IPTC 16685

Renovation of Old Well Logs Run in Boreholes Drilled with Oil-Base Mud
Tarek Ibrahim Elkewidy, Petroleum and Energy Engineering, the American University in Cairo
Copyright 2013, International Petroleum Technology Conference

This paper was prepared for presentation at the International Petroleum Technology Conference held in Beijing, China, 2628 March 2013.

This paper was selected for presentation by an IPTC Programme 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 International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not
necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor
Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International
Petroleum Technology Conference 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, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-
952-9435


Abstract

The problem of correcting for oil-base mud filtrate invasion has been resolved using modern well logging technology of
tools and interpretation techniques. However, many well logs from old wells remain uncorrected. Old interpretation
assumed no oil base mud filtrate invasion. The consequences may vary between unnecessarily perforating a water-
bearing zone to even worse by completely by-passing a hydrocarbon formation.

Lau et al. (1989)

developed a correction for oil-base mud effects on neutron and density logs, however the
standard formation evaluation techniques from the Dual Induction Resistivity Log, DIL, relies on knowledge of the
resistivity of the invaded zone, R
xo
. Since no electrode-type tool can work in oil base mud to measure R
xo
, a synthetically
derived R
xo
from the Electromagnetic Propagation Time (EPT) or the Thermal Neutron Decay time (TDT) logs is used.
In the absence of these unconventional EPT or TDT logs, interpretation is performed assuming no oil mud invasion and
the deep induction resistivity, RID, is reading the true formation resistivity, R
t
. However, it has been proven that oil mud
filtrate will invade the formation sometimes to a diameter greater than 120 inches.

This invasion will greatly affect R
t

masking the hydrocarbon potential of the reservoir to the extent that a water zone may appear as hydrocarbon-bearing.

Without proper consideration to the oil-base environment surrounding the logging tools, essential petrophysical
parameters such as true formation porosity and resistivity cannot be accurately measured. Techniques and concepts such
as crossplotting log versus log R
t
or log R
xo
, BVW, shale zonation index, fracture partitioning coefficient, etc. may not
be all conducted. Evaluation of properties such as S
w
, S
xo
, S
mo
, m, n,
ma
,
f
, FII, etc. will not be reliable. Consequently,
zonation of a heterogeneous reservoir into its hydraulic units cannot be accomplished.

In this study, a new practical and cost effective technique is introduced to correctly evaluate formations with
deep oil-base mud filtrate invasion of old well logs in the absence of EPT, TDT, logs, i.e. without a prior knowledge of
R
xo
or R
w
. This will allow characterization of old reservoirs drilled with oil-base muds using the available old
conventional well logs without the need for running new expensive well logs.

Introduction

From the view point of wireline logging, a borehole which is kept in gauge is undeniably better for accessibility and
accurate log response than a washed-out or caved hole caused by an inferior mud. However, the introduction of this non-
conductive mud which yields oil filtrate into the formation had a great impact on well logging techniques, precluding the
use of certain logging tools and affecting the response of others. No spontaneous potentials can be measured as the non-
conductive oil on the electrode will prevent the origination of the current responsible for the SP. Significant density
changes prevailing in oil-base drilling fluids with temperature and pressure may affect the neutron and density sonde
readings. Oil-base muds are not compatible with the pad-type resistivity tools or those operating at low frequencies (e.g.
microresistivity devices, spherically focused log, dual laterolog, etc.). This limits the conventional resistivity
measurements to the induction sondes. Furthermore, the invading non-conductive oil phase filtrate may significantly
affect the true readings of the induction resistivity tools. The consequences may vary between unnecessarily perforating
a water-bearing zone to even worse by completely by-passing a hydrocarbon formation.

Naturally, the use of such non-conductive muds will have a great effect on logging programs. Electrode type
tools will not work in such muds and the changes in oil-base drilling fluid density in the bore hole and formation pore
spaces will affect nuclear sondes. Indeed, no spontaneous potentials can be measured as the non-conductive oil on the
2 IPTC 16685
electrodes will prevent the origination of the current responsible for the SP. Therefore, a gamma ray is usually combined
with the DIL/Sonic. A negligible effect on GR can be expected due to the change of the bore hole media from water to
oil-base mud.

Porosity Tools Response

Porosity measurements from the conic and FDC/CNL (pad type tools) can be done easily in wells drilled with
oil-base mud because the bore hole is generally in good condition. However, significant density changes prevailing in
oil-base drilling fluids with temperature and pressure may affect the neutron and density sonde readings. Figure (1)
shows an example of FDC/CNL logs for two wells drilled in the same formations. The left well was drilled with salt-
saturated mud, whereas the right well was drilled with oil-base mud. The caliper shows that the well is oversize in the
salt-saturated mud case, and in gage with the oil-base mud case. Also, a poor pad contact of the density log can be seen
in the lower part giving an erroneous gas departure using the salt saturated mud. A decrease in the departure for the
upper gas sand is noted due to the oil filtrate invasion in the case of oil-base mud.

The compensated neutron tool response is determined mostly by the abundance of hydrogen in the environment
investigated. Invasion by the oil component of the mud can cause a hydrocarbon effect even in water-bearing formation.
The changes in hydrogen density of the fluids in the borehole and formation pore spaces will affect the CNL. Because of
the shallow radius of investigation of the CNL, the borehole environment affects to some extent the tool response. A
correction for the borehole environment effect on CNL in oil-base mud systems has been developed.
8
The corrected
neutron porosity,
nc
is expressed by

nc na bh temp pres
= + + + A A A (1)
where
A
bh na bh
d = 4 7090 0 0268 0 5013 . . . (2)

A
temp na
T = + +

41475 01453 2 384 10
2
. . . (3)
and

( )
A
pres na na
P = +

exp . . ln . ( . ) 0 096936 0 9898 3 088 10 14 7

5
(4)

In these expressions

A
bh
= borehole correction, %
A
temp
= temperature correction, %
A
pres
= pressure correction, %

na
= limestone porosity read from the log, %
d
bh
= borehole diameter, inches
T = formation temperature, F
P = formation pressure, psia

For the Formation Density Compensated log, FDC, the tool response is related to
b
, the bulk density. Porosity
can be determined from the FDC using



=

ma b
ma f
(5)
where
ma
= matrix density, g/cc

b
= density log reading, g/cc

f
= fluid density, g/cc

Pore space fluids density
f
can be expressed as

( )
f w xo mf xo
S S = + 1 (6)

where
w
= formation water density, g/cc

mf
= oil-base mud filtrate density, g/cc

In deep wells at elevated temperature and pressure, oil-base mud filtrate density can be significantly different
than at the surface conditions. This may affect the calculation of the porosity from FDC. Oil-base mud filtrate density
IPTC 16685 3
may be expressed as


6 4
0.867343 0.30240 10 3.4069 10
mf mud
P T

= + +
(7)

where T = formation temperature, F
P
mud
= mud hydrostatic pressure
On the contrary, the sonic devices are the least porosity tools affected by invasion of the oil component of the
mud.
4,6,1,2,5
If there were any influence at all, filtration by oil would tend to increase the transit time close to the borehole
wall.

Resistivity Tools Response

Oil -base muds are not compatible with the usual pad-type resistivity tools or those operating at relatively low
frequencies (microresistivity devices, spherically focused log, dual laterolog). This limits the resistivity measurements to
the induction sondes, the EWR (2 MHz electromagnetic wave resistivity) sonde, and the dielectric sondes. Moreover, oil-
base muds were assumed not to invade the formations. Consequently, interpretation programs were run assuming that
induction logs ere always reading the true formation resistivity, R
t
. However, it has been proven that oil mud filtrate will
invade the formation and R
t
reading will depend on the depth of invasion and the resistivity profile.

Since oil-base mud is non-conductive, the bore hole signal is nil. Thus, when there is no invasion the dual induction
readings Ild and Ilm should be the same. With invasion, the same situation occurs in front of a hydrocarbon bearing
zone. However, in case of water or even transition zone, a separation may or may not be observed depending on the
depth of invasion and the amount of formation water present. Thus, the effect of the depth of invasion can be classified
into two cases.

a) Case of shallow invasion

With shallow oil mud filtrate invasion, the response of the Dual Induction Log can be used to distinguish
between pay zones and water zones. In porous water formation, Ilm (the medium induction) reads higher than the Ild
(the deep induction). This indicates an invasion by the oil mud filtrate because formation water is saline and R
t
can be
read directly from Ild. In oil or gas-bearing sands, the medium and deep curves will lie on top of each other because
hydrocarbons are simply replaced by oil. A field example from Tuscalusa trend of South Louisiana is shown in Figures
(2) and (3). Note that in Figure (2) the two induction curves give the same readings through the gas-bearing zones, e.g.
zone 16461-16500 ft. Whereas, Figure (3) shows a separation between the two curves in front of water-bearing zones,
e.g. zone 18027-18170 ft.

b) Case of deep invasion

The main problem arises from the situation where there is a deep oil mud filtrate invasion in a transition and
even a water zone. In this case, filtrate invades the formation so deeply that the separation between Ilm and Ild decreases
and they may have the same readings of high resistivity. Water zones may exhibit the resistivity look of a hydrocarbon-
bearing. A field example is shown in Figure (4). The increasing depth of invasion was due to the increase in the
differential pressure. This figure shows two logs recorded in the same well. The left log was first recorded with a 17.1
ppg mud, while the right log was recorded with the mud weight increased to 17.5 ppg. About 275 bbl of mud was lost.
Note that, a large resistivity increase was recorded in front of all the porous zones especially around 15920 ft which was
fractured. Since the resistivity was high at 15850 ft in the first run, the well was cased and perforated. It produced
water.
12

True formation resistivity can not be obtained directly in this case of deep invasion unless a correction is made for Ild to
take care of the invasion effect. The effect of deep oil-base mud filtrate invasion on the induction resistivity log will be
discussed in detail in the following section.

Effect of Oil-Base Filtrate Depth of Invasion on the Dual Induction Log

The dual induction resistivity log is usually used to measure the formation resistivity in wells drilled with non-
conductive fluids (e.g. oil-base mud, foam, air). An induction log is obtained by a device consisting, in its conceptual
form, of two coils:
1. A generating coil, fed with alternating current from an oscillator, and
2. A receiver coil tied to an amplifier.

The induction sonde does not require direct contact between the coils and the formation. Thus, it works in holes
full with oil-base mud and, even, empty holes.
5
The magnetic field generated by powering the transmitter coil with a
4 IPTC 16685
20Khz induces eddy (foucault) currents in the surrounding medium. These currents flow in circular ground-loop paths
coaxial with the tool. The induction currents give rise to a secondary magnetic field, which induces signals in the
receiver coil. These signals are amplified and recorded. The strength of induction currents is essentially proportional to
the conductivity of the formation. With proper calibration the instrument records a conductivity log.

The Concept of Resistivity Geometrical Factor

A simple non-rigorous theory of the induction resistivity logging can be described by considering the signal
arriving to the receiver coil as the sum of the signals generated by each elemental loop of the formation. The signal from
each elemental loop will be proportional to its conductivity, weighted according to a geometrical factor which depends
on the dimensions and the position of the loop relative to the sonde. The geometrical factor is usually referred to as J or
G. G
A
, G
B
, ..., G
N
are the integrated geometrical factors of zones A, B, ..., N, respectively. Their magnitudes are
function of the diameter d
i
and the volume of the zone they represent. They simply represent the fraction that each zone
contributes to the total signal assuming a uniform conductivity within each zone and symmetry of volumes revolution
about the sonde.

The magnitude of the signal in conductivity units is the product of the geometrical factor and the conductivity of
the material. The total signal sensed by the tool is, thus, the sum of these products for all volumes within range, so

C C G C G C G
T A A B B N N
= + + + ... (8)
Note that
G G G
A B N
+ + + = ... 1 (9)

It is clear from the geometrical factor concept that, an oil-base mud filtrate filled zone of diameter d
i
will alter
the total conductivity signal received by the sonde. A correction must be done to the total conductivity signal. The mud
filtrate properties and its depth of invasion should be considered in order to estimate the true resistivity of the undisturbed
portion of the formation so it can be evaluated. Service companies usually provide charts for their induction tool
geometrical factor. Figure (5) shows the integrated radial geometrical factor values G
d
and G
m
for the Schlumbergers
deep (Ild) and medium (Ilm) induction logs, respectively.

Standard Interpretation of the Induction Resistivity Log in the Non-Conductive Environment

The standard interpretations of the induction resistivity log are made assuming a step profile of invasion. This
assumption is supported by the fact that oil-base mud filtrate will invade the formation with good sweep efficiency since
the mobility ratio, M, between the oil filtrate and the formation water is less than 1. Utilizing the concept of geometrical
factors, the conductivity signals received by the dual induction sonde can be expressed as
4


1 1
R
G
R
G
R
Ild
d
xo
d
t
= +

(10)
and

1 1
R
G
R
G
R
Ilm
m
xo
m
t
= +

(11)

In the above two equations, R
t
, R
xo
, and (G
m
, G
d
), which are function of the diameter of invasion d
i
, are
unknowns. In oil base mud, since no elctrode-type tool will work, usually a synthetic R
xo
calculated from the
electromagnetic propagation time, EPT, or the thermal neutron decay time, TDT, logs is used to solve for the three
unknowns R
t
, R
xo
, and d
i
. The Tornado Chart of Figure (6) provides means of the graphical solution.

If EPT or TDT logs are not available, then interpretations are performed assuming the deep induction resistivity
R
Ild
equal to the true formation resistivity R
t
. This assumption was supported by the early belief that oil-base muds do
not deeply invade the formations. However, it has been proven that oil mud filtrate will invade the formation, in some
cases so deeply that Ilm and Ild will have the same readings of high resistivity next to water and transition zones.
12
This
will result in an error in determining R
t
and consequently the fluid content of these zones. A new method to correct for
the effect of invasion without the need of a synthetically deriven R
xo
will be discussed in the following section.

Correcting the Effect of Filtrate Invasion on R
t
and Determination of S
w
Without a Synthetic R
xo

It can be noted from Figure (5) that for a diameter of invasion, d
i
, less than 40 inches, the Ild will have a
negligible geometrical factor. Consequently, the Ild signal will not be affected by the oil mud filtrate invasion and Ild
readings will be equal to the true formation resistivity, R
t
. On the other hand, for a diameter of invasion greater than 120
inches, the contribution of the uninvaded zone to the total Ild signal will become very small compared to the contribution
IPTC 16685 5
from the invaded zone. In this case, the deep induction is of no use in R
t
determination. Thus, for practical applications,
the range of invasion to be of concern is between 40 and 120.

Parallel Ilm and Ild geometrical Factors

Now, as shown in Figure (7), assuming no resistivity skin effect, both the Ilm and Ild geometrical factor curves
can be generally seen as almost running parallel to each other in the range of 40 < d
i
< 120. An average difference
factor of

G G
m d
= 0 3 . (12)

can be calculated. Figure (7) shows the assumed parallel geometrical factor curves. The system of the two Equations
(10) and (11), as both Ilm and Ild are affected by the same depth of invasion, can be rewritten by subtracting Equation
(11) from (10) to yield
( )
1 1 1 1
R R
G G
R R
Ild Ilm
m d
t xo
=
|
\

|
.
| (13)
then

1 1
0 3
1 1
R R R R
Ild Ilm t xo
=
|
\

|
.
| . (14)

Furthermore, the same observation can be concluded about the geometrical factor of the relatively new High
Resolution Induction, HRI, (Halliburton) or the Phasor Dual Induction (Schlumberger). These resistivity tools offer
higher vertical resolution and deeper depth of investigation than the conventional induction tools. Beds as thin as two
feet and sometimes thinner can be resolved using this technology. Figure (8) is an example showing the resolution power
of the HRI compared to the conventional dual induction resistivity log. The integrated radial geometrical factor of the
HRI tool is illustrated in Figure (9) which indicates a deep depth of investigation.
13
In this case, the diameter of invasion
to be of concern is approximately between 30 and 120. Schlumberger Rint-12 provides correction for the effect of oil-
base mud filtrate invasion on the phasor dual induction readings
11
. However, to use that chart both the conventional dual
and phasor induction logs have to be run.

Since only the non-conductive continuous oil phase of oil-base mud will filtrate and invade the formation, then
the only electrolyte in both the invaded and the virgin zones is the formation water which has a resistivity of R
w
.
According to Archies law, the resistivity of the invaded zone can be, then, expressed by replacing R
mf
with R
w
as

R
FR
S
xo
w
xo
n
= (15)
and that of the virgin zone as
R
FR
S
t
w
w
n
= (16)
Dividing Equation (15) by (16) yields for the case of n = 2


R
R
S
S
xo
t
w
xo
=
|
\

|
.
|
2
(17)
The tornado chart of Figure (6) provides a mean of graphical solution to evaluate the ratio R
xo
/R
t
. The chart is
to be entered vertically with the ratio Ilm/Ild read from the induction log measurements and horizontally with the ratio
R
xo
/R
Ild
. This requires a prior knowledge of R
xo
which may be synthetically deriven from the EPT or TDT logs. In the
absence of these two logs the evaluation can not be accomplished using Figure (6). Commonly, R
t
is assumed equal to
R
Ild
which may yield significant errors and R
xo
will not be obtained. However, as indicated by Equation (17) the ratio
R
xo
/R
t
can be evaluated from the right hand side term, ( ) S S
w xo
/
2
.

Determination of S
xo
from the Neutron Log

Since the neutron has no charge, the mass of the particles will govern its interaction with the formation. The
most efficient parts of the nucleus in matter to slow down the neutron are the protons of the hydrogen nucleus. This is
due to the fact that the mass of the neutron is equal to the mass of the proton. The depth of investigation of the
compensated neutron tool typically does not exceed 12 inches into the formation.
8
It is determined mostly by the
6 IPTC 16685
abundance of the hydrogen nucleus in the environment investigated which, in turn, depends on the porosity and the fluid
type filling the pores.

For the same formation porosity, the depth of investigation of CNL increases when the hydrogen concentration
is small. The neutrons travel farther from the source before being captured. On the contrary, when the hydrogen
concentration of the materials surrounding the neutron source is large, most of the neutrons are slowed down and
captured within a short distance from the source; thus, decreasing the CNL depth of investigation.

For the fluid type filling the pores being either water or hydrocarbon, the quantitative response of the neutron
depends primarily on their hydrogen index. The hydrogen index, HI, of a material is a measure of its hydrogen
concentration. it is defined as the ratio of hydrogen atoms concentration per cm
3
of the material to that of pure water at
75F. Pure water has an HI of 1.0. The HI of water depends on its salinity, temperature and pressure, while the HI of
hydrocarbons depends on molecular type, temperature and pressure.

Hydrogen Index of Salt Water

For a well drilled with oil-base mud, where the invading filtrate is the oil phase, only formation water salinity
needs to be considered. Focusing on the slow-down process, the effect of dissolved NaCl is to take up space, reducing
the water hydrogen density. Schlumberger has proposed the following relationship
11


( ) HI S
w w
= 1 (18)
where HI
w
= hydrogen index of salt water

w
= water density at formation temperature and pressure, g/cm
3

S = water salinity in weight fraction

In case of absence of water salinity, S, the hydrogen index of water HI
w
can be estimated as follows.
Water density at formation temperature and pressure
w
is given by
3




w
wsc
w
B
= (19)
where
wsc
= water density at surface standard conditions
B
w
= water formation volume factor as a function of temperature and pressure only

B
w
can be calculated from
3


B T T P
w s s
= + +

10 12 10 10 10 3 33 10
4 6 2 6
. . . . (20)

where T
s
= T - 60
T = formation temperature, F
P = formation pressure, psia

Schlumberger has proposed an approximated relation between NaCl water salinity and density at 75 and
atmospheric pressure as follow
11



wsc
S = + 1 0 73 . (21)

Substituting Equations (19), (20) and (21) into Equation (18), salt water Hydrogen Index can be expressed as

HI
B
w
w w w
=
173
0 73
2
.
.

(22)

In Equation (22), if formation water salinity is not known, an assumed value of
w
= 1.0 will give a good result
of HI
w
. This assumption takes into account some formation water salinity in the vicinity of 100,000 ppm for the high
temperature deep wells, and the water thermal expansion. Both counteract in the determination of water density. Note
thatn the hydrogen index of water decreases with increasing salinity, temperature and pressure. Figure (10) gives HI of
water at different temperatures and pressures assuming a water density of 1.0.



IPTC 16685 7
Hydrogen Index of Hydrocarbons

The oil phase of No. 2 Diesel mud has a molecular weight of 204.2 lbm/lbm-mol. For hydrocarbon composition
of C
n
H
2n+2
, the average composition of No. 2 Diesel can be approximated as C
14
H
30
. Note that, the hydrogen index of the
hydrocarbon increases with increasing its carbon number, n, temperature and pressure. So, the hydrogen index of No. 2
Diesel oil should be higher than 1.0 at high temperature and pressure (deep well conditions).

Schlumberger proposed a relationship to estimate the hydrogen index of C
n
H
2n+2
hydrocarbons as
11


HI
mf
mf
mf
mf
=

|
\

|
.
|
|
9
4 2 5
16 2 5
.
.

(23)

where HI
mf
= hydrogen index of the oil phase mud filtrate

mf
= density of the oil phase mud filtrate at formation temperature and pressure, g/cc

Diesel 2 fuel density is given by
8



mf
P T = +

0867343 0302420 10 34069 10
6 4
. . . (24)

where T = formation temperature, F
P
mud
= mud hydrostatic pressure, psia

The hydrostatic pressure of the mud can be expressed as

P D
m
= 0 052 . (25)
where
m
= mud density, ppg
D = depth, ft

Figure (11) gives HI of Diesel 2 at different temperatures and pressures.


S
xo
in Terms of the Hydrogen Index

Oil-base mud filtrate will invade the formation with a good sweep efficiency. With a mobility ratio M<1 and a
wetting agent, it can be generally assumed that oil-base mud filtrate will flush the invaded zone. Water or hydrocarbon
bearing formation (gas very likely will dissolve in oil filtrate at deepwell conditions) will flush down to the only
electrolyte being the residual formation water saturation S
xo
. Furthermore, the oil wetting agent in Diesel No. 2 with the
help of a relatively high differential pressure may reduce S
xo
down to the irreducible water saturation in the invaded zone.

For deepwell conditions of high temperature and pressure, oil filtrate will have a relatively high HI. In case of deep
invasion, the environment surrounding the neutron sonde will have an abundance of hydrogen. Thus, the neutrons are
slowed down and captured close to the borehole, resulting in a lesser depth of investigation of the neutron sonde.
Consequently, the neutron sonde will read mostly in the invaded zone.

Neglecting the excavation effect (if there is any) and correcting for any lithology effect, the Neutron-log porosity can be
expressed as
4


( ) ( ) ( )
n w xo mf xo ma
HI S HI S HI = + + 1 1 (26)

When the Neutron tool is calibrated for the rock matrix, the hydrogen index of the rock matrix HI
ma
is zero. Rearranging
Equation (26), the water saturation in the invaded zone can be, then, expressed as


( )
S
HI
HI HI
xo
mf n
mf w
=

(27)

where is the true formation porosity. The above equation may be simplified if the neutron porosity measurement,
n
, is
reading the true formation porosity to become

8 IPTC 16685

mf
xo
mf w
HI
S
HI HI
=

(28)
Invasion Effect on Ilm and Ild

The invasion effect of oil-base mud filtrate on the dual induction log can be summarized as follows. If there is
no invasion the dual induction curves, i.e. Ild and Ilm, will be on top of one another next to water, hydrocarbon or
transition zones. The true formation resistivity can then be read directly from Ild or Ilm. This will not be the case;
however, since invasion of oil-base mud filtrate has been proven to take place. When invasion occurs, the effect on the
readings of Ilm and Ild will depend on how deep the filtrate invades into the formation. Keeping in mind that the only
electrolyte there, is formation water, the effect of invasion depth can be classified according to the type of fluid in the
pore spaces.

In a water zone, when invasion is shallow, Ilm will read higher than Ild. A distinguished separation between the two
curves will indicate a water zone. The formation water resistivity, R
w
, can be then calculated using Archies Law.

R
R
a
w
o
m
=

(29)
When the depth of invasion increases, that separation decreases. Ild will be affected by the invading filtrate. In this case,
a water zone may appear as a hydrocarbon zone and hence, a correction for Ild is needed to get the true zone resistivity,
R
t
, and to evaluate its fluid saturation. In a hydrocarbon zone, the readings of Ilm and Ild curves will depend on the
amount of formation water presents along with the hydrocarbons. If the hydrocarbon zone is at its irreducible water
saturation, Ilm and Ild curves will lie on top of one another no matter what is the depth of invasion. Here, the true
formation resistivity, R
t
, can be read directly from Ild and the formation water saturation can be calculated from Archies
Law as

S
FR
R
w
w
t
2
= (30)

If water saturation in the hydrocarbon zone is greater than its irreducible water saturation, then the case can be described
as with the transition zone. In the transition zone, the amount of the formation water is always higher than its irreducible
water saturation. When oil-filtrate invades the formation, the invaded portion will be flushed down to water saturation
less than that in the undisturbed portion. Thus, for shallow invasion, Ilm will read higher than Ild and a separation
between them can be noticed. When oil filtrate invades deeper into the formation, Ild will become affected and will read
higher than the true formation resistivity. Consequently, the separation between Ilm and Ild will decrease and a
correction for Ild will be needed in order to get the true formation resistivity and to evaluate its fluid contents.

Invasion Correction

To correct for the effect of the deep invading oil-base mud filtrate, the ratio of Ilm/Ild can be evaluated from the
log readings. If Ilm/Ild = 1.0, then the formation is most likely to be hydrocarbon bearing formation. The ratio Ilm/Ild
will increase as the amount of formation water increases for a shallow depth of oil filtrate invasion. A high ratio of
Ilm/Ild might, very much, indicate a water bearing formation. The invasion is not deep enough so that Ild is not affected
by the filtrate (Ilm > Ild) and R
o
can be taken equal to Ild. If the ratio Ilm/Ild is small, then it might be a water zone or a
hydrocarbon zone where the reservoir is at water saturation higher than that in the invaded zone. The invasion is deep
and Ild is affected by the filtrate. As mentioned before, a correction can be done by solving the system of the two
Equations (10) and (11) using a synthetic R
xo
deriven from EPT or TDT.

Now, the new method of estimating the true formation resistivity, R
t
, and the original fluid saturations can be conducted
following the listed iterative steps to solve the system of the two Equations (10) and (11) by:

Step 1 - Calculate the value of the left hand side of Equation (14)

A
R R
Ild Ilm
=
1 1
(31)

Step 2 - Calculate the water saturation in the invaded zone S
xo
using Equations (27) or (28).
Step 3 - Assume a low initial value of the true formation water saturation S
w
and calculate the factor (S
w
/S
xo
)
2
.
Step 4 - Enter the tornado chart, Figure (6), with (R
Ilm
/R
Ild
) and (R
xo
/R
t
) taking equal to (S
w
/S
xo
)
2
. Read the value of
(R
xo
/R
Ild
).
IPTC 16685 9
Step 5 - Calculate the resistivity of the invaded zone as

R
R
R
R
xo
xo
Ild
Ild
=
|
\

|
.
| (32)

Step 6 - Calculate the true formation resistivity from

R
R
R
R
t
t
xo
xo
=
|
\

|
.
| (33)

Step 7 - Calculate the right hand side of Equation (14)

B
R R
t xo
=
|
\

|
.
| 03
1 1
. (34)

Step 8 - If A = B, then the original formation water saturation S
w
is equal to the assumed one, R
t
and R
xo
are equal to
their calculated values.
Step 9 - If A is not equal to B, then the assumed S
w
value should be increased, the new values of A and B should be
compared and so on. Note that, S
w
should not exceed 100%.
Step 10 - If the iterative calculations reached a value of R
xo
/R
ID
greater than that indicated by the d
i
= 40 curve at the
same R
IM
/R
ID
then the depth of invasion is less than 40. R
ID
is not affected by the invading oil and can be
taken equal to R
t
. In this case Archies model can be applied directly to R
ID
.

Figure (12) is a general flow chart of the iteration process.

This cost effective technique allows for calculating the correct R
t
, S
w
, S
xo
, R
xo
, and invasion diameter, d
i
without the need
of a synthetic R
xo
or even R
w
. The true formation resistivity and R
xo
can be, then, used to characterize the reservoir
different hydraulic units. Furthermore, naturally fractured reservoirs, which are usually targeted by horizontal wells
drilled with oil-base muds, may be characterized by expressing the Fracture intensity Index, FII, and consequently the
partitioning coefficient, v, in terms of R
xo
, R
t
and S
xo
as will be explained in the next section.

Calculating Partitioning Coefficient and Fracture Intensity Index for Formations Drilled with Non-Conductive
Mud

Fracture intensity index given in terms of well logging parameters is valid only for 100% water saturated zones
drilled with a water-base mud that yields conductive brine filtrate.
7
For the case of drilling with oil-base mud yielding
non-conductive mud filtrate we can write the following equations taken into account the presence of hydrocarbon in the
system as


( ) 1
1
2
R
S
R
S
R
xo
t xo
w
w
mas
= +
v v
(35)
and

( ) 1
1
2
R
S
R
S
R
t
t w
w
w
mas
= +
v v
(36)

In Equation (35) R
mf
has been replaced by R
w
since in the case of oil-base mud filtrate, formation water is the only
electrolyte present in both the invaded and the virgin zones. Also, water saturation in the uninvaded fracture system is
considered equal to that in the matrix, i.e., S
wf
= S
wma
= S
w
.
Subtracting Equation (35) from (36) results in an expression for FII, in this case of oil-base mud filtrate invasion, as

FII
R
R R
S S
w
t xo
w xo
=

|
\

|
.
|

1 1
(37)

where R
t
, R
xo
, S
w
, and S
xo
can be determined using the new method explained in the previous section. Furthermore, the
10 IPTC 16685
system of Equations used to charachterize naturraly fractured reservoirs can be evaluated in this case of drilling with non-
conductive mud
7
. Once the correct
t
and R
t
have been determined, the concept of crossplotting log
t
versus log R
t
can
be implemented to extract the correct cementation exponent, m.

Example

Figure (13) illustrates an example of a deep well drilled with oil-base mud in Dunn County, North Dakota. The
interval being analyzed in this example is from 12004 to 12120 ft. Figure (13) shows GR and DIL whereas Figure (14)
shows GR and the sonic logs. The compensated neutron-formation density logs are displayed in Figure (15). This well
has been drilled with 9.7 ppg oil-base mud. Bottom hole temperature at total depth of 13660 ft is 269F. The interval is
known to be dolomite with a low porosity ranging from 2 to 8%.

At depth of 12036 ft, the following readings and calculations can be done.

R
ID
= 18 ohmm
R
IM
= 26 ohmm
= 0.043 from the sonic log (At = 47.5 and true is determined using Raymer-Hunt transform)
Temp. = 252F (assuming a geothermal gradient of 1.0F/100 ft)
p = 6071 psia Eq. (25)

mf
= 0.7831 g/cc Eq. (24)
B
w
= 1.04007 Eq. (20)
HI
w
= 0.9451 Eq. (22)
HI
mf
= 1.02502 Eq. (.23)
S
xo
= 0.3131 Eq. (27)

S
xo
were calculated using Equation (27) assuming
N
= since the CNL-FDC log of this well had problems and is not
reliable. The iteration process resulted in

R
xo
= 65 ohmm
R
t
= 14 ohmm
S
w
= 0.68
S
hc
= 0.34
d
i
= 75 inches

Figure (16) shows the recorded R
IM
and R
ID
curves along with the calculated corrected R
t
and R
xo
curves. The
difference between R
ID
curve and R
t
curve is due to the effect of oil-base mud filtrate invasion on R
ID
. Furthermore, R
IM

and R
xo
curves show distinguished separation. If this formation is known to be fractured, then, application of Equation
(37) will result in FII = 0.0023 (for an estimated R
w
of 0.015) which is a very low value indicating that this zone at 12036
is most probably not fractured.

Nomenclature

a = constant

B
w
= water formation volume factor as a function of temperature and pressure only
D = depth, ft
d
bh
= borehole diameter, inches

HI
mf
= hydrogen index of the oil phase mud filtrate
HI
w
= hydrogen index of salt water
F = formation resistivity factor
FII = Fracture Intensity Index
m = cementation exponent
n = saturation exponent
P = formation pressure, psia
P
mud
= mud hydrostatic pressure
R
Ild
= deep induction resistivity
R
Ilm
= medium induction resistivity
R
mas
= resistivity of the matrix saturated with formation water
R
mf
= mud filterat resistivity
R
o
= resistivity of the formation 100% saturated with water
R
t
= true formation resistivity
R
w
= formation water resistivity
R
xo
= resistivity of the invaded zone
IPTC 16685 11
S = water salinity in weight fraction
S
w
= water saturation
S
xo
= water saturation of the invaded zone
T = formation temperature, F
v = partionaning coeffeciant
= porosity

nc
= corrected neutron porosity
A
bh
= borehole correction, %
A
temp
= temperature correction, %
A
pres
= pressure correction, %

na
= limestone porosity read from the neutron log, %

ma
= matrix density, g/cc

b
= density log reading, g/cc

f
= fluid density, g/cc

w
= formation water density, g/cc

mf
= oil-base mud filtrate density, g/cc

wsc
= water density at surface standard conditions

m
= mud density, ppg

References

1. Adams, J., et al.: Advances in Log Interpretation in Oil-Base Mud, Oilfield Review, 1989.
2. Boyeldieu, C., Coentz, A., Pelissier-Combescure, J.: Formation Evaluation in Oil-Base Mud Wells, SPWLA,
1984.
3. Brill, J.B. and Beggs, H.D.: Two-Phase Flow in Pipes, Third edition, Tulsa, April 1979.
4. Debrandes, R.: Encyclopedia of Well Logging, Institute Franais du ptrole publications, 1985.
5. Doll, H.G.: Introduction to Induction Logging and Application to Logging of Wells Drilled with Oil-Base
Mud, Trans. AIME, 1949.
6. Edwards, D.P., et al.: Log Evaluation in Wells Drilled with Inverted Oil Emulsion Mud, SPE 10206, 1981.
7. Elkewidy, T.I. and Tiab D.:" Application of Conventional Well Logs to Characterize Naturally Fractured
Reservoirs with their Hydraulic (Flow) Units; A Novel Approach ", SPE paper No. 40038, 1998.
8. Lau, M.N., Luquette, M.B., and Bassioni, Z.: Correction for Oil-Based Mud Effects on Neutron and Density
Tools, SPE 19611, October 1989.
9. Log Interpretation Charts, Schlumberger Educational Services, Houston (1991).
10. Log Interpretation Charts, Schlumberger Documents, 1986.
11. Log Interpretation, Volume 1, Principles, Schlumberger Documents, 1972.
12. Stevenson, J.A.: Log Evaluation of Wells in the Tuscalusa Trend of South Louisiana, The New Orleans
Geological Society, 1981.
13. Strickland, R., Sinclair, P., Harber, J., and DeBrecht, J.: Introduction to High Resolution Induction Tool,
SPWLA 28th Annual Logging Symposium, June 29-July 2, 1987.

12 IPTC 16685




































Salt-saturated mud Oil-base mud

Figure (1) Example of FDC/CNL logs for two wells
drilled in the same formations.
6

Figure (2) Dual Induction Sonic FDC/CNL Example from Tuscalusa
Trend of South Louisiana (Gas-Bearing Zone)
12

Figure (3) Dual induction/sonic logged in oil base
mud from Tuscalusa trend of South Louisiana (water
bearing zone)
12

IPTC 16685 13










Figure (4) Example of oil base mud filterate invasion effect on resistivity readings
12
Figure (5) The integrated radial geometrical factor
10
14 IPTC 16685














Figure (6) DIL Dual Induction R
xo
Tornado Chart
9

IPTC 16685 15




Figure (7) Assumed parallel geometrical factors
11
















Figure (8) Example shwing the high resolution power of the HRI (courtesy of Halliburton)

Figure (9) The integrated geometrical factor of the HRI
tool
11
16 IPTC 16685




Figure (10) HI of water at different temperatures and pressures assuming a water density of 1.0.



Figure (11) HI of Diesel 2 at different temperatures and pressures.

IPTC 16685 17




Figure (12) General flow chart for the iteration processs




18 IPTC 16685


Figure (13) GR-DIL for a well drilled with oil base mud
(example)




































Figure (14) GR-Sonic for a well drilled with oil base
mud (example)
Figure (15) CNL/FDC for a well drilled with oil base mud (example)

IPTC 16685 19


















Figure (16) Calculated R
xo
and corrected R
t
for a well drilled with oil base mud (example)

Calculated Rxo
ILD
Calculated Rt
ILM