Predicting Two-Phase Pressure Drops in Vertical Pipe

J. ORKISZEWSK!*

I ESSO PRODUCTION RESEARCH CO.

HOI. HTCSN,TEX.

ABSTRACT I!w free-gas phase is small, The gas is present as small

A medmi /.s presented w%ich cm accurately predict,
with a precision oj obour 10 percent, mm-phase presrui-e
● drops in fll?n’il?g o??dga.r-/ijr production wells over a wide
range of welt condiriom. The melbod i,s an exlension of
rhe wwrk tfone hy Grifj’7!h ond Wa[li.r” and was found to
be superior ro five orher published merhod.s, The precision
[~f the tttethod JCaS verified }t,hetz i/s predicted values were
contparedagnins! 148nleasur edpre.rsuredrops. The unique
jca[ures of lhi.s ,Jlerllod over )no.sr o!hers are rha! liquid
h,~ldup iv derived from observed pby.vica [phenonlena, the
pressure gradienr is related 10 lhe geometrical di.srribution
of t!e liquid andgws phase (f70w regimes), and the inethod
provides a goc~d (irralogy of w-ha! happens inside the pipe.
It takes Ies.r tlmn R second 10 obtain a prediction on rhe
IBM 7044 Compt(ter.
INTRODUCTION
The problem of accurately predicting pressure drops in
flowing or gas-lift wells has given rise to many specialized
solutions for limited conditions. but not to any generally
accepted one for broad conditions. The reason for these
many solutions is that the two-phase flow is complex and
difficult to analyze even for the limited conditions studied.
Under some conditions. the gas moves at a much higher
velocity than the liquid. As a result. the down-hole flowing
density of the gas-liquid mixture is greater than the cor-
responding density, corrt:cted for down-hole temp~rature
and pressure. that would be calculated from the produced
gas-liquid ratio, Also, the liquids velocity along the pipe
wall can vary appreciably over a short distance and result
in a variable friction loss. Under other conditions, the li-
quid is almost completely entrained in the gas and has
very little effect on the wall friction loss. The difference in
velocity and the geometry of the two phases strongly in-
bubbles, randomly distributed, whose diameters also vary
randomly, The bubbles move at different velocities de~end-
ing uporr their respective diameters. The liquid moves up
the pipe at a fairly uniform velocity and, except for its
density, the gas phase has little effect on the pressure gra-
dient.
SLUG FLO\\’ (FIG. 113)
In this regime, the gas phase is more pronounced. Al-
though the liquid phase is still continuous, the gas bubbles
coalesce and form stabie bubbles of approximately the
same size and shape which are nearly the diameter of the
pipe. They are separated by slugs of liquid. The bubble
ve!ocity is greater than that of the liquid and can be pre-
dicted in relation to the velocity of the liquid shrg.” There
is a film of liquid around the gas bubble, The liquid ve-
locity is not constant—whereas the liquid slug always
moves upward (in the direction of bulk flow); the liquid
in the film may move upward but possibly at a lower ve-
locity, or it may move downward. These varying liquid
velocities will result not only in varying wall friction losses,
but also in a “iiquid holdup” which will influence tlowing
density. At higher flow velocities, liquid can even bc en-
trained in the gas bubbles. Both the gas and liquid phases
have significant effects on the pressure gradient.
TRANSITION FLOW (FIG. lC)
The change from a continuous Iiquid phase to a con-
tinuous gas phase occurs in this region. The liquid slug
between the bubbles virtually disappears, and a significant
amount of Iiquid becomes entrained in the gas phase. Al-
though the effects of the liquid are significant, the gas
phase is more predominant.
Ub
Do
D

I
fluence pressure drop, These factors provide the basis for 0 b.
categorizing two-phase flow, The generally accepted cate- o 0
gories (flow regimes) of two-phase flow are bubble, slug, 00

(slug-annular) transition and annular-mist.”” They are o~

ideally depicted in Fig. 1 and briefly described as follows. 0
0
BUBBLE FLOY (FIG. 1A) ‘t)
Thepipe is almost completely filled with the liquid and 0°
Original manuseril]t received in Society of Petroleum Engineers office
August 8, 1966. Revised manuscript of SPE 1546 received March 1, 1967. .’O
Paper was presented at 41st Annual Fall .Meeting held in Dallas, Tex.,
Oct. 2-5, 1966. @Cony uight 196; American Institute of Mining, Metal-
0 00
hlrsicd, and Petroleum Engineers. Inc.
,~Presently with International Petroleum Co. Ltd., Talara, Peru, UQ
~tReferences given W end of riauer.
**AU four regimes could concsiy ably exist in the ssme well. An exam- 00
ple would in? a dee]) welt pm?ducing light oil fmm a resmvoir which is
near its bubble point. At the bottom of the hole, with lIttIe free gas BUBBLE A G
meaent, flow would be in the bubble regime. As the fluid mows up tbe TRANSITION
welI, the other m~imes would be encountered because gas continually A B D
comes out of solution, and the pressure continually decreases. Normally, c
however, flow is in the slug regime and rarely in mist, except for con.
densflte reservoirs or stenm-stimulated wells. FIG. 1—GEOMETRICAL CONFIGURATIONS IN YERTICAL FLOW.

829
 .

ANNULAR-MISTFLOW (FIG. lD) quid holdup is determined from mme concept of slip ve-
The gas phase is continuous. The bulk of the liquid is locity (the difference between the gas and liquid veloci-
entrained and carried in the gas phase, A film of liquid ties). The wall friction losses are determined from the
wets the pipe wall, but its effects are secondary. The gas fluid properties of the continuous phase. Four distinct flow
phase is the controlling factor, regimes are considered.
To cope with the complex problem, the many published Of the 13 methods categorized. two from each category
methods were analyzed to determine whether any one were selected for further study, The methods of Poettmann
method was broad enough, or had the ingredients to be and Carpenter,’ and Tek’ were picked from Category 1.
broad enough, to accurately predict pressure drops over a Most of the methods in this category are extensions of the
wide range of well conditions. The methods were first cate- Poettmann-Carpenter work. In the second category. the
gorized. Certain methods were selected from each cate- Hughmark and Pressburg method was selected: the Hage-
gory to predict pressure drops for two selected well cases dorn and Brown method’” was not available at the time
whose flow conditions were significantly different from of the initial screening, but it was included in the final de-
those originally used in developing the various methods. tailed evaluation. There are really only two methods in
Finally, the predicted pressure drops using the more prom- Category 3. The Griffithm and the Griffith and Wallis”
ising methods were compared against known values taken methods are synonymous: the Nicklin. Wilkes. and David-
from 148 cases having widely varying conditions of rate. son method’8 is for special conditions and parallels the
GOR, tubing size, water cut and fluid properties. work of Griffith-Wallis. The other method is that of Duns
and ROS.2
BASIS FOR SELECTING METHODS STUDIED
RESULTS OF THE COMPARISON
Based upon similarity in theoretical concepts, the pub-
lished methods were first divided into three categories. The five methods initiaRy selected. whose results were
From each category certain methods were selected, based hand calculated, were compared by determining the devi-
on whether they were original or unique, and were devel- ation between predicted and measured pressure drops for
oped from a broad base of data. The discriminating fea- the first two weII conditions listed in ~dhfe 1. Fig. 2 com-
tures of the three categories are shown. pares the predictions for Well 1. The results wwe similar
for Well. 2. The most accurate methods (D~ns-Ros and
CATEGORYI (REF’S.1, 3.6, 9) Griffith-Wallis) were then programmed for machine com-
Liquid holdup is not considered in the computation of putation and further tested against 148 well conditions.*
the density. The density is simply the composite density of
Neither method proved accurate over the entire range
the produced (top-hole) fluids corrected for down-hole tem-
of conditions used. Although the Griffith-Wallis method
perature and pressure. The liquid holdup arm the wall was reliable in the lower flow-rate range of slug tlou’. it
friction losses are expressed by means of an empirically
was not accurate in the higher range. The Duns-Ros meth-
correlated friction factor. No distinctions are made among
od exhibited the same behavior except that it fvas also in-
flow regimes.
accurate for the high-viscosity oils in [he low flo~v-rate
CATEGORY2 (REFS. ‘i. 8, 10) range. The Griffith-Wallis method appeared to provide the
Liquid holdup is considered in the computation of the better foundation for an improved general solution al-
density. The liquid holdup is either correlated separately though its predicted values were in greater error (2I.9
or combined in some form with the wall friction losses. percent) than Duns and Ros (2.4 percent). The heart of
The friction losses are based on the composite properties this method, prediction of slip velocity. is deriwd fr’mn
of the liquid and gas. No distinctions are made among physical observation. However. since friction drop ~vas
——. -.——
flow regimes. ‘The data in Table 1 are from ?2 Venezuelan heavy-oil wells. [n addi-
tion to the data presented in ‘ruble 1. the ctaut used are from the rlub-
CATEGORY3 (RfWS. 2, 11.13) Iications of Poettmann and Cnwenter.’ Bnxendell nnd Thomas.” Fa?-
eher and Brmvn,o and Hwwdorn :!nd Brown.~ These rel>resent 121;:IddI-
The calculated density term considers liquid holdup. Li- tiorml pieces of data.
—
TABLE 1—PHYSICAL CONDITIONS AND FLOW RATES OF HEAVY-OIL WELLS STUDIED
Oil Measured Wellhead Flow String
Well Oil Rate GOR Water Wy;;; D~~;h Pressure Diameter Measured
No. (B/D) (scf/bbl) cut (%) (psig) (in.) Jp (psi)
_—
1 320 4020 30 10.3 4360 250 8.76 810
175 6;:: 17 4360 300 8.76 925
$ 1065
1300 252
—
—
1%:
14.6
3825
3940
550
150
2.992
,,
650
850
;;:; 1430 — 14.4 3800 700
,, 550
232 — 14.4 3720 300
,, 900 *
957 — 15.6 4240 700
,, 850
1165
1965 1500 . 13.5 4570 850
#r 650
2700 267 — 15.6 4175 300
,,
1200
855 185 — 12.9 4355 250
,?
1450
,,
2320 1565’ 13.6 4670 910 740
,, 900
2480 858 18.6 4575 650
n 950
1040 472 — 18.6 4400 400
— ,,
1490 341 13.0 4065 500 1:():
1310 335 — 13.6 3705 500 ,)

1350 185 — 12.9 4160 150

,,
1350
,,
788 222 16.0 4210 350 1400
,7
1905 962 ~ 14.1 4487 580 720
,?
967 193 13.3 4766 250 1300
1040 385 — 12.5 4505 250
,,
1;::
— ,,
21 1585 865 12,9 4692 400
22 1850 575 — 18,7 3924 - 700 ,, 800

JO[:RSAL OF PETROLEKM TECKXOLOG1’

 ,

. .

negligible in the work, the metho,d for predicting friction TABLE 2—SUMMARY OF DEVIATIONS BETWEEN
losses is an approximation and therefore open to improve. MEASURED AND PREDICTED PRESSURE DROPS
mmt. On the other hand, the Duns and Ros work in this Prediction Method
range (which they termed plug flow) is presented as a com- This Duns Hagedorn
plex set of interrelated parameters and equations. and is Method
—— and Ros and Brown
therefore difficult to relate to what physically occurs in- Overall Results
side the pipe. (148 Well conditions)
The Griffith-Wallis work was extended to include the Avg. error, percent – 0.8 + 2.4 + 0.7
high-velocity flow range. In modifying the method. a Std. deviation, percent 10.8 27.0 24.2
parameter was developed to account for (1) the liquid dis- Results from Grouped
tribution among the liquid slug, the liquid film and en- Data Sources
trained liquid in the gas bubble and (2) the liquid holdup Table 1 — Heavy-Oil Wells
at the higher flow velocities. This parameter served to (22 Wells, low to medium
better approximate wall friction losses and flowing den. velocities, 10 to 20° API oils)
sity. and was principally correlated from the earlier pub- Avg. error, percent – 1.2 +22.7 +16.4
lished data of Hagedorn and Brown? The data from Table Std. deviation, percent 10.4 18.7 41.4
I were also used to determine the effects of pipe diameter Baxendel[.Thomas3
on the parameter. The details of the parameter evaluation (1 Well, 25 rates mostly high
velocities, 34” API oil)
are given in Appem!ix C and a brief description of the
modified Griffith-Wallis method is outlined in Appendix Avg. error, percent – 2.1 + 2,3 + 8.7
A.
Std. deviation, percent 11.1 20.0 12.7
The rewdts of the study, summarized in Table 2. are Fancher-Brown”
(1 Well, 20 ratea medium to
presented as the deviations between predicted and meas- high velocities, 95 percent
ured values for the modified Griffith-Wallis, the Duns- water cut)
Ros and the then recently published Hagedorn-Brown’” Avg. error, percent J- 0.3 + 1.7 +- 5.4
methods. (The Hagedorn-Brown method was included be- Std. deviation, percent 11.8 32.1 “1O.8
cause of the excellent accuracies reported and the broad HagedornBrown’
data range presented.) Plots of the individual predicted and (1 Well, medium to high
measured values for the three methods are shown in Figs. velocities, 16 water runs,
3 through 5. When the three methods are compared 16 oil runs of 10 to 100 CP oil)
against the various grouped data sources (Table 2), only Avg. error, percent +- 0,1 –16.9 + 1.2
the modified Griffith-Wallis method is sufficiently accurate Std. deviation, percent t3.2 36.6 10.3
(average error) and precise (standard deviation) over the Poettmann-Carpenterl
entire range of conditions. None of the 148 well condi- (49 Wells, low to medium
tions studied were in mist flow or wholly in (annular-slug) velocities, 15 wells high
transition. The breakdown of wells by flow regimes in- water cut, rest
cludes seven partly in slug and transition. 26 partly in 36 to 54° API oils)
slug and bubble. four completely in bubble and 111 com- Avg. error, percent – 1.0 + 5.8 –13.0
pletely in slug flow. Std. deviation, percent 12.0 12.4 22.2
..——
sufficiently evaluated (e.g., flow in the casing annulus and
CONCLUSIONS
in the mist-flow regime). The method’s precision might be
For general engineering work, the modified Griffith-Wal- further improved if the liquid phase distribution could be
lis method will predict pressure drops with sufficient ac- more rigorously analyzed.
curacy and precision over a wide range of well conditions. This method is accurate over a broader range than
I recommend its use, However, the method should be used previous correlations, For a prediction method to be gen-
with discretion for those well conditions which were not eral. it must be expressed in terms of flow regime and
liquid distribution. The other methods, which were not de-

“ ::-
■

o
600 --0 !’

400 –- 3 10% BAND

(114DATA POINTS)
300 -*’-

v
/
290---- -
DATA INSIDE ~10%
❑ BAND NOT PLOTTED
~oo~THP I
I
o 1000 2000 3000 4000 100’‘
100 200 400600 1000 3000
WELL DEPTH- FEET
MEASUREDAP - PSI
FIG. 2—COMPA5USON
OF PRSSSURE PROFILES CALCULATEDBY
VARIOUS
METHODS
FORWELL 1 (TABLE 1 ). FIG.3-THIS WORK(MODIFIED
GRIFFITH
ANDWALLIS
PREDICTION
).

8s1
 ●

veloped in this manner, are only useful in the range of I’ = liquid distribution coefficient, to be evaluated from
conditions from which they were developed. Eqs, C-11 through C-16, dimensionless
p, = viscosity, cp
NOMENCLATURE </D = Moody pipe relative roughness factor (Fig, 7) and
Duns-Ros mist flow factor (Eqs. C-21 and C-
AP = flow area of pipe, sq ft 22), dimensionless
B., = 011formation volume factor. bbl/STB p = density, lb/cu ft
C’,,C, = parameters used to calculate bubble rise velocities
~ = average flowing density, lb/cu ft
from Eq. C-5, dimensionless, to be evaiua ted
~, = friction-loss gradient, lb/sq ft/ft
from Figs. 8 and 9
d,, = hydraulic pipe diameter (4 X A./wetted perime- u = surface tension, lb/secZ
ter), ft SUBSCRIPTS
D = depth from wellhead, ft g = gas
3D = increment of depth, ft L = liquid
} =.. Moody friction factor. dimensionless, to be eval- o = Oij
uated from Fig. 6 r = total
F, = flowing gas fraction, dimensionless
g = acceleration of gravity, ft/ see’
g, = gravitational constant. ft-lb(mass)llb(force)-$ ec’ ACKNOWLEDGMENT
(L.),,= bubble-slug boundary. dimensionless The author wishes to thank the Creole Petroleum Corp.
(L),, = transition-mist boundary, dimensionless for supplying data in Table 1 from 22 Venezuelan heavy-
(L), = slug-transition boundary. dimensionless oil wells.
Nb = 1,488 vMJ,,p,/ }LL,bubble Reynolds number, dinlen-
sionless
,%’,,,.= 1,488 v D,, pip, Reynolds number, dimensionless REFERENCES
p = pressure, psia 1. Poettmann. F. H. and Carpenter, P. G.: ‘“k Lfultipha.e FIOW
4P = pressure drop, psi of Gas, Oil. and Writer Through Vertical Flow String..”. DrilI.
~ = average pressure, psia rm! Prod. Prac., API (1952) 257.
p,,, = pseudo-critical pressure. psia 2. I)nns. H.. Jr.. and Ros, N. C. J.: “Vcrtir=l Flow of (h. ~ncl
p,. = reduced pressure. dimensionless Liquid Mixtures from Itoreholrs”. Proc., Sixth ‘iIrorld I’rt. Con.
gress, Frankfort (June 19.26, 1963) Section II, Paper ?2.PD6.
P = pressure. lb/sq ft
3, BaxendeR. P. B. and ‘I%omas, R.: “The Calculation of Pies.
q = volumetric flow rate. cu ft/sec sure Gradients in High.Ra te Flowing \\’ells”, J. Per. Trch.
q> = oil rate, B/D 10ct.. 1961) 1023-1028.
R= produced GOR, scf/STB
+. Te!i, M. R.: “.lfultiphase Flow of \Vater, Oil, und Nttural Gas
R, = solution gas, scf/STB Through \-ertical Flow Strings”, J. Pet. Tech. ( Oct.. 1961)
TP, ==pseudo-critical temperature. “R 1029-1036.
T. = reduced temperature, dimensionless .5.Yocum, B. T.: “Two-Phase Flow in \l’ell I%wlines... Per. Eng.
~= average temperature, ‘F f Nov.. 1959) B-40,
v = fluid velocity, ft/SeC 6. Ftmcher. G. H., Jr., and Brown, K, t-l: ‘.Prediction of Pressure
w = bubble rise velocity (velocity of rising gas bubble Gradients for Multipbase Flow in Tulring”, Sot. Pe/. Errg. J.
relative to preceding liquid slug), ft/sec IMarch. 1963) 59.69.
v,,, = base bubble rise velocity for Eq. C-9, ft/sec i. Baker, W. J. and Keep, K. R.: ‘“The Flow of Oil and Gas
v. = slip velocity (difference between average Mixtures in R’elk and Pipelines: Some Useful Correlations”,
.- gas
- and J. inst. 01 Pet. (Nlay, 1.961) 47, No. 449, 162-169,
iiquid velocities). ft/sec
8. Hughmark. G. A. and Pressburg, B. S.: “Hold-Up and Pres-
‘9D = q. CVm/@/4 dimensionless gas velocity sure Drop with Gas-Liquid Flow in 8 Vertical Pipe”, AZCM 1.
z = compressibility factor, dimensionless
gas ( Dec., 1961) 7, h’o. .S,677-682.
y = fluid specific gravity. dimensionless 9. Hagedorn, .4. R. and Brown, K, IL: “The Effect of I.iquid l’i.s-

3000
DATA SOURCE

2000 o BAXENDELLet al. __—. .- ...

❑ FANCHERet al. ●
A HAGEDORN et al. ■
~
● POETTMANN et al. ■ ■
● POSTTMANN et al. ● *
■ THIS PAPER —. ■ THISPAPER ,%
~ 1000 t
■ “”’” — .. . . .. ..- .. ●“
ao %
o ●* *
: 400 ---- 0 . . .. . . . ___ _ _ 0 /
❑ :? ‘2/ ‘“
g ~ – --.- ..0 ~10% BAND 5 ●
■

w“’
5 4W
I.u _ (72 DATA POINT5) g ●
g 300 ----- A : t 10%BAND[75
,. DATA,POINTS)
‘1/. A A 300
A ●

:kzffrz3sii’i--
200- . . m.
0
DATAINSIDE~10%BANDNOTPLOTTED

100 200 400 600 100D 3000 ,.,

“

10D 200 400 600 1000 200D3000

MEASUREDAP - PSI AP -PSI
MEASURED
FIG.4-DtiNs AXD Ros PREDICTION? FIG. 5—HAIXDORN A NO BROWs PREDICTION?”

8%4 JO LESNAL OF PETROLEUM TECHiSOLOGY

.
. .

cosity on Two-Phnse Flow”, ]. Pet. Tech. (Feb., 1964) 203. WC= total mass flow rate, lb/see,**
210.
go = gas volumetric flow rate, cu ft/sec.**
10.Hagedorn, A. R. and Brown, K. E.: “Experimental Study of
Pressure Gradients Occurring During Ccmtinnous Two-Phase Whh the above conditions and Eq. A-2, Eq. A-1 may
Flow in Small Diameter \’ertiral Condaits”, J. Pef. Tech. then be expressed in a more convenient form.***
(April. 1965) 475.484.
11. Griffith, P. mul \\lallis. C. B.: “Two.Phaw Slug Flow”. j. Hru(
Transfer; Trans., ASME (Aug., 1961) 307.320.
12.Griffith, P.: “Two-Phase Flolr in Pipes”, Special %lmmer Pm.
Apt =
[ ~1–
1
W,
;+7,
q,/4,637 A,’ ~ I,
AD, , (A-3)

worn. Masmclm:etts Institute of ‘YechnoIogy,Cambridge, Mass. where for average temperature-pressure conditions at in-
(1962). crement k,
13. Nicklin. D. J., TI$lkes, J. O. and Davidson, J. F.: “Two-Phas@
Flow in Vertical Tubes”, Trans., AIChE (1962) 40, 6R6ft. ~ = average fluid density, lb/cu ft.
Ap = pressure drop, psi,
1.!.Stanley, D. \V.: ‘Wall Shear Stress in TWO-PIMWShlg Flow”, ~ = average pressure, psia.
MS Thesis. Massachnwtts Institute of Technology,Cambridge
(June, 1962). Since temperature is related to depth, Eq. A-3 may be in-
15. Moody, I.. F.: “Frirtion Factors in Pipe Flow”. Trrrns., ASME cremented by either fixing 4D and solving for Ap, or vice
i 1944) 66, 671-684. versa. However, since pressure usually has a greater in-
16. Frick, T. C.: Petroleum Production Handbook—Vol. II, R~ser- fluence on the average fluid properties than temperature,
roir Engineering. IklcGrzrw.Hill. New York ( 1962). 4p should be fixed because the change in average fluid
properties would then be more gradual in going from one
APPENDIX A increment to another. The value of Ap should be around
10 percent of the absolute pressure, which is known for
DESCRIPTION OF MODIFIED GRIFFITH AND
WALLIS METHOD* one point in the increment, but should not be greater
than 100 psi for that increment.
The fluid pressure drop in a vertical pipe is the sum ef- Pressure drops can be calculated, using Eq. A-3 in the
fect of the energy lost by friction, the change in potential following manner.
energy and the change in kinetic energy. This energy bal-
ance, which is basic to all pressure-drop calculations, can 1. Pick a point in the flow string (e.g.. wellhead or bot-
be generally written as tom-hole) where the flow rates. fluid properties, tempera-
ture and pressure are known.
– dP = 7,dD + (gp/g, )dD + (pv/g,)dv , . . (A-1) 2, Estimate the temperature gradient of the well.
where P = pressure, lb/sq ft, 3. FIX the Ap at about 10 percent of the measured or
TJ = friction-loss gradient. lb/sq ft/ft, previously calculated pressure, which may be at either the
D = depth, ft, top or bottom of the increment. Find average pressure of
g = acceleration of gravity, ft/sec’, increment.
g. = gravitational consant, ft-lb(mass)/ 4. Assume a depth increment AD and find average depth .
Ib(force)-sec’, of increment.
p = fiuid density, lb[cu ft,
v = fluid velocity, ft/sec. 5. From the temperature gradient, determine average
temperature of increment.
The procedure was credited to Griffith and Wallis because
slug flow occurred in 95 percent of the cases studied. Al- 6. Correct fluid properties for temperature and pressure.
though the mist-flow regime could noi be evaluated, the 7. Determine the type of flow regime from Appendix B.
Duns-Ros method was used because it appeared to be 8. Based on Step 7, determine the average density (~)
more accurate and logical than the Martinelli method rec- and the friction loss gradient (~,) from Appendix C. “
ommended by Gritlith. In two-phase flow, both rf and p
9. Calculate 4D from Eq. A-3.
are influenced by the flow regime type, and all three terms
are functions of temperature and pressure. Therefore, to 10. Iterate, if necessary, starting with Step 4 until as-
use Eq. A-1, (1) the flow string must be incremented so sumed AD equals calculated AD.
the fluid properties do not change markedly within any 11. Determine values of p and D for that increment.
of the increments, (2) the flow regime type and correspond- 12. Repeat procedure from Step 3 until the sum of the
ing variables of p and Tf must be determined for each in- AD’s equals the total length of the flow string. ,
crement and (3) each increment must be evaluated by an
iterative procedure. A detailed example of the above calculated procedure is
given in Appendix D.
The kinetic energy term is significant only in the mist-
fiow regime? In mist flow v,. << v,, the kinetic energy
APPENDIX B
term may be expressed’ more simply (using the gas law).
DETERMINATION OF FLOW REGIME
(pv/g,)dv = – ~dP , . . . . . . . (A-2)
Grititth and Wallis have defined the boundary between
bubble and slug flow,” and Duns and Ros have defined
where A,, = pipe area, sq ft. the boundaries for the remaining three regimes.’ The fio~
. regime may be determined by testing whether the variables
*TMs metlmrlis a com~osite of the following: qJq, or v,,I,, or both, fall within the limits prescribed.
Method Flow’ Regime
——
*.AI1“O:Ume~rjc
(<,, ~“~ ~W~ flo, flOW,a~s ~1.Q
tboseof the DrO-
=ffith:z Bubble duced fluids tlmt are corrected for temperature, pressure and Km SQIU-
Griffith and Wtrllisli Slug (density term) bility.
..*AZ is taken ~~ po~itjve do,,. n,yard. The pr~ure should be made
This work Slug (friction gradient term)
ISuns and Ros: Transition cliscontinuou% with depth should the denominator approach zero, or be-
come negfltive. to estahish the shock front that ckamctcwizes sonic ve-
Duns and Rosz Annular-mist locities.

JUNE, 1967 833

 Limits Flow Regime where I’ is a coefficient correlated from oilfield data.
Griffith and Wallis correlated the bubble rise velocity v,
%/9, < (=% Bubble
by the relationship
%/9: > (-0i7>v,. < (Q8 slug —
(L). > ~. > (L). Transition vb=c,c,~gdk , . 0 . . . . . . . (C-5)
v.~ > (L),, Mist where C, is expressed in Fig. 8 as a function of bubble
Reynolds number (N, = 1,488 v,d,p./~.), and C, is ex-
The above variables are defined as pressed in Fig, 9 as a function of both N, and liquid Rey-
nolds number.
‘UD 9u(4vPL/&?
= d/~P , , , , , . . “ @-1)
(L). = 1.071 – (0.2218 v,’/d,,, with the limit IVu. = 1,488 f3,,~,V,/#L , . . . . . . . (C-6)

(Q> O.13 . . . . . . . . . (B-2) where v~ equals total velocity of liquid and gas (qt/AP),
(,L). = 50 + 36 V,D q,,/q, . . . . . . . (B-3) ft/sec.
Fig. 9 was extrapolated”:* so ihat v,, could be evaluated
(L),, = 75 + 84 (V. ~L/q,~ , . . . . . . (B-4)
at the higher Reynolds numbers. When C, cannot be read
where vo~= dimensionless gas velocity,
*$The p&raIIeI work of N[ck~e;, WIlkas and Davidson~3 provided the
vt = total fluid velocity (q,/A,), ft/see, basis for the extrapolation. It showed that bubble rise veloclty was in-
dependent of iVb in the Reynolds number range of 9 X 10~ to 1 X 10$.
p,, = liquid density, Ib/cu ft, The correlation of bubble rise velocity was found comparable to l%. C-5
when Nb was around 8 X 103. The results were incorporated into the
u = liquid surface tension, lb/sec’. above extrapolation.

APPENDIX C 0.1
009
008
0.0s
EVALUATION OF AVERAGE DENSITY AND 007
0.04
006
FRICTION LOSS GRADIENT . 0.03
- 0.05 002
& 0.01s
In the first four sections of this Appendix, the variables 0 0.04 0.01
c 0.008
~ and Tf are deftned for bubble flow, slug flow, transition a
1- 0.03
0.006
0.004
and mist flow, The second section, while the most complex, g 0025 0.002
is also the most important since slug flow had been en- ~ 0.02 0001
countered in over 95 percent of the gas-lift or flowing z }’j~;
wells studied. The last section of this Appendix describes n 0.013
00002
how T, was developed for the slug flow regime. 0,0001
0,0000!
0.01
BUBBLF. FLOW (REF. 12) 0.009
0,00001
0 00B
The void fraction of gas (FJ in bubble flow can be ex-” 1 234681 234681 234681 234681 2346B1
~x 103+. 104+. 105+. 10+.-. 107+
pressed as
148 SIdv&
REYNOLDS NUMBER Re =
F
+l+&- (1-t-qt/v.
A,Y– ~ 4qg
,
[ v, A,,
. .
4
. . . . . . . . (c-1)
1 FIG,6—MooDY FRICTION
FAcroBY

where v, = slip velocity in ft/sec. Griffith suggested that

a good approximation of an average Veis 0.8 ft/sec.* Thus,
with Eq. C-1, the average flowing density can be computed
as

;=(1 –~,,)P1>+~.Po . . . . . , . Gv
The friction gradient is I J---hu-
9 , . . . . . . , . (c-3)
‘rI = iPLvl.2/4?cA
where
v,, = q,,/[Ap (1 – F,)] o
The friction factor f is obtained from Fig. 6 by using a
Moody relative-roughness factor obtained from Fig. 7.
The Reynolds number is calculated as Af~,= 1,488 ~LdhvI./

where d, = hydraulic pipe diameter, ft, and p. = liquid

II,.;

viscosity, cp.
SI.UG FI.OW (REF. 11)
The average density term is

~= w~ + p,.v~AP
+l’p,, , . . . . . . (c-4)
q, + v~AP
-L t&._&.o- ~.
.,. ,
PIPEDIAMETER- INCHES
*Athough the method is simple, it is reasonaLdy precise: For the four
wells that were wholly in Imbhle flow, the standard devmtlon was 5.1
percent, whereas the deviation was 9.8 pm.mnt for those wells IXWLIYin FIG. 7—EFFECT OF PIPIS DIAMETER AN? MATERIALON
l,ubble and slug. RELATIVEROUGHNESS?

JOURNALOF PETROLEUMT53CIi NOi. CSCY

 L .

. .

from Fig. 9, the extrapolated Vah.3S3 of Vb rnaY be calcu- and when v, >10,
lated from the following set of equations.
v6A ~
r>-— 1 –:..... (C-16)
When N, <3,000, q, + V,AP ()
v, = (0.546+ 8.74X 10-’ ~m) V=, . . . (C-7) The above constraints eliminate pressure discrsntinuities
between fkw regimes.
When N, >8,000,
— TRANSITION FLOW
Va = (0.35 + 8.74 X 10-CNJ w’gd~, . . . (c-8)
Duns and Ros approximated ~ and T, for transition flow.
When 3,000< N, <8,000, The method is first to calculate these terms for both slug
— and mist flow, and then linearly weight each term with
v~\ = (0.251 + 8.74 X 10-0NRJ Vgdk ,
respect to VODand the limits of the transition zone (0s
1 v,,, + 13.59 JL& and (L),,,. The terms v,~, (L)~ and (L), are defined in
Vb=yvbi+ _ ,.. . . (C.9)
Appendix B. The average density term would be
4 p. ddh

F= (L).” – V.. v –(L).

~
[1
~
The wall friction-loss term, which has been independently
derived, is expressed as (L).,, – (L). [1.
,,., + (~;, – (L),
. .
mist
. . . . . . (c-l 7)

1,
fp,.v,’ q,. + v,A. ~ ~
~,. — (c-lo)
2g<d,, [ q, + V,AP “ “ “ “ “ The friction gradient term would be weighted similarly. A
more accurate friction-loss prediction is claimed if the gas
The friction factor is obtained from Fig, 6 and is a func- volumetric flow rate for mist flow is taken as
tion of the Reynolds number given by Eq. C-6 and the f/D
obtained from Fig. 7. The liquid distribution coefficient I’ q, = A .(L).w Ww)-”i . . . . . . . @1@
may be determined by the equation which meets the fol- MIST FLOW
lowing conditions. The average flowing density for mist flow is given in Eq.
Continuous C-2. Since there is virtually no slip between the phases,
F. is
Liquid
Use Equation F, = 1/(1 -i- qL/q,). . . . . . . . , (C-19)
Phase v,
Water ZO C-n Duns and Ros express the friction-loss gradient as
Water >10 C-12
Oil <10 C-13 ~r = / p“vJ’/2gcdA , . . . . . . (C-20)
Oil >10 C-14 where V. is the superficial gas velocity, and j is again ob-
tained from Fig. 6 as a function of gas Reynolds number
I’ = [(0.013 log p,,)/d#S] – 0.681 +0.232 log vt (N.. = 1,488 p,d,,vJp,) and a correlated form of the
– 0.42810gd, . . . . . . . . (C-llJ Moody relative roughness factor f/D that was developed
r = [(0.045 log ~,.)/d#’”1 -0.709 – o.1152log VI by Duns and Ros. In their correlation, they limit &D to
– 0.888 log d,, .- , . . . . . . (C-12) being no smaller than 10-’ but no greater than 0.5, Be-
. tween these limits, #/D is determined from Eq. C-21 if N,.
r = [0.0127 log (P.+ 1)/dJ4”l -0.284 is less than 0.005 and from Eq. C-22 if N,. is greater than
i- 0.167 log V, + 0.113 logd,, . . . (C-13) 0.005.
r = [0.0274 log @.+ I)/d,’”’”l -I- 0.161 &/D = 34 U/(fI,V,:~,,) . . . . . . . . . (c-z])
+ 0.569 log d, - log V, {[0.01 ]og (~’ + 1)/d/”’”1
-t- 0.397-1-0.63 Iogd,,} , . . . . . (c-14) #/D = 174.8 u (NJw’/(p,w,’d,,) , . . . . (C-22)
but is constrained by the limits where N. = 4.52 X 10-’(uOp,,/u)’p./p,, .
r>–o.065vl , . . , ., . . . , , (C-15) DEVELOPMENT OF ~,
FOR SLUG FLOW
A new method was developed to correlate the friction-
0.40 I Ioss gradient for slug flow because neither the GritEth and
Wallis method, nor the Stanley” method (an outgrowth
of the Griffith-Wallis work) proved accurate for the well
conditions studied. (The Griffith-Wallis data were taken
0.30

. u- 1.5
I Al 1 1

*
!4 0,20 -
tk
!5
u

0.10-

1488q, dhp~
REYNOLDSNUMBER- NRe
( APF, )

FIG. 9-GRIFFITH AND V?ALLIS’ C, vs BUSBLE REYNOLDS

Frc, 8-GRIFFITH AND WALLIS’ C, vs EUBBLE REYNOLDS NUMBER. AND REYNOLDS NUMBERS.

JUNE, 1967 8ss

 . >

. .

from low-flow velocity tests in which friction losses were TABLE 3 — FI.OW RATES AND PHYSICAL CONDITIONS
OF HEAVY-OIL WELL 22
minor and liquid entrainment was negligible.) This new
method accounted for the complex nature of friction loss Oil Rate (q,) 1,850 Oil Specific
in slug flow by the introduction of a correlated liquid dis- STB/D Gravity (y.) 0.942
tribution function 1’ (Eq. C-10) which implicitly accounts Pro5uced GOR (f?) 5~~~f/ Gas Specific
for the following physical phenomena. Gravity (y,) 0.75
Total depth (D) 3,890 ft wellhead Pressure 670 psia
1. Liquid is distributed in three places: the slug. the Tubing Area A,, 0.0488
Tu~in~ diameter 0,249 ft
film around the gas bubble and in the gas bubble as en- Sq ft
trained droplets. A change in this distribution will change Temperatures: Dead Oil Viscosity:
the net friction losses. Wellhead 126F at 100F 89 Cp
2. The friction loss has essentially two contributions, one Reservoir 150F at 21OF 8.8 Cp
from the liquid slug and the other from the liquid film. ——— —
3. The bubble rke vclocit y approaches zero as mist flow
is approached. and that the data scatter may be attributable to such addi-
Values of the liquid distribution coefficient were ca[ctt tional parameters as liquid velocity, GOR and interracial
Iatt.xi from the data of Hagedorn and Brown” by using tension.
Eq. A-3 of Appendix A and Eqs. C-4 through C-10 of Ap- APPENDIX D
pendix C. (These data were selected because they covered
a wide range of conditions for each of the four liquids EXAMPLE OF TWO-PHASE PRESSURE
used.) These values correlate with total fluid velocity and DROP CALCULATION
liquid viscosity, Fig.I()shows the results where the fluid
An example calculation of the modified Griffith. Wallis
is water. and Fig. I I shows the results where the fluids method is presented to illustrate the details of the proce-
aI-e oil. Coefficients were also calculated for the heavy-oil dure outlined in Appendix A. In this example we will pre-
wells shown in Table 1. but these values were small and dict the pressure drop for heavy-oil Well 22 (Table 1). The
scattered because 7, was small in comparison to ~ Never- input well data required for the calculation are given in
theless. the results were sufticicntly grouped to show that Table 3. In addition, we will need the following correla-
pipe diameter kl,,) is another independent variable, tions* that correct fluid properties for pressure and tem-
Neither the nor the data scatter as
reversal in slopes perature:
seen in Figs. 10 and 11 can be resolved without additional Gas pseudo-critical properties (Katz er d.) 7,.,, p,,,,
experimental work. it is probable. however, that the slope Gas compressibility (Brown e( al.) z.
reversal may be due to liquid entrained in the gas phase
Live oil viscosity (Chew and Connally) p.
o —-.— . Oil formation volume factor (Standing) B,,.
‘1 !1
,1 0’ Solution gas (Lasater) R..
z ,!
0 For calculational convenience, the temperature-viscosity-
c
3 depth data contained in Table 3 should be plotted. The
C9c .0.11 --- Q+: d O-$$%%=-.--.-:-:
temperature-depth plot is shown in Fig, 12, and log vis-
~g
m-
~u cosity-fog temperature plot is shown in Fig. 13.
;@:’ !’%.~ The detaiIed procedure for the calculation of the pres-
QE
50 ~t)ej.-..... .. .. ... .. . -_.. ... \ ;. sure drop for the first increment (k = 1) is as follows.
z“ ‘0”’
( ““,,~ ;AGEDORN ~ND,BRowN:DATA(,)’ : 1. Based on the 670-psia wellhead pressure, fix Lp at 106
L / mi. Assume M to be 540 ft. The average pressure (~) and
/
1-1/4 “ PIPE depth (~) of increment k is then:
J,,
/ :’, $
.0.3 L --- ,, ,
1 2 3 4 567 S910 20 30 40 60 80100
Lp,
SUPERFICIAL
FLOWVELOCITY- FT/SEC jz = p,. , + ~ ==670 + -~~– = 720 psia
l~l(;. 111.–1{I’’1:(IT(.T OF \l.m(:l’lY [)\ \\ ,ATIK I)ls’rRllR”i’l[l\
(:( M’FFlclK\’l’.

The average temperature (’~). read from Fig. 12 is 127.5F.

—..
‘~Th,se
nw convmi?ntls romul in Frick’s l’drdrMIII Prodw~i~tt lj<IIIcf-
1,00[;, I’d. [I (Ref. 161.

160 r———— —-~

L

1 ~ .. .. ... .
0
,

E 140 — -- -—–4–— ----- –-—

3
a I ~ 0
3 .0.2 -—— .— —-—. —.-— .- —...
HAGEDORN AND BROWN DATA g 120
~ ~-
L 1-K” PIPE I
-0.3, \ 1-
234 6810 20 30 40 6080 )
100.
SUPERFICIAL
FLOWVELOCITY- FT/SEC 0 1000 2000 3000 4000
DEPTH- FT

l~Ic. 12—TI.JI1’V:RATIR I:. w Dmm — N’IH.I. 2?2.

8:s6 JO IJRXAL OF PET ROLEI; M TECH XOLO(; Y

 2. With the condition determined in Step 1, the fluid The corrected densities arc
properties are corrected for temperature and pressure,
p,. = w,./q,. = 7.20/0.129 = 55.8 Ib/cu ft
From Frictc” the following values are obtained.
p., – w.,q,, = 0.565/0.199 = 2.84 lb/cu ft.
R. = 115 scf/bb[ (page 19-9).*
3. The variables described in Appendix B are calculated
B,, = 1.073 bbl/STB (page 19-25). and then are tested against the boundary limits to deter-
p,,. = 665 psia (page 17-6). mine the flow regime.
7’,, =415R (page 17-6). Test Variubles:
p= 18cp** (page 19-40). V, = q,/A,, = 0.328/0.0488 = 6,72 ft/see,
The gas compressibility c is determined as q./qt = 0.199/0.328 = 0.607.
-—..————
~,=~+460 587.5
—-—---=-=1.42
T,,{ 415
V.,D = 0.199

Bwlmfary Limil.f:
[ 1
\10.534 (55.8) /0.0488 = 9.53 , (B-1)

F 720
= 1<08.
p’”=~== (L),, = I .07 I – ~“2~&2:9”73’: . . . (B-2)
find from Frick (page 17-15).
(L),, = –22. Since (L.),, has thu limit of 0.13.
Z = 9.875.
.“. (L),, ==0.13.
The corrected volumetric flow rates arc
(1-), = So + M (9.53) (0.12~)/O.199 = 272 . . [B-3)
q,. = 6.49 X 10-”qr.B,>= 6.49 X 10 ‘ (1,850 (1.073)
= 0.129 cu ft/sec Because q,/q, > (f. ),, anti V.,,,< (1.),, the fluids arc in
s~llg floW,
(7+ 460]
q,, = 3.27 X ltJ ‘Zll,,(/{--1{.)—F-- 4. ‘Ihc cquislion~ given in lhc Slug Ilow swli~~n 1)[
Appendix C art uwd k) calculate ~ oncf ~,.
= 3,27 X 10 ‘ (0.875) (1,850) (57;– 115) Lktcrminc RqvsOlds numhcr. buhhk l?cyII~Ildj number
(5S7.5)/720 and slip vclocity (v,,).
= (J.[99CU tl/scc /v,l,. = 1.48X(55.$) ((),24’))(6.72),’1s 7.720 (c .6)
q, = 0.128 + ().199 = 0.328 CU ft/scc. Simx !hc bubbk riw vclocty i~ a nunlillc:ir uorlcla(iull.
iteratiOn is ncccs!,fily. ‘I’h CiWfL)~C. :t\\Lll~ing u, . I .i’~. bub-
The corrected miss tlow rates are
ble Reynolds numbw is
w,. = q,, (4.05 X 10 “y,, + 8.85 X 10 ‘ y,J<J
N,, = 1.488 (55.8) (0.249) ( 1.75)/1S = 2.0 I(J.
= 1,850 [4.05 X 10-”(0,942) + 8.85 X 10-;
[’, cannot bc read t’rom Fig. 9. ~“hu~ the cx~rapokstion
(0.75) (115)]
equation (Eq. C-7) ‘s used since IV,, < 3.000.
= 7.20 lb/see
v, = [0.546 -1-8.74 ‘ 10 “(7.720)1\’12;2”t[j1249i
IV,,= 8.85 X 10 ‘ q,,y,,(R – f{.)
- I.74 tl/see,
= 8.85 X 10-’(1,850) (0,75) (575 – I I5)
Determine liquid distribution c(wti,icicnt 1’ tind fricliun
= 0.565 lb/see factor /. Eq. C-13 is used to cvalmtc 1’ since v, < 10:
W, = 7.20 + 0.57 = 7.77 Ib(sec. 0.01271cg(18+- 1)
1’ = —–Zo–2~J)i,7,:,------ – 0.284 ~ 0.167 Jog 6.72
—. —.. . —.. -. — . [ 1
,;pa~enthes~.s !,adic,at~. the. ]m~,c number in 10’irk’s bookl’;, wht,w t h!,
various cm twlat ions are found,
+ 0.113 [0~ 0.249 = --0.097.
+~1.ivv oil viscosity. IIwul oil viscosity, a tmramdw in the correlation, Test limiting 1’with Eq. C-15:
is rtmd from Fig. 1:),
—.—— .——— — – (),097 > --0.065 (6.72)
> – 0.436:
therefore, 1’= – 0.097.
The $/D value from Fig. 7 is 0.0006. With this value and
?“H=+=WF=”+=+ the calculated N,,,, of 7,720. a frict”on
read from Fig. 6.
factor of 0.034 is

Evaluate ~ with Eq. C-4:

I ~.— 7.77 -t
——.55.8
— — (1.74) (0.0488)
g 20 I ,, ! ---—–A---L---J 0,328 + 1.74 [0.0488)
II
\l
2 + ( –0.097) (55.8) = 24.9 Ib/cu ft.
8 Ill Ewdluate ., with Eq, C-10:

8 1 I I 1< ~, = 0.034 (55.8) (6,72)’ 0,129 + 1.74 (0.0488) _ 0097

I , 4 I I,I)D 64.4 (0.249) [ .0.328 + 1.74 (0.0488) “ 1
60 80 100 200 300
TEMPERATURE
- “F = 2.26 lb/sq ft/ft.
5. The depth increment from Eq. A-3 is

JUXSI, 1967 837

 ——..
. .=

9 m

~
w, go

AD, = 144 [(
Ap, 1-
4,637 APS~ )]
~+ T, 1 500
\ o MEASURED
7.77 (0.199) — CALCULATED

= 144
[(
100 1 –
4,637 (0.0488)2 (720) )]
24.9 + 2.3
= 529
1
ft.
1000

* 1500
u“ Y
The true value of AD, is near 529 ft. The calculation will
converge very closely to this value even when the assumed
~ 2000
AZ is off by & 10 percent of the assumed value (540 ft) L
because, under these weli conditions, the pressure gradient H
is primarily controlled by the relatively temperature-insert. 2500 1

sitive density head. However, under those circumstances

where the friction gradient, which is temperature sensitive, 3000 -
is significant, iteration would be necessary should the cal-
culated value of AD differ from the assumed value by i I Y-
+ 10 percent.
6. The top of the next increment is fixed at 529 ft and
3500
–11~>
770 psi, and Steps 1 through 6 are repeated for the new 4000.
conditions. 600 800 1000 1200 1400 1600

7. The procedure is continued until 2 AD is equal to the PRESSURE - PSI A

total depth. The calculated pressure” profile is compared FIG. M-CALCULATED w MEASURED PRESSUREDROP — \~rELL 22.

against the measured profile in Fig. 14.

P38 JOURNAL OF PETROLEUM TECHNOLOGY