Advancements in Uncased-Hole, Wireline

Formation-Tester Techniques
A. L. Schultz, * SPE-AIME, Schlumberger Well Services

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

W. T. Bell, SPE-AIME, Schlumberger Well Services
H. J. Urbanosky, SPE-AIME, Schlumberger Well Services

Introduction
The wireline formation-tester (F~) technique was in- ficient accuracy of the recorded pressures (in the range
troduced to provide confirmation of formation- fluid of ± 2 to 3 perc~nt). This, combined with the single-
type, indications of productivity, and 'formation pres- test-per-trip capability, often discouraged the use of
sures. 1 Various improvements have been made in the these tools for recording several pressure measurements
technique and interpretation· methods have been devel- in a well.
oped for best use of the information from the recov- In summary, major limitations of these older tools
ered fluid samples and the· pressure· recordings. 2 were their inability to be repositioned, their single-test
While the technique has been successful locally, it capacity, and the lack of a reliable means for testing the
has not reached its full potential, basically because of integrity of the seal before attempting a sample.
the long rig time required with existing testers for To overcome these limitations, a new formation
multiple-zone testing. Once the tester was set in the tester has been developed.
well, it could not be repositioned at another level in the
zone of interest. Consequently, any test failure caused Principal Features of the New Tester
by a tool setting in an impervious streak or by a The new tester has several distinguishing features as
packer-seal failure resulted in an extra trip in the well. compared with the older tools. Several successive tool
Performance in many unconsolidated sands was not settings are possible without bringing the equipment out
acceptable with these older tools. Techniques to combat of the hole. Combined with this is a "pretest" capabil-
the flow of sand into the tester were never completely ity that permits the operating engineer to ascertain, be-
successful; this- sand flow caused undermining of the fore attempting to take a sample, whether the packer is
packer seal with subsequent mud-sample recovery. sealing properly and, if so, whether fluid flow is
These factors combined to produce an over-all success adequate to obtain a sample in a reasonable period of
ratio of about 70 percent for all formations and about 35 time.
percent for unconsolidated sands. Thus, if the tool is .set and the packer seal fails, or if
Also, tool redressing required between runs was ex- the indications are that the tool.is set in an impervious
tensive. This added to the over-all operating time unless streak, the tool is simply retracted and moved to an-
, additional tools were available at the well. other position in the formation. If both seal and flow
Another limitation of existing FT tools was the insuf- indications dur~ng pretest are satisfactory, a sample is
taken.
* Now with Tesoro Petroleum Corp., San Antonio, Tex. Two separate sample chambers make it possible to
@lTrademark of Schlumberger Well Services. obtain two samples on a single trip into the well. This

Improvements in wireline formation testing have been incorporated into a tool with
multiple-set capabilities. The tool permits pretesting of the formation for permeable regions
and checking of packer seal integrity before sampling. Two fluid sample's can be obtained
on each trip and any number of pressure recordings can be taken during the same trip.

NOVEMBER, 1975 1331

further reduces testing time. The samples may be from TABLE 1·- RFT SPECIFICATIONS
different depths or they may be for a single test at a
Pressure rating, psi 20,000
given depth, in which the fluid produced last is put in a Temperature rating, OF 350
separate chamber and, thus, is segregated from that Minimum hole size, in. 6
produced first. Maximum hole size, in. 143/4
The new tester also provides for sampling in both Basic make-up length
(excluding options), ft 33
consolidated and unconsolidated formations. Averagetool..setting and
Pressure-recording accuracy has been .improved to ± retracting times, minutes 0.6 each
1 percent or better. With special calibration of the ,Formation type hard or soft
equipment; an accuracy of about ± 0.18 percent can be Average sampli'ng rate
obtained. The multiple-set and pretest features permit - no water cushion,
high-permeability formation
any number of pressure tests to be made during a single gal/min
trip in the well. Formation-pressu re, read ings
Finally, upon return to thestirface, very little redress- per trip in hole any number
ing of the tool 'is required. Once the chambersare emp- Sample-chamber sizes,. gal 1,23/4,6, and 12
Strain-gauge transducer
tied of their recovered fluids, the tool may be lowered pressu re-measu ri ng

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

for additional tests. specifications:
Room-temperature *Accuracy resolution repeatability
Tool Description calibration 0.73% 1.0 psi 0.05%
Well-temperature calibration 0.18% 1.0 psi 0.05%
The· general configuration of the new repeat formation
tester (RF~) tool is similar to that of older equipment. *Based on full-scale reading, 10,OOO-psi gauge.
It consists of control panels in the truck, a down-hole
electronic cartridge, a mechanical unit, and sample cham-
bers. General specifications are outlined in Table 1. packer is sealing properly and that the fluid flow. is
The setting section of the mechanical unit is shown in adequate to obtain a sample in a reasonable period of
Fig. 1. The packer assembly. and backup shoes are time. As shown schematically in Fig. 2, a smallpretest
shown in the extended (or set) and retracted (or run- chamber with a 15-cc volume is located between the
ning) positions. The small-area wall-contact points packer and .the valves leading to the sample chambers.
minimize differential-pressure sticking that has troubled When the packer is set, the equalizing valve closes and
existing techniques. Actuation is by means of a hy- the chamber is opened, resulting in one of three possi-
draulic power system in the mechanical unit, which may ble pressure responses: indication of mud pressure if the
be energized on command from. the surface to control seal is lost (Fig. 3a); a very low pressure if the packer is
the setting and retracting of the packer assembly and seated·. on an impervious streak (Fig. 3b); or a pressure
backup shoe, as well as all valving functions. This fea- decrease followed by a buildup to formation pressure if
ture help& to provide the multiple-set provision. the tool is in a permeable zone (Fig. 3c).
The pretest function, incorporated in the mechanical Since the volume of ·the pretest chamber is known,
unit, permits the operating engineer to ascertain that the the rate of fillup provides an. indi.cation of the time that

Fig. 1 - Setting section of the RFT mechanical unit.

1332 JOURNAL OF PETROLEUM TECHNOLOGY

would be required to- fill one of the larger sample cham- When the tool is retracted, the piston within the
bers located below the seal valves in Fig. 2. If probe moves back into closed position in preparation for
all pretes t indications are satisfactory, one of the another test, wiping the filter clean in the process.
seal valves is opened to allow fluid to enter a sample Cleaning the filter is an important part of the multiple-
chamber. These valves may be closed and reopened at set operation. .
any time during the test to obtain a pressure-buildup A strain-gauge transducer is used to achieve accurate
measurement. pressure measurements of high resolution and good re-
On the other hand, if the pretest indications are nega-
tive, the tool is simply retracted. In this event, the
pretest chamber is emptied automatically, and the MUD PRESSURE

equalizing valve is opened automatically as the tester is /,..-~­

retracted. ~ I MUD PRESSURE AFTER
SET INDICATES PACKER
A probe, which can be seen protruding from the ~ SET
(a) en SEAL FAILURE.
center of the packer in Fig. 1, improves testing across a Q)
'-
broad range of formation characteristics - consolidated a..
to unconsolidated. The probe is equipped with a filter to Time

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

restrict flow of loose sand. Fig. 4 schematically illus-
trates probe operation. On setting the tool, the closed MUD PRE'SSURE
'J'"
probe is forced part way through the mud cake. A pis-
ton is then retracted, exposing the tubular filter to the
Q)
'-
:::J
I VERY LOW PRESSURE
AFTER SET INDICATES
(b) ~ SET
formation fluids. If the formation is unconsolidated, Q)
NO FLOW FROM FORMATION-
sand flows into the probe, where its further movement '- IMPERVIOUS STREAK.
CL
is restrained by the filter. Concurrently, the probe
moves into the formation to occupy the void produced Time
by the flowing sand to avoid undermining the packer MUD
seal and subsequent· failure. If the formation is consoli- PRESSURE

dated, the probe does hot penetrate the formation and

only mud cake and formation fluid flow into the hollow
cavity of the probe.
(c) i S{---~X-//"/l-- GOOD SEAL INDICATED.
SOLID LINE REFLECTS
GOOD FLOW RATE ~
~ I ,," FORMATION
DASHED LINE, TIGHT.
CL ,_........ PRESSURE

Time
Fig. 3 - Pretest indications.
PACKER
~ UD CAKE
BACK~
SHOE ~ .. I ....
PACKER ~O::---~

FILTER \. I' ..

FLOW LINE~
.¥'PRESSURE .. UNCONSOLIDATED
GAGE SAND
;'

~---"-1N'lI
'.. PROBE CLOSED
DURING
EQUALIZING INITIAL SET
VALVE
(TO MUD
COLUMN) PRETEST
"
CHAMBER "
-,,-'
- '-
, ". ""

SEAL SEAL VALVE

VALVE (TO UPPER
(TO LOWER SAMPLE
SAMPLE ~--tVY..1 ..PROBE OPEN
CHAMBER)
CHAMBER) 'as SAMPLING
Fig. 2 - Functional schematic of RFT sampling system. Fig. 4 - Packer filter-probe assembly.

NOVEMBER, 1975 1333

peatability (see Table 1). Direct digital pressure read- the success ratio has approached 100 percent.·
out is provided on the cbntrol panel in the truck, with The small wall-contact surfaces featured by. the tool
simultaneous analog and digital· curve recordings on essentially have eliminated sticking of the tool itself and
film. . . the resultant fishing operations. These features also re-
·A typical recording of a "pressure test only" and of lieve the general problems of cable keyseating and
a sampling operation are shown in Figs. 5 and 6, respec- differential-pressure sticking by minimizing the pull
tively. 1p these figures, the left-hand track shows the necessary on· the tool during retraction. Ten fishing jobs
analog pressure recording vs time. The right-hand four occurred during the course of these operations; these
tracks· display the equivalent digital information; the jobs were caused by' cable keyseating.
first of these. tracks gives thousands of pounds per The following examples are typical of the results
square inch, the next gives hundreds, the next gives obtained.
tens, and the last track gives units. Thus, in Fig. 6, the
total digital reading at the end of the test is 4,000 + 300 Example 1
+ 60+ 9, or 4;369 psi. Excessive seal failures have severely handicapped the
On the surface, the time required to empty the sample use of conventional formation~testingequipmerit in
chambers· and prepare the equipment for subsequent many wells. Four conventional wireline tests were at-

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

testing runs into the well has been reduced significantly tempted in the interval from 4,496 to 4,502 ft (Zone· A)
with the RFT. This tool is designed with clean-out ports shown in Fig. 7. Because of seal failures, each of the
that provide easy access to the sampling flowlines and four tests yielded a tool full of drilling mud. The RFT
chambers. A thorough backflush of the sampling system tool was brought to the well and was used to test the
with both water and air is performed by field personnel. same interval. It was set twice in the ·.same zone. The
The result is a quick, effective surface preparation of first setting indicated that· the tool plugged after the
the tool for additional tests. 2%-gal (10,250-cc) chamber was opened. The tool was
retracted and set again. This time, 8.1 cu ft of gas,
. Field Results 2,000 cc of oil~ and 5,000 cc of oil-cut mud were
The RFT technique is currently being evaluated in the recovered.
Gulf Coast area.
One hundred-eighty wells, involving 600 zones, have Example 2
been tested. Fluid-sampling operations were pe.rformed Shallow, unconsolidated shaly sands have presented
on 300 zones and pressure tests only were performed on many testing pro~lems in Gulf Coast formations. The
the remaining zones. An average of more than 10 setl interval from 1,248 to 1,267 ft shown in Fig. 8 has
retract operatiops were performed on each well. been considered a potential· shallow gas pay for several
The success ratio for sampling (number of interpreta- years. However, seal failures and flowline plugging
ble samples obtained compared with those attempted) have prevented this zone from being successfully
has· been more than 90 percent. For pressure tests only, wireline tested. The RFT tool was used recently to test

ANALOG PRESSURE DIGITAL PRESSURE RECORDING- psi

RECORDING - psi a 1000 a 10
a 10000 a 10000 100

o SHORT NORM 10 TEST RESULTS

o AMP SN 2 ZONE A

o R~d 10
---------_ ........ _-- CONVENTIONAL

4 SEAL FAILURES
4 TOOLS FILLED
WITH DRILLING
FLUID

ANALOG PRESSURE DIGITAL PRESSURE RECORDING - psi

I
RECORDING - psi a 1000 a 10 RFT
a 10000 a
~-~I
10000 a 100 1st SETTI NG
~ 1 TOOL PLU'GGED
MUD
PRESSURE SET
/TOOL
. I 2nd SETTING
~FM. 'PRESS
.........
t
--
PRETEST T RECOVERED -
8.1 cuftGAS

I
OPEN SAMPLE 2000 cc OIL
VALVE 5000 cc

~~~~~~~E min
1f OIL-CUT MUD
'"'-
,
i
I
\
t
I
Fig. 7 - Comparison of conventional FT and RFT results
I 4000
L psi
i in same zone. The RFT results here, and in each of the
SUbsequent examples, were achieved during a single
Fig. 6 - Recording of a sample test. trip in the hole.

1334 JOURNAL OF PETROLEUM TECHNOLOGY

this zone. Three attempts were necessary to obtain a was retracted and moved to 6,109 ft. The flowline
packer seal in a permeable interval. A good test Was plugged when the chamber was opened~ Since closing
finally obtained at 1,252 ft. Recovery was 10,250 cc of the tool clears the flowline, the tool was retracted and
fresh water. moved to 6, 109.5 ft, where pretest indications were
positive. Recovery was 44 cu ftof gas and 1,000 cc of
Example 3 mud.
Testing in very thin zones and shaly sands is handled
easily with the RFT technique. Shown in Fig. 9 are two Example 4
zones that were tested with one trip in the well. Zone A A quick check of formation pressures can be made dur-
was tested at 6,494 ft. This interval is only 2 ft thick. ing one trip in the hole in areas where this information
The first setting yielded no pressure on the pretest, indi- is essential for good reservoir evaluation and decision
cating the packer assembly was opposite an imperme- making. As an example, Fig. 10 shows an interval
able zone. The tool was repositioned. On the second where several pressure tests were taken on the same trip
setting 16.2 cu ftof gas and 750 cc of water containing in the hole. Zones F and G are known pressure-depleted
17-percent formation water were recovered. Zone B is sands. The two zones were tested to measure current
indicated by the SP and other logs to be a rather shaly pressures. Zone B shows a somewhat lower pressure
sand. The first test attempt was at 6,110 ft. A slow than Zones C, D, and E; this was because it was being

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

pressure buildup was indicated on the pretest. The tool produced in a nearby well.

Example 5 "
Mud filtrate often masks the results of tests taken with
SP RESISTIVITY the wireline formation tester. This can be quite impor-
10 tant when other means of evaluation indicate the zone to
-H+
--,-.--_._-
5 11 Caliper \5"

i\)
.2 10 20 be' a borderline case, and when the wireline tester re-
covers a small volume of gas or a trace of oil and a
o
I o RFT RESULTS large volume of fluid that appears to be mud filtrate.
I Set:t:f Depth Results
\ The RFT tool's ability to segregate the fluids recovered
I
c~\ I 1251 Low Perm. near the end of a test can help solve this problem in
( 2 1262 Low Perm. areas where test results such as these are common or
I 3 1252 Recovered
(
10250 cc
where the diameter of invasion calculated from logs in-
)
I FreshWater dicates that deep invasion may be a problem. Zone A in
I Fig. 11 was tested at 5,441 ft and the last gallon of the
\
(
\,
I
(
SFL
(
~
~ o SFL 20
, I
(;j o AMP. SFL 4
o
o 0 ~!.-L~ ~Q.
• HYD. ~ REMARKS
PRESSURE
Fig. 8 - RFT results in shallow, unconsolidated sand.
3500 2886

SP RESISTIVITY
20 RFT RESULTS
.2 1 10
Set:l:l:. Depth Results
I----r~-------i]. I 6110 SLOW PRESSURE
BUILD-UP ON
PRETEST

2 6109 TOOL PLUGGED

3 6109-1/2 RECOVERED-
44 cu ft GAS
1000 cc MUD 3593 2606

3600 2810
3613 2800
3626 2800

SP
~

~ 1 6494-1/2 NO PRESSURE
BUILD-UP ON
PRETEST
LOW PERM
3706 821 DEPLETED
6494 RECOVERED- 3710 806 DEPLETED
16.2 cu ft GAS
750 cc WATER
(17% FORM- 3713 3160
ATION WATER)

Fig. 9 - Example of RFT use for multiple tests in thin,

shaly sands. Fig. 10 - Example of a series of pressure tests.

NOVEMBER, 1975 1335

fluids ,recovered by the tool was segregated. The'deci- 2%-gal chamber were 1.8 cu ft of gas, a trace of oil,
sion to segregate was based on previous testing experi- and 10,000 GC of water, the resistivity of which was
ence '. in this area. After setting the tool, the 2% -gal 2.06 ohm-m at 75°P. The filtrate, at the time of log-
chamber was .allowed to fill, and after 7 minutes' it was ging, was 1.59 ohm-m at 75°P. Prom this sample, one
sealed. The I-gal chamber was then opened for' 3' min- might conclude that this zone probably will produce gas
utes; after this, the chamber was sealed and the tool was and some. oil, since there is' no evidence that formation
retracted and, brought to the surface. Recovered in the water· was produced in the test. This conclusion would
be reinforced by the fact that this well is running high
compared with another well that had a good show in the
equivalent geologic interval. However, the recovery
§~ __ 9~_LJf..E_f! __ ~~: o RSFL 10 from the I-gal chamber changes this picture drastically.
Q. __ ~~a J o AMP. R 2
Recovery from this chambe'rwas 0.8 cu ft of gas, a
SP trace of oil, and 2,300 cc of water, the resistivity of
10 o R_IL_d IO_. which was 0.147 ohm-m at 75° (as compared with 2.06
. . ~+ J 'RFT RESULTS
ZONE A ohm-m at 75°P for the water sample in the 2%-gal
2 3/4 GAL CHAMBER chamber) . This resistivity corresponds to more than 90-

Downloaded from http://onepetro.org/JPT/article-pdf/27/11/1331/2225297/spe-5035-pa.pdf/1 by Petrobras user on 12 October 2024

1.8cu ft GAS percent formation water. Therefore, the final· result of
TRACE OF OIL
10,000 cc WATER
the test indicates that water production can be expected
Rrf =2.06 ((i) 75° F from this zone.
Rmf = I. 59 (Q) 75 0 F
These results illustrate the importance of the multiple-
t GAL CHAMBER
set and pretest features for achieving high success ratios
0.8 cu ft GAS
TRACE OF OIL with'minimum rig time. Experience with the RFT tester
2300 cc WATER suggests· that without these features, test data from exist-
R . Rrf = 0.147 (@ 75° F
SFL R w .=$ 0; 14 (Q) 75 0 F ing techniques could be misleading in cases where time
DIGITAL PRESSURE RECORDING -- ps i does not permit the zone to be '''probed'' adequately.
o 1000 10 10 While no results are yet available from tighter forma-
10000 0 100
1---+-1 tions, such. as those in the Mid-Continent, the new tool
is expected, to minimize test time and to imp.rove inter-
TOOL pretability of samples by its capability of seeking out
RETRACTED
~ the more permeable zones in the formation.
Summary'
Substantial improvements in wireline form.ation- tester
techniques have been recently effected through de- ,
velopment of a tool with multiple-set capability. The
device can be set any number of times on. a single trip
in the well, permitting the operating engineer to pretest
I GALLON or probe the formation for more permeable regions and
-+- CHAMBER
OPENED to check the integrity of the packer seal before taking a
fluid sample.

··
Any number of pressure measurements can be made
l rapidly while in· the well with greater accuracy than is
2 3/4 GALLON
CHAMBER .-I
I
I
possible with earlier equipment. Two fluid samples can
be recovered. Two segregated samples can be taken
SEALED
,
I
I
I
from the same zone. Numerous pressure-buildup tests
can be made while sampling. Sampling now can be
done more efficiently in 'unconsolidated as well as in
consolidated' formations.
Pield results indicate success ratios of more than 90
percent, with improved zone interpretation and signifi-
cant reduction in the rig time required for the testing.
References
.,
I
1. Lebourg, M., Fields, R. Q.,and Doh, C. A.: "A Method of For-
mation Testing on Logging Cable," J. Pet. Tech. (Sept. 1957)
2 3/4 :
I
260-267; Trans., AIME, 210.
I
GALLON
i 2. "Formation-Tester Interpretation - Methods and Charts,"
-----CHAMBER " Schlumberger Well Services, Houston (1966). JPT
OPENED
!
i
I
Original manuscript received in. Society of Petroleum Engineers office Aug.
Fig. 11 - Example Where, using the RFT, segregation of the 23, 1974. Revised manuscript received Sept. 23, 1975. Paper (SPE 5035) was first
presented at the SPE-AIME 49th Annual. Fall Meeting, held in Houston, Oct. 6-9,
fluid prod uced last from that prod uced first clarified the 1974. ©Copyright 1975 American Institute of Mining, Metallurgical, and Pe-
interpretation of the zone. troleum Engineers, Inc.

1336 JOURNAL OF PETROLEUM TECHNOLOGY