Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Presented at the 42nd Annual SPWLA Logging Symposium, Houston, Texa June 17-20, 2001
1
• the optimal physical placement of the gas temperatures ran about 290 ° F; at the surface
trap in an environment that is constantly location of the probe (in the shaker box at the
changing due to both changing drilling exit of the mud return line), the mud returns were
conditions and human operator intervention, measured to be 150 ° F.. The distance from the
• the physical size of the equipment, probe, which was inside the shaker box, to the
• water vapor condensing in the suction line detector was approximately 6 feet. Output from
from the trap to the logging unit (a particular the detector, WITS formatted digital ASCII data,
problem when air temperatures are below was sent at six-second intervals through a
freezing) and shielded cable approximately 200 feet to the mud
• the distance the extraction point remains logging unit. In the unit, data was
from the detection instrumentation simultaneously recorded from the Total Gas
(sometimes well over 100m). Analyzer (gas from the trap) and from the
membrane probe.
The new gas probe, shown in Figure 2,
• is small (about 15 cm in length), A comparison of the two data sets is shown in
• is physically robust with respect to Figure 3. The traditional mud log gas
extraction efficiency and calibration, monitoring and analytical system was calibrated
• has no moving parts aside from a low- daily and had undergone a complete QC check
volume air pump, before the comparison test was performed. The
• can be inserted into a mud stream virtually data from this equipment, total gas (HC)
anywhere in the return mud system (either concentration measured with a classic flame-
into a mud return pipe, at the bell nipple or ionization detector, are considered to be of high
at the shaker box), quality and consistent with what is currently
• uses a simple sensor located about 2 meters being acquired in our onshore and offshore mud
from the extraction point, and logging operations.
• includes control, calibration and
measurement electronics in a sealed module Several things were consistently observed during
with a low-voltage digital output (WITS the duration of the test, and can be seen in this
format) compatible with virtually all well- plot. First, the gas probe data gave a consistently
site data logging systems. higher concentration. This is because the probe
was calibrated to read the total hydrocarbon gas
concentration in the drilling fluid. This
The gas probe consists of a very thin polymer calibration was done in the manufacturer’s shop,
tubing wound around a stainless steel mandrel and checked at rigsite, and is found to be
(Figure 2-b). Hydrocarbons are extracted from relatively independent of the mud rate, mud
the surrounding fluid, through the walls of the density, and temperature. The gas concentration
polymer tubing, into a flowing air stream, which measured using the trap, however, is relatively
carries the hydrocarbons to a detector which difficult to calibrate because the extraction
measures the total hydrocarbon content of the efficiency is highly dependent upon mud
stream. The detector output signal is digitized, properties, flow rate, trap placement, and air rate
calibration factors appropriate for the type of which carries the gas to the logger’s unit for
drilling fluid (water-, oil- or synthetic-based) are analysis. Second, the data from the gas probe is
applied, and the result, the concentration of total less noisy. Third, the gas probe measures
hydrocarbon in the drilling fluid, is transmitted changes in gas concentration about 2 minutes
to the data logger. earlier than the trap-extracted stream, because of
the greater distance the sample must travel from
the trap to the logger’s unit (in this case, over
Field Test Results 100 m, compared to less than 2 m for the gas
The gas probe was field tested at a gas well in probe).
South Texas. During the week-long test, the
well was being drilled with a 16 ppg hematite- The field test lasted for six days, with consistent
weighted, diesel-based mud. Downhole mud operation of the probe throughout. Lab
2
calibration substantiated its 2% accuracy SPE/AAPG Western Regional Meeting, 19-23
specification. Analysis of mud samples June, 2000.
collected at rigsite confirmed the measurement
capability of the membrane probe, and provided De Pazzis, L. L., Delahaye, T. R., Besson, L. J.,
an independent evaluation of the gas trap and and Lombez, J. P.: “New Gas Logging System
membrane probe data. Improves Gas Shows Analysis and
Interpretation,” SPE 19605 presented at the 64th
In summary, with the field test of the membrane Annual Technology Conference and Exhibition
gas probe, we see the application of a gas sensor of the SPE, October 8-11, 1989.
based on a technology completely different from
what has been the standard in the oil field for Wright, A. C., Hanson, S. W. and DeLaune, P.
many years. This technology is now available L.: “A New Quantitative Technique for Surface
commercially to measure total hydrocarbon gas Gas Measurements,” SPWLA 34th Annual
concentration in drilling mud returns, as part of a Logging Symposium, June 13-16, 1993.
manned mud logging operation or as an
unmanned, stand-alone monitor. We have seen
that the probe performs robustly in an oil-based About the Authors
drilling mud, is relatively insensitive to changes Dennis Dria is a Senior Research Engineer for
in mud density and flow rate, and, with virtually Shell International E&P, Inc., in their
no attention after installation, performs at least as Technology Applications and Research
well as a constantly monitored, well-maintained Directorate. His areas of current research and
gas-trap type gas logging system. technology implementation include mud logging,
production logging, reservoir monitoring and
In addition to the robust performance with little completions design for sand management. He
maintenance, the membrane-type probe offers has also worked for Shell Western E&P, Inc., as
distinct advantages of simplicity and speed (that a petrophysical engineer, and as a research
is, reduced lag time between gas concentration chemist for the Standard Oil Company. He
changes and the time these changes are measured holds a BS in Physics and Mathematics from
and noted on the drill floor). These advantages Ashland University and a Ph.D. in Petroleum
can yield better thin-bed resolution from the gas Engineering from The University of Texas, and
log, more nearly quantitative characterization of is the author or co-author of 14 U. S. Patents.
a gas show, and safer drilling operations because
of quicker response to gas increases. The results David Stewart is a Senior Technical Associate
of the field test have lead to our deployment of for Shell International E&P, Inc., in their
this gas probe onto Shell rigs in the deepwater Technology Applications and Research
Gulf of Mexico, including one on which we have Directorate, where his current areas of research
logged six months of near-continuous operation include gas analysis, mud logging operations,
with one probe. Under development, as a result electronic instrumentation and the experimental
of the success and acceptance of this technology, study of rock physics.. He holds a BS from Sam
is a gas -chromatographic version of the Houston State University. After joining Shell in
membrane-based sensor, which will give 1978, he worked in the areas of polymer physics
hydrocarbon compositional information, with the and polymer rheology. He is the author or co-
advantages of speed, simplicity and robust author of 9 U. S. Patents in the area of polymer
performance seen in the current total-gas version. processing.
3
Division Petrophysical Engineer for Shell measurement using ion-selective electrodes. He
Offshore, Inc., in New Orleans. holds a BS in Applied Petroleum Engineering
Technology from the Southern Alberta Institute
Aurel Brumboiu is an R&D Engineer with of Technology and a MS in Engineering Physics
Datalog Technology Co. in Calgary. His areas from the University of Bucharest. He has
of interest include the separation of previously worked for the Institute for Project
hydrocarbons from drilling fluids using Automation in Cluj-Napoca, Romainia, and the
semipermeable membranes, gas detection using University of Bucharest.
high-speed chromatography, and H2S
Figure 1: Conventional gas trap for extracting gas from drilling mud.
4
Drilling Fluid Flow
2m Extracted
gas carried
to detector
detector/
electronics
Figure 2-a: Membrane gas extractor positioned in the drilling mud return line.
5
5 3
4.5
2.5
4
Gas Wizard
installed next to
3.5 Gas Trap Inlet
2
3 min
3
gas conc (%)
1
1.5
1
0.5
Figure 3: Performance comparison – conventional gas trap and membrane probe. Membrane probe data
are the lighter points which fall on a continuous curve. Gas-trap extracted data are the small dots which
exhibit a bit more noise.