United States Patent [19]

[11] Patent Number:

[45] Date of Patent:

Sinclair et al.
[57]

[54] COIL ARRAY FOR A HIGH RESOLUTION

INDUCTION LOGGING TOOL AND
METHOD OF LOGGING AN EARTH
FORMATION

[75] Inventors: Paul L. Sinclair, Clear Lake Shores;

Robert W. Strickland, Austin, both of
Tex.

[21] Appl. No.: 604,542

Oct. 29, 1990
Related U.S. Application Data

[63) Continuation-in-part of Ser. No. 474,230, Feb. 2, 1990,

abandoned, and a continuation-in-part of Ser. No.
478,024, Feb. 9, 1990, abandoned.

[51] Int. Cl* ............. G01W 3/28

[52]
[58]
324/338, 339, 344
References Cited
U.S. PATENT DOCUMENTS
2,582,314 1/1952 Doll ..................................... 324/339
4,472,684 9/1984 Schuster ......................... 324/323 X

Primary ExaminerKenneth A. Wieder

ABSTRACT

An induction tool including a coil array having a plural

ity of transmitter coils and a plurality of receiver coils.
Each plurality of coils is symmetrically arranged on an
insulated mandrel about a selected point. The induction
tool also includes an oscillator circuit that provides an
alternating frequency signal to the transmitter coils. A
measure signals indicative of the voltages induced on
these receiver coils to the surface. By varying the spac
ing between the receiver coils and transmitter coils it is
possible to independently vary the vertical and radial
measurement capabilities of the tool of the present in
vention. A dual induction tool having high vertical

Houston, Tex.

[56]

Nov. 12, 1991

receiver circuit is connected to the receiver coils to

[73] Assignee: Halliburton Logging Services, Inc.,

[22] Filed:

5,065,099

resolution is also disclosed. The dual induction tool

includes a pair of transmitter/receiver coil arrays, each

transmitter/receiver coil array having a different trans
mitter-to-receiver coil spacing and substantially indenti
cal vertical resolution. A single receiver coil is located
at a selected point on an insulated Sonde and a pair of
variable symmetrically disposed bucking coils are lo
cated on either side of the receiver coil. Two pairs of
identical transmitter coils are similarly symmetrically
disposed on either side of the receiver coil. In a pre
ferred embodiment, each pair of transmitter coils is
alternately energized and the bucking coils are selec
tively adjusted to compensate for the variation in trans
mitter-to-receiver coil spacing.

Assistant ExaminerWarren S. Edmonds

Attorney, Agent, or FirmWilliam J. Beard

14 Claims, 7 Drawing Sheets

COMPUTER

| ,4AD
Shoulder
SkfM

PoWE&
Stippt. Y

EFFECT
(GRRECTION

16

PASE
SEWSiri VE
t retro

RECORDER H. 3,

U.S. Patent

Nov. 12, 1991

she ori

5,065,099

COMPUTER
SHOULDER
POWER

AWD SKIN

SUPPLY

EFFECT

RECORDER -

34

CORRECTION

&

FIG. 7

^.
$

16

LN

%
%

%z

QS

|=-|-H-H,

$
^

&

S-2

A:
S, - 10

SA || OSCILLATOR

ZS/SS/SS/SS/SS

%
SENSITIVE || K
DETECTOR? t

78

14

40

%&^% S^^

U.S. Patent

Nov. 12, 1991

Sheet 2 of 7

5,065,099

U.S. Patent

707
82.

5,065,099

U.S. Patent

Nov. 12, 1991

she or

5,065,099

DEF/WE :

03 =- 5,

s
:

$3

G.

= G. * ====

329 g

329 b

329 f

YES

329

329 e

YES

3290

COUT = Go 4

5,065,099

COIL ARRAY FOR A HIGH RESOLUTION

INDUCTION LOGGING TOOL AND METHOD OF
LOGGING AN EARTH FORMATION

additional coils are arranged to cancel out much of this

shoulder effect.

It is also possible for a tool to have good vertical

resolution but poor shoulder effect. Such a tool would

This application is a continuation-in-part of co-pend

ing application Ser. No. 07/474,230 to filed Feb. 2, 1990
and co-pending application Ser. No. 07/478,024 filed
Feb. 9, 1990, which are hereby incorporated by refer

be able to accurately define bed boundaries but would

give poor estimates of the conductivities of these thin

beds. Vertical resolution and shoulder effect are two

10

ence.

The present invention relates in general to well log

ging tools and in particular to well logging tools for
Induction well logging tools were originally designed
to provide a resistivity measurement in wells drilled
with electrically insulating, oil-based mud. Since their
introduction, their use has spread to wells drilled with
water-based muds of moderate-to-high electrical con

15

20

induces eddy currents in the formation which are coax

30

35

sate for borehole and shoulder effect. In addition, the

coils must minimize the mutual coupling between trans

mitter coils and receiver coils, as this signal is very large

when compared to most formation signals. In known

coil arrays, the position and strength of each coil con

trols each of these aforementioned effects. Because each

45

ventional coil array designs therefore must be a com

50

55

promise between sharp vertical resolution and deep

radial penetration into the formation.
It would be desirable for an induction logging tool to
permit variation in depth of investigation or vertical
resolution independently without compromising the
other parameter. Previously, such a tool has not been
available.

In most commercial applications, it is also desired to

investigate the strata surrounding a borehole to differ
ent depths of investigation in order to determine the
diameter of invasion of the strata by borehole fluids.
This requires at least two measurements with contrast
ing radial response and ideally identical vertical resolu
tion so that differences in the logs obtained will be due

tion of a tool is a measure of the thinnest bed that a tool

can detect. That is, a tool may accurately estimate the

thickness of beds that are thicker than its vertical

apparent conductivity reading of a thin bed due to the

of these effects may change as a coil is modified, it is

difficult to design a coil array optimized to reduce all of
these effects simultaneously. The different effects inter
act, as one effect is reduced, another is increased. Con

The following are terms of art that are used often to

compare various induction tools. The vertical resolu

can estimate the thickness of the bed. This error in the

typically a correction which yields the true conductiv

ity of a homogeneous formation.
In the case of conventional induction tool arrays,

coils must be positioned to define the tool's vertical

resolution, depth of investigation, as well as to compen

above. Such arrays are generally termed focused ar

thicker than the vertical resolution of the tool, the tool

The foregoing discussion can be understood by as

suming that a tool may be operated at a sufficiently low
frequency so that there is no significant attenuation of
the transmitted signal as the signal propagates through
the formation. In practice, such attenuation cannot be
neglected since it reduces the transmitted signal propor
tionately more in conductive formations. This error is
commonly referred to as skin effect. Prior art practi
tioners generally attempt to design a coil array which
has moderate skin effect at the highest conductivity of
interest in logging situations and then correct for the
skin effect at the surface. The skin effect correction is

served to cancel out the various adverse effects listed

resolution. A tool can also accurately locate a bed

boundary to within the tolerance of its vertical resolu
tion. There is still a significant error in the apparent
conductivity reading in a thin bed due to signals from
adjacent beds; however, so long as the thin bed is

signal comes from the borehole as compared to the

formation. In conventional arrays, coils are arranged to
cancel much of the signal coming from near the tool so
that the depth of investigation will be large and the
borehole effect will be small.

of a single transmitter coil and a single receiver coil

wrapped around an insulating mandrei. The transmitter
coil is driven by an oscillating current at a frequency of 25
a few tens of kilohertz. The resulting magnetic field

rays.

The depth of investigation of a tool is a measure of

how deeply the tool sees into the formation. The depth
of investigation is defined as the radius of the cylinder
from which half the apparent conductivity signal
comes. The borehole effect is a measure of how much

ductivities. The basic element in all multi-coil induction

tools is the two-coil Sonde. The two-coil Sonde consists

ial with the tool. These eddy currents produce a mag

netic field which in turn induces a voltage in the re
ceiver coil. This induced voltage is then amplified, and
the component of the voltage that is in-phase with the
transmitter current is measured and multiplied by a tool
constant to yield an apparent conductivity signal. This
apparent conductivity is then recorded at the surface as
a function of the depth of the tool.
The two-coil sonde has several practical limitations.
Its response is adversely affected by several factors
including the borehole, adjacent beds, and mud filtrate
invasion. Also, the two-coil sonde is difficult to imple
ment because of the large direct mutual coupling be
tween the coils. Even though this mutual signal is out
of-phase with the transmitter current, it is a problem
because a very small phase shift in the electronics can
cause this mutual coupled signal to leak into the ap
parent conductivity signal. For these reasons, it is the
standard practice in the industry to construct induction
logging tools with coil arrays which include additional
coils. Typically, there are several transmitter coils and
several receiver coils. In certain applications all of the
transmitter coils may be connected in series into one
circuit. Similarly, all of the receiver coils may be con
nected in series in a separate circuit. The additional coils

aspects of the vertical focusing of an induction tool coil

array.

BACKGROUND OF THE INVENTION

measuring formation resistivity.

signal coming from adjacent beds is referred to as

shoulder effect. In known induction tool arrays, the

to radial anomalies in the formations such as invasion.

65

Most prior art dual induction tools utilize deconvolu

tion filters to match dissimilar vertical responses of two
induction coil arrays with inherently different vertical
resolutions by smoothing out the response of the array

5,065,099

3
with the sharper vertical resolution and degrading it to
match the vertical response of the second array. This

4
receiver circuit connected to the receiver coil assembly

for receiving signals from the receiver coils. In this

embodiment the receiver circuit is coupled to the oscil
lator circuit to receive a phase reference signal enabling
the receiver circuit to detect an in-phase receiver coil
signal and an out-of-phase receiver coil signal. In this

approach is not desirable in view of the degradation of

vertical resolution which is necessary to match the coil

arrays utilized to investigate the formation.
Thus, a need exists for a dual induction tool capable
of investigating multiple depths of investigation while
maintaining substantially identically vertical resolution
for all coil arrays.
SUMMARY OF THE INVENTION

It is therefore one object of the present invention to

provide an improved dual induction tool.
It is another object of the present invention to pro
vide an improved dual induction tool having high verti

alternative embodiment, each of the transmitter coils

may include two bucking coils symmetrically posi

tioned about a respective transmitter coil, each bucking

10

15

transmitter/receiver coil arrays each transmitter/

20

25

30

35

oscillator circuit also provides a phase reference signal

to the receiver circuit. The receiver circuit in turn de
50

FIG. 5 is a process flow diagram illustrating an itera

tive processing technique which may be utilized in the
FIG. 6 illustrates comparative results of processing

55

with two different formation models;

FIG. 7 is a block diagram of one embodiment of a

dual high resolution induction tool of the present inven

tion positioned in a borehole;

FIG. 8 is a schematic representation of the receiver

coil of the dual high resolution induction tool of the
present invention; and
FIG. 9 is a simplified schematic diagram of one em
bodiment of the dual high resolution induction tool of
the present invention.

insulated body. Also included in this second embodi

ment is a second plurality of transmitter coils. The

transmitter coils are positioned symmetrically about the
selected point of the insulated body and the receiver
coil assembly. This induction tool also includes an oscil
lator circuit that provides a selected frequency transmit
ter signal to the transmitter coils. Also provided is a

junction with the accompanying drawings, wherein:

FIG. 1 is a block diagram of one embodiment of the
induction tool invention positioned in a borehole;
FIG. 2 is a schematic representation of a first embodi
ment of the present invention;
FIG. 3 is a schematic diagram of a second embodi
ment of the present invention;
FIG. 4 is a schematic illustration showing the pre
ferred mode of processing on computer 32 of FIG. 1;
processing mode illustrated in FIG. 4;

this embodiment, the induction tool further includes a

data processor connected to the receiver circuit for

analyzing the receiver coil signals and providing Sioui
der and skin effect computational corrections to com
pute a corrected log signai.
A second embodiment of the present invention in
cludes an induction tool having a receiver coil assembly
positioned symmetrically about a selected point of an

BRIEF DESCRIPTION OF THE DRAWINGS

The novel features believed characteristic of the in

40 tion of an illustrative embodiment when read in con

nected to the receiver coils. In this embodiment the 45

tects an in-phase signal and out-of-phase signal. Each of

the receiver coils may further include two bucking coils
that are symmetrically positioned about a respective
receiver coil and, in one depicted embodiment, the
bucking coils are connected in series and wound in
opposite sense to the respective receiver coil. Also, in

metrically disposed bucking coils are located on either

side of the receiver coil. Two pairs of identical transmit
ter coils are similarly symmetrically disposed on either
side of the receiver coil. In a preferred embodiment of
the present invention each pair of transmitter coils is
alternately energized and the bucking coils are selec
tively adjusted to compensate for the variation in trans
mitter-to-receiver coil spacing.
The above as well as additional objects, features, and

vention are set forth in the appended claims. The inven

tion itself; however, as well as a preferred mode of use,
further objects and advantage thereof, will best be un
derstood by reference to the following detailed descrip

in this embodiment and have the same Sense. These

receiver coils are positioned symmetrically about the

selected point on the insulated body and the transmitter
coil assembly. An oscillator circuit is utilized to provide
a selected frequency transmission signal to the transmit
ter coil assembly and a receive: circuit is similarly con

receiver coil array having a different transmitter-to

receiver spacing and substantially identical vertical
resolution. A single receiver coil is located at a selected
point on an insulated Sonde and a pair of variable sym

advantages of the invention will become apparent in the

following detailed description.

the receiver coils.

In one embodiment of the present invention a cen

trally located transmitter coil assembly may be pro
vided. The transmitter coil assembly is positioned sym
metrically about a selected point on an insulated body.
A plurality of receiver coils may be connected in series

shoulder and skin effect computational corrections to

the receiver circuit signals in order to compute a cor
rected log signal for the tool.

An induction tool may also be provided with multiple

cal resolution.

It is another object of the present invention to pro

vide a new induction tool coil array design wherein
different coils of the array control the vertical and ra
dial focusing, so such an array may possess both sharp
vertical resolution and deep radial penetration without
the compromises present in known arrays.
An induction tool is provided which includes a coil
array having a first plurality of transmitter coils and a
second plurality of receiver coils. The transmitter and
receiver coils are symmetrically arranged on an insu
lated Sonde about a selected point on the insulated
sonde. The induction tool further may include an oscil
lator circuit connected to the transmitter coils to pro
vide a selected frequency transmission signal to the
transmitter coils. A receiver circuit is preferably con
nected to the receiver coils for receiving signals from

coil being connected in series and being wound in oppo

site sense to its respective transmitter coil. A data pro
cessor connected to the receiver circuit may provide

65

DETAILED DESCRIPTION OF THE

PREFERRED EMBODIMENTS

FIG. 1 is a block diagram of the present invention

directed in a borehole 8 surrounded by drilling mud 20.

5,065,099

6
site sense as to the main receiver coil 50. These bucking

5
It should be understood by those skilled in the art that
the dimensions of FIG. 1 are exaggerated to enable a

receiver coils 52 and 54 are so named because their

more clear presentation of the discussion. A logging

cable 28 is preferably connected to induction tool elec
tronics 14. Logging cable 28 preferably includes both
power and signal lines. The induction tool electronic
circuitry includes an oscillator circuit 16 and a receiv
ing phase sensitive detector circuit 18. The oscillator

purpose is to buck out the large mutual coupling from

the transmitter coils 56 and 58 and to cancel any signals

coming from that portion of the formation nearest coil

array 10. The bucking receiver coils 52 and 54 are each
spaced from the main receiver coil by a distance d. In
the embodiment illustrated in FIG. 2, the bucking re

circuit provides a reference signal on line 17 to the

ceiver coils 52 and 54 include fewer turns than the main

receiver circuit 18. The oscillator circuit 16 is also con 10

nected to the transmitter coils 42 that are positioned on

an insulating mandrel 12. The receiver circuit 18 is
connected to the receiver coils 44, also positioned on

and 54 are located closer to transmitter coils 56 and 58,

the insulated mandrel 12. The transmitter coils 42 and

the receiver coils 44 make up coil array 10. An alternat

ing frequency signal produced by the transmitter oscil
lator circuit 16 is coupled to the transmitter coils 42 to
inductively produce an electromagnetic field which

15

causes eddy currents to flow within the formation.

These eddy currents produce a secondary magnetic

field which induces a voltage in receiver coils 44. Re
transmits the receiver coil signals to a computer 32
located at the earth's surface 22 through the logging
cable 23. These received signals are utilized to provide
information concerning the resistivity of the formations
surrounding the borehole.
Computer 32 is connected to a power source 30

cal about the main receiver coil 50. Further, both the

30

35

50, 52 and 54 are symmetrical, each transmitter contrib

utes equally to the total mutual inductance. Thus, the

receiver coils 50, 52 and 54 are mutually balanced to

each transmitter 56 and 58 individually. Assuming that

the coils are small enough to be considered point di
45

50

/or the receiver coils are multiplied by a constant num

induction tool is disclosed in U.S. Pat. No. 2,582,314 to

X. NiMH = 0

j L13;
where Nti is the number of turns on the ith transmitter,

N is a number of turns on the jth receiver, the Lj is a

will be discussed.
55

separation between the ith transmitter and the jth re

mitter-receiver pair in the coil array. In actual practice

ceiver. The summation is carried out over each trans

this formula should be modified to account for the finite

size of coils in the actual tool, but for comparison of the

relative merits of coil arrays, the point coil formula is
sufficient. For the SY coil array, the condition for mu
tual balance is given by the equation:

distance 2D. Each of the transmitter coils 56 and 58

includes N turns that are wound in the same sense. A

main receiver coil 50 with Nrm turns is located at a

the bucking receiver coils 52 and 54 are wound in oppo

poles, the condition for mutual balance for a multi-coil

H.G. Doll. The expression relating to the number of
turns and positions of the coil is:

ber. Therefore only the relative number of turns of coils

coils and the main receiver coil are referred to herein as

the receiver coils. These coils are wound in series but

discussion, it is assumed that the condition of mutual

be omitted from this discussion since the receiver coils

all of the numbers of turns of the transmitter coils and

selected point 51 of the insulated mandrel 12. Also lo

cated on mandrel 12 and disposed symmetrically about
receiver coil 50 are two bucking receiver coils 52 and
54, each consisting of Nrb turns. The bucking receiver

positions will now be discussed. For purposes of this

balance exists. In view of this assumption of mutual
balance, one of the two transmitter coils 56 and 58 may

through it, and the area of its turns. Therefore, when

one coil has more turns than another, the dipole mo

FIG. 2 illustrates one preferred embodiment of the

present invention. A pair of transmitter coils 56 and 58
are preferably disposed symmetrically about a selected
point on an insulated mandrel 12 and separated by a

and 54 are symmetrical about a selected point 51 on the

mandrei 12. This coil arrangement is referred to herein
as an SY type of coil array. This arrangement differs
from the known coil arrays that are reciprocally sym
metrical, that is, the transmitter array is a mirror image
of the receiver array.
The relationship of the numbers of turns and coil

number of turns which will be referred to herein as the

ment of that coil is said to be greater than the other. It

should therefore be understood that the spatial response
of a particular coil array does not change significantly if

receiver circuit 18 receives the phase reference signal 17

in the induced voltage from the receiver coils 50, 52 and
54 to produce an in-phase signal and out-of-phase signal
on lines 60 and 62 respectively.
As can be seen, coil array 10 is completely symmetri
transmitter coils 56 and 58 and the receiver coils 50, 52

log 36 which depicts the corrected log signal versus the

depth of the tool in the borehole.
Each coil (both receiver and transmitter) includes a

relative dipole moment of the coil. The dipole moment

of a solenoidal coil is approximately proportional to the
product of its number of turns, the current passing

making their mutual coupling per turn greater. Oscilla

tor circuit 16 provides an alternating current to each
transmitter coil. The resultant induced voltage in the
receiver coil array (coils 50, 52 and 54) from the forma
tion includes two components. One component is the
in-phase induced voltage and the second component is
out-of-phase induced voltage, where the phase is rela
tive to the current in transmitter coils 56 and 58. The

20

ceiver circuit 18 is connected to receiver coils 44 and

which also provides power to downhole tool electron

ics 14 through the logging cable 28. A winch 24 is pref
erably connected to the logging cable to raise and lower
the logging tool in borehole 8. In the depicted embodi
ment of the present invention, winch 24 provides a
depth signal on line 26 to computer 32. The computer
receives signals from the receiver coil 44 and provides
shoulder and skin effect correction computations to
compute a corrected log signal which is provided to a
recorder 34. Recorder 34 preferably produces a well

receiver coil 50 because the bucking receiver coils 52

65

where Nrm is the number of turns of the main receiver,

Nrb is the absolute number of turns on the bucking re

5,065,099

bucking receivers 52 or 54 to the main receiver 50 as

shown in FIG, 2. By setting t equal to the turns ratio
Nm/Nrb (the number of main receiver turns to bucking
receiver turns) and setting S equal to the spacing ratio
D/d (the transmitter spacing to the bucking receiver
spacing) the condition for mutual balance can be ex
pressed in the following equation:

In thick beds composed of discrete cylindrical zones,

the integrated radial geometric factor is more useful.
The integrated radial geometric factor at a particular
radius is obtained by integrating the radial geometric
factor from zero to the given radius. The integrated
radial geometric factor quantifies the depth of investiga
tion of a coil array. Traditionally, the depth of investi

ceivers, D is the distance of either of the transmitters to

the main receiver, and d is the distance of either of the

gation is the 50% point of the integrated radial geomet

ric factor.
10

The integrated radial geometric factor also provides a

method for evaluating the borehole effect of an induc
tion tool. The plot of the integrated radial geometric

factor on an expanded scale in the borehole region dis-

This equation is all that is necessary to completely spec

ify a mutually balanced SY type coil array assuming

15

point dipole coils. The remainder of this discussion will

address mutually balanced SY type coil arrays but it
should be understood by those skilled in the art that
mutual balance is not a characteristic of SY type coil
arrays. It is easier, in practice, to construct a mutually

was found that the width of the central peak of the

vertical geometric factor curves changes considerably

balanced coil array.

The response of SY type coil arrays to various types

of geological formations will now be discussed.
A wide class of physical formations are cylindrically
symmetric. In such formations, eddy currents induced
by the transmitter coils of an induction tool are all paral

when the bucking coil spacing is varied and the spacing

between the transmitter and receiver is held constant.

However, the integrated radial geometric factor for the

bucking coil spacing variation is approximately the
same. It was also found that the maximum usable bore

lel and coaxial with the axis of the tool. In such forma

tions, the conductivity of the earth through which the

current loop passes is constant. If the conductivity of
such a formation is sufficiently low, the eddy current
loops in the formation do not interact. In such forma
tions, there is no significant skin effect. The geometric

30

factor theory applicable to such formations is disclosed

in the article Introduction to Induction Logging and
Application to Logging of Wells Drilled with Oil Base
Mud, Journal of Petroleum Technology, Vol. 1, No. 6
(June, 1949), pp. 148162, authored by H.G. Doll.
According to geometric factor theory the apparent
conductivity seen by an induction tool due to one loop
of formation of negligible cross-section coaxial with the
tool is the conductivity of the loop times a geometric
factor that depends solely on the geometry of the loop
and the tool. The response of the tool to a formation
composed of resistive loops of formation can thus be 45
expressed as the integral over all space of the conduc
tivity times this geometric factor. The three dimen
sional plot of the geometric factor versus depth and
radius for a particular tool is referred to as the map" of
that tool.

In formations in which the conductivity is constant in

one dimension, different slices of the map may be inte
grated. If the formation is not invaded, i.e. the effects of
the borehole may be neglected, conductivity varies
only with depth, and the vertical geometric factor is
useful. The value of the vertical geometric factor may
be obtained at a given depth by taking a slice of the map
at that given depth and integrating it over all radii. This
vertical geometric factor provides a means for quantify.
ing the vertical resolution of a coil array. The vertical
resolution is the width at half the maximum height of
the vertical geometric factor.
The radial geometric factor is useful in thick, invaded
beds. Value of the geometrical factor can be obtained at
a given radius by taking a slice of the tool map at the
given radius and integrating it over all depths. The
radial geometric factor is useful in thick beds where
conductivity varies continually with the radius,

closes what is termed the borehole geometric factor.

The two coil sonde has a very poor borehole effect.
To gain an understanding of how each coil of an SY
type coil array affects the tools spatial response, the
variation of each of these geometric factors with varia
tions in coil spacing has been evaluated.
From the investigation of the geometric factors, it

50

55

hole diameter is improved by utilizing a small bucking

coil spacing. By changing the bucking coil spacing, the
vertical response of the tool may be changed to a great
extent while making relatively minor changes in the
radial response. However, the borehole effect decreases
with decreasing bucking coil spacing since more of the
signal near the sonde is cancelled out if the bucking
coils are nearer to the main coil.

Fixing the bucking coil spacing and varying the loca

tion of the transmitter coils relative to the receiver coils

while adjusting the number of turns of the bucking coil

to zero the mutual inductance results in the finding that
the central peak of the vertical geometric factors is the
same for each of the different transmitter/receiver spac
ings. However, the depth of investigation is approxi
mately proportional to the spacing between the trans
mitter and receiver coils. Also, the maximum usable

borehole size is nearly proportional to this spacing.

Therefore, in the symmetrical spaced induction tool
of the present invention, the bucking coil spacing deter
mines vertical resolution and the transmitter and re

ceiver coil spacing determines the depth of investiga

tion. In other words, the larger the transmitter and
receiver coil spacing and the smaller the bucking coil
spacing, the smaller the borehole effect. The larger the
transmitter and receiver coil spacing, the deeper the
radial response will be, and the smaller the bucking coil
spacing the sharper the vertical response will be. This
result is quite different and unexpected when compared
with conventional coil arrays in which the spacing of
the transmitter and receiver coils determines both the

60

vertical resolution and the depth of investigation. From

this analysis, the SY type coil array with the best re
sponse is one with the largest transmitter and receiver

coil spacing and the smallest bucking coil spacing that

sensitivity and mechanical considerations will allow.

65

Therefore, those skilled in the art will appreciate that

the advantage of the SY coil array is that the vertical
and radial focusing are controlled by separate spacings
in the array and, because of this, it is possible to control

5,065,099

9
the vertical focusing without destroying the radial fo
cusing and vice versa.
Another unique feature of the SY type array is that
the width of the main beam of the vertical geometric
factor is the distance between the two bucking coils.
This distance between the two bucking coils determines
the vertical resolution of the array and in the depicted

embodiment of the present invention this distance may

be as small as fourteen inches. This is quite different

10
non-linearity of the tool response may be expressed as

GD(ora)F(ora).

If it is assumed that the degree of non-linearity in

duced by the tool on the apparent conductivity ora is the
same as the difference of the Doll model and the higher
order model applied to the true formation conductivity
ort, it may be asserted that
GD(ori)F(or)= GD(ora)F(ora)

from known conventional arrays in which the main

transmitter and receiver coil spacing fixes both the ver
tical resolution and the depth of investigation. There

10

lution smaller than the resolution of any other transmit

ter and receiver coil pairs that make up this array.
FIG. 4 illustrates schematically the preferred pro
cessing which may be used in the computer 32 to deter
mine an output conductivity value orout. It will be ap
preciated that this Figure illustrates processing steps

15

fore, by a proper adjustment of this SY type array, the

array can be configured such that it has a vertical reso

(i)

Since F(or) corresponds generally to the measured

conductivity orm, the only unknown in equation (1) is
GD(or). Equation 1 may be rewritten as

20

and does not show actual hardware structure within the

computer 32. Of course, it is within the skill of the art to

implement the processing steps in either hardware or

The Doll response is easily inverted, thus by inverting

the right-hand side of equation (2b), one obtains
(3a)

software.

Briefly, the R and X value of the signals sent uphole

on lines 60 and 62 to the computer 32 are provided to

processing step 321 which calculates a measured con

and

ductivity value orm in a conventional manner. A con

ventional skin effect correction may then be applied to

the measured conductivity value orm to obtain an appar

ent conductivity value ora.

In making measurements with an induction logging
tool, a complex function maps the formation conductiv

45

ity Ort.

In order to provide a better estimate of the true for

mation conductivity ort, the apparent conductivity ora is
provided to processing step 323 and to processing step
325. Processing step 323 applies the well-known Doll
geometrical factor model GD and processing step 325
applies a high order fast forward model F. Typically,
these models are implemented as filters. Briefly, the
models attempt to approximate the complex function
which maps the true formation conductivity or, into the
tool response. Stated mathematically, or mp3s G. Lo?.)
where GD is the Doll response. More generally, it can

50

55

60

The Doll model used in processing step 323 is a first

order function which, for example, does not account for

the portion of signals detected as a result of non-linear

sponse. Thus, an approximate measure of the degree of

equation (3b) does not require a value for GD(ora), pro

cessing step 323 may be omitted. This initial value -, is
then provided to processing step 329 wherein an itera
tive process is used to obtain a more refined value for
Ort.

The preferred iterative technique of processing step

329 is illustrated in FIG. 5. The step 329 receives as
inputs the values 6-t from processing step 327 and orn

from processing step 321. Additionally, a local integer

counter i may be set equal to 1. Step 329(a) defines
additional local variables. Specifically, a value oro is set
equal to the current value of Gr, a value ori, is set equal
to F(oro); an error value 6 is set equal to ormor?; and a
value q is set equal to oro +6.
By applying the high order model F to the value q,
one obtains

(4a).

panded in a Taylor's series to yield

F(q)=F(oro) + 6 F(oro)

(4b)

wherein only the first two terms of the Taylor's series

interaction of the eddy currents induced in the forma

tion of such non-linear effects on the measured re

forms the calculation of equation (3b) to obtain an initial

value ort for the true formation conductivity. Since

The right-hand side of equation (4a) may then be ex

which maps the true formation conductivity into the

tion. The higher order model used in processing step

325, however, is intended to account for at least a por

Since the value of ort obtained with equation (3b) is an

approximation, in the following discussion this value
will be designated ort.
Referring again to FIG. 4, processing step 327 per

F(q)=F(oro--6)

be stated that armF#n(or) where Ffm is the function

measured response.

or ~ ora + G-Morm F(ora)]

I}

ity into the signals detected by the receiver coils. The 35

measured conductivity orm, and thus the apparent con
ductivity ora, is subject to various non-linear influences
caused by induced eddy currents in the formation and
other environmental effects. For example, the induced
eddy currents themselves act as transmitters which
affect the signals detected at the receiver coils and may
induce further eddy currents in the formation. Al
though skin effect corrections reduce the impact of
some of these effects, the apparent conductivity ora is

only an approximation of the true formation conductiv

(3b)

30

are expressed.
65

Since the initial conductivity value rt is an approxi

mation, the true formation conductivity ort is equal to

6, plus an error value. This may be expressed in equa

tion form as

5,065,099

11
ort- 6 -- A = oro--A

12

(6a)

In cylindrical geometry, the vector potential due to a

infinitesimal magnetic dipole satisfies the equation:

(6b).

where p is the magnetic permeability, k?=jwn (orjw

and, by two-term Taylor's series expansion

orm = F(oro) + AF(oro)

eoe,), eo is the freespace dielectric constant, e, is the

Equation (4b) can be rewritten as

6F(orc)=F(q)F(oro)

relative dielectric constant, or is the conductivity and J

(4c)

is the dipole loop current. When or or e, are functions of

10 (p, z) the solution to equation (9) cannot be put in a
closed form. However, letting

and equation (6b) can be rewritten as

AF(oro)=ormF(oro)

(6c).

where ko-jwuoro
JW pu is a constant and 3 is a function of

By taking the ratio of equation (6c) to equation (4c), 13 (p, z), equation (9) can be written as
one obtains

V*Ad-i-koAd= p?324d,
AF(oro)

orm F(oro)

(10)

(7)

#FF- = Hi-H

20 The symbol oro for homogeneous medium is chosen for

.
Solving for A,

convenience and has no direct relationship to the use of

the same symbol prior to equation (9). Now, consider

A = 8-1- *----:
T * F(q) F(aro)

F(q) or

(8a)

V?Aq,0)-i-koA(0)=p.J

and substituting 6 = or, or

pri

the Solution to following system of wave equations:

'

(11a)

25

V*A*(*)+koAd)- 32.43,07- )

82

( 1 b)

(8b)

A = F(q) or

for n = 1, 2, . . . . Then, the sum

Equation 8b is computed in process step 329(b) to calcu- 30

late error value A.
Process step 329(c) compares the calculated error

4(n)
4 = .204

value with a threshold error value At. If the calculated

satisfies equation (10). The solution to equations (1 ha)

error value is smaller than the threshold error value A,

and 11(b) are

27 ejkoRO
aikof

(12a)

A$"(p. z) = + O - cos(d0)ado
x:

27

kof

A "(p. z) = + ? c J. pndpndzn?pn, z),4' "(pa, zn) ? + cos(5 m)ddn

(12b)

where

Rop, z) = p + pt 2pp; cos(50) + (2 27)

R*(p. 3) = p + p. - 2pp; cos(b) + (; 30
then process step 329(d) assigns output formation con
ductivity orout to be equal to oro-- A. Otherwise, control
passes to processing step 329(e) which compares itera
tion counter i to a predetermined maximum iteration
count imax. If the iteration counter i is less than the
maximum iteration count imax, then the ite: allon

For a small transmitter dipole, it can be assumed that p?

is small and zT=0, and then equation (12a) becomes

55

4%. , ---o--:
(13)
counter i is incremented in processing step 329(f) and
2Ro
t is set equal to oro-- A. If, on the other hand, the maxi
mum number of iterations has occurred, control goes go where Ro- Vp?--p2. For a small receiver located at
from processing step 329(e) to processing step 329(d)
wherein a value is assigned to the output conductivity z = L and radius pr0, Ad) can be simplified as:
O'out.
(14)
Various models are known which may be used for the A?'(pr. L) =
function F(or). Such models include, for example, a
complex set of Maxwell's equations which, for purposes 65
PRPI (1
of real time processing in a well logging environment,
are impractical to solve. Thus, a model better suited to
real time well logging is needed.
-

#3

5,065,099

13

14

-continued
R

oc

g ?

J. dp/dzi?(pi, zi)GD(pi, zi)(1 jkoRo) (1

oc

where the orthogonality property of Bessel function Ji

where R1l = N p R* + (2 L) and GD(p,* z) =

Lp
2R03R13

has been used. Similar expressions for A39), A49) ect.

can be derived. The general expression is

The vector potential at the receiver is approximated

10

(19)

by

-- oc

ec

2:

J.J.

where 3, -

The above expression represents a primary homogene

ous term with a first order pertubation. The expression

20

rt

In particular, when equation 19 is evaluated at the

receiver, =pr and z=zR= L. Letting prand priend to
zero and using small argument expansion for Ji
QpRp.T

represents a propagated geometric factor which has the

advantage of being symmetric (a sufficient but not nec
essary condition for reciprocity). Physically, the expres
sion represents the effect of wave propagation from the
transmitter to a ground loop through otherwise homo
geneous medium followed by propagation from the
ground loop to the receiver through the same medium.
One can go one step forward and add a second order
pertubation to the vector potential expression in equa
tion (15). The corresponding geometric factor is

(20a) -

[Link] T

(20b)

- oc

35

pre:

and Re(v)-30. Then, the vector potential is

Ro? R23

where R1 is evaluated at (p2, z2). The integral in the

above expression involves radii pland p2, which are not
small and hence, the approximation used in equations
(13) and (14) cannot be used here.
Fortunately, for a horizontally stratified media with
no radial variations, it is possible to derive higher order
vertical geometric factors even though the correspond
ing 2-dimensional geometric factors cannot be ex
pressed in closed forms. For this, the expressions for
vector potential in equations (12a) and (12b) are used.
From the tables of integrals

27 egkorn

J. Rn

(16)

Ji (Ap).J1(Apn)da

where j=VA2ko. Using equation 16, equations

(12a) and 12(b) can be rewritten as

40

45

50

For a homogeneous medium of conductivity or different

from oro the above sum converges if | Cr-Oroi is gro.
Several features of equation (205) are noteworthy.
The expression represents the interaction of n arbitrar
ily placed layers in sequence, each layer being excited
by the currents in the previous layer. It is a function of
Zn, which is the absolute length of the path that the
signal took while traversing n layers of the formation
from the transmitter to the receiver Finally, , Adn),
n=2, 3, can be considered as correction to Ad;(0.

)+Ad(). In particular, the terms for n=4, may be ig

nored, then

The second term in the above expression represents the

(17a)

contribution due to perturbation of the homogeneous

Now assuming that the formation is strictly layered i.e.

32(p, z)={32(z), substitution of equation (17a) in equa
tion (17b) for n = 1 and integration over p1 yields

medium and the third a correction due to the interaction

between two layers and so forth. Previous research

indicates that the first three terms used in equation (21)

5,065,099

15
are sufficient for modeling the response of an induction

tool. However, in presence of highly conductive layers

all four terms may be necessary.

The expressions in equations (20a) and (205) can be

converted to conductivity units by dividing with the
tool constant

10

where to distinguish equation (24) from equation (25)

g2 has been replaced by hy. Further, it is assumed that
the modified second exponential term varies slowly
with respect to A as compared with the rest of the inte
grand. Using arguments similar to those used for the

steepest descent technique, equation (25) becomes

-vOzT-zirizi1849)
O

15

J3

dA

To ensure that the above expression gives the same

value as does equation (24) in a homogeneous formation

it is multiplied by a factor
20

-- x:

where gn is an n-dimensional propagated geometrical

25

factor defined as

The second order contribution to conductivity is

gn(zi, . . . . 2n) =
30

The apparent conductivity ord is

Crn =

35

The above sum converges subject to the same con

straints as equation (21).
The expression in equation (23b) for n = 1 is a simple
convolution, however, for n=2 we have higher dimen
sional convolutions which would require a significant

42

amount of time for evaluation. It becomes essential to

approximate the convolutions for n=2 in such a way

that they can be evaluated fast yet accurately. Consider

equation (19), for n=2, with p1 and p small, z0 = z7 the

transmitter location and z}=zR the receiver location,
Japl.

45

tion. A comparison between cr) obtained by using g.

exc

e.

50

joul

(z1, z2) and h7(z1, z2) is shown in FIG. 6. These are the
2nd-order responses of the conventional deep induction
tool to the Oklahoma Formation. The two results are

which can be rewritten as

g:(zi,z) = ;

Equation (27) is a product of two one-dimensional con

volutions, with the computational complexity of order
O(2N) instead of O(N2) for a two-dimensional convolu

remarkably close.

A3

+.

It is interesting to give a physical interpretation to

According

to

the

Triangular

inequality

|z1 z2|| = | |z2z R |z1 ZR |. For ZR small the

right-hand side of the inequality is of the same order as
the left-hand side if z1 and z2 are of the same sign. In
reality, z1 and z2 may not be of the same sign. Neverthe
less, we proceed by replacing ||z2ZR |z1ZR || by
|z1z2|, Equation (24) now becomes

60

65

equation (27). The integral with go term represents the

attenuation and a phase shift suffered by a plane wave as

it traverses conductivity purtubations of a homogene
ous formation from a point z1 to z2 and back. In a sense
it modifies the conductivity at z1, to include the second
order effects, which then is convolved with another

geometric factor to account for the effects of the adjoin

ing regions. Higher order geometric factors can be
similarly modified to give

5,065,099

17

18
for shoulder effect by applying the filter for the R signal

as disclosed by U.S. Pat. No. 3, 166,709.

The logging system corrects for most of the shoulder

where go has been defined earlier and

effect. Thus, the burden for correcting the shoulder

effect does not fall on the placement and relative
strength of the coils. The fundamental advantage of SY
coil arrays, therefore, is that the depth of investigation,
vertical resolution, and shoulder effect are each con
10

The foregoing formation response model is intended

to provide the most accurate conductivity values given

computational constraints associated with real time well
logging. Other models may also be applied. Addition
ally, techniques performed away from the well site may
employ powerful main frame computer and, thus, im

need for compromise as is the case in known induction

tools. Therefore, it is possible to design an SY type

array that can resolve thin beds and see deeply into the
15 formation.

There are several other alternative embodiments of

the SY type coil arrays that have spatial responses iden

tical to the responses of the design illustrated in FIG. 2.
A second implementation which is identical in result to
the implementation of FIG. 2 discussed above can be
obtained by substituting receiver coils for transmitter
coils and vice versa. Additionally, the bucking coils of
this embodiment would become transmitter bucking
coils as opposed to receiver bucking coils as depicted in
FIG. 2. This coil array would be operated in the same

plement relatively complex response models.

Another possible method of determining an improved
output conductivity value by correcting for shoulder
effects at the surface in the logging truck computer has
been disclosed by T.D. Barber in the article Introduc
tion to the Digital Induction Tool in the Transactions
of the 58th Annual Technical Conference and Exhibition
of the Society of Petroleum Engineers of AIME, SPE

manner as the array disclosed in FIG. 2. While the

12049, 1983. Also, U.S. Pat. No. 4,471,436 and U.S. Pat.

absolute number of turns of the transmitter and the

No. 4,467,425 disclose possible alternative corrections

of this shoulder effect. These references disclose a

method termed Phasor Processing which uses both

the components of the receiver voltage that are in-phase
and out-of-phase with the current in the transmitter.

30

receiver coils might have to be adjusted because of

instrumentation considerations, this would not change
the tools spatial response. By reciprocity, those skilled
in the art will appreciate that this implementation would
have the same response as the implementation illus

These two components are referred to as R and X (sig

trated in FIG. 2.

nal lines 60 and 61 of FIG. 2). U.S. Pat. No. 4,455,529

discloses one method of measuring the R and X signals.

The R and X signals are each filtered with their own
spatial averaging filters. The spatial averaging filters
may be applied in a manner disclosed in U.S. Pat. No.

trolled by different tool design parameters. Since these

design parameters are separate and do not interact ap
preciably, each can be optimized separately without the

35

A third alternative implementation is illustrated in

FIG. 3. In this implementation, there is one transmitter
coil 120 and two main receiver coils 124 and 130, with

four bucking receiver coils 126, 128, 132 and 134. The

3, 166,709. The filtered conductivity signal at a particu main receiver coils 124 and 130 are located symmetri
lar depth is obtained by taking a weighted average of cally about the transmitter coil 120 with the bucking
the readings surrounding the depth of interest. The 40 receiver coils 126, 128 and 132, 134 symmetrically ar

ranged about the main receivers 120 and 130. Again, the
main receiver coils 124 and 130 are spaced by a distance

weights in the filter for the R signal are chosen to cor

rect for the shoulder effect in resistive formations (in the

limit of no skin effect). The X signal weights are chosen

to match the X signal to the error in the filtered R signal
caused by skin effect. The level of the filtered X signal

D from the transmitter 120 and the transmitter 120 is

45

is then corrected so that it matches the level of error in

located at a selected point in the center portion of the

insulated mandrel 112. Also, the bucking receiver coils
126, 128 and 132, 134 are symmetrically positioned at a

the R signal due to skin effect for a homogeneous for

mation. This possible correction technique for skin ef

distance d from their respective main receiver coils 124

and 130 as shown. As can be seen, array 110 is symmet

R signal. The phasor processing algorithm can be sum

is connected to oscillator circuit 102 and that the two

marized by

bucking receiver coils 126, 128 and 132, 134 are con
nected to two separate receiver circuits 104 and 106 that
combine the recorded signal with the phase reference

ric about the transmitter coil 120. The schematic dia

fect is disclosed in U.S. Pat. No. 3,147,429. The filtered 50 gram
of FIG. 3 illustrates how the transmitter coil 120
and level adjusted X signal is then added to the filtered

rt

r:

ori

main receiver coils 124 and 130 with their respective

55

where or 4 is the corrected log, or R is the R conductivity

signal, orx is the X conductivity signal, the coefficients

ajare the weights of the R signal spatial averaging filter,

the coefficients by are the coefficients of the X signal

60

special averaging filter, and q is the skin effect correc

tion function. This phasor processing may be accom
plished in the computer 32 of FIG. 1.
Another method to correct for shoulder effect and 65

skin effect would be to operate the induction tool at

sufficiently low frequency such that skin effect is not a
problem. In this case, the log signal could be corrected

signal on line 103 produce the in-phase and out-of-phase

signals on lines 140, 141, 142 and 143 respectively. In
the preferred embodiment, these receiver circuit signals
are digitized and then set to the computer 32 (see FIG.
1) at the surface. Since these two receiver coil arrays
are at different depths, they will respond to different
portions of the formation at different times as the tool
moves through the borehole. The in-phase and out-of
phase signals from the upper receiver coil 124 are re
corded and delayed in depth by the computer at the
surface so that these signals will align with the measure
ments from the lower receiver coil 130. The spacing
between the two receiver coils 124 and 130 is 2D. The

5,065,099

19
depth-shifted measurements from the upper receiver
coil 124 is then averaged with the measurement from
the lower receiver coil 130. According to the principle
of superposition, the averaged measurement is identical
to the measurement made by the receiver coil in the first
implementation illustrated in FIG. 2. This is because, in
the first implementation, the receiver array is symmetri

20
into the formation than will be experienced with trans
mitter coil pair 246 and receiver 248. In this manner,

bed 240 may be investigated to two different depths of

investigation, resulting in measurements of invasion
zone 238 and uninvaded zone 239.

The voltages induced in receiver coil 248 are coupled

through phase sensitive receiver circuit 218 to com
puter 232 located at the earth's surface through logging

cal and so the effects of the transmitter coils can be

considered separately. The conditions for mutual bal

ance and the responses are identical to those of the

10

implementation depicted in FIG. 1.

Of course, a fourth implementation could also be
derived from the implementation of FIG. 3 by including
a single receiver and two main transmitter coils with
each main transmitter coil including two bucking trans
mitter coils. This embodiment is more difficult to imple
ment however, because the two arrays of transmitter
coils must be multiplexed in some manner so that the

cable 228. These received signals are utilized to provide

information concerning the resistivity of the formation

surrounding borehole 208. Computer 232 is preferably

coupled to a power source 230 which provides electri
cal power to downhole electronic package 214 through
logging cable 228.
15

A winch 224 is also utilized in conjunction with log

ging cable 228 to raise and lower dual high resolution
induction tool 210 into and out of borehole 208. In a

preferred embodiment of the present invention, winch

224 provides a depth signal on line 226 to computer 232
to provide a depth correlation for a log plot. Computer
232 receives the output of receiver coil 248 and is uti
lized to provide shoulder and skin effect correction

signal depicted by the single receiver due to each of the

20
arrays of transmitters can be measured separately.
These third and fourth implementations of the SY
type coil arrays are particular cases of any of several
different implementations in which the coil arrays are computations as discussed above or in any other manner
separated into different combinations consisting of well known in the art to compute a corrected log signal
transmitter-receiver pairs. The portion of the signals 25 which is then coupled to recorder 234. In the depicted
from each of the constituent coil pair is measured sepa embodiment of the present invention, recorder 234 pref
rately, and then all the measurements are combined
erably produces a well log 236 which depicts the cor
(after possible depth-shifting) in order to recreate the rected log signal versus depth of the tool in borehole
total SY coil tool response. Therefore, all such imple 208.
mentations should give identical responses to the em 30 FIGS. 8 and 9 together illustrate one preferred em
bodiment illustrated in FIG. 2.
bodiment of the present invention. Two pair of trans
With reference now to FIG. 7, there is depicted a mitter coils 244 and 246 are preferably disposed sym
block diagram of dual high resolution induction tool metrically about a selected point on an insulated sonde
210 of another embodiment of the present invention. 212. Each of the transmitter coils includes N, turns that
Those skilled in the art will appreciate that the dimen 35 are wound in the same sense. A main receiver coil 248
sions of FIG. 7 have been exaggerated for the purposes with Nrm turns is located at a selected point on insulated
of illustration. As can be seen, insulated sonde 212 and
Sonde 212. Also located on mandrel 212 and disposed
downhole electronics package 214 are suspended in a Symmetrically about receiver coil 248 are two bucking
borehole 208 by means of a multiconductor logging receiver coils 250, each consisting of Nb turns with a
cable 228. Logging cable 228 is preferably coupled to tap located in each bucking coil 250. These bucking
downhole electronics package 214 and preferably in coils are wound in series with receiver coil 248 and are
cludes both electrical power and signal conductors.
wound in opposite sense to receiver coil 248. Bucking
Downhole electronics package 214 includes transmit coils 250 are so named because their purpose to buck
ter 216 and phase sensitive receiver circuit 218. Those out the large mutual coupling from transmitter coil pair
skilled in the art will appreciate that transmitter 216 will 45 244 and transmitter coil pair 246 and to cancel any
typically include an oscillator circuit which will pro signals coming from that portion of the formation near
vide a phase reference signal to phase sensitive receiver est the coil array. Bucking coils 250 are each spaced
circuit 218 via line 217. Transmitter 216 is alternately from the main receiver coil 248 by a selected distance,
coupled via selective control 242 to either transmitter which, as discussed above, will control the vertical
coil pair 244 or transmitter coil pair 246. As those 50 resolution of dual high resolution induction tool 210.
skilled in the art of induction tools will appreciate, an Unlike known bucking coils, bucking coils 250 are de
alternating frequency signal generated by transmitter picted including taps which may be utilized in conjunc
216 and coupled to transmitter coil pair 244 or transmit tion with electronic switches (not shown) to carry the
ter coil pair 246 will inductively produce an electro effective number of turns in each coil. This is necessary
magnetic field which will cause eddy currents to flow 55 due to the variance in mutual signal present whether
within the formation. These eddy currents produce a utilizing transmitter coil pair 244 or transmitter coil pair
Secondary magnetic field which will induce a voltage in 246. In this manner, bucking coils 250 may be accu
receiver coil 248. Bucking coils 250 are utilized, in a rately varied to optimize the isolation of receiver coil
manner which will be explained in detail herein, to 248 whether utilizing transmitter coil pair 244 or trans
isolate receiver coil 248 from the large mutual signal mitter coil pair 246 and as a consequence, receiver coil
which might otherwise be coupled from transmitter coil 248 may be alternately utilized in conjunction with two
pair 244 or 246.
different transmitter pairs. The output of receiver coils
As will be apparent to those practitioners in this art, 248 is then coupled as depicted in FIG. 5 via all or part
the spacing between transmitter coil pairs 244 or 246 of bucking coils 250 and shielded line 256 to amplifiers
and receiver coil 248 will directly effect the depth of 65 252 and 254 to produce a deep receiver signal (DR) and
investigation demonstrated by dual high resolution in a medium depth receiver signal (MR).
duction tool 210. Thus, transmitter coil pair 244 and
Those skilled in the art will appreciate that bucking
receiver coil 248 will result in a deeper investigation coils 250 will typically include fewer turns than re

5,065,099

21
ceiver coil 248 because bucking coils 250 are located
closer to transmitter coil pair 244 and transmitter pair
246, making their mutual coupling per turn greater.
Thus, as transmitter 216 provides an alternating current
to each transmitter coil the resultant induced voltage in
the receiver coil array (coils 248 and 250) from the

22
this type are rationetric and it is necessary therefore to

have a reference voltage against which to measure the

received signal.

Calibration/zero switch 270 is also controlled by

formation will include two components. One compo

ment is the in-phase induced voltage and the second

component is the out-of-phase induced voltage, where

the phase is relative to the current in transmitter coil
pair 244 or transmitter pair 246. Phase sensitive receiver
circuit 218 receives the phase reference signal via line
217 from transmitter 216 and produces an in-phase sig
nal and an out-of-phase signal for recording and analy
SiS.

As can be seen, the coil array of the dual high resolu

tion induction tool 210 of the present invention is com
pletely symmetrical about main receiver coil 248. Fur
ther, both transmitter coil pair 244 and transmitter coil
pair 246 are also symmetrical about receiver coil 248.
This arrangement differs from generally known coil
arrays which are reciprocally symmetrical, that is, the
transmitter array is a mirror image of the receiver array.
The basic theory of this symmetrical array in a single
induction tool may be readily understood upon refer
ence to the aforementioned referenced application.

10

Measurement/calibration switch 274 may then be uti

15

20

25

With reference now to FIG. 8, there is depicted a

simplified schematic diagram of one embodiment of

dual high resolution induction tool 210 of the present
invention. As can be seen, dual high resolution induc
tion tool 210 preferably includes a processor 258 which
is utilized to control and order the various operations

30

lized, as will be obvious to those skilled in the art upon

reference to FIG. 6, to couple one of four inputs to the
receiver electronics: a calibration voltage signal; a zero
voltage signal; a deep receiver signal (DR); or a medium

depth receiver signal (MR).

In the embodiment of the present invention disclosed
in FIG. 8, processor 258 may then be utilized to alter
nately couple one of these four signals to the receiver
electronic circuitry. Those skilled in the art of induction
tools will realize upon reference to the foregoing that
alternate embodiments of the present invention are pos
sible in which the deep transmitter coil pair and medium
depth transmitter coil pair are simultaneously energized
at two different frequencies. Similarly, the deep trans
mitter coil pair and medium depth transmitter coil pair
may be energized at identical frequencies with a modu

lation signal incorporated to permit the receiver cir

cuitry to distinguish the source of the transmitted signal.
The receiver electronics circuitry of the present in
vention is then utilized to couple the receiver signal

through a low noise preamplifier 276. Next, low phase

necessary to operate dual high resolution tool 210. Pro

cessors 258 will control, in a manner well known in the

art, waveform generator 260 which generates the alter

processor 258 and is utilized to provide a calibration

signal or zero signal to the input of the receiver elec
tronics in a manner which will be explained in greater
detail below. As can be seen, the deep receiver signal
(DR) and the medium depth receiver signal (MR) are
alternately coupled to the receiver electronics via
DR/MR switch 272 under the control of processor 258.

35

nating frequency signal which will be utilized to induc

tively produce an electromagnetic field which will in
turn cause eddy currents to flow within the formation.
Waveform generator 260 is also utilized to provide
phase reference signals fe and fe--T/2 which will be
utilized in conjunction with the phase sensitive receiver
circuit.

The output of waveform generator 260 is then cou

pled via high-power isolation switches 264 through
amplifiers 262 to transformer 266. Transformer 266, in 45
conjunction with high-power isolation switches 264, is
then utilized to couple the output of waveform genera
tor 260 to either transmitter coil pair 244 or transmitter
coil pair 246. High-power isolation switches 264 are
utilized to ensure that cross talk will be minimized by 50
holding either transmitter coil pair 244 or transmitter
coil pair 246 at an open circuit when not energized. This
is necessary when utilizing alternate measurements of
two different depths of investigation to prevent cross
talk caused by large circulating currents which may be 55
induced in one transmitter coil pair when the other
transmitter coil pair is energized. In actual practice, the
applicants have discovered that transmitter coil pair
244, the deep investigation coil pair will induce large
circulating currents in transmitter coil pair 246 when
energized. Therefore, one alternate embodiment of the
present invention may utilize a high-power isolation
switch 264 only to isolate transmitter coil pair 246 dur
ing energization of transmitter coil pair 244. Also cou
pled to the secondary of transformer 266 is a winding 65
including reference resistor 268 which is utilized to
generate a reference voltage. Those skilled in the art
will appreciate that all measurements utilized in tools of

shift band pass filter 278 is utilized to further prepare the

received signal for analysis. Amplifiers 280 are then
utilized to further amplify the received signal and the
in-phase and out-of-phase components are then applied
to phase sensitive detectors 282 which, in conjunction
with the phase reference signal provided by waveform
generator 260, will output a signal having a DC compo
nent and a large number of harmonics. Low pass filters
284 are utilized to attenuate these harmonics to an insig
nificant level. Low pass filters 284 are preferably fast
settling filters which permit rapid sampling of received
data. The D.C. output of each low pass filter 284 is then
applied to an analog-to-digital converter 286 and the
resultant digital signal is coupled via data bus wires 288
to processor 258. Processor 258 may then, in a manner

well known in the art, couple the measured data

through an interface module 290 onto logging cable 228
for transmission to the earth's surface for correction and

computation by computer 232 (see FIG. 7).

Thus, it should be appreciated by those skilled in the
art that the applicants have developed a dual high reso
lution induction tool which utilizes only seven coils and
which permits investigation of a formation to multiple
depths without degradation of the vertical resolution of
either coil array. When compared with known dual
induction arrays which utilize ten or eleven coils the
reliability and accuracy of this tool will be far superior.
Although the invention has been described with ref
erence to a specific embodiment, this description is not
meant to be construed in a limiting sense. Various modi
fications of the disclosed embodiment as well as alterna

tive embodiments of the invention will become appar

ent to persons skilled in the art upon reference to the
description of the invention. For example, it is possible
to implement the dual high resolution induction tool of

23

5,065,099

10

24
being connected in series with said main coil,
wherein said first pair of coils are wound in op
posite sense to the main coil, one of said first pair
of coils being positioned a distance d above the
main coil and the other of said first pair of coils
being positioned the distance d below the main
coil; and
an outboard coil system having:
a second pair of coils, each coil of said second
pair of coils having the same number of wind

15

positioned a distance D above the main coil,

and the other of said second pair of coils being
positioned the distance D below the main coil,
wherein the distance D is greater than the

the present invention by exchanging receiver coils and

transmitter coils. Thus by utilizing a single transmitter

coil with symmetrically disposed variable transmitter
bucking coils and multiple symmetrically disposed re
ceiver coil pairs it will also be possible to implement
two transmitter receiver arrays which will have differ
ent transmitter-to-receiver coil spacing (and thus differ
ent depths of investigation) while maintaining substan
tially identical vertical resolution for both arrays. It is
therefore contemplated that the appended claims will
cover any such modifications or embodiments that fall
within the true scope of the invention.
What is claimed is:
1. An induction logging tool comprising:
a mandrel having an axis of elongation;
a central receiver coil disposed on said elongated
mandrel and wound about the axis of elongation;

a pair of bucking coils connected in series with said

receiver coil, said bucking coils being symmetri
cally positioned about said receiver coil on said

ings, one of said second pair of coils being

distance d,

(c) a transmitted signal source coupled with one of

said central coil array and said outboard coil Sys
tem; and
20

mandrel at substantially equal distances from Said

receiver coil and wound about the axis of elonga
tion in opposite sense to said receiver coil;
a pair of transmitter coils symmetrically positioned

coil array and said outboard coil system.

8. An induction logging tool comprising:

an insulated Sonde;

about said receiver coil on said mandrel at substan

tially equal distances from said receiver coil, said

bucking coils being positioned between said re
ceiver coil and said transmitter coils;

a transmission signal source operatively connected

with said pair of transmitter coils for coupling a
transmission signal to said pair of transmitter coils;

30

and

receiver means coupled to said receiver coil for de

tecting a voltage induced therein and producing an
output detection signal related to said induced volt
age.

(d) a receiver for coupling to the other of said central

a coil array including a plurality of coils symmetri
cally disposed about a selected point on said insu
lated sonde wherein identical pairs of coils within
said coil array are disposed on opposite sides of said
selected point at substantially equal distance from
said selected point, said plurality of coils compris
ing
at least one central coil disposed at said selected
point,
a first set of three coils arranged above said at least
one central coil, the coils of said first set of coils

35

being wound in series and disposed symmetri

cally about one of said first set of three coils, and
a second set of coils arranged below said at least
one central coil, the coils of said second set being
wound in series and disposed symmetrically

2. The induction logging tool of claim 1, wherein

each of the pair of bucking coils has the same number of
windings.
about one of said second set of three coils;
3. The induction logging tool of claim 2, wherein 40
a transmission signal source coupled to said at least
each of the pair of transmitter coils has the same number
of windings.
one central coil; and
4. The induction logging tool of claim 3, wherein
receiver means coupled to at least one of said first set
of coils and said second set of three coils for detect
each of the pair of bucking coils has the same number of
windings.
45
ing voltages induced therein.
9. A method of logging an earth formation with a
5. The induction logging tool of claim 1, wherein said
transmission signal source provides a periodic current Sonde adapted to move through a borehole comprising
signal and wherein said receiver means includes means the steps of:
for detecting the phase of said periodic current trans
positioning the sonde in the borehole;
mission signal source and further includes means for 50 applying a first excitation signal to a first pair of trans
providing an output including an in-phase detection
mitter coils positioned symmetrically on a principal
signal and an out-of-phase detection signal.
axis of the borehole, wherein each of said coils are
6. The induction logging tool of claim 1, further com
spaced a first, equal distance from a reference point
prising a data processing means for receiving detected
on the Sonde;
voltages from the receiving means and providing an 55 detecting the electromagnetic effect of said first exci
output signal representative of formation characteris
tation signal at a receiver coil array to obtain indi
cations of characteristics of the formation at a first
tics, and recording means coupled to said data process
ing means for recording the output of said data process
radial depth of investigation from the borehole,
ing means.
said receiver coil array including main windings
60
and bucking windings, said bucking windings being
7. An induction logging tool comprising:
(a) an insulating mandrel having a vertical axis of
wound in an opposite sense from the main windings
elongation;
and positioned symmetrically about said reference
point;
(b) a symmetrical coil system disposed on said insulat
ing mandrel, said coil system including:
selectively coupling a receiver means for detecting
65
a central coil array having:
voltages in the receiver coil array to equal numbers
of Said bucking windings spaced an equal distance
a main coil; and
on either side of the main winding, the number of
a first pair of coils, each coil of said first pair of
windings of said bucking coils being selected to
coils having the same number of windings and