s), as well as the opening and closing of control valves. There are various ways to get commands
to this system. These include, but are not limited to, RFID tags injected into the flow stream; wired drill
pipe (Note: Wired drill pipe transmits data, it currently does not supply power to operate equipment.);
Pressure pulse (flowing); Mud‐Pressure pulse (closed chamber); Unique chemical signal(s) from fluid
pumped through the tool; external and/or internal wire line and/or fiber optic line; external, internal,
and/or concentric drill pipe with more than one flow path from the surface that allow hydraulic fluid to
be pressured up to signal the computer, and/or power the sleeves open and/or closed; and on board
timers that tell the computer when to act and what to do.
The advantages to the advance crossover tool system is it gives the operator more options, e.g. allows
reaming in a liner while pumping in the reverse mode, not just placing cement. The disadvantages are
the system uses batteries that have a limited life and are temperature limited. The system is slow to
operate, e.g. a low volume high pressure pump moves fluid slowly – typically 1 cubic inch of fluid per
minute at 5,000 psi differential pressure. To shift a sleeve 5” may take 10 minutes. Unless wired drill
F‐4
pipe, electric/fiber optic lines, or hydraulic pressure is used there is no feedback to the surface that the
tool has shifted to a particular position.
Crossover tool with mechanical separators in work string
The minimum crossover tool is relatively simple to operate with balls and/or darts. It has the basic
switching function from conventional to reverse and back to conventional. A standard liner hanger can
be set with this tool.
The disadvantages of this system are the pack‐off will not allow pipe movement after the pack‐off is set.
The system will only switch once from conventional to reverse cementing. When the system is used on a
liner the system must switch from reverse mode to conventional mode, and the operating balls and/or
darts must be move out of the way so the work string can be used to pressure up and set the liner
hanger. In conventional cementing, darts are used to latch into plugs located inside the casing and
shear them and pump them to bottom. Since plugs will not be usable with RCPC, any darts and balls will
be used for the work string section only. Operating balls will have to pass through the crossover tool or
be used to function that tool and conventional wiper darts run in front and behind the cement may
present a problem with the operation of this tool.
In summary the crossover tool system run on top of the liner can be a simple ball/dart operated system
that will shift a sleeve to a given position when the ball/dart lands with positive feedback to the surface
the tool has shifted. Or a more advance system can be developed that will allow multiple shifts of the
tool and allow the operator more options on how to cement the well.
This system places a port collar below a liner hanger or casing hanger. The casing is first hung off by
setting the liner hanger or hanging it off the preinstalled casing hanger. With the casing hung off and the
running tool released from the casing, the running tool is then used to shift open a port collar. The port
collar may or may not have a pack‐off above it that is set with pressure. The pack‐off would force all
fluid pumped down the work string to go down the liner annulus. The casing hanger or the liner hanger
may have a built in pack‐off to do the same function. Returns are taken from the ID of the liner and the
returns exit to the annular area above the liner hanger or casing hanger and the work string. A float
collar with a flapper valve is located near the port collar that’s held open with a stinger on the running
tool.
Some of the advantages of the port collar system is the casing is already hung off before any cement is
pumped. Cementing takes place immediately after the liner is in place. The port collar can be run lower
in the string, if desired. Conventional balls and/or darts can be used to separate the fluids, because the
balls and/or darts will be caught in a dart/ball catcher located in the work string, before or/just after the
cement exits the work sting into the annulus area.
F‐5
Some of the dis‐advantages of the port collar system is the sleeves in the port collar tend to have a
lower collapse rating than the casing. The seal system(s) used in some port collars may not meet the
operator’s needs for permanently sealing over the life of the well. Also, a hole in the casing below the
liner hanger is not acceptable to some operators.
This system allows the operator to do a conventional cement job, where the cement is pumped down
the work string into the casing ID, and exit the bottom of the casing and travels up the annular area. The
port collar(s) allow the operator to squeeze off the annular area of the liner, after the primary cement
job. Once the liner is hung off and the primary cement job is complete the work string along with the
liner running tools is retrieved. The port collar operating tool system is then run in the hole on a
separate run of the work string.
The advantage of the port collar system is it has flow openings in the casing where cement can be
squeezed into, and the port collar(s) can be located at strategic locations, and then permanently closed.
The use of the port collar(s) to squeeze cement into the annular area can be done after the primary
cement job has been fully evaluated, e.g. cement bond logs are run. Some systems require that only one
port collar can be squeezed through on a single run in the hole. Other systems offer the opportunity of
operating multiple port collars on a single trip in the hole. The biggest advantage of this system is it uses
existing technology, can be deployed relatively easily, and tool personnel are available that are familiar
with the system.
The biggest disadvantage of this system is the rig time it takes to execute the squeeze cementing of the
liner. Some of the other disadvantages of the port collar system is the sleeves in the port collar tend to
have a lower collapse rating than the casing. The seal system(s) used in some port collars may not meet
the operator’s needs for permanently sealing over the life of the well. Also, a hole(s) in the casing below
the liner hanger is not acceptable to some operators.
F‐6
Any number of ideas can be explored for diverter tools and for crossover tools. As in all applications the
ideas must be developed into a feasible tool systems that can be applied within the parameters of the
requirements for a cementing system that will effective isolate the casing from the open hole.
Another consideration is restricting the inlet ID at the bottom of the liner with floating equipment will
increase the ECDs seen on the annulus side of the casing when cement is being pumped in the reverse
direction.
Conventional floating equipment is seen by some operators as having a dual purpose: primarily to hold
cement in place after the primary cementing job, secondly acting as an aid in down hole well control
while the casing is being run in the hole. (Floating equipment is not a down hole blow out preventer. It
will restrict some flow into the casing, primarily liquid – not gas. It is not intended to exceed the rated
designed working pressure – generally 5,000 psi or less.)
To meet the requirements for reverse cementing a flapper valve type float shoe/collar can be run at or
near the bottom of the casing. The flapper valve is then opened with a stringer run on an inner string
run below the running tool. The float collar may have opposed flapper valves with the upper valve being
held open with a sleeve that is released by the stinger, and the stinger holds the upper and lower
flapper valves open when it stings through both valves. (The opposed flapper valve float equipment
keeps fluid moving from either direction after the stringer is removed from the float collar.) The float
collar will require at least one joint of casing below it to allow for the stinger length.
In order to keep the ECD to a minimum during cementing the float collar may be run higher in the
casing, e.g. just below the liner hanger/casing hang off collar. To meet the requirement for a down hole
well control valve a pump out float valve or solid plug could be used. The pump out float valve will be
pumped out with a dropped ball after the liner is below the blow‐out preventer BOP with drill pipe in
the BOP. The solid plug is simply pumped out, after the liner is below the BOP. Both the pump out float
valve and the solid plug should be designed so that they will not interfere with the running of the liner
further in the hole, and have no chance of re‐plugging the shoe if the shoe is on bottom.
These are viable floating equipment options determined during the Deepwater RCPC project. Some
operators may choose to run no floating equipment at all in order to reduce the ECD’s to a minimum. A
minimum recommendation is running a float collar with a flapper valve just below the liner
hanger/casing hang off collar. Float equipment used for reverse applications are still independent of the
F‐7
liner hanger running tool and the operator can make the decision on‐site where the float collar is
located based on well conditions.
1.9) Reamers
If the liner is to be reamed into the hole then a full open reamer shoe may be considered. The reamer
shoe will allow the liner to go through ledges and oval sections of the drilled hole more easily.
1.10) Simulations
Currently, commercially available simulators can be used up to a point for deepwater RCPC operations.
Temperature estimates of BHCT can be obtained through workarounds within the software. Other
information may be estimated, depending on the simulator capabilities. Future versions of
commercially available simulators may have extended abilities. Simulators should be used as a tool to
determine if there will be an ECD advantage to using reverse placement for each job, since this is
dependent on the location of the critical point and weak zones downhole.
Initial Input Data Needed: Depth, formation lithology and pore/frac gradients, hole geometry and
deviation, BHST, pilot fluid information, location of crossover, floats or other flow constrictions.
Simulation Outputs: BHCT, first and last sack temperatures vs time, temperature differentials between
the top and bottom column of cement, WOC temperatures, u‐tubing effects, acceptable pump rate
ranges, ECDs, pressures on surface, pressure at weak zones downhole, centralization, displacement
efficiency, spacer contact time.
During displacement, the volume of cement taken into the casing prior to cessation of pumping is
dependent on the volume difference between the widest and narrowest portions of the annulus at the
shoe. The cement on the widest side of the annulus arrives at the float shoe first. Cement displacement
in the narrow portion arrives some time and volume later. The difference in volume between heads of
the fastest and slowest profile will be the minimum volume that will be returned to, and will remain
inside the casing. The larger the stand‐off at the shoe, the less volume will be required to be circulated
inside the casing to fully cover the shoe annulus. Simulations using the expected standoff and annular
geometry at the shoe can be used to determine the estimated displacement needed to ensure full
coverage at the shoe.
F‐8
Key aspects and best practices for thickening time, free fluid, fluid loss, gel strength development and
slurry stability are expected to be the same as for conventional slurry design. The most significant
design modification for RCPC will be the change from the conventional density hierarchy. It is
recommended that the density variation of cementing fluids (including spacer) be kept minimal from the
mud density. While fluids of increasing density could be pumped, the density differential increases the
risk of fluid swap as fluids are pumped down since reverse placement in the annulus is assisted by
gravity. Rheological hierarchy of fluids will be critical for RCPC slurry designs, and as with conventional
placement, fluids should be more viscous than the preceding fluid to aid displacement.
Use Spacers and/or Washes ‐ Sufficient volumes of spacer and/or wash must be pumped for effective
removal of drilling fluid. Rule of thumb: Pump Spacer / Wash volume which provides 8 to10 minute
contact time in annulus. Increasing the contact time may be more beneficial. If the hydrostatic in the
annulus does not fall below the minimum needed to control pore pressure, a large volume of Newtonian
fluid will be beneficial.
1.13) Procedure
Prepare a job plan that includes at least the following:
a) Rig up procedure, to include reverse circulation tools.
b) Safety concerns.
c) Pressure testing procedure.
d) Spacer type, density and volumes to be pumped.
e) Wiper dart or ball dropping sequence and procedure, if used for work string
f) Downhole equipment functioning: (crossover/floats)
g) Cement slurry formulation(s) densities and volumes.
F‐9
h) Conversion factors for sacks per barrel of slurry and barrels of slurry per barrel of mix water.
i) Pumping schedule indicating rates, volumes and time for each fluid including displacement.
j) Total job time including time to operate tools and flush lines
k) Anticipated job pressures during pumping
l) Hydrostatic pressures in the hole of each fluid after placement.
m) Personnel requirements for the job.
3) Personnel
3.1) Training
Preparing personnel for the reverse operation will be specific to the type of equipment that is used
and/or developed for the operation. The company that supplies the specialized equipment can provide
the personnel to oversee the running and operation of the equipment. Possible contingency situations
need to be reviewed and plans should be covered as part of personnel training/pre‐job procedure. The
complete function of RCPC tools and related operations should be discussed with the client and all
personnel involved in the operation prior to the job. This information also should extend to personnel
indirectly involved with the reverse cementing operation (i.e. involved rig personnel).
F‐10
of: operator, service company and rig personnel required are expected to be the same as for a
conventional cement job. Additional specialists may be required on location to operate tools and
equipment specific to a reverse‐cementing operation.
4) Hole Conditioning
There are two potential flow paths for running casing and hole conditioning. The first is conventional
circulation and the second is reverse circulation. If the casing is run and the mud is conditioned
conventionally, then the mechanical tools for a reverse cementing operation will need to undergo
conversion to allow for the change in flow directions. The second potential flow path is to run casing
and circulate in the reverse direction. In this configuration, conversion of equipment is not necessary
once reverse‐circulation equipment has been run in hole.
As with a conventional primary cement job, two bottoms up circulation is recommended to clear the
annulus of as much debris as possible. Total circulation is dependent on the prevalent hole conditions
encountered. Hole conditions will most probably dictate when to switch circulation directions. Static
time should be minimized when switching circulation between conventional and reverse flow directions.
Up to one additional full hole volume circulation in the reverse direction may be recommended to
ensure the flow path is unrestricted in that direction.
Another option is reverse circulating with the casing as it is being run in the hole, and continuing to
condition in the reverse flow configuration. Because the casing normally will contain more volume per
foot of depth than the annulus, the flow rate inside the casing may not be able to carry the cuttings as
efficiently, especially if the hole angle is high at the point of initial reverse circulation. A developing
volume of cuttings that remain inside the casing due to a lower fluid carrying velocity is currently not
expected to have an adverse effect on circulation, or running casing, unless conventional circulation
becomes necessary for some reason. If conventional becomes necessary, then there is the possibility
that cuttings, unable to be removed due to a lower flow velocity inside the casing from reverse‐
circulation operations, can block float equipment. In this case, precautions should be taken to ensure
that the hole has been thoroughly cleaned, or that there is a down hole mechanical apparatus capable
of preventing the cuttings from reaching the floats.
For the most part, stationary cutting deposits are anticipated to be relatively shallow, and the amount of
captured material may not pose a significant problem. Planned periodic circulations of 50 to 100 barrels
as conducted in conventional operations will most likely prove to be effective in keeping the annulus
clear, and the pipe free. Tight spots, pack‐offs, and similar problems will have to be dealt with as they
occur.
The hole should be conditioned in the conventional manner for better cleaning out the cutting or any
sediment in the hole since the fluid will not have the velocity up the casing to carry the cuttings to
surface. After hole is clean then operate the diverter tool and pump one more bottoms up to confirm
you are ready to perform your cement job as planned.
F‐11
4.1) Considerations for high deviation wells
With conventional hole cleaning, circulation is conducted from the bottom of the hole to surface and
the force of gravity is working against cuttings removal. Mud carrying capacity and circulation velocity
due to hole geometry, will need to be taken into consideration. In conventional operations a potential
issue caused by in‐hole debris is stuck pipe, or a plugged annulus when running casing. The same will be
true with reverse circulating operations.
Lost circulation, particularly near a previous shoe, can present a significant problem in obtaining
adequate coverage over critical areas. Full losses at the previous shoe indicate a wet shoe risk; if the
cement is lost to the formation at this point during reverse placement lower zones in the annulus,
including the shoe area, may not be covered. This is a concern especially if there are producing zones
below the loss zone. In this situation a contingency switch to placement through conventional
circulation would be recommended; thought the TOC will be at the loss zone and may be lower than
designed, at a minimum the critical producing zones will be covered. Overall, placement method
selection will be dependent on the location of the loss zone and critical area of coverage and having the
option to pump conventional or reverse circulation based what the loss zone dictates may be beneficial.
If the flow path becomes blocked during conventional circulation, one option is to apply pressure to
pump out the debris. The maximum pressure to be applied should be determined beforehand, and
must not exceed the casing collapse or burst pressure, equipment pressure ratings, or formation
fracture pressure.
If the flow path becomes blocked during reverse circulation, one option is to apply pressure to pump out
the debris. The maximum pressure to be applied should be determined beforehand, and must not
exceed the casing collapse or burst pressure, equipment pressure ratings, or formation fracture
pressure. If a switchable crossover is used, then the flow path can be converted back to the
conventional direction to attempt to dislodge the debris by pumping in the opposite direction.
If circulation cannot be regained, the blocked tools must be pulled out of the hole and inspected. If
cuttings are expected to be an issue with the reverse flow direction, mechanical screens or similar types
of equipment could be installed prior to running casing.
F‐12
5) Running Equipment In hole
5.1) Centralizers
Depending on the type of centralization selected, a specialized service person may be needed for the
proper installation of the centralizers. Installation of centralizers will be the same as for a conventional
cement job.
If RFID tags are to be used then the launcher to inject the tags into the system needs to positioned so
that it can perform its function, but not interfere with other operations on the rigs. If the tags are pre‐
loaded, then verification of the proper tag and sequence needs to be done.
If chemical triggers are to be used then the proper volume of each trigger needs to be verified, and the
way the trigger(s) will be pumped into the flow stream.
If hydraulic lines are used then the running procedure and the proper way of attaching it to the OD of
the pipe needs to be reviewed.
If balls and/or darts are to be launched to operate the crossover then the proper launcher needs to be
available, and the loading of the balls and/or darts needs to be verified.
If electric wire line and/or fiber optic line is to be used then the installation procedure, if run on the OD
of the work string, needs to be verified. If the line is run in the ID of the drill pipe then the proper
methods of running it inside the drill pipe need to be reviewed.
F‐13
If mud pulse is to be used then the proper pumps needed to generate the pulse need to be identified.
5.7) Others
A full open reamer shoe could be used on the end of the casing to allow reaming into the hole. These
can be purchased as a saw tooth shoe, Texas pattern, tiger tooth pattern, etc. These shoes allow the
operator to get ledges and oval sections of the hole more easily than simple circulation and push
through the restrictions.
6) Rig Up
The rig’s permanent equipment lay out, will be the primary factor in deciding how and what additional
equipment is necessary to rig up for the cementing portion. Rig up to perform a reverse cement job will
still entail all of the same lines and valves required for a conventional job. The cement and spacer will
still originate from the cement pumping equipment or from the rig pits if the spacer(s) were mixed there
and may be pumped by the rig. All of the flow reversal will be performed by the tool(s).
One possible flow path for reverse circulation is through the BOP stack lines and down the annulus with
returns taken from the drill pipe. However, due to safety considerations, use of this flow path through
the BOP stack was not considered as an operational possibility during the course of this project.
7) Pre‐Job Procedure
1. Prior to performing the cement job, obtain up‐to‐date well information and cementing objectives
from the operator’s Drilling Supervisor. This should include, hole size with caliper data or pre‐
determined and calculated excess factor, casing length, size and weight, drill pipe length, size and
weight shoe track dimensions, required length of tail cement, desired top of cement, liner hanger
configuration, offset well data, and/or any other pertinent information. If no caliper is run on the
section, an arbitrary excess volume of cement may preclude getting cement all the way to the
shoe, or place a larger than anticipated volume of cement into the casing. Empirical data from
drilling, or from off‐set wells may allow a given or arbitrary percentage to be used. There are also
the times when a caliper will go to maximum extension and the actual dimensions of the hole will
be in question. A trade off situation may be necessary.
2. The pre‐job meeting shall include all service personnel (mud loggers, etc.) that can provide
information or add value to the process.
3. Check equipment and material requirements and ensure that all required equipment will be on
location for the job.
4. Review cement and spacer formulations with cementing coordinator/drilling engineer.
5. Review lab test results from the rig samples with the cementing coordinator paying special
attention to the thickening time (available pumping time) and the required WOC (wait on cement
time, time to a minimum of 500psi).
F‐14
6. Calculate cement volumes in barrels, in cubic feet and by sacks of cement required for the lead
and tail formulations. This should include a breakdown of these volumes based on cased hole,
open hole and shoe track volumes.
7. Calculate mix water requirement, additive requirements and the resultant mix fluid volume for
both lead and tail cement slurries.
8. Calculate spacer volumes and material requirements, taking into account available mixing space.
9. Check available water supply and ensure that sufficient quantities will be available for the job.
10. Verify that the suction rates required for the job can be achieved on both drilling fluid and water.
11. Calculate displacement volumes for casing or liner and drill pipe as required.
12. Develop a pumping schedule based on the cement job simulator output.
13. Determine if the pumping time, as indicated by the laboratory thickening time test result, is
sufficient for the planned job.
14. Cementing Contractor equipment will be used for all mixing, pumping and displacement.
8) Job Execution
Depending on the operator the floating equipment will be placed at the desired location in the casing.
Generally near the liner hanger if ECDs are a major concern, or near the bottom, if well control is a
concern. The type of equipment recommend is a flapper valve that can be opened with a stinger run on
the bottom of the running tool that attaches the liner to the work string. The floating equipment can be
opposed flapper valves so that when the stringer is removed it will have shut off in both directions. The
flapper valve(s) can be held on by tubes inserted in the ID of the flapper valves before the tube(s) are
removed by the stinger. If well control is desired then the lower flapper is run in the hole in the closed
position. If surge reduction is desired them the tube allow auto fill of the casing. The stinger will also
allow surge reduction if desired. (A surge reduction tool can be run above the crossover tool, as long as
the closing ball and/or dart, if used, will go through the ID of the stinger and exit to the bottom of the
hole. The floating equipment will remain open during the entire time the cement is pump, and will only
close after the running tool has been released from the liner and the stinger is extracted from the
floating equipment.
F‐15
to allow reverse flow, a pressure profile should be acquired by circulating at a slow rate to determine a
kill rate.
F‐16
8.5) Additive proportioning
In general, additive proportioning and delivery considerations for a reverse‐circulation primary cement
job on offshore rigs are expected to be the same as those for a conventional primary cement job. The
unique perspective of reverse cementing may allow for variances in the cement program not practical in
a conventional cementation. Liquid additives from drop tanks, or an automatic liquid additive injection
system, will make it possible and practical, depending on the individual set up of the offshore unit.
Computer controlled liquid additive systems meter a precise amount of chemical into the displacement
tank. The most sophisticated utilize process control units that are tied to the mixing rate and
automatically meters and inject chemicals directly into the mix water feed line in the correct proportion
to match the rate at which the cementer is mixing the cement. This type of automatic additive unit
would simplify operations if the cementer were to mix multiple slurry systems with staged additive
formulations for RCPC.
In general the spacer/wash should be pumped a high rate for mud removal, and the wash should be
pumped in turbulent flow if possible. During cement mixing, density control should not be sacrificed for
pump rate. Depending on the cement characteristics, the last 25 bbls of pumping could be at the lowest
rate possible to remove the maximum amount of annulus fluid in complete plug flow (Possibly at .5 to 1
BPM). Design criteria will be critical for mud and spacer removal. This will involve holding backpressure
on the casing to control the flow rate. Just prior to cessation of pumping, the maximum pressures will
be placed on the formations. The method used to shut down pumping may be critical to the success or
failure of the operation.
Batch mixing
Batch mixing the shoe cement may prove to be critical to obtain a competent cement sheath. If no
batch mixer is available, it is advisable to batch the largest size possible on each tub before going down‐
hole.
Displacement
F‐17
In reverse cementing, the displacement volume is expected to typically be less than the volume required
for the same job in a conventional setting, since the entire casing volume will no longer need to be
displaced. A spacer should immediately follow the last sack of cement to separate fluids and prevent
mud contamination. The use of wiper plugs or balls in the drill string may be an option to aid in fluid
separation, and to clean the drill pipe ID. Volumes can be estimated by simulation, and verified by
caliper logs, or the use of a dye tracer to calculate the annular volume.
One expected challenge with deepwater RCPC displacement is to accurately identify when cement
reaches the casing shoe. With a conventional job there is a pressure increase (“bump”) when the plug
reaches the float collar, indicating that displacement is complete. With a RCPC job, float equipment may
need to be manually activated without a positive indication of cement placement when cement has
reached the float shoe or some other predetermined position inside the casing.
In conventional primary cement jobs it is critical to not over‐displace, with fluid compressibility and
pump efficiency need to be accounted for. The calculated displacement volume should not be exceeded
by more than 50% of shoe track volume in order to minimize the risk of a wet shoe. However, with
RCPC, over displacement into the casing may be the only method that will give a good casing seat at the
shoe. The maximum volume of over‐displacement, if any, should be agreed upon by all parties prior to
starting the job.
Return Flow
When the rig has the capability to dedicate a pit to take returns, the volume in and volume out can be
measured and compared. Marker pellets or fluids can be pumped and the time for return noted and
rate and volume can be calculated and/or hole size inferred. This method has been used for land‐based
RCPC operations. Down hole sensors may be developed that can give positions of the different fluids in
real time. Such sensors could differentiate between different fluids and send back a wealth of
information. Temperature, rate, and the type of fluids passing the sensor could be noted and recorded.
The sensors could be attached to a special casing sub or attached to the casing either internally or
externally.
F‐18
the bottom of the running tool that attaches the liner to the work string. The floating equipment can be
opposed flapper valves so that when the stringer is removed it will have shut off in both directions. The
flapper valve(s) can be held on by tubes inserted in the ID of the flapper valves before the tube(s) are
removed by the stinger. If well control is desired then the lower flapper is run in the hole in the closed
position. If surge reduction is desired them the tube allow auto fill of the casing. The stinger will also
allow surge reduction if desired. (A surge reduction tool can be run above the crossover tool, as long as
the closing ball and/or dart, if used, will go through the ID of the stinger and exit to the bottom of the
hole. The floating equipment will remain open during the entire time the cement is pump, and will only
close after the running tool has been released from the liner and the stinger is extracted from the
floating equipment.
If the crossover is battery operated tool system then RFID tags or Chemical Tags can be pumped down
to the tool to cause it to shift to the conventional circulation direction, then the RFID tags can be
pumped down a half barrel to 1 barrel of fluid more to cause a flapper valve to close below the running
tool and allow the pressuring up of the work string to set the liner hanger.
Another method of getting a liner hanger setting ball down to the crossover tool without disturbing the
cement just placed behind the liner is the use of a circulation valve above the crossover. The circulation
valve will be opened for circulation with RFID tags or chemical tags. This will allow the ball to be
launched from the surface. When the ball gets to the circulation tool it will have RFID tags attached to
close the circulation tool and allow the ball to fall or be pumped the ½ to 1 barrel to a seat in the
running tool.
(Note: All the above can be accomplished with wired drill pipe. It should be further noted that the
computer(s) on board the crossover tool systems are capable of obtaining operation signals RFID tags,
chemical tags, wired drill pipe, mud pulses (flowing system), mud pressure pulses, and internal timers
that can be preset with signals received prior to the event occurring.)
F‐19
Converting crossover
If there is an issue with converting the crossover tool from a conventional to reverse flowpath, one
option is to complete the cement job conventionally. In this case, there should either be a contingency
cement spacer and slurry design using the additives already on location. Another option would be to
pull the tool for inspection, repair or replacement.
Port‐collar (if used)
In the case where port‐collar is run below the liner hanger to divert flow into the annulus, potential
issues are that the sliding sleeves do not open/close as required. Another consideration is that typically
the port collar has a lower pressure rating than the surrounding casing and will be a weak point in the
string. The contingencies for difficulties with port‐collars will be similar to plans used for conventional
cement jobs.
Pack off at float/shoe area
This approach is similar to if the pack off occurs during pre‐job circulation. One option is to apply
pressure to pump out the debris. The maximum pressure to be applied should be determined
beforehand, and must not exceed the casing collapse or burst pressure, equipment pressure ratings, or
formation fracture pressure.
If a switchable crossover is used, then the flow path could be converted back to the conventional
direction to attempt to dislodge the debris by pumping in the opposite direction.
If circulation cannot be regained the cement job cannot be continued. To avoid cement left in the work
string, it should be released (if possible) and reversed out.
Setting Liner Hanger
The contingency approach for difficulties with setting the liner hanger will be similar to plans used for
conventional cement jobs.
Drill Pipe Stinger (if used)
If a stinger is used in conjunction with flapper type floats to hold the valves open during placement, it
will need to be pulled out at the end of the job to close the floats and prevent fluid flowback. If the
stinger is unable to be removed, this contingency approach will be similar to plans used for conventional
cement jobs.
Cement in Crossover Tool
If channeling or poor displacement occurs during the job, channeling of fluids may occur. One possibility
is cement channeled and remaining in the crossover tool at the end of the job. Cement slurries are
designed to incorporate a safety factor into the required thickening time. The time it takes to release
the tool at the end of the job and circulate out any cement should be included in the slurry design so
that cement will not set up inside any tool.
Floating equipment does not hold
If the floats do not hold at the end of the job, the contingency approach will be similar to those used for
conventional cement jobs.
F‐20
8.12) Contingency: Unplanned Shutdowns
As with a conventional job, unplanned and prolonged shutdowns could have a negative effect on the
quality of the cement job. While static, muds will gain gel strength. If this gel strength development is
significant, it will create difficulties when pumping is resumed since the cement and spacer will now
need to break the mud get strength to continue placement. Long periods of shutdowns can increase the
chance of channeling and poor displacement. The thickening time of any cement downhole should also
be taken into account. The risk of unplanned shutdowns occurring can be reduced through regular
equipment maintenance, pre‐job checks of all equipment, materials and procedure reviews. For each
occurrence of a shutdown, the maximum allowable shutdown time should be agreed prior to the job.
The rig pump is normally connected to the cementing line to save time if the rig has to displace or
circulate out cement in the event of a cementing failure. This contingency configuration is expected to
be the same for RCPC.
F‐21
hanger during the reverse cementing process. It is also assumed that there is a fluid entry line below a
lower ram, and a separate return line above that lower ram. This configuration will be needed to
circulate and take returns. A lower ram will have to be closed to allow pressure to be applied to the
annulus for reverse circulation. The inward flow will be down the drill pipe after the hanger is landed
and through the cross over tool. During reverse circulation with mud, the returns can be taken up the
riser.
Other considerations will be testing the casing and the shoe before drilling out. If floats are placed
higher up in the casing for RCPC, multiple casing tests may be needed: 1) Casing pressure test with
floats intact and before drill out. 2) At an acceptable depth after drill out to ensure anticipated
pressure(s) will be withstood, either from continued drilling, or from production/completion operations.
As with conventional cementing, shoe cement integrity and the existence of a wet shoe can only be
verified after the float is drilled out. If there is the possibility that there is no shoe integrity and the
casing will not test, a retrievable packer will need to be used to test and identify the location of the
F‐22
pressure leak. If it is the production string, setting a sump packer may solve the problem as it will be part
of the production package, and will also meet the casing test requirements as the sump packer is
normally set right above the TOC.
F‐23