An Evaluation of Critical Multiphase Flow Performance Through Wellhead Chokes
An Evaluation of Critical Multiphase Flow Performance Through Wellhead Chokes
An Evaluation of Critical Multiphase Flow Performance Through Wellhead Chokes
Discussion
Various developments have been published- that
present theory and correlations for describing simultaneous liquid and gas flow though a restrictive orifice.
The correlations of Poettmann and BeckS were
intended to aid in the prediction of gas-liquid flow
through chokes. Their development followed the
original presentation by ROS1and was derived for
an average orifice discharge coefficient.
Poettmaun and Beck considered the polytropic
expansion of the gaseous phase of the fluid expanding through the choke. The polytropic expansion
theory was used successfully by Ros and is perhaps
the most rigorous development in the application of
orifice flow theory to oilfield conditions. Basically,
the derivation of any orifice relationship is dependent
on two main criteria. First, an expression must be
Mritten relating the flowing fluid specific volume and
velocity to the mass flow rate. Second, an independent
equation must be written inco~orating the behavior
of the gaseous phase of the fluid with pressure. The
above stipulations are met by the following relationships.
The energy balance a:ound a fluid flowing through
an orifice may be written as
144
Iv
vdv
V, dp +
*Now
V2
with
Mane
vi
Grand.
t?c
o.
(1)
The polytropic expansion equation relating the specific volume of the gas, (Vj VZ), to the confining
pressure, p, the polytropic expansion constant, b, and
the ratio of specific heat at constant pressure tc
specific heat at constant voiume, n, is
p(v, vl)~=b.
qm=
c.+
(2)
.(3)
%.,.....
vf*
where C is the orifice discharge coefficient, and Subscript 2 denotes downstream orifice throat conditions,
The solution of Eqs. 1 through 3 is given in Appendix A. The results are summarized below.
For critical orifice flow, the critical pressure ratio,
ec, defined as the ratio of the upstream pressure to the
downstream choke pressure, occurs when
e=e~;
iqm
= O; e. =
de
(4)
Vanezuela.
+R(p,Z)ec
-~2(n_11=n
1
.
_l
(5)
A multiphase flow equation describing the behavior of orifice flow may be used directly
to evaluate well performance as a function of choke size; upstream choke pressure;
choke temperature; producing and solution GOR; gas, oil, and water gravities; and a
discharge coefficient. The coefficient compensates for nonideal factors excluded in the
development of the equation and relates theoretical oil production rates through chokes
to field-measured rates.
This equation is applicable to that case of a singlephase gas since at R(p, 7)~ w
2Z
n-l.....
()
&c=
(6)
/1+1
For the purpose of simplicity and facilitation of
the solution of the energy balance, Ros reports that
values of n ands maybe taken as approximately 1.04
and 0,544, respectively. These values reflect average
flowing conditions based on numerous measurements
made in the oil field. The inherent errors created by
incorporati~g the values suggested by Ros may be
absorbed in the correlations of the discharge coefficient. The proper expression for the free-gas/liquid
ratio I?(P, T), in Eq, 5 is given by
PscT,zx
R(p,T)=x(R-RJ&
(7)
Pg + Fwop,o
9.
qTF
0 + 5.615
~. + (R5-J)
T;ZI;:C + ~wo
(8)
8C
qm=@h+(R-l?8)~~
[
198.6
+WOR#&
, . , . . (9)
1 3
Correspondingly, the orifice equation for multiphase flow, with oilfield units incorporated, becomes
90
(lo)
Because of the nonidealities associated with Eq. 10
and the use of the single-phase gas c~itical pressure
rat io, C may be calculated for any given combination
of fluids. The resulting values may be correlated with
the choke size for subsequent predictions involving
the same fluids or fluids with like properties.
Eq. 10 has been used to develop appropriate discharge coefficients from flowing well tests with different-sized surface chokes. The computations are.
summarized in Appendix B. The results show good
agreement in that the constant C is close enough to
1.0 to justify the theory used in developing the relationship. An example problem is also presented in
Appendix B to illustrate the application of the procedure,
Once C has been fairly well defined for a given
production province or operation, the equation may
be used in a conventional fashion to evaluate (1) flow
rates arising from changes in choke size; (2) wellhead
pressures arising from changes in choke size; and (3)
choke sizes necessary to achieve a given wellhead
844
Nomenclature
A
~:
B. =
c=
DC =
F.. =
gc =
n=
p, =
P2 =
pa, =
% =
go =
qTF
R,
R(p, T) =
{[ TIzI(RR,
)+ 151 P,](Y.+ .000217 Y,R, +FWOYW)}M
[~,z,(R
R.) + 11lpl] (yO+.000217y,R +FWOyW)
.
where C and Dc are the actual orifice discharge coefficient and orifice diameter, respectively. If C is
unknown, a value of 1,0 may be used to obtain a
reasonable estimate of orifice performance,
R=
c DC2p,
1.53 (B. + Fwo)%x
,,
T, =
T,c =
v
J/f~
VI =
z=
&=
P9
p.
plo
-fg
y. =
y,. =
1=
2 =
References
1.Ros, N. C. J.: An Analysis of Critical Simultaneous
Gas/Liquid Flow ihrough a Restriction and Its Application to Flow Metering, Applied Science Research
( 1961) 9, !k3C, A, 374.
2. Gilbert, W. E.: Flowing and Gas Lift Well Performance, Drill. and Prod. Prac., API ( 1954) 126.
3. Poettma.nn, F. H,, and Beck: R. L,: New Charts Developed to Predict Gas-Liquid Flow Through Chokes,
World Oil (March 1963 ) 95-101.
JOURNAL OF PETROLEUM
TECHNOLOGY
APPENDIX
(2)
V2
-@-
Vr)p,
1
[
n =
+ 2V,P,(I
-:)}%
where VZis the choke throat fluid velocity. The approach velocity, VI, has been ignored, since VI< <w.
The mass flow rate through the choke can be written as
qm=cAL.
e..
Vf,
e...
(3)
(A-2)
. VI ~=
2p,
()
n -1
{RYT[l
-(;)7]+(1
-:)}%
--
1
n
l+
R(p,
T)
atl&c,
PI
R =
.
AUGUST,
1974
pl
.
(4)
(n+n) = R(P,T)n
2(n1) 1 =
. . . . . . . (5)
&P-w
1
p, T,C ) 5,615
T8C
.
(L~-~
R, =
~..
(7)
V1 =
po+
(A-6a)
gR + Fw.pu
5.615
---------.
, ,-----------0.-
-+
where
P~*
~ R(p, T) + 0.76 ~ ,
R(p, T) + 0,56
. . . . . . (A-4)
p..
Tlzl
(A-3)
If we let the right-hand skie of Eq. A-3 equal a dimensionless flow rate, q, and note that the conditions for
critical flow occur such that
4=0,
de
R(p~.
()
(6)
eC= 0.544.
(A-1)
,..
1.04,
-(m
&
n-l
(v,,
n1
()
n-1
----
VI
-. -
,.
P2,
-----
f
----------
#-
---*
..
V*
(Throat
Conditions)
----------
84S
V1 =
(A-6b)
+ FWOP,C
0 + 5T5
p,
5.615
+ F,,. q.] 86,400
or
qTF =
(R R,)
q.
B.
5.615
+ F,rO
86,400
Fro == q,r,
1
.
(8)
TABLE l-CHOKE
yolI
Test
No.
(in
1?/4
in.)
(scf!STf3) _
(scfp6TB)
e g)
1.53
q. =
{[ TIz,(RR,)+
RJ+
C D, p,
(Bo -t F,oo)% x
(lo)
APPENDIX B
pg
-t Fwop,o
o + 5.615
.
p,
[T,z,(R
B. (R R,) T1z1P8.
P,TCC
5%3+
[
T,z*
198.6
qm
14,7
520 X 5,615
T,z,
q.Bo + qO(R R&)
(pRa)
&
Wate? =
Example 1
Assuming that the values for Test Well 3 (Test 3 of
Table 1) are valid, the discharge coefficient may be
evaluate-das follows:
PERFORMANCE
Air%
1
-iii
q.
(bbl%TB)
Fwo
Measured
1.0
1.01
1.01
nil
nil
1,010
230
0.885
0.865
0.885
1.0
1.0
1.0
1.01
1.01
1.01
nil
ni!
nil
1,065
1,065
nil
nil
iii-
505
120
120
owl
0$44
180
180
173
nil
nil
nil
325
465
665
120
UN
120
c,
(WI)
Cai.ulated
Discharge
Coefficient
900
229
1,477
1,142
704
1.019
1.042
1.023
1,392
1,078
618
0.963
0.979
0.955
1.122
1.004
1
2
3
32
16
32
24
16
1
2
3
32
24
16
363
337
341
nil
nil
nil
425
575
775
120
120
120
0.867
0.867
0.867
1.0
1.0
1.0
1.01
1.01
1.01
nil
nil
nil
1,505
1,190
720
1,340
1,055
590
1
2
3
32
24
16
118
107
108
nil
hit
nil
375
525
740
120
120
120
0.883
0.883
0.883
1.0
1.0
1.0
1.01
1.01
1.01
nil
nil
nil
2,088
1,752
1,068
1,977
1,516
877
1.056
1.156
1.218
1
2
3
32
24
16
127
120
102
nil
nil
nil
100
125
225
120
120
120
0.882
0.882
0.882
Lo
Lo
1.0
1.01
1.01
1.01
nil
nil
nil
370
350
290
576
412
336
0.842
0.850
0.863
1
2
3
32
24
16
197
185
160
nil
nil
nil
215
265
295
120
120
120
0.901
0.901
0.901
1.0
Lo
1.0
1.01
1.01
1.01
nil
nil
nil
725
575
415
879
675
353
0.825
0.852
1.176
1
2
3
32
24
16
377
391
290
nil
nil
nil
115
135
245
120
120
120
0.883
0.883
0.883
1,0
Lo
1.0
1.01
1.01
1.01
nil
nil
nil
305
295
190
376
249
229
0.811
1.185
0.630
36
407
0.092
1,215
100
0.863
0.72
1.091
nil
3,686
2,820
0.765
36
392
0.092
1,235
100 _
0.863
0.72
1.091
nil
3,898
3,060
0.785
10
40
410
0.092
1,115
100
0.883
0.72
1.091
nil
4,576
3,802
0.831
11
40
410
0.W.2 1,165
100
0.863
0.72
1,091
nil
4,728
3,905
0.826
561
0.092
100
0.863
0,72
1.091
nil
2,661
2,539
0.954
2,038
0.790
2,026
0.793
-i--
-i
12
30
13
30
483
0.092
1,265
100
0.863
0.72
1.091
nil
2,580
14
30
457
0,092
1,235
100
0.883
0.72
1.081
nil
2,555
846
1,175
R, scf/STB
341
o
775
scf/STB
~, psia
R,,
120
580
0.867
1.0
1.01
o
y.
Yo
Yw
F *O
R,) +
+ 111 X775=2.62X105.
YO+ 0.000217YJ?S + F~~Y~= 0.867
YO+ 0.000217YOR+ F@~ = 0.867 + 0,000217
X1 X341 +0 =0.941
1.53 X 162 X 775
1,53 x Dc x p,
BO%
=
l,ol~
= 3.0205 X 10.
qo = 3.0205 X 105 (
q. = 618 II/D.
Step 3 Evaluation of C
The theoretical rate is 618 B/D. Therefore the value
of C is given by
c
=
q. measured
qOcalculated
590
= 0.955
618
Example 2
Assuming that the orifice discharge coefficient of
0,955 is valid for orifice diameters larger than %
in., the increased capacity of the subject well for rhe
same flowing tubing pressure would be as follows:
Orifice Diameter
(% in.)
20
24
32
Capacity (B/D)
590 X (20/16)2 = 922
590 X (24/16) = 1,328
590 X (32/16) = 2,360
Example 3
For constant producing GORS a wellsflmvingtubing
AUGUW, 1974
7:0
775
900
1,000
1,100
%590
400
300
200
Step 2
From the appropriate curve in Appendix C compute
the orifice size required to support each rate with the
actual flowing tubing pressure.
a. The proper curve is for Rs = O; Fwo = O.
b. For Rate 1, the flowingtubing pressure is 700 psia
mid the rate is 700 B/D, This rate represents a
choke larger than % in. However the equivalent capacity for a %-in. choke with a 700-psia
flowing tubing pressure is 600 B/D. This value
represents an equivalent capacity because the
orifice curves contained in Appendix C reflect
a %-in. orifice. The 600 B/D rate represents
a starting point in the computation from which
a new orifice size corresponding to tlie 700 B/D
rate may be computed.
c. The actual choke diameter for a flowing tubing
pressure of 700 psia is computed from the foUowing relationship:
1..53DO*plc
90 =
(B. + FWO)Mx
q02
Dc,
847
256 X 700 ~
600
)
Dc, = 17.28= % in.
(Rounded to the next largest?& in.)
Dc,
>CO
,W
APPENDIX C
, ,CO
Fig. 5-Solution
ZCca
...
..
j
t
:
~
t
~
co.
,W
, .Ca
me
X0
,-
Fig. 6-Solution
N.(
,.M
mm
xc+
.
-
, >m .,
:
:
,*
!
~
>
X0
,W
.,.
,-
Fig. +Solution
848
,m
, we
. . . .
. . . . .
. . .
JOURNAL
OF PETROLEUM
TECHNOLOGY
---
Discussion
lNTERCOMPResourceDevelopmentand Engineering,Inc.
T. L. Gould, SP13-A.IME,
I would like to congratulate the author for an excellent contribution to the literature on multiphase flow
through chokes. The theoretical development will be
of great interest to those doing research in this area.
The capacity curves are of immediate value to those
involved in operational problems.
However, there is one point that was not stressed
in this paper that will prove to be of considerable
value in the future. That is, the critical pressure ratio
has been defined for multiphase flow through a choke
where the gas is expanded polytropically. When dry
gas is flowing, the critical flow pressure ratio is shown,
as expected, to be a function of the gas heat capacity,
When the gas/liquid ratio decreases, so does the
critical pressure ratio. The following discussion will
amplify and clarify the point.
To put this work in perspective, we must first
discuss the work presented by Fortunate, That paper
reports that, on the basis of field experien~, the
critical pressure ratio through a multiphase choke will
decrease with decreasing gas volume fraction. Fig.
D-1 is a redrafting of Fortunatis original figure and
shows values down to a gas fraction of 0.40, with
a corresponding pressure ratio of about 0,25 for
critical flow.
300~-
Values below a gas fraction of 0.40 are not available from Fortunatis work. However, Wallisz has
shown that the sonic velocity of a homogeneous mixture will pass through a minimum at a gas fraction
of 0.5, as can be seen from the following equation:
.
.
FLOW
Fig.
AUGUST,
0.3
04
OS
06
07
Preseure
Rahu,
WP, (with Pt.
a8
19,88
09
1974
-l
WDfm+IS*
A.ob0811
b eeImImlOn
001
01
da)
D-l-Velocity
of gas-oil mixtures through
chokes (from Fortunate),
FLOW
.~.
02
(D-1)
C = sonic velocity
CRIIKAL
0.1
where
U-CS171C4L
.-
10.
{0
n,qm-lwd
100
KOO!OOOOI
Fig. D.2-Comparison
correlations
849
Conclusions
are severid impcrtant conclusions to be drawn
from Ashfords Eq. 3 and Fig. 2, Currently a great
deal of research is being aimed at developing critical
and subcritical flow correlations for multiphase
chokes. In addition to measuring flow rates and pressures, some work should involve locating experimentally the boundary between critical and subcritical
flow as a function of gas/liquid ratio and fluid properties. This boundaxy must be determined before subcritical flow correlations ale developed.
In addition, Eq. 3 implies that the critical pressure
ratio approaches zero as the gas/liquid ratio approaches zero. This result is of course not reasonable,
but is a direct result of the assumptions used to
There
850
JOURNAL
OF PETROLEUM
TECHNOLOGY