Vous êtes sur la page 1sur 9

PRESSURE

ANALYSIS

Analysis of Pressure Build- Up and Flow Test Data


C. S. MATTHEWS SHELL DEVELOPMENT CO.
MEMBER A/ME HOUSTON, TEX.

Abstract Pi = pressure in well after infinite closed-in time in


infinite reservoir, psi (see Nomenclature for
Methods of using pressure build-up or flow tests to esti-
more general definition),
mate formation permeability, formation pressure and well q = oil-production rate at surface conditions, BID,
damage are reviewed. A number of phenomena which
fL = oil viscosity, cp,
cause actual pressure build-up behavior to differ from the B = oil formation volume factor,
idealized case, such as the effects of boundaries, skins, k = permeability of formation to oil, md,
inhomogeneities, partial penetration and two-phase flow, h = sand thickness, ft,
are discussed. It is concluded that the same method of t = flowing time of well, and
analysis of build-up curves can be applied with slight modi- At = closed-in time of well.
fication to oil reservoirs, gas reservoirs and reservoirs pro-
ducing both oil and gas. The calculations can be carried Derivation of the factor 162.6 is given in the Nomencla-
out on a form sheet, a copy of which is included in the ture. A graph of this equation is shown in Fig. 1, along
paper. Examples are worked out for four different types with actual field data for a new well in an oil reservoir. For
of reservoirs. this case, the theory and practice agree well. To analyze
this curve, note that the slope of the curve is equal to
the coefficient of the logarithm term in Eq. 1. Therefore,
Introduction
kh =
162.6 qfLB (2)
Although the basic theory of pressure build-up behavior slope in psi/cycle
in wells was developed many years ago, important con- Extrapolation of the straight-line section to an infinite shut-
tributions since that time have extended the original ap-
in time, [(t + b.t)/At] = 1, gives the value for Pi' as
plicability to a much wider variety of situations. The pur- shown.
pose of the present paper is to summarize the present
status of pressure build-up theory and of its applicability. Bounded Homogeneous Reservoir
The approach in this paper will be to start with the The pressure build-up behavior of a well in a bounded
simplest type of pressure build-up curve and to show how homogeneous reservoir is given by
reservoir rock properties, reservoir fluid properties and
p" = Pi -
qfLB[
162.~ log~+
t + b.t
well bore conditions tend to distort the idealized picture.
Methods for taking these distortions into account and for
determining values of reservoir formation properties from 1320r------------------------------------------,

build-up curves will then be considered.


1300
C>
Build-up Curves 'iii p,= 1280 psig
':1280 '~
Infinite Homogeneous Reservoir w
a:
:::>
/
"
The theory for the pressure build-up behavior of a well ~ 1260
producing a single, slightly compressible fluid from an w
a:
Q.
infinite homogeneous reservoir was presented by HorneL'
l;! 1240
According to him, the equation for build-up when oil is o
III
the only phase flowing is I
j 1220
162.6 qfLB t + b.t w
Pw = Pi - kh log ~ . (1) 3:
1200
where Pro = pressure in well during build-up, psi,
Original manuscript received in Society of Petroleum Engineers office 10
Nov. 3, 1960. Revised manuscript received June 15, 1961. Paper pre (t+At)/At
sented at SPE Formation Evaluation Symposium, Nov. 21-22, 1960, in
Houston. Fig. I--Pressure build-up for a nearly ideal reservoir
'References given at end of paper. (from Horner').
SPE 1631-G
862 JOURNAL OF PETROLEUM TECHNOLOGY
Y(t + M) - Y(~t) - Y(t)] boundary fluid flows toward that well, and on the other
2.303 . (3) side toward another well. For some time after a well is
This equation differs from Eq. 1 only by the addition of closed in, it can be treated as if its drainage boundary
the term Y (t), which may be called the boundary effect still exists. Thus, a well surrounded by other wells will
and is a function of the shape of the drainage boundary have a build-up curve qualitatively similar to that of one
of a well and of the production time. For a square drain- well in a bounded reservoir. For very long closed-in times
age boundary, the boundary effect causes the pressure this is not true (see "Interference Tests").
build-up curve to bend over as shown by the field example' Values for the average pressure p for reservoirs which
in Fig. 2. The extrapolated value from the straight-line have not been closed in long enough for the "bend over"
section is called Pi; in a bounded reservoir, Pi is best shown in Fig. 2 to be observed may be obtained by the
defined as this extrapolated pressure (see Nomenclature). method of Ref. 3.
The final static value of the pressure is called 'p the average Wellhore Damage
pressure. A plot of (Pi - p)/(70.6qfLB/kh) is given in As just mentioned, the effect of production at other
Fig. 3 for one w,ell in the center of a square drainage area. wells is to flatten the build-up curve at long times. Quite
It may be seen that the difference between Pi and in- p logically then, the effect of well bore damage would be
creases as production time increases. to distort the curve at early closed-in time, as shown in
The function Y (t) is related to the function plotted in Fig. 4 (from Ref. 4). The effect of wellbore damage on
Fig. 3 by pressure distribution is to cause an extra drop right at
the wellbore, as illustrated in Fig. 5 (from Hurst'). This
Y(t) =
0.00331kt _ Pi + P extra drop in pressure has been called a skin effect. Shortly
CP.fLcA 70.6 qjJ-B/kh after shut-in, the well pressure should rise by the amount
The quantity (p - p)/(70.6 qfLB/kh) is plotted in Ref. ~P'kin shown on Fig. 5. Thus, the order of magnitude of
3 for a num'- of cases. In that paper it is called (P* - the skin effect should be evident from the difference be-
p)/(qfL/ 47Tkh ). tween flowing pressure and the pressure shortly after shut-
If there are other wells in a reservoir, the effect of pro- ting in. Quantitatively, the skin effect S as defined by van
duction at the other wells is to cause a well to be sur- Everdingen4 can be calculated from
rounded by a drainage boundary. On one side of this S = 1.151[ Plho",' - PI
6.p (1 cycle)
log (_k_._.
CPfLcr"
w
)+ 1.07]. (4)
3oo,.--------------------,,..
and the pressure drop in the skin or damaged zone near
the well by
1200
6.p = 6.p (l cycle) X 0.87S
S';'1 (5)
.~ 1100 The value for Pi hon>' must be taken on the straight-line
w .6 t I hours
a: 100
~ 1000 I 10
en
w
a:
a. 900
w
--'
o
7 800
~
o
f-
f-
om 700

600

50~~04~~~~~IO~3~~~-~IO~2~-L~L--I~oULLL~L--J

(t + L:.t)/L:.t
Fig. 2-Pressure build-up showing boundary effect (from
Mead2 ) .

7r--------------------~

(t+l>!l/l>t
6
Fig. 4-Pressure build-up showing effect of wellbore dam-
age and after-production.
:C5
.>t: BORE
"-
m
;'4 ------=~-_p~~~~IJRE

'"0
....
~3
G SKIN
PRESSURE IN
FORMATION
la. OR
I ZONE OF
52 DAMAGE
p PRESSURE DROP

~
t; 0

ski n ACROSS SKIN

.1 1.0 10 FLOWING PRESSURE


.000264 kt / CPfLCA Fig. 5-Pressure distribution in a reservoir
Fig. 3-Correction to Pi for well in square drainage area. with a skin (from Hurse).

SEPTE1I1BER. 1961 863


pomon of the pressure build-up curve. If the build-up Fissured limestone reservoirs also show this type of
curve is not straight at one hour, it is necessary to extrapo- build-up." In fact, it appears that most reservoirs with
late the straight-line portion backwards, as shown in Fig. two distinct types of permeability can show such behavior.
4. Note that a scale for (t + t::..t)/t::..t and also a scale for The tail is greatest when the permeability difference is
t::..t are used to facilitate extrapolation to P, bon" A meas- greatest, as shown by the example for a fissured lime-
ure of the efficiency of completion can be obtained by stone in Fig. 9.
comparing the actual productivity index J and the ideal
Gas Wells
(no skin). The ratio of these two quantities is
Aronofsky and Jenkins" have shown that gas wells be-
. J actual Pi - Pi - t::..p 'kin (6) have in a manner very similar to oil wells. With acceptable
Flow Effi clency = --- = ~---=-'-----=---
J id"n Pi - Pi accuracy, the build-up equation is given by
This is quite similar to the condition ratio of Gladfelter,
P = Pi 162.6 qy/LgBYlogt -I- t::..t (7)
et al: Although the use of Pi in this equation is strictly k,J7 t::..t
correct only for an infinite r,eservoir, the error introduced
where
is small.
Well Fill-up Effect B ,,= Z TT [ P"
-~--, - -J =
gas f ormatIOn
. volume factor
" (p, T p) /2
at arithmetic average reser-
The idealized theory, as exemplified by the straight line
on Fig. 1, assumes that a well is closed in at the sand face voir pressure.
itself and that, after closing in, no production at all enters Note that Bo is computed at the arithmetic average reser-
the wellbore. In practice, howev,er, a well is closed in at voir pressure. This av,erage changes during build-up, so the
the surface, and for an instant after this time fluid con- slope of the curve of P vs log [( t + t::..t) / t::..t] should also
tinues to flow into the well bore at an unchanged rate. Only change slightly. This change will usually be negligible, and
after this fluid accumulates in the bore is the effect of it will usually be satisfactory to approximate P by Pi in the
closing-in at the surface transmitted to the formation. For equation for Bg. When the gas ,equation is used in the form
this reason there is a lag in the build-up at early times, of Eq. 7, its analogy with the oil equation (Eq. 1) is em-
as shown by Fig. 6 (adapted from Ref. 7). If both gas phasized. In Eq. 7, the gas flow rate needs to be expressed
and oil are flowing into the well bore , the rise of gas to in barrels per day at standard conditions. Trube13 has
the top of the well during shut-in and the fall of oil to the giv,en curves for gas compressibility Cg An example of a
bottom can cause the pressure to "hump", as shown in gas-pressure build-up curve is shown in Fig. 10. The form
Fig. 7 for an example well from South Texas. s One of of the gas equation used by many authors may be obtained
the ways of decreasing these well bore effects is to use a by substituting for Bg in Eq. 7.
recently developed tool which closes in the well at the 7 0 0 , - - - - - - - - - - - - - - - - - - - - - - - - - - - - -__________- ,
bottom" This tool allows interpretation of a pressure
build-up curve after a much shorter closed-in time.
600
Effect of Reservoir Heterogeneity
A reservoir consisting of stratified layers of different '"0.500
permeability was shown theoreticallylO to have a "tail" a:-
=>
o
to the build-up curve. An example of this for a well ..J 400
:;
which is completed in two strata of different permeability <D

is shown in Fig. 8. The pressure build-up takes place in l:! 300


:::>
both layers at early times, giving a straight-line section (f)
(f)
as shown. The slope of this straight-line section gives the ~ 200
total kh for both layers. After build-up is complete in the Q.

most permeable layer, the less permeable layer - which 100 ~-


is at a higher average pressure - begins to feed fluid into
the more permeable layer. This causes the rise above the
straight-line section. Finally, an equalization will occur, O~I-===~~ilil~O--~~~~IO~O~L-~~~IO~O-O--L-LL~I~~OOO
CLOSED-IN TIME, minutes
and the curve will flatten as indicated by the dotted line.
Fig. 7-Hump due to rise of gas in tubing after closing in.
2ooo,-------------------------______________-,

1800
3000 -
~

o "GAS-FILLED WELL BORE '"a. STRAIGHT-LINE SECTION"


'" 1400
':
w
a: __-"::0----
:::>
w :il2900
a:
=>
~ 1200 '"
a:
Q.
w w
a: ..J
Q.
o
1000 :t:

~ 2800
f0-
800 b<D

600 'J FLOWING PRESSURE


,0001 ,001 2.01 ,I
,000264 kl1t1</:>fLcre 2700~1--~~~~10~-L-L~~~IOO~-L-LLL~I~0700~~--~I~~000
SHUT-IN TIME, minutes
Fig. 6-Well fill-up effect (after Miller, Dyes and
Hutchinson'). Fig. 8-Build-up in a two-layer reservoir.

8M JorR:\',\L OF PETIIOU;t'M TECI/:\,QLOGY


Two-Phase Flow Effect Fall-off Curves
The shape of the pressure build-up curves when both In water-injection wells it is natural to determine forma-
oil and gas are flowing is very much the same as when tion properties from a fall-off curve determined by closing
only one phase flows. The interpretation is also much the in the well." The methods already discussed in this paper
same. Eq. 2 is also applicable during two-phase flow, as for analysis of build-up curves can also be used for fall-off
was justified by Perrine." The skin effect and wellbore curves if the mobilities (k//-t) of water and oil are about
damage can be calculated from Eqs. 4, 5 and 6, provided the same. However, when the data for a fall-off curve are
that the total mobility (k o / /-to + kg//-tg + kwl/-tw) is used plotted in the usual manner, it is often difficult to deter-
instead of k o / /-to and provided the total compressibility of mine just where the straight-line section begins and ends,
oil, gas and water is used instead of that for oi!." Similar as in the example in Fig. 2. A better method for analysis
methods can be used for obtaining average pressure. of these curves is to use the log (p - p) -vs-!::..t plot, as
At high drawdown, the flow of gas can introduce an shown in Fig. 12. This same plot may be used both when
apparent skin effect even when there is none. An example oil and water differ widely in properties and when they
of a pressure and a saturation distribution curve for a do not.
reservoir in which both oil and gas are flowing is shown The method of plotting these curves is similar to that
in Fig. 11. A flow efficiency of about 70 per cent is calcu- for layered reservoirs. It depends on finding, by trial and
lated for this well, when actually there is no skin at all.
The higher gas concentration around the well bore causes
p
error, a value of which makes the fall-off curve linear at
large time. This value is shown to be 230 psi in Fig. 12.
the apparent damage, as was shown by Perrine." This type of plot works at large times because the well
Partially Penetrating Wells
pressure is approaching the average pressure exponentially.
Values of kh and skin factor can be obtained from the
Nisle" and Brons and Marting" have studied the pres- slope {3, and the intercept b,. Ref. 17 should be consulted
sure behavior of such wells. Their work shows that the for details of the method.
effect of partial penetration is to introduce an apparent
well bore damage. The amount of extra pressure drop Interference and Flow Tests
caused by partial pcnetration can be found from Ref. 16.
Interference Tests
When one well is shut in and its pressure is measured
2000~------------------------------------~--, while others in the reservoir are produced, the test is
termed an interference test. This type of test can give in-
1900 formation on reservoir properties that cannot be obtained
iii from pressure build-up tests. In particular, values of the
Co

Ii 1800 porosity can be obtained from these tests, whereas they


::>
(I)
cannot be obtained from pressure build-up. The simplest
(I)

l:! 1700
method of interpretation is based on superposition of the
0.. effects of each of the producing wells at the well in ques-
'"c!x 1600 tion. When the boundary of the reservoir is near, it also
needs to be taken into account (for example, by imaging
2 the producing wells). Collins" has described a method for
~ 1500
b analysis of interference tests. Another method is given as
ID follows.
1400
When the outer boundary effect can be neglected, the
1300L-~-L~~~--~-LLU~--~~~~L-~~~~W
pressure at the observation well is given by
.1 1.0 10 100 1000 162.6q/-t B I t +!::..t 70.6/-t B
CLOSED-IN T1 ME, l> t, hours
P = Pi - kh og~+kh
Fig. 9-Build-up in a fissured limestone reservoir (from
Pollard")
[q, E{ ~O~~~:;)+ q, E{~O;;; ~::)+ 1(8)
6t, hours
.B2~-------------------------------,

.BO - 1400

~.7B 1200
;:: iii
l.)
<I 0-
:::76 1000 _
UJ
n:
:::>
(f1

BOO ~
n:
0..

-- 600

400

1.0 5 10 50 100 500


10" RADIAL DISTANCE FROM CENTER OF WELL, It
(1+61)/61
Fig. II-Pressure and saturation distribution when both
Fig. IO-Build-up in a gas well. oil and gas are flowing.

!'jEPTEMRER, 19fil
where q, is the production rate of Well No.1, q, the pro- Interference tests require considerably mo.re time than
duction rate of Well No.2, etc.; t, and t, are production pressure build-up tests, especially in low-permeability reser-
times of these wells; a, and a2 are distances from Wells 1 voirs. Note that a minimum of two mo.nths was required
and 2 to the observation well. Variations in rates q, and q, for the case shown in Figs. 13 and 14. This reservoir had
may be taken into account by the principle of superposi- a permeability o.f a few millidarcies.
tion. ' Eq. 8 is strictly applicable only to single-phase flow
Flow Tests
and, thus, is best used when reservoirs are above the
bubble point. The log term in this equation gives the effect Flow tests can be interpreted in much the same manner
of producing and shutting-in the o.bservatio.n well itself. as pressure build-up tests. In particular, the slo.pe o.f the
The Ei terms give the pressure drop at the distance a1 or a, curve o.f flowing pressure vs flo.wing time is the same as
fro.m the pro.duction Wells 1 o.r 2. Elkins" has presented the slope o.f the curve o.f shut-in pressure vs shut-in time.
a go.o.d example of u&e o.f interference tests. In tight rocks Wellbo.re damage can also. be determined in a similar
it may be necessary to. shut-in a well fo.r a Io.ng time to. manner. A metho.d fo.r determining reservoir size from a
observe the eflects o.f interference. An example is sho.wn flow test has been described by Jo.nes." The method de-
in Fig. 13. The do.tted line in this figure was o.btained pends o.n determining the point at which the plo.t o.f pres-
by extrapolating the linear portio.n o.n the log [(t+ M)/b.t] sure vs Io.g (flowing time) departs from linearity. The
plo.t, as shown o.n Fig. 14. To. determine the magnitude o.f method is useful for estimating minimum size of o.il ac-
cpp.c/k by this proposed metho.d, it is necessary that the cumulatio.ns without sacrifice o.f pro.duction, as is necessary
difference between the extrapolated curve and the observed if a well is clo.sed in fo.r determinatio.n of average pres-
points be measurabIe with reasonable accuracy - the sure.
magnitude o.f this diflerence is equal to. the interference
terms (the Ei terms) in Eq. 8. If the difference is zero, Discussion
then the co.ntribution o.f the Ei terms is zero and cpp.c/k Time Required for Build-up
canno.t be determined. A difference o.f 40 to. 50 psi is and Interference Tests
necessary in order for .cpp,c/k to. be determined with some A well should be clo.sed in fo.r pressure build-up Io.ng
reliability. enough to. allow the straight-line section indicated o.n Fig.
The quantitycpp.c/k can be o.btained by assuming values 6 to. be clearly delineated. The longest time will be re-
fo.r this quantity, substituting in Eq. 8 and computing the quired fo.r deep, low-productivity, pumping wells no.t
o.bserved pressures. The value o.fcpp.c/k which gives the equipped with packers, because in these a Io.ng period o.f
best fit to. the observed pressures is assumed to. be ap- "after-production" will be needed to. fill the wellbore with
plicable to the reservoir. From known values o.f p, and c, liquid and co.mpressed gas. Only after the influx into. the
a value fo.r cp/k can be o.btained. Since kh can be o.btained
from the slo.pe o.f the pressure build-up curve, a value fo.r
cph can be o.btained as the product o.f these two quantities.

'"a.2000
1000~----------------------------------------.
W
800
!3
I/)
1900
I/)

~ 1800
600 Q.
--
42 psi

400
b, = 360 psi

i3, =
(log 360-log 21) 2,303 .."',., ."' ."',
I

t!-
1
.."' ."'
,
,
"'"', "'
'?
~
250 x 10' T
m &; 2 ~ .'..
= 1.14 x 10- 0 sec-'

200 0,

" '0,
0, 40 80 120 160 20e
x,, CLOSED-IN TIME, days
0,
'00 x, '0 Fig. 13-Interference test in a low-permeability reservoir.
a. , ....
..::
,a. 100
x,, 0
I
a.
x, 'O'Q 2000r-----------------------------------------,
80 'x,, '0'0 CURVE B
.... 0 p = 200 psi 1900
60
'x\ "'0
"'0,0 EXTRAPOLATED BUILDUP PRESSURE .-
\
, '"'b'o
0>
'in
a. ~ ."""","-
x,, W 1800
, Q: ,,"
40 x,, :::>
I/)
.-"
, ~ 1700
x, Q:
Q.

1\ W
c5 1600
\ :r,
\
20 :::;;
X
\
, i? 1500
\
A
5
m
X CURVE
\
\ Ii = 260 psi
\
I
10
0 50
X
100 150 200 250
CLOSED-IN TIME, thousand seconds (t+~t)/llt
Fig. 12-Pressure fall-off in a water-injection well. Fig. 14--Interference test in a low-permeability reservoir.

866 JOURNAL OF PETROLEUM TECHNOLOGY


well bore becomes small can the simple theory of pressure TABLE I-EXAMPLE OF SUBSURFACE PRESSURE CALCULATIONS FOR
RESERVOIR ABOVE BUBBLE POINT.
build-up be applied. A simple method of determining when Test Data:
the rate of influx has become small is to determine casing- Test Dale Jan. 4, 1951 Company Shell
head pressure (CHP) and tubing head pressure (THP), Producing :OFo~r~m--:aI7io~n--------;D::-o'-:-lo~m-;Cit:-e Lease Lend
as well as bottom-hole pressure (BHP) during build-up. riole Size (in.) . 4 3/, Well No. 1
Cum. Prod. N p oo(b-o-b~I)----------:-14~2cc,OoOl~0 Field Center
The difference between BHP and THP or CHP is directly
~~:~:~~z:~r~~~I~i:~Otdih:~~s~I~~200-4O-:-N:-p"/q-~1:-::3C-=,!:-::~~~
State Texas
proportional to the mass of fluid between these two levels.
A curve showing (BHP-THP) and (BHP-CHP) can be
I. Calculation of kh (md-ft) and k (md)
plotted. When the slope of this difference curve falls to, say,
only 10 per cent of its initial value, the influx is probably kh = 162.6 q/L B . k = J<"-
tlp (1 cycle) , h
small enough that the simple theory can be applied. At h 69.0 It /L 0.80 cp
this time, the influx rate into the wellbore is only 10 per q 250 B/D B 1.136
tlp (1 cycle) _ _ _ _ _--'7c.:0'-'p:.:.:si
cent of the production rate at time of closing-in.
kh = 162.6 X (250) X (0.80) X (1.136)
Miller, Dyes and Hutchinson' have shown that boundary (70) = 527.75 md-It;

and interference effects begin to distort the straight-line k = (527.75) = 7 65 md.


(69) .
section of a pressure build-up curve at a time 1:"1 iI =
0.000264kt:..t/1>fLCr', of the order of 0.1. This gives an II. Calculation of Skin Effect S; and Pressure Loss Due to Skin .6.p skin
(psi)
upper limit to required shut-in times. Most of the points
which will be used in pressure build-up analysis should fall 5 = 1.151 [p;ph~lurCY~I~1 ~ log (l'>/Lc r2
k
J + 1.07].
in the range of dimensionless time t:..t D from 0.005 to 0.1. !J.p 'kin = !J.p (1 cycle) X 0.87 S.
k 7.65 md rw 2.375 in.
Interference tests will normally require that the observa-
I'> 0.039 P ho"r 4295 psig
tion well be closed in for a time t:..t D = 0.2 to 0.3.
1

0.80 cp PI 3534 psig


6.8 X 10 "vol/vol/psi !J.p (1 cycle) _ _ _ _ _ _7~0~p'--si
Selection of Wells for Production Stimulation
5 = 1.151 X
As discussed by Gladfelter, et al," there are three main (4295) ~ (3534) I (7.65) 1
causes of low-productivity wells - (1) a "skin" near the [ (70) ~ .og (0.039) (0.80) (0.0000068) (5.64) + 1.07
wellbore, (2) low permeability throughout the reservoir = 5.91.

and (3) lack of pressure. By taking a pressure build-up !J.p ,k;n = (70) X 0.87 (5.91) = 360 psi.

on a well prior to a well-stimulation treatment, it is possible III. Calculation of Productivity Index (BID/psi) and Flow Efficiency
to determine which of these three are causing the low J (actual) = -~q-. ) (;"'" I)
q
= -,-----C- - ; - -
productivity. If the trouble is a skin, then a wash treat- Pi - PI (Pi - PI) - ~p skin
~p skin 360 psi Pi 4585 psig
ment or a small-volume fracture treatment is in order. If q 250 B/D PI 3534 psig
the formation has a low average permeability then a deep- (250)
penetrating, large-volume fracture will be required. If the J (ootu.1) = (4585) ~ (3534) = 0.238 B/D/pSl
difficulty is due to lack of pressure, some type of injection (250)
) (;de.1) = (1051) ~ (360) =
.
0.362 B/D/pSl
such as of water or gas will be required to expel the oil.
. J (actual) 0.238
Because they aid in diagnosis of the difficulty, pressure Flow EffiCiency = J (i(Ical) = 0.362 = 0.657
build-up curves greatly aid those involved in selecting wells
for production-stimulation treatments. To check on whether the correct section of the build-up
curve was used, the maximum dimensionless time of shut-
Relative Importance of Various Types of Build-Ups in t:..t D is calculated.
So many variations from ideal behavior have been dis- 0.000264 (7.65) 136
cussed that the reader may begin to feel that the un- t:..t D = 0.039 (0.80) 6.8 X 10- 6 (2,640)" = 0.19.
tangling of real situations is hopeless. This is not true.
Most pressure build-up curves will look like that shown Since most of the points used in the analysis are within
the interval 0.005 to 0.1, it appears that the correct portion
in Fig. 4 and can be analyzed very simply on a form
sheet which will be explained in the next section. Far less of the build-up curve was analyzed. Casing and tubing
frequent are the "hump" build-ups shown in Fig. 7. This pressures were not taken during build-up, so it was not
type usually occurs in deep, permeable flowing wells equip- possible to compute by this method whether the correct
ped with packers. The heterogeneous type shown in Fig. 8 build-up section was analyzed.
may be mOT>e common than is generally supposed. To ob-
serve the tail, it is often necessary to shut-in the well for a
considerable portion of time. If reservoir pressures seem
to give incorrect results when used in the material balance,
it may be helpful to close-in one well for a week or two
to see whether low-permeability, relatively un depleted lay-
ers are present which may give a significant tail to the
build-up curve. The example calculations shown in the
following section and tables should help clarify the hand-
ling of actual pressure build-up curves.

Example Calculations
Oil Reservoir
A pressure build-up curve for an oil reservoir above the
bubble point is given in Fig. 4. After the slope of the curve
is measured and other pertinent data are entered on the
example calculation of Table 1, the calculations for k, S {I + L>tl / L>t
and flow efficiency can be carried out as shown. Fig. IS-Pressure build-up in a heterogeneous reservoir.

SEPTEMBER, 1961 867


Two-Type Permeability Reservoir Gas Reservoir
The type of pressure build-up curve shown in Fig. 15 The form sheet used for oil reservoirs can also be used
cannot be analyzed completely by the foregoing method. for gas reservoirs. It is only necessary to convert the gas
In particular, the curve cannot be extrapolated to give p, rate in cubic feet per day to barrels per day by dividing
as was the build-up on Fig. 4. This extrapolation should by 5.615. The gas formation volume factor is obtained
be made by using the type of plot shown on Fig. 16. A from
value for the average pressure p is estimated, and log
B = Z~ po
(I) - p) is plotted vs t::.t the closed-in time. The value of p y Ta (p, + pf)/2
which gives the best straight line is the value for the average The method of obtaining Bu, C g and fly is shown in Table
pressure in the drainage radius of the well. Note that, for 3A. The rest of the analysis for Fig. lOis straightforward.
values of p which are too small, the plot of log (p - p) For this example,
vs t::.t bends down. For values which are too large, the 0.000264 (6.9) 24 = 0.0056.
curve bends upward. Sometimes this upward bend is diffi- t::.t D = 0.16 (0.02) 3.5 X 10-' (2,640)'
cult to detect, and all the plots appear to be straight when Although many of the build-up points are outside the
p is above a certain value. In such case it is usually satis- recommended interval, a study of Fig. 10 leaves little doubt
p
factory to take as the lowest value for which the curves that the correct section was analyzed. In cases such as
become straight. A value of kh for both layers is obtained this it is often valuable to plot (BHP-THP) vs time to
from the slope of the earlier straight-line section indicated ensure that the influx rate is small at the end of build-up.
in Fig. 15. An approximation for the skin effect is ob-
Reservoir Producing Below Bubble Point
tained from the same equations as for a homogeneous reser-
voir. Calculations for these reservoirs can be handled on the
same form sheet as that for the oil reservoir. The only
An example calculation for a two-type permeability
reservoir is shown in Table 2. For this case Pi and pare TABLE 2-EXAMPLE OF SUBSURFACE PRESSURE CALCULATIONS FOR
approximately equal, but this is often not true. It should be TWO-TYPE PERMEABILITY RESERVOIR.
Te!o;t Data:
emphasized that the skin effect and flow efficiency obtained Hole Size (in.) 5
by the method shown are only approximately correct for Cum. Prod. N" (bbJ)------ - ---------------2390
the two-type permeability reservoir. The maximum dimen- Stabilized Daily Prod. q(bb-I) 130
sionless shut-in time is Effective Prod. life t (hours) = 24N p /q 441

0.000264 (23.7) 49 I. Calc.;Irtion of kh (md-ft) .md k (md)


t::.t D = 0.12 (0.37) 1.5 X 10' 2,640), = 0.066. h 27 It iJ. 0.37 cp
~ 130B/D B 1.39
This is within the interval 0.005 to 0.1, so the correct IIp (1 cycle) 17 psi
portion of the build-up was used. ~h = 162.6 X (130) X (0.37) X (1.39) = 39 .f. k = (639.5)
(17) 6 .5 md t, (27)
= 23.68 md.
100,---------------------------------------------,

0.37 cp PI 2620 psig


------;I-.5;--:-X-;-c:l0;;-;c5-vo""I~/vc-:o-;.I/'p~si llr:> (1 cycle) 17 psi

_ [(3723) - (2620) (23.68) + 1 07]


S -- 1.51 (17) 109(0.12) (0.37) (0.000015) (6.25) .
= 68.14.
IIp "kin = (17) X 0.87 (68.14) = 1008 psi.

III. Calculation of Productivity Index (BID/psi) and Flow Efficiency


q
J (actual) =- ---q--. J (i tl"l1 1) ,-jII
= :-(P-i---P""f)~_----C.:lC--p-,-_
Pi - PI

!lp skin 1008 psi P 3775 (used as pi! psig


-----------~ PI
q 2620 psig
(130) .
J (actual) = (3775) _ (2620) = 0.1125 B/D/pSl
(130)
J (idcal) = (1155) _ (1008) 0.8844 B/D/psi

Flow Efficiency = J (actual) = 0.1125 = 0.1272


J (id('al) 0.8844

TABLE 3A-~XAMPLE
SHOWING CALCULATION OF
p ~ 3775 Bg , /La and c" FOR A GAS WEll
T p. Ta = 520 o R.
20 Bo = Z - - - - - -
Ta Pi + PI pa = 14.65 psia.
---2--
Pseudocritical temperature Tc = 420oR.
PseudocriticaI pressure pc = 663 psia.
Reservoir temperature T = 200 ;- 460 = 660<'R.
Pi + PI = (2910) + (2422) = 2666 psig = 2681 psia.
2 2

Til = T = 660
To 420 = 1.57 ~
Z = 0.809 (from Ref. 22).
pl/ = (Pi + P/)/2 =
2681 = 4.04 I
pc 663 ,
660 14.65
1o0L-------".O------,-'-0------,L- - - - - 0
O 4 50 60 70
Bo = 0.809 X 520 X 2681 = 0.005611.
t::. t, hours Us.ing TR and PR and a gas gravity of 0.8, J.Lg i'i obtained from Ref. 21, figs.
4 and 6, as /Lo = 1.7 X 0.01185 = 0.0201 cpo Also, Co is obtained from
Fig. 16--Pressure build-up in a heterogeneous reservoir. Trube'" as 0.23/663 = 0.000347 psi- ' .

8611 JOURNAL OF PETROLEUM TECHNOLOGY


tABle 3-5U8SURFACE PREs~URE CALCULATIONS FOR A GAS whl. 1; corresponds to infinite shut-in time in
Test Data:
an infinite reservoir, psi
Hole Size (in.) ______________7_ _ _ _
Cum. Prod. N" (bbl) 1.[38 X 10' (6390 MMcf)
pw = pressure in well, psi
Stabilized Daily Pwl. q (bbl) 536,900 (3.01 MMcf/D) P1 hou, = pressure read from linear portion of pres-
Effective Prod. Life t (hour) - 24N plq 50.8 X 10' - sure build-up curve at one hour closed-in
r. Calculation of kh (md-ft) and k (md)
time, psi
h 84 fl I'" 0.0201 cp PI = flowing pressure prior to closing-in well
q--536.900--=_-B/D By 0.005611 for a build-up, psi
!J.p (1 cycle) 17 psi
p= average pressure in drainage area of a well,
kh = 162.6 X (536,900) X (0.0201 )~.005612l 579.2 md.ft; psi
117)
~_7~2) = 6.90 md.
j,p (1 cycle) = slope of linear portion of pressure build-up
184) curve in psi/logw cycle
Tht.., remainder of tie analysis is straigj,!forword. j,p ,kiu = pressure drop in "skin" region next to well-
change necessary is to calculate the total mobility and total bore
compressibility. Calculations for the curve shown in Fig. q = production rate prior to closing-in well for
17 are made in Table 4A. The total mobility and the total a build-up, B/D at suface conditions
compressibility are then used on the form sheet in calcu- r", = well bore radius, in.
lasting S. These steps are exemplified in the example calcu- re = drainage radius, ft
lation of Table 4. In this case, S = skin factor
0.000264 (92.9) 24 t = time of flowing, hours
6.t IJ = --- -- --------------- = O.OIS. j,t = closed-in time, hours
(0.15) S.5 X 10' (500)-
6.t II = closed-in time, dimensionless
Since this time is in the interval 0.005 to 0.1, the correct 0.000264k6.t
portion of the data was used in build-up analysis.
CPfLc r/
Y (t) = boundary effect
Conclusions cP = porosity, fraction
It appears that a single form sheet can be adopted with
slight modification to the analysis of nearly any pressure - Ei ( - x) = J" Cds
s
build-up curve. Pressure build-up behavior now appears to
be reasonably well understood.
TABLE 4A-EXAMPLE SHOWING CALCULATION OF TOTAL MOBILITY Ik/I-') ,
AND TOTAL COMPRESSIBILITY C1' FOR A RESERVOIR BELOW BUBBLE POINT.
Nomenclature
( J<..)=~+!<.u.+~.
A = drainage area of well, sq ft '" JLo JLy
I J1.w

{/" a, = distance between observation well and pro- ( -;) t = l!J.p~~~~~e) h [ Boqo + B'I ( qa - qvR, ) + Bwqm 1
duction well (No.1 or No.2), ft
162.6 [
B = oil formation volume factor = (135) 120) 1.2271924) + 12.9 X 10- 3 12.740 X 10 - 924
By = gas formation volume factor,
X 53.1) + 0] = 2159.
z-T ------p"
Bo dR, 1 dBo 1 2.9 X 10 3 1
T a (Pi+Pf)/2 c" = BodP - S;; dP = 1.227 (0.0455) -- 1227
C = compressibility, psi' (0.0001425) = 0.0003622.
h = formation thickness, ft Cg
0.55
= 644 = 0.000854 psio- 1
IRef. 131 .
.I = productivity index, B/D/psi
Cw = 10- 6 pSi-I.
k = formation permeability, md ct = S"C() + sgey + 5wCw.
p, = pressure obtained when linear portion of 01' = 0.546 (0.00036223) + 0.20410.000854) + 0.25110- 6 ) = 0.000372
pressure build-up curve, P vs log [( t + TABLE 4-SUBSURFACE PRESSURE CALCULATiONS FOR RESERVOIR
6.t)/6.t], is extrapolated to (t + 6.t)/6.t BELOW BUBBLE POINT.
Test Data:
6t, hours Hole Size (in.) 12
I
Cum. Prod. N p Ibbl) 33.300
Stabilized Daily Prod. q (bbl) 924 oil, 15.38 MMcf gas 12.740 MM bbl gas)
Effective Prod. Life t (hour) = 24N p /q 865
J. Calculation of kh (md-ft) and k (nHI)
h 20 It Ba 12.9 X 10-3 bbl/bbl
q 92~. B!'.D R, 298 cu ft/bbl or 53.1 bbl/bbl
1-'0 0.675 cp
80 1.227
!J.p 11 cycle) 135 psi
h _ 162.6 X 1924) X (0.675) X 11.227) 1922)
k -1135) 922 mdft; k '
120)
= 46.1 md.
II. Calculation of Skin Effect oS; and Pres~ul"e Loss Due to Skin
..'ll' skill (psi)
!-/I-' ~ ______ --"'<J fro 6 in.
~0~.1~5_____________ PI hour 1195 psig
cp PI 240 ~-=-- psig
0.000372 vol/vol/psi !J.p 11 cycle) 135 psi

s = 1 151 [(1195) - (240) 12159)


100,000 10.000 1000 100 . 1135) - log 10.15) 11) 10.000372) 136) + 1.07
It + 6t) It ~ 2.43 .

Fig. 17-Pressure build-up in a reservoir when both oil !J.P,kin =


1135) X 0.87 (2.43) 285 psi. =
and gas are flowing. Tho remainder of the analysis is~jghtforword.

SEPTEMBER. 1<;61 869


Subscripts 10. Lefkovits, H. c., Haz.ebroek, P., Allen, E. E. and Matthews,
a = 0
atmospheric (60 P, 14.65 psia) C. S.: "A Study of the Behavior of Bounded Reservoirs Com
posed of Stratified Layers", Soc. Pet. Eng. Jour. (March,
f = flowing 1961) 43.
i = infinite time ll. Pollard, P.: "Evaluation of Acid Treatments from Pressure
0, w, g = oil, water, gas BUild-Up Analyses", Trans., AIME (1959) 216, 38.
12. Aronofsky, J. S. and Jenkins, R.: "Unsteady Radial Flow of
Dimensionless Quantities Gas Through Porous Media", Jour. Appl. Mech. (1953) 20,
Dimensionless time in the Darcy System of Units (darcy, 210.
sec, cp, em, atm) is t D = kt/</>p,c r2e. In the system of units 13. Trube, A. S.: "Compressibility of Natural Gases", Trans ..
in this report (md, hr, cp, ft, psi), this quantity is t D = AIME (1957) 210, 355. .
0.000264 kt/<pp.c r'e. The constant 0.00105 in Eq. 8 is 4 14. Perrine, R. L.: "Analysis of Pressure BuildUp Curves", Drill.
(0.000264). In the Darcy System of Units, the flow rate is and Prod. Prac., API (1956) 482.
usually written qp./47rkh, when it is desired to express it in 15. Nisle, R. G.: "The Effect of Partial Penetration on Pressure
Build-Up in Oil Wells", Trans., AIME (1958) 213, 85.
units of atmospheres. In units in this report, this quantity
l6. Brons, F. and Marting, V. E.: "The Effect of Restricted Fluid
is 70.6 qp.B/kh. Also, Entry on Well Productivity", Jour. Pet. Tech. (Feb. 1961)
qp.B 17~ ,
70.6--rhloge t = l7. Hazebroek, P., Rainbow, H. and Matthews, C. S.: "Pressure
Fall-Off in Water-Injection WelIs", Trans. AIME (1958) 213.
250. "
References 18. Collins, R. E. and Kolodzie, P. A., Jr.: "Effect of Interfer-
ence on Pressure Buildup - A Method of Calculating
1. Horner, D. R.: "Pressure Buildup in Wells", Proc., Third
Porosity.Thickness Product", Paper 1333-G presented at 34th
World Pet. Congo (1951) Sect. II, 503. Annual Fall Meeting of SPE in Dallas (Oct. 4-9, 1959).
2. Mead, H.: "Another Concept for Final Buildup Pressure",
19. Elkins, L. F.: "Reservoir Performance and Well Spacing,
Paper Illl-G presented at 33rd Annual Fall Meeting of SPE Spraberry Trend Area Field of West Texas", Trans. AIME
m Houston (Oct., 1958).
(1953) 198, 177. '
3. Matthews, C. S,' Brons, F. and Hazebroek, P.: "A Method 20. Jones, P. J.: "Gulf Coast Wildcat Verifies Reservoir Limit
for Determination of Average Pressure in a Bounded Reser- Test", Oil and Gas Jour. (June, 1956) 54, 184.
voir", Trans .. AIME (1934) 201, 182,
21. Carr, N. L., Kobayashi, R. and Burrows, D. B.: "Viscosity of
4. van Everdingen, A. F.: "The Skin Effect and Its Influence on Hydrocarbon Gases Under Pressure", Trans., AIME (1954)
the Productive Capacity of a Well", Trans., AIME (1953) 201, 264.
198, 171.
22. Standing, M. B. and Katz, D. L.: "Density of Natural Gases",
5. Hurst, W.: "Establishment of the Skin Effect and Its Impedi-
ment to Fluid-Flow Into a Well Bore", Pet. Eng. (Oct., 1953)
Trans., AIME (1942) 146, 140. ***
25, B-6.
(). ('iadteltcr, R. E., Tracy, G. W. and Wilsey, 1. E.: "Selecting
Wells Which Will Respond to Production-Stimulation Treat-
r"lenls", Drill. and Prod. Prac., API (1955) 117.
~\liller, C. c., Dyes, A. B. and I-Iutch;nscn, C. A., Jr.: "Esti-
mation of Permeability and Reservoir Pressure from Bottom C. S. MATTHEWS, f Jrmerly chief res-
Hole Pressure Build-Up Characteristics", Trans., AIME (1950) ervoir engineer for Shell Oil Co.'s Tech-
189, 91. nical Services Div., is now senior reser-
g, Stegemeier, G. 1. and Matthews, C. S.: "A Study of Anomal- voir engineer with Shell Development
ous Pressure Buildup Behavior" Trans. AIME (1958) 213 Co. in Houston. He joined Shell Oil as
44. " ,
a chemical engineer immediately fol-
9. Pitzer, S. S.,
Rice, J. D. and Thomas, C. E.: "A Comparison
lowing his graduation from Rice U.
of Theoretical Pressure Build-Up Curves with Field Curves
Obtained from Bottom-Hole Shut-In Tests" Trans. AIME with a PhD degree in r:hemistry.
(1959) 216, 416. ' ,

870 JOil.tNA OF PETROLEUM TECHNOLOGY