SPE 110831
Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir
Pressure in Carbonates Using a Pump-Out Formation Tester with an Oval Pad
Chris Jones, SPE, and Wandi Alta, JOB Pertamina-Hess Jambi Merang; Jorawar Singh, Bob Engelman, Mark Proett, and
Bob Pedigo, SPE, Halliburton Energy Services
Copyright 2007, Society of Petroleum Engineers
This paper was prepared for presentation at the 2007 SPE Annual Technical Conference and
Exhibition held in Anaheim, California, U.S.A., 11–14 November 2007.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any position
of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE
meetings are subject to publication review by Editorial Committees of the Society of Petroleum
Engineers. Electronic reproduction, distribution, or storage of any part of this paper for
commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.
Abstract
Collecting representative, single-phase retrograde gas samples
from heterogeneous fractured carbonate formations is
challenging, especially for cases in which the reservoir pressure
narrowly exceeds the dewpoint. Conventional wireline
probe/packer assemblies are undesirable in this situation
because their small diameter probes increase the pressure
drawdown, compared to other available intake configurations.
Furthermore, the likelihood of establishing hydraulic
communication with an open fracture or vug is low with a ½-in.
diameter probe, compared to setting straddle packers spanning a
1-m formation interval. The straddle packer intake geometry
reduces the drawdown pressure required to establish flow, thus
maintaining the flowing pressure above the dewpoint and
preventing the onset of unwanted condensation.
Inflatable straddle packers can be run with pump-out test
tools, but they have temperature, pressure, and operational
limitations. The new Oval Pad* design, an elongated oval-
shaped probe/pad assembly, possesses many of the advantages
of dual packers without their limitations. Like a straddle packer,
the Oval Pad tool is effective when testing thinly laminated and
tight formations as well as fractured or vuggy carbonates. The
oval shape enables the pad to encompass a significant vertical
length of the borehole while establishing a pressure seal with
the formation.
A retrograde gas reservoir is contained within the vuggy
carbonates in the Jambi Merang area of South Sumatra. Here
the reservoir pressure narrowly exceeds the dewpoint. Pressure
must be maintained above the dewpoint during sampling
operations to prevent unwanted phase separation. The RDT
pump-out tester was fitted with Oval Pad and nitrogen-charged
sample bottles to collect single-phase samples and to keep the
phase intact until PVT properties can be analyzed in the
* A mark of Halliburton Energy Services
laboratory. These samples were necessary to determine
laboratory PVT properties and to define a production strategy.
This paper presents the details and challenges of the
successful collection of single-phase samples in this
environment.
Introduction
The Jambi Merang block is situated in the north of the south
Sumatra basin (Fig. 1). The block was awarded in February
1989; and since then ten wells have been drilled, of which six
were exploration and four were delineation (Fig. 2). The
carbonate in the South Sumatra basin comprises build-ups on
carbonate platforms along with the Marine Shale equivalent.
The platform limestones are micritic and tight, while the build-
ups form the reservoirs. Porosity is predominantly created by
meteoric waters at the top of the build-ups, hence the traps are
typically stratigraphically controlled by the diagenetically
developed porosity distribution. The platform carbonates
generally form on the shelf of the Sunda Craton with the build-
ups developing on the platform. In the Jambi Merang block, the
carbonate platform is isolated from the palaeo-coastal fringing
platform limestones and is situated high on the Merang Block.
On this isolated carbonate platform, there is a series of build-
ups, one of which is Sungai Kenawang. 1
Many operators have worked in this area, which is mainly a
gas and condensate producer with the gas transported mostly to
Singapore LNG Plants via undersea pipe lines and in west Java
Power Plants. The carbonate reservoir is a 700-1000 ft thick
retrograde gas reservoir with reservoir pressure very close to
the dewpoint and with a bottom aquifer drive. JOB Pertamina-
Hess Jambi Merang drilled two exploratory wells with the
primary objective of enhancing formation evaluation. The
evaluation program involved openhole logs, core analysis, and
sampling. Cores were taken for laboratory studies of the
carbonate reservoir rock. Openhole logs were run including
standard resistivity, density, neutron, sonic, and spectral gamma
ray in addition to electrical micro imaging.
Pressure testing, sampling, and fluid analysis were an
important part of the formation evaluation program. Samples
were required from the gas as well and single-phase gas sample
was necessary for future production planning. It was anticipated
that the dewpoint would be no more than 50 psi below the
formation pressure. Obtaining representative gas and
condensate samples is always challenging; but, in this case, the
2 Chris Jones, Wandi Alta, Jorawar Singh, Bob Engelman, Mark Proett and Bob Pedigo SPE 110831

plan was to keep the draw down differential pressure at the sand advantages of a probe. As shown in Figs. 3 and 5, the new
face to only 20 psi below the formation pressure. Reservoir Description Tool (RDT*) Oval Pad can expose
nearly 10 inches of vertical (or longitudinal) distance which is
approximately one-quarter of a typical SPS interval.2
JOB PERTAMINA – AMERADA HESS
JAMBI MERANG

Pulau
Gading

Gelam
Sungai
Kenawang
S. Lalang

JOINT OPERATING BODY

PERTAMINA – AMERADA HESS
JAMBI MERANG

JAMBI MERANG AREA

LOCATION MAP
Arch. No. : SS - 20 – 2666a Author : I.K.
Date : 12 January 2005 Drwn by : S.D.S.
AH/2609-May 2003

Fig. 1 – South Sumatra location map where the wells were drilled in
the carbonate.
32
38 3 103° 45' 104° 00' 104° 15'
1
109
52 2
SU NGAI GELAM
8
7
SG-4
SG -5 5
3 2

KETALING 2 SE
1
1
3 MB 01° 45'
4
MUAR A SABA K-1 IL A
5
EA ST KETALIN G NG
PL
H

SU NGAI
EP AT
HIG

M EDAK
NO RTH-1
DE FO
RM
G
G

IN

CE
LIN

SU NGAI
L

MP
TA

ME DAK-1
ED
KETA

AK
EAST KE

H
IG

ME RANG -1
PULAU
H

SU KAJAY A-1
1
GADING G
AN
3
SU NGAI 2
1 ER ME RANG Basement High Oil Field
BAYUNG M Basement Deeps Gas Field
02° 00'
3 L INCIR
2
1 Surface Highs
2
3 6 Non Depositional Areas of TAF
4 1
KA

1 ST1&2 Initial Jambi M erang PSC Area Boundary

3
2
LI

ER 5 Rem ain Jambi Merang PSC Area Bounda ry

B

AU 1
HI G
H N G DE EP
JOINT OPERATING BODY

W. BERA U-1 M ERA GELAM PERTAMINA – AMERADA HESS JAMBI MERANG

KAL I BERAU-5 KUKULAM BAR-1

JAMBI MERANG BLOCK
MAIN TECTONIC ELEMENT
Fig. 3 – The RDT dual-probe oval pad (right) is compared with the
0 5 10 15 KM S
SUNG AI
KENAWANG
(SCHEMATIC) conventional dual-probe pads (left). The opening of the oval pad
Arch. No.

Date
: SS - 20 - 2523

: 26 June 2002
Author

Drwn by
: DJH/ Adt/ JC/ BR

: S. D. S.
that is exposed to the formation is nearly 10 inches long. Notice the
Fig. 2 – Detail location map of Jambi Merang block. Red blocks are two ports at either end of the pad which are the same size as the
the gas / condensate field. probes. These ports are screened much the same as a
conventional probe and either one or both can be used for
sampling.
Formation Testing and the Oval Pad
Modern pumpout wireline sampling tools were introduced in Job Planning for Pressure Testing and Sampling
the early 1990’s and included single or multiple probes (Fig. 3). Based on the system simulations similar to that shown in a
Most of these wireline formation testers (WFT) offer probes in previous paper (i.e., Zefzaf, T. El, et al.), comparisons can be
the ½ to 1-in. diameter range, and some probes are as large as made between sampling systems over a range of operational
two inches. But sampling in highly heterogeneous formations conditions.2 This analysis can assist in the planning of WFT
still represented a challenge for WFT; and soon after the pumpout operations. Typical parameters for this analysis are
pumpout tools were introduced, straddle packers were adapted. shown in Table 1 with the simulation results in Table 2. Dual
Dual straddle packer systems (SPS) offer advantages over probes are shown because the RDT can use two closely spaced
probes in low permeability applications as well as probes (7.25-in. spacing) that can sample from the same zone.
heterogeneous environments (Fig. 4). In carbonates, thinly Notice in this case that the SPS would take only 30 minutes to
bedded sands and naturally fractured reservoirs, most of the inflate the packers, pumpout a sample and deflate, while the
production occurs from small features. Such features make Oval Pad would take 1.5 hours. However, a single probe would
sampling and reservoir characterization difficult with a probe. take nearly 10 hours. Figs. 6 and 7 show the variations in
The probe is more likely to be placed in a location that is pumping rates and pumpout times with mobility changes while
characteristic of the rock matrix, which usually results in a tight holding the pumping differential to less than 25 psi. Because
test. The SPS typically isolates an interval of 1 meter which is the SPS can pump at nearly 1 GPM at very low pressure
normally ample to characterize heterogeneous rock. differentials, it usually has a lower pumping time. But straddle
The primary advantage of an SPS is its ability to cover a packers have operational difficulties and are generally more
vertical interval where a probe is a pinpoint by comparison. In expensive to run. This planning analysis can help to put into
this paper, we introduce a new probe concept that is able to
traverse a vertical interval but still retains the fundamental
SPE 110831 Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir Pressure With an Oval Pad 3

context the tradeoffs needed when deciding which system to

use.
Considering the objectives of the testing plan, the Oval Pad
was selected for this reservoir. One of the primary reasons for
this selection was that the pressure survey planned was
extensive. In this carbonate reservoir, WFT probe tools had
proven to be problematic; but the Oval Pad had been used
successfully in other carbonate reseroirs.2 On the other hand,
straddle packers can only take a limited number of pressure
tests before they become fatigued and unusable. They also have
a higher risk of differential sticking. As shown in Table 2, a
single probe can flow at 0.6 cc/sec while the Oval Pad can flow
at 3.75 cc/sec while maintaining a 25 psi pressure differential
(i.e., ΔPpout= Pformation - Pflowing). This means the Oval Pad can
flow over 6 times faster than a probe under these conditions.
Also, because it traverses nearly 10 inches of well bore, both
pressure testing and sampling are significantly more
representative in these carbonate formations.

Fig. 5 – These photos show the Oval Pad in the retracted and
deployed positions.

Fig. 4 – This photo shows a Dual Straddle Packer Section (SPS) for
the RDT that is typically used in low permeability, carbonate
formations.
4 Chris Jones, Wandi Alta, Jorawar Singh, Bob Engelman, Mark Proett and Bob Pedigo
Pumpout Flow Rates
100
10
Flowrate(cc/sec)
1
0.1
0.1 1 10
Mobility (md/cp)
Set Single Dual
Fig. 6 – Pumpout flow rates versus formation mobility using a
Single Probe, Dual Probe, Oval Pad and Straddle Packer System
(SPS) and the inputs from Table 1. The set line is determined by the
permeability and viscosity inputs in Table 1.
Pumpout Time
10000
Tpout (minutes) ())
1000
100
10
0.1 1 10
Mobility (md/cp)
Set Single Dual
Fig. 7 – The pumping time comparisons are based on pumping a
total of 50 liters at the rates shown in Fig. 6. The SPS curve is
shifted due to the additional volumes required to inflate the
packers and a portion of the fluid between them.
Formation Pressure Testing & Sampling Case Study
As noted above, two wells were drilled in Jambi Merang block
to evaluate carbonate reservoir quality and hydrocarbon fluid
type. Wireline logging, including pressure testing and sampling
along with conventional coring through the potential reservoir
sections, were used extensively to accomplish formation
evaluation objectives.
Previous WFT fluid sampling attempts in Jambi Merang
condensate reservoirs yielded multi-phase contaminant-bearing
fluid and, in these cases, the inlet flowing pressure was allowed
to drop below the dewpoint, a drop which induces unwanted
liquid condensation in the saturated Jambi Merang reservoirs.
Allowing liquid condensation to occur in the pore system
reduces the relative permeability to gas, which in turn induces
additional pressure drop and liquid condensation. Collecting a
gas sample above the dewpoint avoids both the unwanted phase
separation and biasing the sample towards the higher mobility
light-end fluid components, thus assuring the collection of
representative samples which contains all reservoir fluid
components. Liquids can drop out in the pore space due to
SPE 110831
ongoing condensation, especially in low mobility reservoirs
where increased pressure drop is necessary to induce flow.
Extensive pre-job planning was conducted to optimize the tool
string for achieving low differential pressure during pumping
and collecting single-phase samples (see previous section).
The RDT consists of multiple sections that can be connected
in numerous configurations for testing and sampling
applications (Fig. 8).3,4 A typical configuration consists of a
Hydraulic Power Section (HPS), Quartz Gauge Section (QGS),
Flushing Pump Section (FPS), and Multi Chamber Sections
(MCS).
100 1000 10000
HPS
Oval SPS
DPS
Oval Pad
By-Pass Flow
Continue Pumping
QGS While Stationary
Fluid Remains in
MRILab
FPS
MCS
Normal Flow
Continue Pumping
While
Fluid Flows thru
MRILab
MRILab
MCS
100 1000 10000
Oval SPS
Fig. 8 – RDT Tool configuration used showing the normal through
flow path and the bypass method. Pumping down allows the fluids
to naturally segregate rather than become highly mixed when the
pump is placed above the probe.
During normal flow pumping, the fluid is routed through the
MRILab and expelled to the borehole through the expulsion
valve in the downstream MCS (lower MCS in pump down
case). The bypass flow method was developed for recording
NMR T1 measurements on stationary fluid in the MRILab
without affecting pumpout flow geometry. Pumprate is
maintained during bypass flow, and the fluid is expelled
through the MCS expulsion valve on the upstream side of the
MRILab. Bypass flow is used periodically to improve NMR
signal to noise and confirm T1 response (Fig. 8).
A pump down mode was selected because it has proven to
be the most effective configuration for sampling light oil and
gas. This is due to the fact that the heaver contaminants tend to
segregate naturally while flowing down rather than mix when
pumping up. This phase segregation also improves the
determination of fluid type and contamination estimations. Oval
Pad was selected to overcome common challenges associated
with pressure testing and sampling heterogeneous and vuggy
carbonate reservoirs with conventional probes which include
SPE 110831 Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir Pressure With an Oval Pad
establishing pressure seals with the formation and avoiding
tight tests.
X,928.0 ft X,931.0 ft X,934.0 ft X,937.0 ft
Fig. 9 – Photograph of the core taken in Well 1 showing the
heterogeneity of the carbonate at the depth where gas sampling
was attempted. One core image is 3 ft long. An image of Oval pad
is shown to scale of the core to show the depth at which it was
sitting while taking the sample.
Sample Contamination Analysis. A Magnetic Resonance
Imaging Laboratory (MRILab*) tool section was used for
contamination analysis and fluid probability determination while
pumping. A popular approach to fluid ID is based on the optical
properties of the fluids entering the tester.5 The MRILab Fluid
Analyzer6, on the other hand, offers an alternative based on the
NMR properties of the fluids.7 The MRILab makes Saturation
Recovery (SR) T1 measurements during the pump-out process,
and contamination can be estimated in either time or relaxation
domains. A wealth of reservoir fluid information is available
from the RDT/MRILab combination including viscosity,
diffusivity, and hydrogen index (HI).8 During normal flow
pumping, the fluid is routed through the MRILab and expelled
to the borehole through the expulsion valve in the downstream
MCS (lower MCS in pump down case). The bypass flow
method was developed for recording NMR T1 measurements on
stationary fluid in the MRILab without affecting pumpout flow
geometry. Pumprate is maintained during bypass flow, and the
fluid is expelled through the MCS expulsion valve on the
upstream side of the MRILab. Bypass flow is used periodically
to improve NMR signal to noise and confirm T1 response
(Fig. 8).
5
Well 1 Open Hole Logs and Borehole Imaging (X,931 ft)
Well 1 is a delineation well drilled to determine the extent
of the carbonate structure and to determine reservoir pressure
and fluid properties. A photograph of the core recovered from
Well 1 shows the heterogeneity of the carbonate at the depth
where a condensate sample was collected. An image of the
Oval pad is shown to scale on the core to show the depth at
which it was sitting while taking the sample (Fig. 9). A close up
of the core section where the Oval Pad sample was taken is
shown in Fig. 10 where a dense tight section of rock was
spanned by the Oval pad. This close up also shows the
carbonate heterogeneity on a smaller scale with vugs, inclusions
and micro fractures in the rock.
9.6 in.
Fig. 10 – Photograph of the core adjacent to the Oval Pad at the
depth the sample was taken (X,931 ft). Notice the Oval Pad spans a
dense section of carbonate that would have been detected as tight
and a sample may not have been recovered with a conventional
wireline probe tool.
The Dual Laterolog, Density, Neutron, Sonic and Spectral
Gamma Ray recorded through the reservoir section in Well 1
are shown in Fig. 11. Collecting and retrieving single-phase
retrograde condensate fluid samples was established as a
priority. Because the reservoir pressure could be equal to or
slightly above the dewpoint pressure, a plan was developed to
maintain the pressure differential between the reservoir and the
inlet flowing pressures to less than 20 psi while pumping to
clean up and collect a clean representative sample.
6 Chris Jones, Wandi Alta, Jorawar Singh, Bob Engelman, Mark Proett and Bob Pedigo
X900
RDT
X950
Fig. 11 – A Quint log recorded in Well 1 within the Baturaja
Carbonates interval. The Quint log consisted of dual laterolog,
density, neutron, sonic and spectral gamma ray. The point where
RDT Oval Pad was positioned and pressure test and sampling were
undertaken is shown by the red dashed line.
High resolution electrical borehole images indicate a
heterogeneous carbonate depositional environment where a
condensate sample was collected in Well 1 (Fig. 12 and 13).
The pressure drop through the Oval Pad is less than the
equivalent drop would be through conventional DPS probes at
the same flowrate. The pre-job planning based on anticipated
mobility, borehole overbalance pressure, and maximum
allowable drawdown pressure influenced the selection of the
Oval Pad for this well over dual probes or straddle packers. As
previously mentioned, the tool was configured to pump down.
This configuration enables the heaver WBM filtrate to gravitate
downward through the pump and flow line and be cleanly
flushed out to the wellbore.
X900
RDT
X950
Fig. 12 – High Resolution electrical Imaging log recorded by
Extended Range Micro Imager (XRMI) showing the heterogeneity of
the carbonate reservoir in Well 1.
SPE 110831
Gamma Ray Dip (Bed) GR
Static Image 2D-Static Image Depth 2D-Dynamic Image
(ft) 0 - 90º AVEMIRES
X RDT
X
Fig. 13 – High resolution electrical imaging log recorded by XRMI
showing the heterogeneity of the carbonate at depth where
condensate sampling was attempted in Well 1.
Pressure Testing Results. Thirty nine pressure tests were
recorded in Well 1. A gradient analysis of selected pressure
data shown in Fig. 14 yields a gas gradient of 0.087 psi/ft with
a high degree of accuracy (±4% standard deviation). The
gradient accuracy was determined using a statistical analysis
method recently developed by Collins et al.9 This analysis
enables the gradient accuracy to be compared with expected
results. For example, it was assumed that pressures could be
recorded with a precision of ±0.1 psi and a depth precision of
±2 ft (standard deviations). The precision is basically the
repeatability of a measurement, and these values are typical for
probe WFT in a high quality reservoir. This level of precision is
shown by the red lines representing the boundaries shown in
Fig. 14. The blue lines represent the measured variations of the
gradient pressure points. Because the blue lines are near the
boundaries represented by the red lines, it can be concluded that
the gradient accuracy is within the high standards expected of a
standard probe WFT in a high quality reservoir.
Fig. 14 – Statistical gradient analysis for Well 1 reveals a gas
gradient of 0.087 psi/ft with a high degree of accuracy (± 4%).
Residual plot shows the variations of the pressure points from the
gradient.
Sealing efficiently was also comparable to a standard
probed WFT. There were five seal failures and two tight tests.
Six samples were collected at four different depth stations.
Three sampling stations were gas-bearing, and one was water-
bearing. The Oval Pad pressure sealing efficiency on this well
was 87 %, a value which is consistent with 85% sealing
efficiency attained throughout Indonesia.
SPE 110831 Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir Pressure With an Oval Pad 7

The primary objective of reducing the inlet pressure performed in Well 1. Initially, FluidXpert predicts high water
drawdown during pumping was achieved, namely to maintain probability as shown by the blue shading in track 4 plot of
the fluid above the dewpoint pressure. Fig. 17. This track is a fluid probability track that is based on
MRILab measurements. After the fluids captured in the
Well 1 Sampling and Contamination Analysis. A plot flowline during the previous pumpout first pass by the sensors,
depicting the entire pumpout sequence is shown in Fig. 15. the fluid probability is nearly 100% water, a value which is
Inlet pressure, outlet pressure, and pump rate are plotted vs. expected when pumping a formation drilled with water-based
time. Initially the outlet pressure indicated some flowline mud. Immediately after new fluid arrives, mud fines first
plugging, which was remedied by modulating the pumpout rate appear and introduce a significant increase in surface area
and selecting different outlet ports. A closer look at the causing a reduction in the T1 log mean measurement. T1 log
pressure vs. time data indicates the pressure differential mean subsequently increases as mud fines clean up. Clean-up
between formation and inlet pressure was maintained at results in an increase in characteristic T1 log mean. As
approximately 15psi. This difference was sufficient to maintain pumping continues, the fluid properties change from initial
single-phase conditions and not cause the pressure to fall below filtrate toward pristine reservoir fluids.7,8 In this particular
the dewpoint (Fig. 16). pumpout, the T1 log mean and hydrogen index approach values
computed for gas (T1 log mean and HI are computed for gas
5300 50 based on flowing pressure and temperature), so the fluid
5100
Over Pressure Samples
45
probability approaches 100% gas as indicated by the red
shading in the track 4 plot. Two water slugs were observed
40
4900
during pumpout, an indication that water accumulated at the
Outlet Flowline Plugging

4700
35
bottom of the tool string, where the sensors are located, due to
Pumpout Rate (cc/sec)

Pump Outlet Pressure

pumping at a relatively low rate to maintain the inlet pressure
Presssure (psi)

30
4500

25
above the dewpoint. This accumulation of water was specially
4300 seen during the first By Pass which helped to identify that there
20
Equalize Flowline Pumpin Oval-Pad Pressure was still some filtrate in the Fluid Pumped.
4100
15

3900
10

Oval Pad Pretest Pumpout rate

3700 5

3500 0
0 1000 2000 3000 4000 5000 6000 7000
Time (sec)

Fig. 15 – Sampling pumpout sequence for Well 1 showing the inlet

and outlet pressures. Initially the outlet pressure indicated some
flowline plugging, which was remedied by modulating the pumpout
rate and selecting different outlet ports.

3980 40

3975 35

3970 30

Pumpin Oval-Pad Pressure 15 psi Control Limit

Pumpout Rate (cc/sec)

3965 25
Presssure (psi)

3960 20

3955 15

3950 10

Pumpout rate
3945 5

3940 0
500 1500 2500 3500 4500 5500 6500
Time (sec)

Fig. 16 – It was necessary to maintain the pumping differential

pressure near to 15 psi to maintain single-phase and not cause the
pressure to fall below the dewpoint.

Real-time contamination estimates are determined using Fig. 17 – FluidXpert plot for Well 1 showing fluid probability and
FluidXpert.* FluidXpert is an analysis method that incorporates contamination during a pump-out sequence. The bypass method
several sensor measurements in combination to estimate is used for making T1 measurements on stationary fluid in the
MRILab without disturbing pumping flowrate and the clean-up
contamination. The measurements included are NMR T1 log process.
mean, HI, mobility index, resistivity, and capacitance The
FluidXpert log in Fig. 17 shows the analysis of a pumpout
8 Chris Jones, Wandi Alta, Jorawar Singh, Bob Engelman, Mark Proett and Bob Pedigo SPE 110831

To maintain the drawdown pressure to less than 15-20 psi while

2770 40
pumping, the flow rate needed to be accurately regulated by the Pumping Oval-Pad Pressure
testing engineer at 4-5cc/sec. A clear transition from water 35
2765
based mud filtrate to gas was observed using the FluidXpert log
shown in Fig. 17. Fig. 17 contains plots of the pump flow rate, 30

inlet and outlet pressures, fluid type probability, contamination 2760

Pumpout Rate (cc/sec)

25

estimation, and T1 NMR data. The FluidXpert analysis of this 20 psi Control Limit

Presssure (psi)
Oval-Pad Pretest
pumpout determined that the contamination was reduced to 6% 2755 20

(by volume at reservoir conditions) before collecting the first

15
sample at sequence 150. In this particular case, each MRILab 2750

sequence or NMR experiment is 32 seconds duration, so 10

sequence 150 is 80 minutes after pumping initiation. 2745

Pumpout rate
5

Well 2 Sample Gas Sample Case Study (X609.1 ft) 2740 0

RDT configured with MRILab was used to collect another 200 400 600 800
Time (sec)
1000 1200 1400

condensate sample in Well 2. The dual laterolog, micro-SFL, Fig. 19 – The pumping pressure, pumpout rate vs. time curves
density, neutron, sonic, and spectral gamma ray logs are plotted while sampling gas at X609.1 ft in Well 2. Two samples were taken
in Fig. 18. The density caliper log indicates poor hole quality at this depth, and one of the samples was taken in a nitrogen
charged chamber to maintain the single-phase.
at the depth where the Oval Pad was set and the sample was
collected. In fact there were very large washouts above and
blow the sample point of x609.1ft. The inlet pressure versus
pump rate and time is plotted in Fig. 19. Pump rate was
initially set at a low rate to minimize pressure drawdown. In
order to maintain less than 20 psi drawdown and avoid
unwanted condensation in the pore space, the maximum rate
was set at 8cc/sec at approximately 1000 seconds. The
FluidXpert plot indicates a high probability of water flowing
through the tool and past the sensors before the arrival of new
fluid from the current pumpout station (Fig. 20). The
expectation is that the new fluid type is probably gas. Clean-up
is relatively quick; the plot indicates contamination is less than
5% by volume at experiment 35 which was taken
approximately 18 minutes into the pumpout.

Fig. 18 – The Quint log recorded in the Well 2 within the Baturaja
Carbonates. The Quint log consisted of dual laterolog, density,
neutron, sonic and spectral gamma ray. The point where RDT Oval
Pad was positioned and pressure test and sampling were
undertaken is shown by red dashed line.
Fig. 20 – The fluid Identification was determined using MRILab
while sampling gas at X609.5 ft in Well 2. Two samples were taken
at this depth, and one of the samples was taken in nitrogen
charged chamber to maintain the single-phase.
Down Hole Production Sample. After Well 1 was completed,
a DST was performed. The well was produced for six days and
then shut in for multirate production log (PL) test. A bottom
hole sample (BHS) was taken during this initial flow period of
the DST before the PL test was run. The PL was run under
SPE 110831 Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir Pressure With an Oval Pad 9

stationary condition and two different rates to determine the hydrocarbon liquids were found when the sample was
volumetric contribution of fluid from the carbonate section transferred at reservoir conditions from its original container.
perforated from X965 ft to - X085 ft in a 7-in. liner. Production These results tend to contradict the compositional analysis
volumetric results, both surface and down hole, rates are shown which shows a lower dewpoint at reservoir conditions. Also in
in Table 3 and Fig. 21 below. The PL density curve, Pdens the second RDT sample the C7+ composition went up, an
determined from the pressure curve, Pres (see Fig. 21), shows increase which should lower the dewpoint. The sensitivity
an abrupt density change near the center of the perforated analysis shown in Fig. 22 illustrates the difficulty in testing
interval. This change corresponds to an apparent transition from condensate reservoirs. In this case, the lower dewpoint of 4020
a fluid to a gas at this point. psi at 279.5 F may still be higher than the actual value but is
It is also interesting to note that the volume flow rates in the probably within the expected error band. But clearly, the higher
3rd and 4th tracks reveal that the there is very little production dewpoint of 4521 psi at 304 F is a less likely result and is not
below the gas to liquid transition. The PL volumetric summary supported by the other dewpoints recorded or by the
of two perforated intervals in this well also shows the majority compositional-based PVT simulation.
of the production from the upper interval with both intervals The BHS was taken during the DST under flowing
producing gas in the production tubing near the perforations condition. The sample was taken inside the tubing ~120 ft
(Table 3). This observation strongly suggests that the above the top of perforation (X965 ft). A dewpoint of 4093 psi
formation gas is condensing in the well bore due to the pressure at 304 F was determined along with a lower 1.8 Mol% of C7+.
drop and is collecting below this fluid-to-gas transition point. Because the BHS was taken in tubing above the perforation
This pattern of condensation means that the well bore is acting interval when the well was flowing, there is a high probability
as a separator and the downhole sample taken above the the sample was taken when the heavier component already fell
perforations may not be representative of the total production. in the well. This conclusion is supported by the PL log shown
in Fig. 21.
Summary of Sampling Results. The samples taken for this Table 4 – Down Hole Sample Summary
case study are summarized in Tables 4 and 5. The results Formation Mud Inlet Outlet
shown in Table 4 are basically the conditions under which the Well Sample
Deg F psi psi psi psi
samples were taken. Table 5 shows the conditions at the surface
when the samples were recovered and also shows laboratory 1 RDT - 1 279.5 3969.6 4307.8 3975.9 5200
1 RDT - 2 279.7 3973.4 4341.7 3981.9 7500
testing results including the dewpoint and heavy hydrocarbon
1 DST BHS 298.7 3960 NA NA 8000+
composition (C7+ Mol %). As shown in Table 4 two RDT
2 RDT - 1 266.1 2762.6 2929.1 2771.7 6980
samples were taken in Well 1 (X,931ft), and one BHS was +
at 86 Deg F
taken during DST above the perforated interval (X,964-
X,979ft). An additional surface stock tank sample (STS) was Table 5 – Surface Sample Summary
taken, and the results are shown in Table 5. In Well 2 a single Surface Lab Results
sample was taken (X609.5 ft).
Well Sample Dewpoint C7 +
Using the detailed compositional analysis of the RDT
Deg F psi Deg F psi Mol %
sample in Well 1, an equation of state (EOS) simulation was
developed using PVTsim. The results are shown in Fig. 22.10 1 RDT - 1 76 2300 279.5 4020 2.27
To test the sensitivity of the sample to the heaver carbon 1 RDT - 2 76 2800 279.7 4513 2.34
content (i.e., C7+, typically liquids) two additional PVT 1 DST BHS 70 7500 304 4093 1.84
1 DST STS 118 510 304 3833 NA
simulations are shown; one assuming 1% by molecular weight
2 RDT - 1 73 4000 NA NA 3.53
additional C7+ content and the other assuming 1% less. From
this analysis it is clear that small changes in the heavy carbon
6000
content can dramatically affect the dewpoint. For this reason, it EOS - 1% Mole C7+
Gas RDT EOS
is critical that extreme care be taken to preserve a sample’s EOS + 1% Mole C7+
integrity when obtaining a sample as well as during subsequent 5000 Formation
RDT-1 Sample
handling. RDT 1 Dew Point
DST-BHS Sample
Pressure (psia)

4000
DST BHS Dew Point
Table 3 – DST Production Log Flow Summary
Zone Rate Qt (res) Qo (res) Qg (res) Qo (Surface) Qg (Surface)
Ft B/D B/D B/D STB/D MMscf/D 3000
Gas & Liquid
X965 - X979 6,103 0 6,103
Q1 238 8.65
X979 - X024 2,242 0 2,242
X965 - X979 11,644 0 11,644 2000
Q2 523 18.6
X979 - X024 6,432 0 6,432

1000
Because two RDT samples were taken in Well 1, dewpoint
tests were performed on each. The results are shown in Table 4,
0
with the RDT-1 sample having a dewpoint of 4020 psi at 279.5 -50 50 150 250 350 450 550
Temp (F)
F and RDT-2 sample a dew point of 4513 psi at 279.7 F. Only
gas was recovered from the sample chamber at reservoir Fig. 21 – Phase diagram using composition of RDT sample. Sample
conditions and dewpoints are plotted for the RDT sample and DST
conditions, and the laboratory reported that no water or bottom hole sample.
10 Chris Jones, Wandi Alta, Jorawar Singh, Bob Engelman, Mark Proett and Bob Pedigo
Fig. 22 – Production log showing contribution mainly from the top
part of the perforated interval.
During the PL-DST, a gas liquid stock tank sample (STS)
was taken which consisted of gas and liquids at surface
conditions. The dewpoint for this sample was found to be 3833
psi at 304 F after the sample was elevated back to reservoir
conditions and the liquid and gas recombined. The separator
samples are not as representative of the reservoir sample,
particularly for a retrograde gas reservoir. Separator samples
typically lack the heavy component which tends to fall in the
well as a condensate. The lower dewpoint suggests separation
has happened to some degree with this STS when compared to
the RDT and BHS sample dewpoints and compositions.
An RDT sample was taken in Well 2 using a nitrogen
compensated chamber allowing it to be recovered at the surface
at a higher pressure (see Table 5). Such a procedure is not
normally necessary with gas samples since it is standard
laboratory practice to recombine the condensate with the gas at
reservoir conditions. But it is critical that the recombination be
done before the sample transfer to insure all the fluids are
included. This sample had a much higher C7+ content which
was evident by the fact there was 5cc of liquid present in the
chamber (0.5% by volume). It is not possible to compare this
sample with the ones taken in Well 1 since it was at a different
depth and location. The Location of Well 2 is known to have a
higher condensate yield compared to Well 1 and this was seen
in the RDT sample C7+ contents.
Conclusions
1. This case study shows the results of successful application
of the new wireline pumpout tester Oval Pad technology in
a difficult, carbonate, retrograde gas reservoir (the Jambi
Merang block situated in the north of the south Sumatra
basin Indonesia).
2. Previous attempts to obtain representative WFT samples in
this field were not successful because the pumping
drawdown was not maintained to within 20 psi of
formation pressure.
3. Careful job planning made it possible to review alternative
sampling methods; and, as a result, the RDT Oval Pad was
chosen as the best alternative.
SPE 110831
4. The Oval Pad operates with a lower pressure differential
than probes, and the RDT tool offered the rate controls
necessary to maintain the flowing sample to within 20 psi
of reservoir pressure.
5. Samples were taken in two wells including an RDT
sample, a production logging DST bottom hole sample, and
a stock tank sample.
6. The RDT MRILab and FluidXpert were used to estimate
contamination and determine when clean samples were to
be taken.
7. Laboratory analysis of the samples revealed that the RDT
samples were comparable with the PL DST bottom hole
samples.
8. An EOS PVT simulation based on the composition of the
RDT sample demonstrated that the samples were within a
reasonable error band.
Acknowledgements
The authors would like to express their acknowledgement to
their respective management teams for granting permission to
develop this paper and exhibit the results displayed within. We
would also like to acknowledge Mickey Pelletier of Halliburton
for his contributions to the development of the discussion of the
PVT simulations and figures.
References
1. Clure, J., Fiptiani N., “Hydrocarbon Exploratoin in the Merang
Triangle, South Sumatra Basin,” Paper IPA01-G-10, presented at
the Indonesian Petroleum Association, Twenty-Eighth Annual
Convention & Exhibition, October 2001.
2. El Zefzaf, T., El Fattah, M. A., Proett, M. A., Engelman, B.,
Bassiouny, A.: “Formation Testing and Sampling Using an Oval
Pad in Al Hamd Field, Egypt,” Paper SPE 102366, presented at
the SPE Annual Technical Conference and Exhibition held in San
Antonio, Texas, U.S.A., September 24–27, 2006.
3. Proett, M. A., Chin, W. C., and Mandal, B.: “Advanced Dual
Probe Formation Tester with Transient, Harmonic, and Pulsed
Time Delay Testing Methods Determines Permeability, Skin, and
Anisotropy,” Paper SPE 64650, presented at the SPE
International Oil and Gas Conference and Exhibition held in
Beijing, China, November 7-10, 2000.
4. Proett, M. A., Gilbert, G. N., Chin, W. C. and Monroe, M. L.:
"New Wireline Formation Testing Tool with Advanced Sampling
Technology," Paper SPE71317, SPE Reservoir Evaluation &
Engineering Journal, April 2001.
5. Mullins, O. C., and Shoroer, J. J.: “Real-Time Determination of
Filtrate Contamination during Openhole Wireline Sampling by
Optical Spectroscopy”, Paper 63071 presented at the SPE Annual
Technical Conference and Exhibition, Dallas, TX, Oct. 1-4, 2000.
6. Prammer, M. G., Bouton, J., and Masak, P. C.: “The Downhole
NMR Fluid Analyzer”, SPWLA-2001-N Paper presented at the
SPWLA 42nd Annual Logging Symposium, Houston, TX, June
17-20, 2001.
7. Masak, P. C., Bouton, J., Prammer, M. G., Menger, S., Drack, E.,
Sun, B., Dunn, K.J., and Sullivan, M.: “Field Test Results and
Applications of the Downhole Magnetic Resonance Fluid
Analyzer”, SPWLA-2002-GGG Paper presented at the SPWLA
43rd Annual Logging Symposium, Oiso, Japan, June 2-5, 2002.
8. Akkurt, R., Fransson, C., Witkowsky, J. M., Langley, W. M.,
Sun, B., and McCarty, A.: “Fluid Sampling and Interpretation
with the Downhole NMR Fluid Analyzer,” SPE Paper 90971,
presented at the SPE Annual Technical Conference and
SPE 110831 Collecting Single-Phase Retrograde Gas Samples at Near-Dewpoint Reservoir Pressure With an Oval Pad 11

Exhibition held in Houston, Texas, U.S.A., September 26-29,

2004.
9. Collins, C., Proett, M. A., Storm, B., and Ugueto, G.: “An
Integrated Approach to Reservoir Connectivity and Fluid Contact
Estimates by Applying Statistical Analysis Methods to Pressure
Gradients,” SPWLA-2007-HH Paper presented at the SPWLA
48th Annual Logging Symposium, Austin, Texas, June 3-6, 2007.
10. Pedersen, K. S., and Christensen, P. L.: Phase Behavior of
Petroleum Reservoir Fluids, Cat. #: DK4661, CRC Press, Nov. 1
2006.