The most frequently asked questions about PNN System

Following are the answers to the most frequently asked questions about PNN
system that will help you better PNN System understanding; from both technical,
and interpretation/application point of view

1. It looks as though you are marketing the tool as outperforming the

regular PNL Capture tools in low salinity and low porosity. The capture cross
section of fresh water is the same as oil so how low a salinity can we go to and
still tell the difference between oil and water? In the past Ive used [(sigma water
sigma hydrocarbon) x porosity] to give a sensitivity.

A1. PNN is thermal neutron decay type of the tool and this is the fact. In other
words this is as you say thermal neutron capture tool or sigma tool like most are
used to say. So you may say that it is similar tool as standard gamma ray type of
pulsed neutron tools (PDK, TDT etc..). However there is a lot of advantages and
lot of differences in approach to measurement and processing of measured data
which, we believe (and so far have lot of proofs and success), makes this tool
much more advanced and possible to use in much lower salinitys of formation
water and in lower porosities. It would be easier to explain all to you live and
showing real examples but we will try to explain this through some explanation
lines and some real examples.

Thermal decay time measurement theory:

Neutron Generator is emitting very high energy neutrons (14.1 MeV).
The path the neutron takes as it scatters, moves it a given distance, sometimes
toward the source; but on the average, it gradually moves away from the source.
With each interaction, the neutron typically loses some of its kinetic energy. This
continues until it has a value just above thermal energy (0.025 eV). The energy
of the thermal neutron is now about equal to the thermal vibration energy of the
formation nuclei. Subsequent collisions will, on the average, maintain the
neutron's energy in equilibrium with the thermal energy of the formation nuclei.
During this thermal energy phase, each neutron can be thought of as a gas
molecule, which undergoes repeated collisions and changes in direction. This
process eventually terminates when the neutron is captured by one of the
formation elements.
The nucleus where neutron is captured becomes excited and it tends to become
ground again. In order to become ground it emits some gamma rays (). So there
are two ways how to record or measure thermal neutron decay function or
thermal neutron decay function. Either to measure thermal neutrons that are still
not captured itself (kind of direct way) or to measure gamma rays. More or less
these two ways are theoretically equivalent and should finally give similar results.
However there are significant advantages using neutron measurement, as it will
be further explained.

1
The biggest difference between PNN and standard PNL tools, which also
results in most of advantages, is that PNN is measuring thermal neutrons that
are still not captured and not gamma rays.
In late 60s and early 70s when big companies like Dresser Atlas and
Schlumberger started with pulsed neutron measurements and systems, they
made a lot of laboratory and field tests in order to obtain relevant information
about what is the better way to measure thermal neutron die-away spectrum. At
that moment they come with solution that better is to use gamma rays for
different reason. One of the reasons was that efficiency of thermal neutron
detectors at that time was very low around 16%, and gamma-rays efficiency was
much better 30-40%. Now-days situation significantly changed and detectors that
are used in PNN tool are highly efficient (>95%) while in mid-time gamma-ray
detectors technology has not improved much.
Typical gamma ray represented thermal-neutron decay and real PNN thermal
neutron decay can be seen at following figures.

Gas

Water Oil
Shale

Left figure shows typical GR represented thermal neutron die-away distribution

where it is clearly shown influence of background gamma-rays that interfere with
thermal neutron die away spectrum. Although this background gamma-ray is
measured and removed from capture gamma rays this removal is newer perfect
and also influenced by statistics and further more part of decay below this
background in count rate and on the right in time can not be used. This part of
decay may sometimes be of essential importance for detecting different
saturation zones especially in cases of small difference between capture-cross
sections of formation waters (low salinity) and hydrocarbon (heavy oil).
Right figure shows, for comparison, PNN measured real example (four different
zones are found in one well) of decays in different lithologies and saturations. It
can clearly be seen that there is no any background and any part of measured
decay can be used for processing and interpretation.
When formation capture cross-section becomes very low (<20 to 25 c.u.) PNN
measurement shows much higher count rate and better statistics than GR type of
tools:
Why? Low capture more neutrons remained for measurement and in same time
less gamma rays from capture. This results in better statistics for PNN in such

2
conditions. This is in most situations with low salinity formations and in tight or
low porosity situations.
From the other side high capture cross section formations high salinity
formation water shale it is opposite, but because of high difference for capture
cross section between water (high salinity) and oil this is not a big issue still PNN
can be compatibly used in this conditions.
Schematics of difference between PNN and GR type of tool decays in different
capture cross-section formation can be seen on following figure:

PNN Decay
TD Log
High Sigma

Medium Sigma

Low Sigma

tim
This implies following schematics for interpretation between different zones:

PNN GR Type Theoretical Formation

Oilheoretical Log
Oil
O
Low salinity
water
Low
Oil
O
High salinity
water
High

3
If sufficient statistics even small differences in water and oil saturated zones can
be interpreted with sufficient confidence. Of course in such cases after
quantitative interpretation these results should be judged qualitatively.

It is not enough only to have statistically satisfying curve but also precise and
resolute curve in order to make good interpretation. Also we have to have curve
that is minimally borehole influenced and maximally formation influenced and
also diffusion influence should be removed. Unique way of PNN processing and
interpretation enables us to make different visualization of processed data and to
remove these influences. One of most important steps is creation, visualization
and examination of sigma matrix files which enables us to see borehole effect
and to remove it and also to see if present diffusion effect and to remove it from
final curve. Example of sigma image with different extracted zones can be seen
on following figure:

1. Late Thermalization Influenced Zone.

2. Borehole Influence Zone.
3. Formation Reponse Zone.
4. Statistical Influenced Zone.

4
11 2 3 4
2 3 4
Borehole influenced zone 2 can be recognized-seen and therefore removed.
Sigma should be calculated after this zone and as far as better higher formation
response but if we go too far then statistical influence can be seen. This borehole
influence can be different and in some cases it can be lower sigma then
formation and differently expressed and in this case it is under higher possibility
of diffusion effect influence. Also it can be different in different zones of logged
interval. There is no automatic way of recognizing and removal of borehole
influence. Example of different borehole conditions in one interval and process of
removal of borehole influence can be seen on following figure.

There is also another method of improving statistics of PNN measurement and

this is performing of stop-check measurement at selected points (in suspected
contact zone). PNN tool can stay in one point and measure thermal neutron
decay as long as necessary (usually 5 minutes) and to get capture cross section
of formation which is virtually with zero statistical influence (following figure). In
most difficult cases it is essential. This kind of measurement is impossible with
GR type of tools because of activation.

All of this previously described enables us to get correct, precise and resolute
formation sigma curve which enables us to make interpretation of saturation in
much lower formation water salinitys and lower porosities then it is possible with
standard GR types of tools. And so far we have very good results from very
tough conditions in Maracaibo lake Venezuela where salinities are as low as 3-
5000 ppm.

5
To come back to the fresh waters and oil small difference question. When you
say fresh water what do you mean (0 ppm). Although there are some rare
examples in world where formation water is several hundred ppm in oil reservoirs
in most conditions we consider low salinity reservoirs for under 15000 ppm and
very low salinity but still not fresh as in Venezuela Maracaibo lake 3-5000 ppm.
From theoretical point of view capture cross section of water can be from 22
138 capture units depending on salinity. 22 would be for distilled water (0 ppm)
and 138 for oversaturated waters.
For hydrocarbons sigma can be:
5-16 cu for gas depends of composition and pressure and temperature.
16-22 cu for oil depends mostly about GOR, composition, density of oil.

So if you have situation of reservoir that has 0 ppm formation water and
extremely heavy oil (bitumen or asphalt) than we would say that you have
situation 22:22 and there is no winner and no way to say how much water and
how much oil in reservoir. Also when this is not so but little bit in both directions
little bit higher salinity and little bit lower oil density situation still is not brilliant.
But in most cases that we encountered so far oil capture cross section goes from
18-20 and water capture cross section in most difficult cases 24-25. In this kind
of situations and in moderate porosities 15-20% we can say that we can get good
results with PNN tool and we have good references.

2. Most of the wells we would be running this log on would be porosities of

30% or greater, salinities of 50,000 ppm or greater (usually greater) and a lot of
gas wells. How well does the tool work in this environment? Also in many cases
we are running near salt domes in salt saturated formations. How does it work in
this environment?

A2. To continue from previous answer all other conditions all higher salinitys
and porosities are just easier to solve and results with more confidence can be
provided. The answer is: yes in this kind of conditions PNN can be used with
very good success. Also because this is neutron detectors tool the ratio and
separation between long and short spacing detectors count rates makes this tool
also very good hydrogen index indicator and therefore low porosity or gas zones
can be very easily recognized and with much higher confidence then with GR
types of tools. In order to support this we have a lot of examples but one well is
Oil Gas Water Gas Separation Shale

Sand

0 GR OH [GR] 100 100 Sw[SWPNN] 0 50 Porosity [POR] 0 0 Lithology [VSH] 100

0 GR OH [GRPNN] 100 100 Sw OH [SW] 0 50 Porosity Wate [PORWPNN] 0 100 Porosity [PORT] 0

50 Sigma [SG] 0 6 LSN Pnn [LSN] 156

0 SwGas [SWPNNG] 100 0 SSN Pnn [SSN] 1000

2000

6
presented for you that is measured together with TDT (customer wanted to
compare results) in one Egyptian Red Sea well.
In this example gas oil contact can be seen at 2007 m but small sand interlayer
1999-2003 is identified and interpreted as gas saturated from us by PNN. This
was not case with TDT interpretation. The customer believed Schlumberger and
perforated and of course, got gas what didnt want.

3. I looked at the tool timing. Why dont you increase the tool pulse
frequency? As best as I can tell you pulse it for 3 microseconds and count for 1.8
msec. This leaves 48.2 msec of dead time. If you could pulse it quicker you could
increase the logging speed or are you recording the backround for some reason?

A3. This is a low frequency but high neutron output neutron generator.
Physically it is not possible to fire this generator higher then 30 Hz. We have tried
with different frequencies and we came to conclusion that 15 20 Hz is the best
enabling sufficient statistics for measurement and saving generator lifetime. Even
in easier cases gas and higher porosities tool can be run up to 3 m/min (10
ft/min). But also because of no special calibrations before and after logging spent
rig-time is much less then with other similar tools. Also measurement in shut-in or
flowing conditions no matter which kind of fluid in borehole additionally saves
workover operations. On the following figures you can see repeatability of
thermal neutron decay measured in flowing and shut-in conditions, which means
that count rate is sufficient. Also, one example of main and repeated logs is
given.

0 CCL [CCL]
10000 0 GR PNN [GRPNN] 250 50 Sigma [SG2329] 0 77 LSN Pnn [LSN] 277 0 Lithology [VSH] 100

0 GR PNN_Rep 250 50 Sigma_Rep [SG2329_C] 0 200 SSN Pnn [SSN] 1200 100 Porosity [PORT] 0

16 LSN Pnn _R 216 0 [CSND] 100

-100 SSN Pnn_R 900 0 [CLIM] 100

2625

2650

7
 4. Do you have a way of telling gas from low porosity tight streaks on the
log?

A4. Low porosity or better say zero porosity tight formations are recognized
from PNN according to low hydrogen index (HI) measured through ratio between
long and short spacing detector count rates (total or partial). Gas saturation
expresses lack of hydrogen also as tight so there is no sure way how to say
solely from PNN weather something is gas saturated or tight zone. There are
some differences although they are small and require high experience and also
knowledge about local situations to differentiate. However by PNN and consulting
other openhole logs (basic or expanded set) we have procedure and we are able
to differentiate in any well tight from gas saturated zone in more then 95% cases
(having extended set of openhole in 100%) of cases.

5. Is the reason for this tool being advertised as better in low salinity/ low
porosity simply better statistics?

A5. Processed formation sigma curves, without any borehole and diffusion
effect, also very resolute and with minimum and if necessary zero statistical
influence enables differentiation between oil and low salinity formation waters.
On the following figure is one example where there is 4000 ppm formation water
differentiating from Oil and even on sight without calculation you can see
difference between water and oil.

There on the log can be noticed the differences on the sigma curve between
originally water saturated zone, depleted zone and zone where is still good oil
saturation. We have number of wells measured in Venezuela with 3000-5000

8
ppm with good results and most of wells measured with PNN are of salinities of
up to 20000 ppm.

6. How are you calculating shaliness? Gamma Ray?

A6. Shaliness can be calculated and this is most common from Gamma Ray
curve from PNN. PNN tool has a GR section away from generator in order to
measure natural radioactivity for correlation purpose but can be also used for
shaliness calculation in the case there is no significant amount of precipitated
radioactive minerals in producing zones. In this case the other curves should be
used for shaliness calculation. Sigma can be calculated from PNN, but it is
influenced by saturation. There is special method of different parameters sigma
curves which gives processed curves that overlay in shale zones. Also Ratio can
be used for shaliness calculation.
The best way to calculate shaliness in the practice is from openhole curves (GR
if suitable or other) if available. Theoretically it would be best to use customers
petrophysical interpretation for PNN interpretation as another input. When this is
not available openhole curves should be supplied and we can make
petrophysical interpretation and use this as input to PNN. If nothing is available
we are forced to calculate everything solely from PNN and in past we have done
it with great success and on satisfaction of customer, but the best is to have
independent petrophysical interpretation at least shaliness and porosity.

7. How good is the porosity measurement? Is it a calibrated porosity and

is it necessary to calibrate the tool at the shop in a fixture for either porosity or
sigma?

A7. Like it is said in answer on question 2, PNN measurement with two

differently spaced neutron detectors makes this tool among the other excellent
hydrogen index (HI) indicator and therefore excellent porosity tool (much better
then from GR type of tool). Unfortunately this tool is not calibrated (we didnt
have access to any calibration utility and there was not need in past because we
rarely use tool for porosity itself) but calibration for some field can be made on
similar well where there is openhole compensated neutron measurement and
from this correlation on all other wells in the same field it can be used. The other
way is to go to calibrate tool in some calibration facility and to make etalons for
calibration of all tools in laboratory.
Of course neutron porosity is influenced with gas saturation and this should be
considered. On the following figure is comparison and correlation of PNN Ratio
curve with openhole compensated neutron porosity curve. Very good correlation
is noticed and it proves that PNN measurement of HI is excellent porosity
indicator. On particular field some correlation can be made with other porosity
curves and this can be used as porosity.
Also below (on right picture) is one example of correlation of CN and PNN Ratio
at openhole well.

9
 6 Caliper 16 60 Neutron Porosit [CN] 0

0 GR OH [GR] 100 26 PNN Ratio [RAT1736] 3

20 GR PNN [GRPNN] 110

10150

10200

10250

10300

10350

10400

10450

8. Do you have any problems with communications from the tool to

surface in plastic coated pipe or are there any other situations such as nickel
cable where the communications can be affected?

A8. Plastic pipe is not influencing communication but may influence CCL
curve. Considering Nickel cable so far we did not have any chance to try but
according to our knowledge it should be no problem. The communication can be
easily adjusted by logging engineer to the particular logging line. The longest
logging line the communication was successfully tested was 42000 ft long 3/16
diameter single conductor line (the two drums with 21000 ft line connected in
serial).

9. How does PNN sigma compare to other Sigmas from capture tools? If
possible send me some examples of PNN Sigma versus Conventional Sigma
measurements in the type of environments that we will be running in. In many
cases we will be running a second or third log behind Schlumberger or Atlas to
evaluate movement over time. So comparing Apples to Apples can be
important.

A9. Processing of PNN can be set to perfectly match and compare Sigma
curve processed from other TDT or PDK or any other sigma measured

10
previously. In several cases we had measured after TDT and curves were very
comparable and we assure you that PNN can be used to continue time lapse

technique after any other pulsed neutron measurement. In the figures above is
one available log we have to compare. In this log TDT is measured 3 years after
openhole and PNN 15 years after TDT. Notice Shale part above reservoir and
matching of Sigma curves, and change of reservoir saturation in perforated zone.
The picture on the right shows the interpretation with the same parameters.

10. Is the real time Sigma accurate or is post processing required for a
usable sigma? In many instances the engineers just want a real time Sigma and
need to make a decision rather quickly on what they want to shoot or if the zone
is any good at all? Time is of the essence.

A10. Currently we are operating PNN with two surface equipments. One is with
small surface panel, and another is with 32 bit WARRIOR STIP panel.
With small panel, and with software for this, you will see that there is no
automatic way to predict what will be the borehole effect as it is influenced by
several parameters; fluid in borehole, casing, cement, tubing, fluid between
tubing and casing Therefore in most of the cases final Sigma will be after
processing. However in one field in similar conditions parameters for this field
can be established in a manner that after processing sigma curve and field

11
Sigma curve are very similar and, if we guess parameters well, maybe at the
same.
With 32 bit WARRIOR STIP panel, situation is little bit different. Using high
processing capabilities of WARRIOR hardware and software, during real time
logging we have enough time to apply some algorithms which improves real time
Sigma calculation and in most cases this Sigma is almost the same as Sigma
after processing. Explanation for this algorithm you will find in attachments in
Appendix 2.
Anyway, in cases of site emergency to get results in field an interpreter can be at
well site and make immediate processing and interpretation. In one hour good
information can be supplied. Also, in processed sigma curves mainly absolute
value is changed between field and processed sigma curve but changes of
amplitudes should be similar.
We always force after-processing because, although in most cases will be made
not big mistakes, they are possible and repercussions for mistakes are more
costly than one day rig-time.
Important is that the parameters can be established for good real time
sigma processing.

11. How do you quantify hold up?

A11. It is strongly advised to judge borehole sigma and count rates only on
qualitative basis. But if you want to do so only using same well data you should
calibrate for 100% water or any other fluid and you should know holdups above
all intervals than you may quantify.

12. What kind of velocities do you need to calculate water flow?

A12. If you suspect on behind casing water flow than the water presence
between the casing and the formation can be confirmed by activation logging
down. Also by stop check measurement you can calculate velocity of possibly
moving fluid, but for quantities you should know cross section. This is mostly
effective for detection of flow and in the present setup of PNN just for flows from
down-up.
Additionally, with WARRIOR 32 bit system, we have specially designed software
module for flow estimation. Acquisition logging screen is shown on figure xx-1.
Main characteristic are:
1. Window 1 is logging, up to 1000 sec of data might be observed, and
stored;
2. Window 2 is stationary, for fine observing time GR counts.
3. Adjustable activation time,
4. Automatic time from generator to GR detector calculation.
5. Can be applied stationary in up-down flows, with cable speed correction
also and dynamically.

12
 Figure A12-1: Logging screen for PNN flow estimation

13. Can you place a Gamma Ray Section below the source section to
detect down flow of water?

A13. At the present setup and tool construction it is not possible. We had some
ideas to set this but so far we did not have economical justification to set tool this
way.

13
 14. Did you ever think of making a gamma ray section with 2 detectors
that would detect the gamma rays of capture and give a conventional pulsed
neutron response? You could even place a longer spaced third detector on the
tool for detection of gas. In this way you could use the tool however you wanted
to use it. Lower salinities use the NN section; higher salinities use the GG
section. There are even some clients that would run both tools on the same well
to see what each tool would tell them. It seems to me the tools are made and all
you would need is an added detector section. For this kind of tool I would even
be willing to share some of the development costs for exclusive use.

A14. Any tool that is of interest for customer may be interesting to be built and
we would be ready to work on it if really there is economical justification. Anyway
as we discussed with our colleges technically it seems that it is easy just instead
of neutron to put GR detectors but as we understand it and from our point of view
there are some significant difficulties. Definitely it will be time to discuss about
this with our designing people during our visit when our possible projects start
more seriously. But we need to say that from our point of view (interpreting) PNN
is completely compatible tool with GR type of tools for working in higher salinitys
and we already have had measurements in oversaturated fluid in borehole
together with very saline formation water and with good results. And this is really
the worst case for PNN tool because in very saline fluid in borehole count rate is
significantly decreased because of thermal neutron capture before they come to
detector.

Log examples of, decays and Images in high salinity formations and salts follows:

Salt

Normal
Formation

14
Salt layers and domes will be easily recognized and measured with very high
Sigma values while zones and reservoirs without salt may be evaluated standard
way. Only problem will be if there is precipitation of salt inside pore space of
reservoir rock. In this case interpretation is questionable

15. Are there any published technical papers on the PNN?

A15. Unfortunately we do not have so far published papers. But we have

papers presented on different petroleum or geological Congresses in different
countries. As well as presentations made and organized by petroleum engineers
associations.

16. What do you advertise as the depth of investigation as well as the

vertical resolution of the tool?

A16. Depth of investigation is hard to calculate but in our mostly experience and
testing depth of investigation is between 8 and 12 depending also about
different conditions (borehole, formation). Vertical resolution would be 1.5. But
for intervals as thick as 3 we are sure we can get good results.

17. Salinity:
What Salinity Range influences PNN result? How big is the influence? For
example: from ...to...is the best

A17. Salinity of formation water is one of main factors or parameters in

interpretation of water saturation from PNN measurement.
From theoretical point of view capture cross section of water can be from 22
138 capture units depending on salinity. 22 would be for distilled water (0 ppm)
and 138 for oversaturated waters.
5-16 for gas depending of composition and pressure and temperature.
16-22 for oil depending mostly about GOR, composition, density of oil.

If there is a reservoir which has 0 ppm formation water and extremely heavy
oil(bitumen or asphalt) than we would say that you have situation 22:22 and
there is no winner and no way to say how much water and how much oil in
reservoir. Also when this is not so but little bit in both directions little bit higher
salinity and little bit lower oil density situation still is not brilliant.
But in most cases that we encountered, oil capture cross section goes from 18-
20 and water capture cross section in most difficult cases 24-25. In this kind of
situations and in moderate porosities 15-20% we can say that we can get good
results with PNN tool and we have good references.
All of this previously described (questions 1 to 15) enables us to get correct,
precise and resolute formation sigma curve which enables us to make
interpretation of saturation in much lower formation water salinity and lower
porosity then it is possible with standard GR types of tools. Although there are
some rare examples in world where formation water is several hundred ppm in oil
reservoirs, normallt we consider low salinity reservoirs for under 15000 ppm and

15
very low salinity but still not fresh as in Venezuela Maracaibo lake 3-5000 ppm.
So far we have very good results from very tough conditions in Maracaibo lake in
Venezuela where salinities are as low as 3-5000 ppm.

From the other side when we talk about high salinities situation is better because
difference between water and oil is bigger and it is easier to make differentiation.
However PNN tool because of the fact that is measuring neutrons has higher
statistical influence but still satisfying.

Some examples of PNN log measured in different salinity environments are

shown below for better answer to your question:

16
Oil Sat. Oil Displ. Water Sat. Shale Sandstone

Gas Sep. Gas Sat. Casing Perforation

0 CCL10000 0 SP 100 100 Sw PNN [SWPNN] 0 50 Porosity [POR] 0 0 Lithology [VSH] 100

0 GR OH [GR] 150 100 Sw OH [SW] 0 50 PorWaterPNN [PORWPNN] 0 100 Porosity [PORT] 0

0 GRPNN 150 50 Sigma [SIGMA1] 0 72 LSN Pnn [LSN] 472

0 So PNN [SO] 100 300 SSN Pnn [SSN] 2300

50 PorWaterOH [PORW] 0

50 PorOil [POROPNN] 0

5350

5400

5450

5500

5550

5600

Figure 17.a.

PNN interpretation example in low formation water salinity (2-4000) ppm.

17
 9100

9150

9200

9250

9300

9350

9400

9450

9500

9550

9600

9650

9700

9750

9800

9850

9900

9950

10000

10050

10100

10150

10200

10250

10300

10350

10400

10450

10500

10550

10600

10650

10700

Figure 17.b.

PNN interpretation example in high formation water salinity (90000-100000) ppm.

18
 18. Porosity:
Water Sat. Water Sat.

Oil Saturation Gas Saturation

How big difference from PNN porosity result to OH
0 CCL10000 60 TempOuter 75 porosity result?
10 PERF 0 0 GR_OH 100

A18. Details about PNN Porosity measurement can

be found in Answer on question 7.

Differences in our, PNN or openhole, and customers

openhole porosity interpretation may come only
from different approach to porosity interpretation
and of course using different parameters for matrix or similar.
Even using same matrix parameters there could be differences in porosities in
case when different approach to the influence of dispersed or laminar shales
inside sandstones on porosity is used.
This is the only reason why porosities on this well differ between our calculation
and customers calculation.
With normal and same approach there should be no differences.
Hydrocarbon Sat. Shale Sandstone Gas Separation Displaced Hydrocarbons

Water Sat. Perfs.

0 CCL 5000 0 GRPNN 150 100 Sw PNN Std [SWPNN] 0 60 Porosity [POR] 0 0 Lithology [VSH] 100

10
Perf. [PERF] 0 20 SP [SP] 50 60 Sigma [SIGMA] 0 60 PorWaterPNN [PORWPNN] 0 100 Porosity [PORT] 0

100 100 SWOpen

Temp
Hole [SW] 0 6 LSN Pnn [LSN] 120
206

0 SSN Pnn [SSN] 1000

60 Porosity [POR_ET] 0

2370

2380

2390

19
As it can be noticed in the figure above; in clean zones there are no significant
differences in porosities and only high differences are in shally zones. Openhole
and customers porosity interpretation is green porosity curve in porosity column
and our interpreted porosity is red curve.

These differences are even higher in thin zones like in following example.

Hydrocarbon Sat. Shale Sandstone Gas Separation Displaced Hydrocarbons

Water Sat. Perfs.

0 CCL 5000 0 GRPNN 150 100 Sw PNN Std [SWPNN] 0 60 Porosity [POR] 0 0 Lithology [VSH] 100

10Perf. [PERF] 0 20 SP [SP] 50 60 Sigma [SIGMA] 0 60 PorWaterPNN [PORWPNN] 0 100 Porosity [PORT] 0

100 Shale Influence 100 SWOpen

Temp
Hole [SW] 0 6 LSN Pnn [LSN] 120
206

0 SSN Pnn [SSN] 1000

60 Porosity [POR_ET] 0

2460

Notice similar porosity in clean and in shally zones difference. Porosity input for
quantitative PNN interpretation should be effective porosity. This should not be
porosity influenced by shaliness.

19. If the Salinity range isn't suitable for PNN, is there any way to deduct
the influence by using some parameters during analysis?

A19. Salinity, if very low and combined with very low porosity, should be carefully
interpreted and good interpreted input parameters like porosity, shaliness and

20
matrix minerals volumes used. Small incorrectness of these parameters may
influence final water saturation interpretation. Local experience and development
of different interpretation modules may help getting much more correct
parameters.

20). How to distinguish formation matrix for a well had 3 times production
recovery using Polymer?

A20. Matrix should not be influenced by polymer injection and recovery. What is
influenced is pore space. Or in different words; part of pore space that is filled
with polymers should be considered not any more as pore space but as loss of
pore space and finally it can be considered as matrix that means pore space is
reduced. Influence of different polymers depends of polymer chemistry and this
should be known. Also some modeling of polymer distribution may help in the
parameters selecting. Anyway, we did not have too many measurements in
zones with extensive polymer usage in production. The polymers final influence
may be different from case to case. Any local experience in such cases is of vital
importance for good interpretation.

21). What is the procedure for Sigma processing, correction and

calibration ensuring correct and repeatable results

A21. In answer on question one there can be found a lot about PNN system
description and some small issues about processing of data.

Capture Cross Section

The rate of thermal neutron decay in the formation can be measured and used to
differentiate between hydrocarbons in the pore space and water especially salt in
water. A mineral's ability to absorb thermal neutrons is defined as its capture
cross section, sigma.
Sigma is measured in capture units (c.u.). The higher the value of sigma, the
greater is its ability to capture thermal neutrons. By measuring the rate at which
thermal neutrons are absorbed into the formation (either by measuring the
thermal neutrons themselves or by measuring the prompt gamma rays given off
during thermal neutron absorption), the capture cross section of the formation
and formation fluids can be predicted.
The capture cross-section tables list many of the elements and compounds found
throughout the world. Formations, and even more importantly, formation fluids
containing chlorine atoms are the most effective capturers of thermal neutrons.
This is because high thermal neutron absorbers such as cadmium are rarely
found in oilfield geology (Boron is more commonly encountered, but is still just a
drop in the bucket when compared to salt water). From the table, it can be seen
that rock salt has a high sigma value due to its chlorine content. Salt-water
values may range from 25,000 ppm (30 c.u.) to 275,000 ppm (130 c.u.)
equivalent chlorides.
Sigma Processing from PNN measurement Theory
After correct log is acquired, processing of the measurement is the next step. By
using different processing programs different parameters can be calculated:

21
correct tau (), sigma (), and ratio (r) for different channels. With correct values
of the parameters listed a qualitative and quantitative interpretation according to
measurement requests can be preceded.
Processing principles.
As it was mentioned before, PNN system measures die-away population of
thermal neutrons on two detectors. Population of neutrons is measured in 60
time channels by both detectors. Every channel is 36s wide. Measuring of the
thermal neutrons itself towards measuring gamma rays that are product of
thermal capture, has some advantages as it is explained in Answer 1.

7.1.1 Exponential decay.

The formation is subjected to a pulse of high energy neutrons (14 MeV) from
neutron generator. This pulse is repeated at fast rate. The thermal neutron
population is sampled in the two detectors with 60 time channels each. The
whole measurement with the time distribution over all 60 channels is recorded
and available for processing at any time after the logging job is finished. This is
one of the PNN system advantages, because it might be that some additional
information about formation and distribution of the fluids inside reservoir rock is
missed with real time processing and data removal, and later will be available in
reprocessing.
Emitted neutrons are rapidly slowed down to thermal state by collisions with the
nuclei of elements present in the formation. They are eventually captured by
nuclei at the rate that depends on the product v * abs , where v represents
thermal neutron velocity (constant at a given temperature), and abs is
macroscopic capture cross section of the formation at that point. If neutron
capture is the only phenomenon occurring, the number of neutrons decays
exponentially.

Gas

Water Oil

Shale

Figure 21-1. Decay rate difference between shale and oil, water or gas saturated
reservoir rock.

Figure 21-1 demonstrates how the decay rate increases when oil is replaced by
water in the pores of a rock, because water has a higher than oil. The number
of neutrons (count rate) at time t1 is given by:

22
 N1 = N0e(-v*abs*t1) (eq. 1.)

where
N1 - number of thermal neutrons remaining per unit volume at time t1;
N0 - number of thermal neutrons per unit volume at an arbitrary time
t=0.
t1 - elapsed time since an arbitrary time zero.
abs - sum of the atomic cross sections of all the atomic nuclei in unit
volume of formation for an average neutron velocity of v=2200 m/s (at
75oF) It is measured in cm2/cm3 (or cm-1).

If we repeat measurement at time t2 we will obtain:

N2 = N0e (-v* abs*t2) (eq. 2.)

Where:
N2 - number of thermal neutrons remaining per unit volume at time t1;
T2 - elapsed time since an arbitrary time zero.

The decrease of the thermal neutron population density can be quantified by a

comparison of the two measurements:

N2=N1e-v abs(t2-t1) (eq. 3.)

From which it can be derived as follows:

abs=1/v*(t2-t1)*ln(N1/N2) (eq. 4.)

Using logarithms with base 10 and v=2200 m/s, t in s and in cm-1 abs is:

abs=10.5/ t * log10(N1/N2) (eq. 5.)

Since the decay is exponential, it can be represented it in another way by

introducing an intrinsic decay time, int, corresponding to the time necessary for
the initial number of neutrons N0 per cm of formation to decay by 63%, or to 37 %
of it's original number (which is 1/e):

int
Nt=N0e-t/ (eq. 6.)

Where Nt is number of neutrons per cm3 at time t. Using both, eq. 1 and eq. 6 it
comes out:

int=1/v*abs (eq. 7.)

23
The decay-time is almost not temperature dependable. It is also known as the
neutron "die-away" time, or "life time". With t in s, and v=0.22 cm/ s, equation 6
becomes:

int=4.55/ abs [cm2/cm3] (eq. 8.)

is more conveniently scaled in units of 10-3cm-1 [capture units c.u. in sigma units
s.u.]. Equation 8 becomes:

int=4550/abs [c.u.] (eq. 9.)

This consideration is taken into account with programs for as most as possible
accurate calculation of die-away time () and bulk capture cross-section ().

As mentioned previously PNN measures on both detectors density of thermal

neutron cloud in 60 time channels as shown on following picture:

Figure 21-1 PNN tool timing sequence.

There are 60 time channels and each channel is 36 s wide. During acquisition
all raw data acquired by sensors are stored without any filtering and removal of
the data. This ensures not to be able to skip any potentially good data which may
help later during processing.
Logging Linear Scale Processing Logarithmic scale

Counts
Counts

Time
Time
Figure 31-3 and 4 PNN measured decay during logging and during processing.

24
Figures 21-3 and 21-4 are showing PNN measured thermal neutron decay during
logging in linear scale and later on in processing phase on natural logarithmic
scale. This decay is analyzed and from it is processed Sigma curve. All
calculations are made according to above mentioned mathematical and physical
considerations and formulas. Important is to make selection of parameters for
processing in a way to remove all effects connected with:
- borehole influence,
- diffusion of thermal neutrons and
- statistical influences
and to make processing in zone with maximal formation influence to get
formation capture cross-section curves. In real conditions may exist number of
different borehole zones with highly different borehole influences, but most of
them may be divided in three main influences with all variations between them.
1.) Borehole Sigma much higher than formation Sigma as on
following figure.

Figure 41-5 Borehole Sigma higher then formation.

2.) Borehole Sigma same or similar as formation Sigma as on

following figure.

Figure 51-6 Borehole Sigma similar as formation.

25
 3.) Borehole Sigma lower than formation Sigma as on following
figure.

Figure 61-6 Borehole Sigma lower then formation Sigma.

First case is most common due to casing and other different equipment in
borehole. Sometimes like in shown figures it is easy to make processing because
it is very easy to recognize this. After recognizing that borehole effect it is easy in
processing software to make definition of parameters and to get formation Sigma
without borehole influence.
Second case is least problematic because it is not so sensitive to parameters
selection.
Third case is probably most difficult because is no result only of borehole fluid but
significant neutron diffusion is present as well, and then with starting channel it is
necessary to go much further in time. This can also be recognized and
parameters selected to get formation influenced sigma.
After this we may select one of four processing modes to make Sigma
calculation. There are more different modes because some of them differently
react on selection, but consequently all of them should give same result if similar
and good parameters for removal of unnecessary influences are selected.
All of these effects may be present in one well with different degrees and PNN
processing software enables to have full control over the data and if sometimes
necessary every depth level may be reviewed and examined to enable selection
of parameters. To look at any level decay may be confusing that we developed
part of software processing where we can review overall depth-time image of
different parameters like: count-rate (short or long), ratio between short and long
count rates, Sigma images and also water saturation images.
In order to see these images it is necessary first of all to create so called matrix-
files. Explanation of these matrix files can be found in Appendix A.
By reviewing these images it is possible to see much better picture about well
and it is easier to recognize different zones for better parameter selection.

26
In following figures one sample Sigma Image will be reviewed and selection of
parameters will be changed in order to see differences in good and bad
selection.

It is easy to recognize several different zones with completely different borehole

conditions at this example. At least 6-7 different borehole condition zones are
selected. At this step Sigma is calculated simply from fixed gates through whole
zone. This gives good processing parameters for most bottom zone at this well

27
while in all other zones parameters are not good and Sigma with too high
borehole influence is processed. In this case it is necessary to make different
zones processing and patching final calculated curves. There is better way to do
this and this is to use our Sigma auto-processing module which does lot of job
automatically and calculates best possible and not borehole or diffusion
influenced Sigma curves. However even this is not good in more complex
situations involving multiple completions and presence of different equipment and
different fluids in borehole like in following situation.

These processing parameters ma pass in this bottom zone as well although with
higher statistics in higher Sigma zones.

28
In complete other part of the well, parameters are wrong and too high Sigma
processed is calculated. That is why even this auto-processing module is
possible to strongly control by setting rules suitable for certain situations: like
high salinity fluid in borehole or in space between tubing and casing, or in high
hardware presence zone like packers or similar, or gas in borehole. All of these
cases can be solved excellent and without problems like in following parameters
selection example:

29
Now it is possible to be confident in good processing results. Any existing auto-
processing routine for Sigma processing will not give good results no meter what
we do, if we can not see effect of selected parameter set on real Sigma Image
where borehole influence is visually seen. Hotwells PNN Processing package
enables this kind of visualization and ensures that in all possible cases formation
Sigma may be processed without borehole and diffusion influences.
This is the reason that there is no need for calibrations on well-site. What is
necessary is to have quality control during tool production and to ensure good
timing for time channels. This is ensured with our ISO procedure for production of
equipment. Further more all tools are tested after production in standard
conditions in our standard well where all responses of formation are known. This
is checking procedure for every tools timing sequence.
Second thing which is necessary to be done during PNN tool is at usage is to
ensure good count-rate which will ensure repeatability of measurements without
too high statistical influence. That is why tool needs to be checked for generator
output and count-rate in some standard conditions. This needs to be done in
simple water tank, every 1-3 months depending of tool activity.
This procedure needs to be followed in order to ensure good measurement and
processing as a predisposition for good quantitative saturation interpretation.

22). can you easily discriminate tool response in the borehole and
reservoir response. Reservoir response from pulsed neutron type TDT is often
influenced by reading in the borehole. For these tools, to have the best survey,
recommendation is to have borehole filled with high salted water in order to have
a quick capture of all the neutrons located in the borehole. Does that mean that
the best scenario for a good acquisition with PNN is to have the borehole full of
air? (Question by: Dominique ANDRE, Reservoir Engineer, Total)

A22. As stated in the question this fact is known to us as well that TDT in most
cases needs to have borehole filled with high salinity water in order to have good
measurement. If you carefully read the answer to question 21 you will find how
PNN processing is dealing with different borehole conditions. Here it can only be
stated that PNN can be performed in any fluid in borehole. Because of count rate
the most unfavorable conditions for PNN measurement would be borehole filled
with oversaturated fluid in borehole. However PNN is equally usable even in
these kind of conditions as it was proved in number of cases worldwide. PNN will
have best count rate in gas or air filled wells but it does not automatically means
that these are best conditions for measurement. That is why we recommend
always to customer to leave fluid as it is in well during measurement. PNN
processing technique enables recognition and removal of all borehole influences
weather in borehole is water, oil or gas. Similarly high influence of neutron
diffusion can be recognized and removed.

If you have any other question, which can help better understanding and making
decision dont hesitate to contact us:

By e-mail:

30
z.markovic@hotwell.at Zoran Markovic
m.novak@hotwell.at Mijo Novak
k.knapp@hotwell.at Kresimir Knapp
analysis@hotwell.at
office@hotwell.at

Call:
tel: +43 2687 48058

Send a fax:
fax: +43 2687 48059

Or write to:
HOTWELL
Oedenburger str. 6
7013 Klingenbach
Austria

Yours truly,

HOTWELL PNN team

Appendix A
Explanation of Matrix Files
Matrix Files Explanation

Matrix files are binary files that represent three-dimensional matrixes.

One dimension of such matrix file is depth increments from top of some
interval to bottom of some interval. Typical depth increments are 0.1 m for
metric units and 0.5 ft for English units.
Second dimension of such matrix file is time represented through 60 time
channels at each depth level. Each time channel is 30 s wide which
means that time sequence of 1800 s or 1.8 ms is covered.
Third dimension actually is value placed at particular depth into particular
time channel. This third dimension can actually be next values: neutron
count rate in long or short spaced detector, ratio value between short and
long spaced detector, capture cross section ( ) value or calculated water
saturation value. Particular matrix files will be in more details explained in
following chapters. Graphically presented matrix file would look like on
following listing:
Depth Ch1 Ch2 Ch3 Ch4 Ch5 Ch6 Ch7 ........... Ch56 Ch57 Ch58 Ch59 Ch60
2284.5 39 34 63 69 66 66 50 ........... 0 0 0 0 0
2284.4 35 31 51 69 67 58 58 ........... 0 0 0 0 0
2284.3 36 36 51 55 58 53 45 ........... 0 0 0 0 0
2284.2 47 48 69 77 76 75 80 ........... 0 0 0 0 0
2284.1 27 28 52 52 47 51 49 ........... 0 0 0 0 0
2284.0 35 37 67 67 66 58 57 ........... 0 0 0 0 0
2283.9 32 40 62 59 69 59 50 ........... 0 0 0 0 0
2283.8 32 48 51 55 63 60 55 ........... 0 0 0 0 0
.
.

31
.
.
2221.2 37 44 52 51 19 10 9 ........... 0 0 0 0 0
2221.1 43 56 58 57 39 9 11 ........... 0 1 1 1 0
2221.0 44 50 52 53 28 16 9 ........... 1 1 0 1 0
2220.9 46 58 70 46 37 16 5 ........... 1 0 0 0 0
2220.8 39 43 52 41 28 15 8 ........... 0 1 0 0 0
2220.7 53 47 62 52 43 23 9 ........... 1 0 0 1 0
2220.6 46 48 59 49 33 25 14 ........... 0 1 1 0 0
2220.5 34 45 46 42 28 15 7 ........... 1 0 0 0 0
2220.4 42 49 56 42 34 17 7 ........... 2 1 0 0 0

In HWProc processing modules there are several ways how to view or

present and how to examine and to explore matrix files extracting different
curves.
There are several types of matrix files: like Count-rate matrix files(short or
long spaced); Ratio matrix files as representation of porosity, Sigma matrix
files and water saturation matrix files and all of them may be examined
and reviewed as images.
There are also some special analysis types of matrix files connected to
Fourie analysis of Thermal Neutron Decay function or time spectrum.
Some of matrix file images are shown on following figures:

32
Count Rate Images (Short and long spaced detector). On these images curves
from certain channels may be extracted and used for log presentation and some
differences between Sands or shales or different lithologies may be seen.
Sometimes it is possible to see gas saturated zones with high count rate on long
spaced detector.

33
 1. Late Thermalization Influenced
Zone
2. Borehole Influenced Zone
3. Formation Response Zone
4. Statistical Influenced Zone

1 2 3 4
On Sigma Images 4 zones may be extracted. Dark colors are higher Sigma
values and yellow lower. Borehole influences may be recognized and removed
using examination of Sigma matrix files.

34
Ratio Matrix files are interesting because different ratio curves may be extracted
these curves are measure of Hydrogen Index and hence good porosity indicator.
Especially good porosity indicator is curve extracted from zone of maximum
thermalization because this is closest possible to be porosity at border of thermal
and epithermal neutron porosity measurements. Gas saturated zones may easy
be recognized at Ratio Images. These are the lightest zones on shown Image.

35
Water Saturation Images are good because straight from them Water saturation
curves can be extracted with recognized borehole influenced zones.

Other type of matrix files are used for special processing involved in some cases
for extraction of gas saturated zones or discrimination of tight zones from gas
saturated zones. These are images got with Fourie analysis of PNN thermal
neutron decay curves.

36
Appendix B

Figures AB-1 and AB-2 are showing PNN measured thermal neutron decay
during logging on short (left) and long (right) detectors, in linear scale, with
Warrior software.

Figure AB-1: Sigma Calculation from Figure AB-2: Automatic detection Sigma
fixed Channel. calculation.

For any type of formation, for any fluid in well or formation, there is on common
characteristic point maximum in the first channels which is related with diffusion
effects. After this maximum, decay is starting. On figures AB-1 and AB-2 more
ideal case is shown. Figure AB-1 show position for Sigma calculation from
channel number 10, and figure AB-2 show position from channel 17.
In reality, after maximum, decay curve can be broken in few lines with different
slopes. Sigma calculation (using any model) basically starting from the fixed time
(eg. 300us) or fixed channel (eg.10) and last again to the fixed time or channel.
However, this fixed gate calculation might be accurate only in cases like on
figures above. But in the cases where is decay broken in segments, logged
sigma will be far away from formation Sigma, because in the calculation will be
included all influence from borehole, fluid,
Because of that, we are using algorithm which give us not fixed starting and
ending point for calculation, but floating starting and ending calculation point.
1. Software search first maximum on decay curve.
2. From this point, it moves 3 to 5 channels right, to be sure that we are on
the begging of the decay.
3. Calculate First Tao from this point up to point where is count rate less than
some value (can be adjusted, usually count less than 2). In this First Tao
calculation, all borehole influences are encountered.
4. In logging software, after this point, software goes back to channel where
is maximum value detected, moves to right for number of channels which

37
is related with this First Tao value, and calculate again Tao and Sigma
value. In processing software, it moves 3 times First Tao value from
begging of the time Spectrum. Calculating Sigma on this way, we
dynamically change position for Sigma calculation taking a care to be
always out of borehole influence region.

38