~pE

SPE 14686

A Combined Perforating and Well Testing System

by P.J. Westaway, * Schlumberger Technical Services Inc.; 1. El Shafie, Belayim Petroieum Co.;
and M.J. Wittman, ● Schlumberger Overseas S.A.
●SPE Members

Copyright 19S5, society of Petroleum Enginaara

This paper wae praaanted et the SPE 19S5 ProductionTechnology Symposiumheld in Lubbook,Texas November 11-12, 19S5. The material ia eubjad
to cormdon by the author. Perrrbaion to copy ia restrictedto an aberracfof not more than S00 worda. Write SPE, P.O. Sox SS3S3S,Riihardaon, Texaa
750KM836. Telex 7S09S9 SPE DAL.

ABSTRACT INTRODUCTION

Underbalanced perforating is widely used in well When a prospective pay-zone is perforated with
completions and is considered by many operators to pressure underbalanced, the sandface is suddenly
be an effective method of obtaining improved well exposed to the lower wellbore pressure and flow is
productivity. initiated immediately. The correct underbalance
should lead to effective clean-up of the formation
A measurement of downhole pressure before, during near the wellbore, with minimal formation breakdown
and after perforating can be made by installing a or sandingq*2. The well will eventually either pro-
pressure gauge on the gun-string. duce to surface, or die if there is Insufficient
static reservoir pressure.
By using a wireline, the added capability of real-
time read-out on eurface allowa the entire opera- This initial flow provides an opportunity to con-
tion to be monitored ‘live’. Correct underbalence duct a transient pressure test baaed on the
can be accurately established prior to shooting, ~=a=d==aez~ & the dQW_hOIe flowing pressure (back-
there is an unambiguous shot indication, and a pressure) while the well is filling up. This is
pressure tranaient analysis can be made during the analogous to the “slug-teat” of conventional drill-
initial flow or fill-up period. From this we can stem testing3 , where flow is induced by the opentng
obtain an estlmste of permeability, skin damage of a production valve in the DST string rather than
and, possibly, static reservoir pressure, which is by perforating.
.
a useful supp~ement to -1-
cne
- UUaL-LE
_L. - * L...4 1.4,...
UUAAUWY a..=.=.-
“slwl+~

which usually follows if flow reaches surface. Any For this to be possible, the sensors must be in
subsequent conventional well-test can of course be place and recording when the perforators are fired.
planned without the need to’ retrieve or run in Equipment is being developed for connection in-line
additional equipment since the pressure gauge ia with the normal types of wireline conveyed guns.
already in place. The experimental system described in this paper waa
teeted in conjunction with tubing conveyed perfo-
The advantagea of perforating and testing in one raters (TCP).
run have proven particularly attractive in low
energy or low transmissivity oil-fields. Here, In its final form, the system will serve the
testing of moat wells by conventional methods ia following functions :
difficult, time coneuming, or impossible because
flow either does not reach surface or takes a very - depth control with gamma-ray and caaing-collar
long time to do so. locator;

The measurement of pressure on a perforating run ia - checking correct underbalance before perfo-
not a new idea, although real-time read-out of data rating;
ia less common. The purpose of this paper is to
describe the system used, and to present a method aa+,m.t+q ~f the nt2rfora~0r/pOaitiVe
r----- shot
= ‘=’’’u-’””
of interpreting the data which will have a wide indication;
range of applications, particularly if a simulta-
neous measurement of the downhole flow-rate la - continuous monitoring on hu,t{ace of down-hole
available. The examples demonstrate clearly the pressure, flow-rate and, optionally, tempera-
validity of this approach. Tubing-conveyed guns ture and fluid deneity, during the operation;
were used, but the technique is suited to any type
of underbalanced perforating equipment. interpretation of the data at the well-site.

References and illustrations at end of paper.

 A COMBINED PERFORATING J ) WELL-TESTING SYSTEM SPE 1468(

In a low energy resenoir which will not produce to Flow must not have reached surface. The theory
fmlrfsrc=.
--.----> Qr vhe3-e low transmissivity causes an --1....,.-...4
Vu.y .le..e
.UU=AU=.= F1..’IA
a r%iag ..... ,.al,,nm
..... .
extremely slow fill-up, the initial flow data is of
particular significance. Since production from The reservoir static pressure must be Icno&.
wells of this kind usually requires installation of This can not be measured before the teat be-
a PUmPS advantage ia taken of the flow induced by cause the pressure gauge ia exposed to fhe
underbalance to obtain, by a comparatively simple wellbore cushion pressure. It can be obtain~d
method, an early estimate of reservoir permeability from a subsequent conventional shut-in teat ,
and total akin factor. These are needed to predict or, in a low energy reservoir, by waiting for
productivity and most suitable pump characteristics. the well to die. Alternatively, eatimetes from
Should the well die, the gauge will eventually elsewhere in the field, or from a wireline
register static reservoir pressure. formation tester, can be used.

If the well flows to surface, the teat can be con- The radius of influence of the pressure dis-
tinued with a drawdown/shut-in sequence as in any turbance is relatively small, but may extend
conventional testing programme, but without the to several thouaaxid wellbore radii if the
necessity for any further trip into or out of the dimensionless storage constant is large
hQl~ , --_..-L3
cXM.?ugll .
~e=ti~~~ fro= 4.1.eaa
L,,=m=
*a.+.
k=-.-
m“,,
,Jnwy
tha---
-..=..=

fore be strongly influenced by conditions near

The examples in this paper are from the Gulf of the wellbore.
Suez, and show how the technique ia applied In a
low energy reservoir producing a low gravity, vis- It ia aaaumed that fluid density, inside pipe
cous crude. b interpretation procedure for the diameter and angle of deviation are constant
initial fill-up period, where the increasing back- (and known). These terms appear in the expres-
pressure contlnuoualy reduces the flow-rate, ia sion for the storage constant.
presented based on variable-rate theory. The paper
demonstrates how a direct measurement of flow-rate Kabir et a19 propose a method for testing low
would remove some of the uncertainties usually transmissivity pumping wells by monitoring annular
inherent in this form of analyaia, enabling it to fluid level after turning off the pump. This in-
be applied in a wider range of conditions than is volves a variable-rate analysis incorporating down-
----4K1- ..4-I. . . ..4-*4..- --4.L...4-
pu15uUlLe WLLLL =AL5!.UI& U=L118JUV.
hole pressure and flow-rate estimated by tagging
the rising annular fluid level with an acoustic
THEORETICAL BACKGROUND well-sounder. They comment that final static pres-
sure is approached more rapidly by allowing the
Several references discuss interpretation of fill- column to rise against atmospheric pressure than
UP or ‘slug-test‘ data in the context of drillstem againat a closed surface valve.
testing. Kazemi 4 emphasizes the importance of
allowing for any variation in flow-rate during A number of the sources of uncertainty outlined
fill-up when working on the aubaequent shut-in data. above can be alleviated by direct measurements.
He describes a method of estimating the flow-rate Ideally, the measuring system would acquire aimul-
from pressure data, but doea not analyse the flow- taneoualy the pressure, temperature (for gauge
ing period itself. Development of a type-curve correction), flow-rate (spinner), end fluid density
approach5,&*7 culminated with the work of Ramey, (gradiomanometer).
Agarwal and Martin3 who presented a aet of curves
from which permeability and akin factor could be In principle, the instantaneous flow-rate could be
obtained provided the porosity-compressibility pro- used in a variable-rate analyais whether the well
duct and wellbore storage constant were known. The ia shut in or flowing at surface, or even non-
flow-rate is implicit in these curves, and does not radial in the formation. This is the eventual
have to ba calculated. objective of the technique to be described. The
data we present at this stage conaiat of the firat
There are several limitations to ‘slug-teat’ two of the above-mentioned measurement a. The
analyais in its current form : examples do, however, clearly demonstrate the
potential of the method.
Flow must not be ‘critical’ (i.e. in excess of
acouatic velocity) across any restriction in
the testing/completion sting. This might occur In the abaence of a direct measurement, flow-rate
at high rates through a downhole production can be approximated by equating an increase of
valve, and is sometimes observed in drillstem back-pressure with the corresponding increase in
testing of high tranemiasivity wella3. fluid column height (ignoring frictional losaea) :

Flow in the formation must be radial and Darcy.

15p
The usual constraint pertaining to the analy- 61 = ..........(1)
ais of the flow of a slightly compressible 0.4335p Cosa
fluid through a porous medium also apply8 .
This representa a volume of produced fluid given by:
The well must be open at surface. Calculation
of the wellbore storage constant for fill-up
against atmospheric pressure $s quite straight- nd;bl ..........(2)
forward, whereaa the compression of cushion 6$”= delddes.fjls
liquid or gas cauaed by shutting in the well
at surface introduces an uncertain and varying
storage term.
3PE 14686 PETER J. WESTAWAY, IBRAH1 EL SHAFIEI, M.J. WITTMANN

If this occurs in a time interval dt, the average

flow-rate during this period is : S=l.1513 -–log k
[ m @ct7-:
()
w 0.00224dz6p
. . . . . . . . . (3)
9 = z= 24 pcosa z + 3.23 .,
......... (69
1
This fs in fact just a restatement of the general
equation for flowrate when there is wellbore
where q, ia the reference flowrate, and RHP(0) is
storagefo :
the intercept of the line at zero time-function,
i.e. the value of the rate normalized pressure at
6p
●.**.*.**(4) Zmlqn = o.
9 ‘24CE
In ref (7), Ramey et al demonstrate the use of
the first term on the right in (3) being the their ‘slug-test’ type-curves with several examples.
storage constant C for a moving fluid columnlO. We will first show that rate convolution givee
comparable results by enalysing their Example A.
Equation (3) tells us that when the back-pressure Their data is reproduced here in Table 1. In order
ia increasing at 1 psi/rein- 60 psi{hour in the to convert the dimensionless pressure function they
... .-.”.lm*tal” used for the type-curve into actual pressures, we
units of the equation - cue
“ rate is -PPrvaA-.%.,
29 Bbl/D for water in a vertical 3%” tubing (I.D. have takan for simplicity a value of 2100 psi
2.99”). Drilling and completion data should be for initial cushion pressure (P.) and 2200 pai for
referred to when applying equation (3) over a long static raaenroir pressure (Pi),aince these are not
hole section, to take account of changes in pipe specified in the reference (actual values will not
internal diameter and deviation. affect the answer). Intermediate teet pressures
were then calculated from the dimensionless valuea
It is obvious from this that a direct measurement in Table 1. The flowrate was calculated using
of flow-rate would be far more satisfactory. equation (3); in this caee 1 psi/rein ia equivalent
to about 116 Bbl/D in the 6 in. I.D. casing. Pres-
There are two approaches to handling the variable sure and flow are plotted in Figs. 1 end 2.
rate analysis. Both require the computation of a
time-function based on the flow-rate, prior to The rate-convolution plot in Fig. 3 exhibits a fair
making the semi-log (or other) interpretational amount of scatter, arising both from the original
plots. The clasaical method 6’11 approximates the data end the effect of calculating the pressure
flow schedule as a series of eteps and is the more derivative from widely spaced points. (There is,
suitable for hand calculation. The alternative is equally, some uncertainty associated with finding
to represent the continuously varying flow-rate by the best type-curve match.) However, from the
a series of straight line chords (ref. 12, Fig. 2) slope and intercept at zero time function, we esti-
rather than steps. This method is better sufted to mate, using equations (5) and (6) end a reference
computer analysis and the subsequent computation is flowrate q, of 1860 Bbl/D :
usually referred to as ‘rate convolution’.
k = 588 md and S = -0.17
Although the mathematics involved In the two
methods are different in detail, the end result which compares very well with Ramey et al’s semilog
should be the same. For the convolution method used type-curve reeults which were :
in this paper, a plot is made on semi-log scales of
the instantaneous ‘rate normalized pressure draw- k= 611mdand S=-O.36
down’ :
P, - P.Y(4J EQUIPMEHT AWL)PROCEDURES
9.
Underbalanced perforating and the operation of tub-
versus the ‘rate normalized time-function’ : ing conveyed perforators are discussed in refs. (1,2
and 13).
%/9.
Large diameter casing guns are connected below a
The term Z. is defined in the AppendIx. Rate norma- packer and run in on tubing. The guns are position-
lization simply involves dividing by q. = q(t.)/q.~ ed at the precise depth by running a casing collar
where’ q, iS WI arbitrary fixed reference flOW-rate, end game-ray (or neutron) log and correlating with
and q(tn) the instantaneous rate at time tn. an open-hole reference log. The packer is then set.
A production valve is next opened below the packer
If flow ia radial in the formation, the plot will to establish communication between the casing (rat-
be a straight line, from which kb end skin factor hole) end tubing. Underbalance IS achieved by

can be derived in the ueual waylz,ls from equations having the tubing only partially filled; It is
(5) and (6) : either run in partly empty, or excess liquid is
swabbed or gas-lifted out.

~=E24&t! ● 0........ (5) Several methods are available for detonating TCP
guns. In the present examples, the guns were fired
electrically. The single conductor wireline used to
run in tha preseure gauge after the packer had been
set also served as the carrier of electrical with a fluid cushion consisting of diesel over oil-
current to detonate the guns by means of a ‘wet- based mud. Guns used were 3-3/8 in. diameter at a
connect’ adaptor. The firing system, plus preaaure density of 4 shots per foot over a 28 ft interval.
gauge and thermometer, is shown in Fig. 4.
Fig. 7 is the complete pressure record. The data
This system operates by running the female wet- have been edited down from a sampling rate of 1 ‘per
----—
secund. . .
t%s tes~ ... +-.
W== ...+aa.+aa
.=.~*--.=- +*-Y ab~~~ ~ h~l~rae
a..-.
connect adaptor on wireline, down onto the firiiig-
head pin which is part of the gun assembly. This just before returns were observed at surface.
eatabliahes electrical continuity and the guns can
be detonated at will. A J-latch mechanism locks the Ater clean-up, the well produced 21°API oil of 3 cp
tool onto the pin to prevent its being blown uphole viscosity at 54 Bbl/D. A conventional surface shut-
after perforating. The production valve can be in test was later performed, and with this example
r---w- by f~~ rQQl as it passes%
nnonod Or by a prior run we have the opportunity to compare the type-curva
ith a slickline, depending on the type of valve and rate-convolution analysea of the fill-up period,
installed. and a Horner analysis of the final shut-in.

The quartz crystal pressure gauge and temperature The downhole flow-rate (Fig 8) waa derived using
sub are mounted in-line with the firing adaptor as equation (3), taking into account that the well was
shown in the Figure. A shock-absorber or slip-joint vertical, tubing ID waa 2.99”, and produced oil
protects the pressure gauge from damage when the density 0.88 gfcms. We have assumed that the forme-
TCP string (which can be several hundred feet long) tion immediately produced oil, whereas it may have
is fired. flowed drilling mud (OBM in this caae) or a mixture
during clean-up. Also, under large drawdowna the
On surface, a computerized logging unit displays situation may be complicated by the evolution of
the down-hole measurements on a screen, and records dissolved gas from the oil. In the preeent case, it
data on tape and film. The data can be processed at was fairly safe to assume that the physical pro-
the wellsite if required, using a variety of diag- perties of the viecous, low gasfoil ratio oil did
nostic plota. not change much during the test. As long as we have
to &&c~ flow-rate, we are eubject to the same
Fig. 5 is a schematic of the set-up showing only the errors as the type-curve method regarding fluid
principal components. density.

Fig. 6 shows the sequence of events as they were We see a aemilog straight line on the rate-
observed on the screen of the logging unit at the convolution plot after about the firat hour of data
beginning of a TCP well-test. Following depth cor- (Ffg.9). From the slope and intercept (equatione
relation, setting of the packer, and opening of the (5) and (6)) we estimate :
production valve, the wireline tool has been latch-
ed on to the firing pin and the gauge allowed to k = 3.5 md and total ekin factor S = -1.8.
stabilize. The cushion pressure ia checked at
1624 psi. (This measurement is particularly useful This is corroborated by the Ramey et al type-curve
when underbalance has been established by gas- match (Fig. 10) on the ‘CDezS = 100’ curve, from
lifting the tubing contents, and it is not certain which we obtain388 :
just how much fluid is left).
From the Horner build-up plot of the final shut-in
Next there is a clear indication that the gun has (Fig. 11) we find : k = 2.9 md with S = -1.6,
detonated, or, rather, that the formation haa been assuming a constant rate throughout the clean-up
perforated. The pressure oscillate for several and flow periods.
minutes after perforation and continues to build up
(at a gradually decreasing rate) aa the perfo- In this case then, a good estimate of permeability
~~~i~as prQduce into Ehe ‘Ubi~~. and akin appears to have been obtained from the
fill-up data.
Looking at equation (3), it can be appreciated that
the steep pressure buildup observed at the begin- EXAMPLE B :
ning corresponds to a high initial aurge caused by
the large underbalance to which the formation ia This well was perforated with tubing-conveyed
suddenly subjected - in this example over 3000 psi. 3-3/8 in. guns at 4 shots per foot density. A
cuehion pressure of 2097 psi once again established
EXAMPLES a large underbalance relative to the 5000 psi
reservoir pressure.
The following examples were chosen from a number of
tubing-conveyed perforation jobs executed in a low The recorded data and derived flow-rate are shown
energy, viscous, low gravity oilfield. Few of the in Figs. 12 and 13 (lower curve). Initial rate waa
wells in the field are able to flow to surface, end considerably higher in this example, declining to a
conventional well-testing is rarely possible. The few hundred Bbl/D by the end of the test, at which
purpose of recording backpressure while the tubing point returns were observed on surface.
fi~~ed UP after pe~fo~a~in~ was tO aasist the

operator in the choice of appropriate pump design. Much of the oil-baeed mud and sea-water used for
the cushion were recovered before the well, which
EXAMPLE A : was flowing saline formation water, died. Unfortu-
nately, the test had already been terminated end no
The beginning of this test is shown in Fig. 6. The final downhole measurement was made to confirm the
underbalance of approximately 3000 psi was obtained stattc reservoir pressure.
 PETER J- WESTAWAY. IBRAHIM EL SNAFIE:, M.J. WITTMANN c
LU ------ .

The rate-convolution produces a well-defined aemi- 3) In low energy or low tranamissivity wells
log trend (Fig. 14) and we estimate : which do not flow to surface or which would
take a very long time to build up against sur-
k=4mdandS=-1.5 face shut-in, it is a means of estimating
permeability, skin factor and static reservo$r
compared to 5 md and -1.0 from the type-curve match prassure in a relatively short time, with
(not shown). The relatively low viacoaity (0.5 cp) prompt wellsite processing of data if required.
of the formation water and thicker pay (87 ft)
explain why the flow-rates were so much higher than 4) It is a useful supplement to a conventional
in Example A, despite the similarity of initial well-test where the latter is feasible.
drawdown, permeability and skin factor.
We have interpreted data recorded after tubing-
In view of the high production rate at early time, conveyed perforation, using the Ramey et al type-
a tentative correction waa made for pipe friction curves and the variable rate-convolution method and
using a Moody friction factor of 0.02 for turbulent obtained similar results.
flow in the Faming equation’, and allowing for
the continuously increasing column height. After Where downhole rate haa to be deduced (it is
correcting for the pressure loss caused by pipe implicit in the type-curves, and calculated from
friction, a slightly higher flow-rate schedule was pressure derivative for convolution), the well must
computed (upper curve in Fig. 13). be open on surface. This la principally because the
wellbore storage constant can be estimated with
This straightens out some of the early rate- more confidence for a moving fluid column than for
convolved data (Fig. 15, compare Fig. 14). Although compression against a closed valve.
this is only an approximate correction, it demon-
strate that friction losses along the pipe wall The rate-convolution approach presented in this
must be considered at high rates unless flow is paper has the advantage that, with a direct simul-
measured directly. Pressure losses across the pro- taneous measurement of flow-rate, the technique can
be applied more generally to open or shut-in situa-
duction valve will, of course, always appear in the
final plot (unless the gauge ie in communication tion, and many of the uncertainties involved in
with the rat-hole instead of the tubing) - this ie attempting to calculate the downhole flow-rate can
analogous to a rate-dependent skin and may partly be removed.
explain the curvature at the beginning of Figs. 9
and 15. Clean-up of the perforations and formation
NOMENCLATURE
near the well-bore may also contribute to thie constant
c- wellbore storage (bbl/psi)
curvature.
cm . dimensionless wellbore storage conetant
“

There is, however, no evidence of critical flow3. .

Ct total formation compressibility (l/psi)
EXAMPLE C : d . internal diameter of tubing (in.)
h . reservoir thickness (ft)
In this example underbalance was adjusted by gas-
lifting out the appropriate amount of tubing fluid k . effective permeability (red)
after setting the packer. Progress of gas lift was I! . length of fluid column (ft)
monitored in the well-site computer unit to verify
that correct cushion pressure had been attained m . slope of line on rate convolution plot
before perforating. Fig. 16 shows the complete *
Pi initial reservoir preesure (psi absolute)
pressure record from the etart of the gas-lift,
through to the actual gun detonation when the P. . cushion pressure (psi absolute)
desired 800 psi pressure had been reached. . well flowing pressure (psi absolute)
Pwf
q . in-situ flow-rataof produced fluid (bbl/D)
CONCLUSIONS
i . vellkmre radius (ft)
This paper has reviewed the recording and Interpre-
tation of downhole pressure data in underbalanced Sw . total skin factor
perforating operations, using a wireline-conveyed .
t time (hours)
testing tool. Coupled with simultaneous surface
read-out facilitated by the use of a wireline v . in-situ volume of produced’ fluid (bbl)
system, the measurement offers a number of advan- a . angle of deviation of tubing (degrees)
tages :
* . poroeity (decimal)
1) Verification of cushion pressure before perfo- e . produced fluid in-eitu density (g/cmg)
rating is extremely useful, especially where
tubing contents are gas-lifted out to establish 2. . rate-convolved time function at. time t“
the desired underbalance.
(see Appendix)
2) It provides an unambiguous indication of per-
SUBSCRIPTS
foration and allows the progress of the entire
test to be mnitored.
j,n = data sample nnmbere
r= reference
 A COMBINED PERFORATING ND WELL-TESTING SYSTEM SPE 14686
ACKNOWLEDGEMENTS 13. Cone, E.A.,Jr. : “A CompilationTechnique for
Overcoming Formation Damage,” paper SPE 7009
The authors thank the Belayim Petroleum Company presented at the 1978 SPE Symposium on
(I?etrobeijfor their %iiitiative =G6 Cmperstim i~ ~~p=~i~n Daaage, Lafayette, Feb 15-16.
testing the technique presented in this paper, and
for permission to publish the examples. 14. Begga, H.D., and Brill,J.P. : “Two-phaae Flow
in Pipes,” University of Tulsa, 1975.
REFERENCES
15. Amin A. : “Application of Downhole Flow end
Pressure Measurements in Well Testing of
1. Bell, W.T. : “Perforating Underbalanced -
European and North Sea Resenroirs,” paper SPE
Evolving Techniques,” J.Pet.Tech. (Ott 1984)
14457 presented at the 1985 SPE Annual
1653-1662.
Technical Conference and Exhibition, Las
Vegas, Nevada, Sept. 22 - 25.
2. Bell, W.T. : “Perforating Techniques for
Maximizing Well Productivity ,“ Paper SPE
10033 presented at the International APPENDIX
Petroleum Exhibition and Technical Symposium,
The full drawdown equation for radial flow,
Beijing, Cinina,liarcii18-26, 1982.
derived in ref. 12, is :
3. Ramey, H.J.,Jr., Agarwal, R.G., and Martin, I.
: “Analysis of ‘Slug-Test’ or DST F1OW Period
Data,” J.Can.Pet.Tech. (July/Sept 1975) 37-47.

4. Kazemi, H. : “Damage Ratio from Drill-Stem

Testa with Variable Back Pressure,” paper SPE
1458 presented at the 1966 California
-3.23 + 0.875S
1 (A-1)

Regional Meeting in Santa Barbara, Nw 17-18.

where the dimensionless qm = q(t.)/qr, and q~ la
5. Cooper, H.H.,Jr., Bredehoeft, J.D., and an arbitrary reference in-situ flow-rate.
Papadopuloa, 1.S. : “Response of a Finite
Diameter Well to an Instantaneous Charge of The rate convolved time-function, ~. , at time t.
Water,” Water Resources Research (1967), after the beginning of drawdown is given by :
VO1.3, No.1, 263-269.
Xm =’5iq;+,
j-l - q)V.- wgK - t,)
6. Kohlhaas, C.A. : “A Method for Analyzing
Pressures Measured During Drillstem-Teat Flow
Periods,” paper SPE 3695 presented at the SPE
42nd Annual California Regional Fall Meeting,
+q;t.logt.- q-loge (A-2)
in Los Angeles, Nov 4-5, 1971.
where:
7. Ramey, E.J.,Jr. , end Agarval, R.G. : “Annulus % - 9j-#
ff; =tt
Unloading Rates as Influenced by Wellbore $- j-f
Storage and Skin Effect,” Soc.Pet.Engrs.Jour.
(Ott 1972) 453.
A DIOt Of the rate normalized pressure drawdown
8. Eerlougher, R.C.,Jr. : “Advances in Well Test
(tie left-hand term in (A-1) versus the rate
Analyaia,” SPE monograph (Henry L. Doherty
normalized time-function Z. Iqn should give a
series), Vol. 5, 31-33.
straight line if flow is radial in the formation.
Permeability and skin factor can be calculated
9. Kabir, C.S., and Kucuk, F. : “Well Testing in
from the slope m and intercept at zero abscissa
Low Tranamissivity Oil Reservoirs,” paper SPE
using equation (6) and (7) in the text.
13666 presented at the 1985 California
Regional Meeting, Bakersfield, March 27-29.

10. Ramey, ?l.J.,Jr. : “Won-Darcy’F

low and Wellbore
Storage Effects in Pressure Build-up and
Drawdown of Gas Wells,” J.Pet.Tech. (Feb
1965) 223-233.

11. Odeh, A.S., and Jones, L.G. : “Pressure Draw-

down Analysis, Variable Rate Caae,” paper SPE
1084 presented at the Production Research
Symposium, Tulsa, May 3-4, 1965 (or SPE re-
print eeries No. 9, 161-165).

12. Stewart, G., Wittmann, M.J., and Meunier, D. :

“Afterflow Measure~nt and Deconvolution in
Well Test Analysia,” peper SPE 12174
presented at the 58th Annual Technical
Conference, San Francisco, Ott 5-8, 1983.
 WE 146816

‘MBLE 1: I&&&sm#e FimliemA ofRsmsY,AgsrwsI

RESERWRANDmlltlA
Fornut ion thickness, h . . . . . . . . ~2~qft
I-...--,...A ................... . ...
ryusl!y,v
vlscoslty,~ ..................; .~~D. i_l
~-~-l
,“,.$,
-—.
e“,
acre;
p,. ===,
h; li t,,
ul o!.
C
,,-, . . . . . ---- -- ~-

Fluid density, p ... .. .. .. .. . 1 9/om3

WELL DATA
2200
Wel!bore rodius, rw . . . . . . . . . . . 0.25 ft formation pressure
Cos!ng Internal radius, r ~.. .,25ftft
mm
■■
lwX TB1’ DATA mD
,m

P1-l%f .99
t (sees) P“f n
PI–PO .+
m
■
0.0 E ■
●

3.0 w
6.0 ■
9.0 ~ 2150- ●
12.0 9
~
●
m
M
M ■
L

Z,ojy:npress; ,0 j.

TIME ( sees )

15-

10 m

● 6
● m = 1.595
●
●
APmo = 10.14
● ✎✘
.*. *e ●
‘@e. e
[ , I 1 I

-’3 -2 -1
20 40 6U

TIME ( sees ) RATE–CONVOLVEDTIME: Z/A;~

~~mf=tihr~l.
Fl@410wmlSdukd-~hT-1.
-,)
and VALVE
&
Cable

Ill
— Calb2e head
.= ‘- Diesel

1 IIIll .-
.-
—!111
Ml_
lhbing
%bing
CCL/C(zmma ray
-
.,

Ill
.—and
Ill
PRESSURE GAUGE
THERMOMETER
,.
,,

—
[
lhbing Sho,ck absorber ., ‘- Oil-base mud
\
.swed.oe\ ‘,.
.- A
I -K
?I]

1[”
.,
Wet-connect ,.

adaptor Casirq
,“.
.,
Wet-connect
pin ----
f .“
J1 —Packer
:.,
.,
0 1
- — Production
Fill-joint< : .:
valve
-.. - 0 ‘-,, PRESSURE
GAUGE
,’L
,“.c=
TCP Guns
“.E~
.’6_
GUNS
.,

h
i
,. 3
Fig. 4–WlmIin@w’et-comeetorfifing hati with presewd .,.
temperaturesub (echemetic). .:[ . .

Fig. 5-The TCP/pmeeureWet conflgumtlon.

rT=
---–.--––-––. –m–amT _____ 300
1800
16
PGP(Ps1a)
I Pressure
-—~ lkrnperature
Time
listin~g ,/ -
18:50 1631.9 I
.8 m
16e9’.6 I ■ ■
I .8
18:52 ■■
16?7.6 I ●m
.9
CUSHION PRESSURE ~ 3000
181S4 1626,.0
, i
m

I-t-l-M
~
f I m
18:S6 SHOOT *:;:; g I m 9
I ka
. lx
18!s8 1646,.7 t I & 2000 ,=
16701.00
~~ BACK-PRESSURE
19:00 <+ Cushion pressure
169!.2

19:oe 171!.2
—
10001“ ~~ ■
1730. e 61
n 2
19:04
1749.1 T:[ME (4hours )
176;7.51 IWI. 7-Exampk A-PI-W - dudng 1111-uP
PUM.
19:06
/1[

1500

1000

500
m = 323.5
APz.o = 641.5

0
04-—-~-~ -2 -1
o 4
TIME ( hours ) RATE-CCINVOLVEDTIME: Z/Aq~
w 6-3XWIIPI. A-varhblm tie mdyals.
FIII. 6.-Sxmoh A —down6wb 6!aw m*. .
 1 :)0001

n
..
me””
0.8 450 c1
.4**
,...
.. ...* ...0

✎...” ●

●
CDe2’ = 100 tp = 41 hrs
0.6 4ooc1 ●

● m = 15.83
●

b
●o P(lhr) = 4178 psia

0.4 ●
●
P* = 4770 psia
:350C) ●

L
0.2
:300C)
2 1 )
,~g tp+At
c -n
At
7--’---0.1 1 10 100 1
F@.
13-Eumpk d-ltomw pl.ald WbnqwlltWti stlut-h wm.
tD/cD

10000
6000

8000

,mmm
>
.9
,8 ) FRICTION CORRECTED
,m
6000
■
■
●
■

m
●
● 40(D0
‘%
●m
m 20(D0
m
(
I
~ Cushion pressure
UNCORRECTED
o , ,
2000 , ~r—
0:2 0.’6 0:8
0:2 0<.6
T;;E ( ho~fs ) TI!$E ( hours )
.
ng.l~~fw-.
 f

m = 47.37
m
m = 49.49 /
APz.o = 149.9
m APx.o = 159.7

/ ‘/
o! I I I

-4 0 2
RA%-CONVOLVEDTIME: X/Aq~
Fig.
l-lll@OB-~ ---