Note: Descriptions are shown in the official language in which they were submitted.
7~7
20 ,1671
BACKGROUND OF THE INVENTION
2 ''.
8 ' This invention relates to well logging methods and a~pa-
4 Iratus, and more particularly, to methods and apparatus for pro-
6 ',ducing measurements from multiple transducer arrays and combining
6 ¦!them to provide compensation for variations in instrumentation and
7 ~Iborehole conditions.
8 jl It is well known in the art of acoustic logging that some
9 Idegree of compensation for variations in travel time introduced I -
'by changes in borehole diameter may be provided by a borehole tool
11 ''that includes two receivers and one transmitter. (Such arrays will
12 ''be abbreviated hereinafter by using a "T" to represent a trans-
13 Imitter and an "R" to represent a receiver with the relative posi-
14 'Itions at the T's and R's indicated by the sequence, the hyphen
i~ .
,"-" separating the transducers indicating a common signal path.
16 I'Thus a T-RR array specifies a transmitter on one side of two re-
17 '¦ceivers, with the receivers having in common the signal path be-
18 'Itween the transmitter and its nearest receiver.) Unfortunately,
19 ,such a T-RR arrangement does not compensate for the tilt of the
Itool with respect to the axis of the borehole. To overcome the
21 ,tilt problem an additional transmitter may be provided to form a
22 ,tool that has a T-RR-T array. As described in U.S. Patent No.
a3 '3,257,639 i-ssued to F. P. Kokesh on June 21, 1966, each of the
24 two transmitters may be selectively operated and the travel time
26 'to each of the two receivers measured. The individual travel time
26 measurements may then be combined to produce an average travel
27 time for the interval between the two receivers. That average
28 time has the advantage of being compensated for both changes in
29 borehole diameter and tilt o~ the tool. ~ -
81 - 2 -
æ
, 20.1671
.1 ,
1 ~¦ As with many different types of measurements under condi-
2 ¦tions varying non-homogeneously in a direction radial to the bore-
8 jhole, acoustic measurements appear to vary with distance between
4 'transmitter and receiver or, more appropriately for two-receiver
6 larrays~ with the distance between the transmitter and a point
6 ; midway between the receivers. It is for this reason that the
7 borehole-compensating type tools have two transmitters located
8 equidistant from that mid-point.
As recognized ln U.S. Patent No. 3,312,934 issued April ~,
0 , 1967 to A. A. Stripling, one reason why the acoustic velocity may
ary with different T-R distances is that different signal paths
~2 l¦may result with the longer distance having a path somewhat farther
13 ; from the borehole and deeper into the formation. This deeper path
~4 ~ may be less affected by factors which radially alter acoustic
, properties when drilled or exposed to the borehole fluid, such as
16 `I hydrophilic shales which tend to swell. This altered zone may
17 ,l exist deep enough into the formation to cause a short T-R dis-
18 ,~ tance to measure, at least in part~ properties representative of
19 this altered zone, rather than the desired unaltered formation.
` Longer T-R distances, such as 8 or 10 feet, are preferred to
21 overcome this particular formation alteration problem.
22 ,¦ Longer T-R distances require longer tools, and in the
23 ¦older two-receiver type arrays, i.e., those of the T-RR type, an
24 '¦increase from 3 to 8 feet in T-R distance requires a 5 foot
i longer tool. However, in the T-RR-T borehole compensating tools,
26 llsuch a substantial increase in T-R distance results in undesir-
27 lably long tools since the T-R distance occurs twice. Longertools
28 iiare undesirable since their length makes them more expensive and
29 Idifficult to transport and increases the problem associated with
lgetting them down crooked or inclined boreholes.
81 l
32 ' ~ 3 ~
c~ o./~/
~-3-&~
~ 7 ~ ~
Borehole compensating type arrays are also employed in
sidewall devices such as disclosed in U.S. Patent No. 3,849,721,
issued to T. J. Calvert on November 19, 1974. Here longer T-R
distances in the prior art T-RR-T array increase skid length,
which unfortunately decreases the chances of keeping the skid
in continuous contact with the borehole wall.
Previous approaches to providing at least a partial
compensating system without unduly long tool lengths, such as
described in U.S. Patent No. 3,207,256 issued to R. B. Blizard
on September 2, 1965 or U.S. Patent No. 3,330,374 issued to
D. E. Broussard et al. on July 11, 1967, require memorization
of at least two different measurements for at least two
different distances. This requirement leads to additional
memory costs and more vulnerability to depth positioning
problems such as introduced by a yo-yo motion of the tool.
Further, the compensation for tool tilt is not always complete.
A further problem with either the T-RR-T or its
reciprocal, the R-TT-R array, is that because of the large
distances between the T's in the T-RR-T array or the R's in the
R-TT-R array, the operating conditions for transducers located
at the ends of the array may be quite different, resulting in
significant differences in the received signals which are
presumed to be equal. For example, if severe tool tilt places
one of the outer transducers in a substantially eccentered
position while the like corresponding transducer at the other
end of the tool remains more centered, signals associated with
these outer transducers could vary considerably and, in turn,
could affect both the travel time and the amplitude
measurements.
.
'7
When a T is between a pair of R's or an R between a
pair of T'S, there is often a problem with electrical noise, as
for example with electrical cross-talk from an electrically
noisy transmitter circuit into receiver leads which must pass
close to the transmitter or still worse, from a transmitter
firing lead having high voltage and current transients, as in
the case of acoustic tools, which pass by one of the receivers
or received signal amplifiers. For example, the firing pulse
leads going to the bottom transmitter in the T-RR-T array must
pass by both receivers. A further appreciation of the
electrical and mechanical problems introduced by transmitter
leads passing receivers may be found in U.S. Patents 3,734,233
and 3,712,410. It would be highly desirab]e to have a
compensating array where all receivers could be isolated from
all transmitters and further, where no high voltage pulse leads
pass anywhere near a receiver, its associated amplifier or
receiver signal lines.
In prior art compensation type arrays and in some two
receiver arrays, it was not possible to obtain measurements
over some parts of the borehole. For example, in the T-RR-T
array, the tool might not operate properly with the upper
transmitter inside the casing and the remaining transmitter and
both receivers still out in the open hole. Similarly,
measurements of the formation could not be made in the critical
bottom part of the hole for a distance corresponding to at
least one transmitter-receiver distance. It would be desirable
to be able to log as close to the bottom as possible even if it
were necessary to temporarily forego the compensation feature
for this interval.
,~ .
In acoustic tools which may be required to operate at
some distance from the borehole wall, the acoustic signals
arriving at a given receiver effectively leave the borehole
wall at a point ahead of the receiver, the displacement of the
point varying with the approach direction. This gives rise to
what is known as a refraction error. This error and one
correction technique for compensating type arrays, as described
in U.S. Patent No. 3,304,536, issued February 14, 1967, to
F. P. Kokesh, and U.S. Patent No. 3,524,162, issued August 11,
1970, to F. W. Zill, involves the use of an additional receiver
with each of the two existing receivers. Each additional
receiver is spaced from each existing receiver by a small
distance corresponding approximately to twice the displacement
introduced by the refraction error--one displacement for each
of the two different reception directions. Thus, four
receivers are used, two for each reception direction.
Furtber, in the T-RR-T type array, omnidirectionally
receivers are required since each receiver must anticipate
signals arriving from either the upper or lower transmitter.
Highly desirable directional receivers cannot be used unless
four receivers are employed, as in the above refraction
correction approach; i.e., two receivers directed towards each
of the transmitters. Another approach would be to use the
R-TT-R type array, but now desirable directional transmitters
cannot be used unless, as with the four-receiver approach, four
transmitters are employed. Needless to say, these extra
transmitters add considerable control complexity and expense.
Two different T-R distance investigations are
desirable and, as described in the above discussed 3,312,934
patent, it is possible the close comparison of the different
--6--
. .
investigations may even lead to a direct indication of the
presence of hydrocarbons when it occurs in the form of gas, or
in some cases, to estimate the degree of shaliness as suggested
in U.S. Patent 3,096,502, issued to C. B. Vogel on July 2,
1963. It should be apparent that in order for measurements
having different T-R distances to be useful in these
applications, the measurements must be as accurate as
possible. The accuracy of the T-RR type of measurements
illustrated in the above patents is often such that the
observed differences in these different investigations may
actually be due to uncompensated tilt or system measurement
errors, rather than radial differences in the acoustic
properties of the formations. At least two measurements are
required for this application, and it is important that both of
these measurements be borehole compensated.
When using prior art compensation type arrays to
obtain the different T-R distance investigations, two
additional outside transducers at an additional distance beyond
those usually provided and a large number of additional
measurement subcycles beyond the four normally employed would
be necessary. Furthermore, the tool length would be increased
by twice the desired difference in distance. Such requirements
for additional transducers and tool length render the second
measurement impractical under many circumstances, since the
second measurement is usually redundant to the first
measurement. However, if it could be provided without such
costly complications, this second measurement would increase
the value of the primary measurement by providing substantial
assurances that at leas the longer T-R distance was adequate
for altered formations and when favorable conditions did occur,
would provide a direct indication of the presence of gas.
7!~t7
.SUMMARY OF THE INVENTION
It is therefore a genera]. object of the present
invention to provide method and apparatus which retain both the
advantages of l.ong T-R distances and borehole compensation
without requiring unduly long borehole tools.
The foregoing and other objects are obtained, in
accordance with one aspect of the invention, by comprising a
method of producing measurements adapted for determining a
compensated measurement of a physical characteristic of
subsurface media near a borehole penetrating the earth
employing multiple transmitter and receiver type transducers
supported on a support member elongated in a direction
generally parallel to said borehole for movement through said
borehole, comprising the steps of: providing a first pair of
transducers of a first type positioned at a preselected
separation along said member; providing a second pair of
transducers of a second type positioned at said preselected
separation along said member and located on one side of said
first pair of transducers in the direction of said elongation;
producing a first measurement of said physical characteristic
of the subsurface media when two of said transducers are
positioned at a selected position in said borehole; selecting
the transducers of said first and second pair to have,
respectively, substantially the same operating characteristics;
storing said first measurement for combination with a later
measurement of said physical characteristic of the subsurface
media; and producing said later measurement when two other of
said transducers are effectively positioned in said borehole at
-8-
: .
.
. . ~ ~ ~ . . .
~1'7~7
said selected position for combining with said first
measurement to produce a measurement compensated for
misalignment of said support member with said borehole and/or
var;ations of said borehole.
Another aspect of the invention is attained by an
apparatus for determining a physical characteristic of
subsurface media near a borehole penetrating the earth which
employs multiple transmitter and receiver type transducers
supported along a support member adapted for movement through
said borehole and elongated generally along a direction
parallel to its direction of movement through said borehole,
comprising: a first group of transducers of a first type
supported for movement through said borehole with adjacent
transducers of said first group being separated from each other
by a preselected separation along a line generally parallel to
the elongated direction of said support member; a second group
of transducers of a second type supported for movement through
said borehole and located on one side of said first group in a
direction therefrom parallel to said elongated direction with
adjacent transducers of said second group being separated from
each other by said preselected separation along said line; the
transducers of said first and second group, respectively,
having common operating characteristics; means for producing
measurements of said physical characteristic of subsurface
media at different depths of the support member in said
borehole; and means for combining said measurements taken at
different selected depths of the support member in said
borehole to provide compensation for variations in the borehole
and/or misalignment of transducers therein.
_g_
1~17.97
In accordance with a more specific aspect of the
invention, methods and apparatus are provided for using a
borehole tool to produce measurements of physical
characteristics of subsurface media near a borehole penetrating
the earth and to obtain improved measurements that are
compensated both for changes in the borehole diameter and for
misalignment of the borehole tool with the borehole. A number
of transducers are positioned along either the tool or a
support member included within the tool and are supported in a
line generally parallel to the movement of the tool when in the
borehole.
As used herein, the term "transducer" means a device
that is capable of either transmitting or receiving a
particular type of signal. For example, in acoustic
measurements the transducer may be either an acoustic
transmitter or an acoustic receiver, the transmitter serving to
convert electrical energy into mechanical or acoustic energy
and the receiver to convert acoustic energy into electrical
energy. Similarly, in electromagnetic wave measurements, the
transducer may be an antenna or radiator of electromagnetic
waves while the receiver may be an antenna for detecting
transmitted electromagnetic waves.
A number of transducers for a first type, such as
transmitters, are separated from each other along the tool by a
preselected separation, and a number of transducers of a second
type, such as receivers, are separated from each other by the
same separation and positioned a preselected distance on the
tool from the transducers of the first type.
--10--
.
.
- ` 20.1671
~ 7
., i
1 The distance between the two same-type transducer groups
2 may be as long as desired. An array constructed as above using
8 , transducers of the ~irst type capable of operating as Tls and
4 ~ of the second type as R's may be described as a TT-RR array.
~ ll Measurements produced at selected borehole depths between dif-
6 ~ ferent transmitter-receiver combin~tions as the transducer array
is moved through the borehole may be combined to produce compen-
i! sated ~easurements-
9 ,; For example, with the above novel '~T-RR array, ~wo measure-
i' ments with the same T-R spacing that use~ different T-R combina- !
~1 ` tions are possible, since the separation between each receiver
~2 ¦ pair equals the separation between each transmitter pair. If
13 ~ one measurement is made with a ~irst T-R pair at a selected
14 'j borehole depth and a second measurement is made when a second
T-R pair has moved to the same depth, the two measurements may
~6 be combined to provide a measurement that is compensated for
17 !. variations between the characteristics of the transducers and
18 other systematic errors.
i9 ~ Furthermore, di~ferential measurements between one trans-
mitter and two receivers when the two receivers are adjacent
21 a selected interval in the borehole at a selected depth can be
22 repeated when the two transmitters are adjacent the interval
23 and all of the measurements combined to produce a borehole
24 compensated measurement for the interval; i.e., a measurement
26 ~ th~t is compensated for tool tilt, borehole eccentricity, etc.
26 Moreover, due to the arrangement of transducers in the
27 array, different measurements that are compensated for borehole
28 errors can be obtained for two different T-R investigation
29 distancesj i.e., a long ~-R measurement and a short T-R measure-
ment. In each of the long- and short- T-R measurements, each
81 ~ transducer in one group respectively spaced the longest and ~r~t
æ
-- 11 --
~` 20.1671
.1 1 .
,i i
1 distance from the other group is used, and in such use, may be
2 I regarded separately as the long and short spacing transducer in
8 ~l each group, that is, the transducers in each group that are,
4 I respectively, the longest and the shortest distance from the
6 j~ other group.
6 ¦~ Since the groups of same-type transducers are closely
7 , positioned on the tool, the operating environment, direction of
8 j signal propagation and the refraction error are essentially the
9 '' same for either of the transducers in a given group during a
~! given measurement As a result, directional receivers and
11 li transmitters may be used, thereby improving the quality of the
12 1 measurements obtained.
13 ll While acoustic and electromagnetic transducers are illus-
14 , trated in centralized and sidewall skid configurations, this
Il invention applies as well to other types of measurements using
16 ~l at least four transducers, operated either in a pulsed or con-
lq ~' tinuous mode while making one or more measurements such as
18 ~ travel time, phase angle, amplitude ratio or attenuation-like
19 measurements.
,
21~ ,
22 ''~
25 ,,
2~ i .
26 .
2q
28 ,
29
31 - 12 -
32
7~
8RIEF DESCRIPTION OF THE DRAWINGS
Further features and advantages of the invention will
become more readily apparent from the following detailed
description when taken in conjunction with the accompanying
drawings.
FIG. 1 shows, in representative block form, apparatus
in accordance with the present invention for acquiring, storing
and combining measurements of physical characteristics of
subsurface media near a borehole;
FIGS. 2A through 2C show the measurement acquisition
sequence using the transducers included in the apparatus of
FIG. l;
FIGS. 3A and 3B show surface and downhole circuitry in
accordance with one embodiment of the present invention; and
FIG. 3C shows the corresponding timing diagram.
FIGS. 4A through 4D show the effect of borehole
conditions such as misalignment and tilt of a transducer
support on the techniques of the present invention;
FIGS. 5A and 5B show different borehole operating
environments that are compensated for in accordance with the
present invention;
FIG. 6A shows a prior art skid mounted borehole
compensation array and FIG. 6B shows the prior art array
modified in accordance with the present invention;
FIGS. 7A and 7B show further advantages of a
transducer array modified in accordance with the present
invention;
FIGS. 8A and 8B show alternative circuitry for use
with circuit 24 of FIG. 3A;
FIGS. 9A and 9B show the relationship between
different measurements of physical characteristics of
subsurface media; and
FIGS. 10 and 11 show further embodiments of apparatus
in accordance with the present invention.
-13A-
20.1671
~ 7 ~7
1 ~i DESCRIPTION OF THE PRE~ERRED EM3ODIMENT
.1 .
8 '~ Refe~ring to FIG. 1, there is shown in representative
4 ~ block form suitable apparatus in accordance with the present
5 ii invention for acquiring, storing, recording and combining
6 1~ measurements of physical characteristics of subsurface media
7 '~ near a borehole that penet~ates an earth formation. The appara-
8 ¦ tus includes a borehole tool 10 with a transducer array having
g j four transducers numbered 1 through 4. ~he array may be in-
0 ¦I cluded in a tool that is either a mandrel type adapted for
~ centralized or eccentered operation or in a skid member, with the
12 " transducers located on the skid for operation in close contact
13 with the borehole wall.
4 1 ~he explanation which follows assumes that the tool has
1~ " been run to the bottom of the borehole so that it can then be
~6 ll retrieved slowly towards the surface under mechanical control
of a logging wirelin~2 wound on a winch 14 at the surface, which
t8 ~ also provides signal and control communication between the tocl
19 ` and surface control 20. In this manner, movement of the
20 !~ tool may be directly related to the movement of thewirelineat the
21 " surface.
æ, The surface controller 20 acts as a progra~med trans-
23 ~ mitter and receiver selector, which communicates through a
24 slipring collector 16 on the winch 14 to the logging.wi~e~ ~ 22~ ` and downhole to subsurface control 11 in the tool 10. Synch-
26 ronously with the wire~mOvement, incremental depth pulses are2q provided to both the controller 20 and to the measurement
28 storage apparatus 22 through any appropriate mechanical or
29
31 - 14 -
æ
~ ` 20.1671
7 ~ 7
., I
i
1 ll electrical connection 18 and, if present at the well site so that~
2 the measurement processing may be done at the same time, to the 1 -
8 jl measurement selection and combination apparatus 24. In this
4 ,I manner, the transducer selection and corresponding measurements
6 il, are synchronized as described hereinafter.
6 ~ It is understood that the actual measurement selection
7 l¦ and combination need not be done in con~unction with the acquisi-
8 , tion o~ the individual measurements since these measurements
9 may be provided at any later time from conventional analog or
0 , digital storage facilities for processing at a site remote from
~ the borehole. It is important, however, that incremental depths
12 ` corresponding to the movement of the tool in the borehole be
13 I recorded in conjunction with the measurements, since it is
14 il necessary, as described hereinafter, to accurately relate the
15 ~, measurements to one another on a depth basis.
16 " As the tool 10 containing the four transducer array
17 l is moved upward through depth positions I, J, K and L, various
18 ' transducers are selected in a systematic manner such that a
19 sequence of measurements is made at regular depth increments.
1 It is customary that a particular point on the tool be selected
21 as a reference point so that measurements taken with various
22 transducers can be related to one another and to the depth of
æB the tool in the borehole as recorded at the surface. Although
24 any point may be selected, FI~. 1 shows and the description
hereinafter is based on the selection of the depth reference
26 point as being the location on the tool lO at the upper-most
~7 transducer; i.e., the transducer that is closest to the surface
28 of the earth as the tool advances through the borehole.
29 '
31 - 15 -
æ
`~ 20.1671
~7 ~ ~
: I
1 In order to describe the sequence of measurements, reference
2 i~s made to FIGS. 2A through 2C, which show the four transducer
8 array of the tool 10 as Tl through T4. For descriptive purposes,
4 the letter T with a subscript will be used to indicate transducerq,
either receivers or transmitters. Further, it will be assumed
6 that the two upper-most transducers, T1 and T2, operate as
7 receivers a~ that the two lower-most transducers, T3 and T4,
8 operate as transmitters.
9 It is desirable that transducers of a particular type, such¦
as those operating as receivers, be grouped or paired together
11 in the tool and that groups of transducers move laterally and
12 vertically in the borehole in a coordinated fashion. Further-
13 more, for reasons which will be apparent hereinafter, the pre-
14 selected separation between transducers in each group should be
the same; i.e., the separation between Tl and T2 along the
16 length of the transducer support member of the tool should be
17 the same as the separation between the transmitters T3 and T4.
18 ; The distances between the groups of different types of transducerl,
19 for example, the distance between receiver T2 and transmitter T3,
may or may not be the same as the separation between same-type
21 transducers depending on the physical characteristics of the
22 earth formation being measured, the depth of investigation into
23 the earth formation desired, and other factors.
24FIGS. 2A, 2B and 2C each show the transducer array Tl, T2,
T3 & T4 in two separate positions indicated by the depth level
26 indexes at the top of each transducer array. These indexes, I
27 through L, are referenced to the top transducer Tl. In FIGS.
28 2A and 2B, these positions are I and L; i.e., the top transducer
29
82
` 20.1671
7~7
.,
1 Tl at depth levels dI and dL, respectively. In FIG. 2C, the
~ i' two positions are labeled I and J because the top transducer
8 Tl is at depth levels dI and dJ, respectively.
4 l As the array is advanced from position I to L in FIGS. 2A
6 and 2B and I to J in FIG. 2C, the array moves up the borehole
6 ¦I from depth dI through dL, using the Tl as the depth reference
q 1i point. A signal is generated by transmitter T3, which will
8 ~I propagate uphole towards receivers T2 and Tl. Each of these
g ~I receivers will convert the received signal into a corresponding
o electrical signal which may be processed into a measurement m.
11 i1 Since it is normally expected that a signal traveling from
12 '^ T3 toward T2 and Tl will arrive first at T2 and then Tl, the
13 ~! T3 - T2 measurement will be designated as ml and T3 - Tl as m2.
14 jl ~easurements ml and m2 may then be combined to obtain a measurement
16 l~ of a subsurface physical characteristic in a manner that will
16 1, depend on the characteristic being measured.
17 ~ For example, if T3 is transmitting an acoustic pulse,
18 ` measurements ml and m2 will represent travel time through the
~9 formation and media surrounding the borehole from T3 to T2 and
~~ Tl, respectively, and then may be combined to determine the
21 interval travel time between T2 and Tl, called ~t.
22 ' At some short time separation from the generating o~ a
23 signal by transmitter T3, a signal is generated by transmitter
24 ~ T4, as shown in FIG. 2B, which is received by receivers T2 and
26 ~ Tl and there converted into measurements m3 and m4, respec-
26 tively.
27 A complete sequence of measurements at depth dI would
28 include, therefore, all of measurements ml, m2, m3 and m4.
29 Hereinafter, m will designate individual measurements in general,
irrespective of typej ml being made while operating T3 with T2
81 and m2 with Tlj and m3 operating T4 with T2 and m4 with Tl-
æ - 17 _
20.1671
~ 17~7
., I
1 Since the four measurements may be acquired in a very short
2 period of time relative to the tool movement, they may be
8 , considered as essentially acquired at the same depth. For
4 jl example, acoustical transmitters may be pulsed on the order of
6 ,l 20 times per second. This rate provides at least five complete
6 il sequences per second during which a very small tool displacement ¦
7 1l would take place at normal logging speeds. The four measure-
8 1¦ ments are transmitted uphole and stored for later use as shown
9 i~, at 22A in FIG. 1 and as will be described in greater detail
1~ hereinafter.
11 1' At some later time, when the tool has advanced through
12 '' the borehole to depth dL, as shown in FIGS. 2A and 2B, a second
13 ,, sequence o~ measurements ml, m2, m3 and m4 may be taken and may
14 ,I be used in accordance with the present invention for co~pensat-
~ ing for borehole effects on the individual measurements.
16 l For example, when T3 is an acoustic pulse transmitter,
17 ' the interval travel time, ~t, between T2 and Tl will be in
18 1 error if the portions of the signal propagation paths that are
19 , located in the borehole are of different lengths at the two
receivers. Such a difference would occur in the case of tool
21 , tilt.
22 Prior art borehole compensation techniques in acoustic
23 logging tools use separate transmitters located on opposite
24 sides of the receivers in a T-RR-T array to obtain two ~t~s
having reversed near and far receiver relationships.
26 In accordance with the present invention, that type of
27 borehole compensation is possible with an arra~ having a
28 significantly shorter overall length. By combining a first set
29
:
31 - 18 -
~2
~ ` 20.1671
7 ~7
., `
1 IOf measurements, ml and m2, taken at depth dI (see FIG. 2A at I)
2 with measurements ml and m3 taken at depth dL (see FIG. 2A at L)
8 ~ a novel combination of measurements from transducers having a
4 iI reversed near and far relationship is obtained that provides the
6 j~ desired borehole compensation.
6 ,I Furthermore, a second borehole compensated measurement can
7 jl be made simultaneously with and over the same interval in the
8 1I borehole as the borehole compensated measurement described above.
9 li Such a second measurement cannot be obtained with the prior art
1 T-RR-T array. Referring to FIG. 2B, by combining a second set of~
11 ~ measurements, m3 and m4 taken at depth dI (see FIG. 2B at I) withj
12 i~ measurements m2 and m4 taken at depth dL (see FIG. 2B at L), a
13 ,i second borehole compensated measurement is obtained, but here
14 ! having a longer T-R distance than the first measurement. This is
~ because this second set of measurement~ is referenced to trans-
16 1~ ducers more distant than in the first set.
17 ¦ A further advantage of the transducer array of the present
18 invention relates to the use of the array to compensate for
19 1 statistical or systematic errors in the measurements taken and
i may be described in connection with FIG. 2C.
2i ' Note that measurement m2 at depth dI is essentially
22 repeated by m3 at dJ, when T2 replaces Tl and T4 replaces T3 as
23 the tool advances through the borehole. Under perfect measure-
2~ ~ ment conditions, therefore, m2 should equal m3. However, under
26 I typical borehole measurement conditions there are several known
26 ~ reasons why this may not occur. Even if small statistical varia-
27 i tions may be expected, for example when acoustic interval -transit
28 ' time measurements are being made, an improved measurement is
29 ~, obtained by averaging m2 at dI and m3 at dJ to provide a
-
31
- 19 -
` 20.1671
i7~t7
,, ,
1 measurement that is compensated for such statistical variations. ,
2 While comparable statistical compensation might be accomplished
8 by repeating the measurement at dI, such repeat measurements
4 ~' cut the duty cycle of the tool by half. In contrast, no increase
I in duty cycle is required to obtain this result by combining the
~ l~ already available m2 and m3. Further, as will be explained
7 I hereinafter, there are other reasons why it is preferred to use
8 , different transducers and even different tool positions to
9 til obtain measurements for such combinations.
, While not shown in FIG. 2C, it will be appreciated that
1l ji other measurements may also be combined advantageously to
12 ii compensate for random noise or different transducer effects and
13 ~~ their relative positions in the borehole. For example, m2 at
14 ~ depth dK may be used with m3 at dL.
lB ~¦ In some cases measurements such as m2 and m3 may also be
16 ~ compared to detect borehole distortions, such as tool tilt. A
~7 ~,l comparison of such measurements can give an indication of the
18 borehole compensation being applied to the basic measurements
19 and, thereby, an indication of the reliability of the borehole
compensated measurements.
21 ' As described above, all four measurements in each sequence
22 are not essential to provide one compensated measurement, nor
23 ~ is it necessary to make each measurement after individual
24 transmitter firings as described. However, as shown in FIGS.
26 2A - 2C and summarized below, each individual measurement will
26 be used at least twice in different combinations to provide
27 two different borehole compensated measurements of a selected
28 borehole interval, corresponding to two different transmitter-
29 receiver investigation distances:
81 ~ - 20 -
æ
-- ~0.1671
1~ ~ 17
, ~ I
1 T A 3 L E
2 1) m2 and ml at dI (interval T2 to Tl operating T3)
8 ~ 2) ~ and m1 at dL (interval T3 to T4 operating T
4 i, 3) m4 and m3 at dI (interval T2 to Tl operating ~4)
6 i1 4) m4 and m2 at dL (interval T3 to ~ operating Tl) ¦
6 ~ 5) m2 at dI (T3 to Tl) and ~ at dJ (~4 to ~2)
~ I 6) m2 a~ dK (~3 to Tl) and m3 at dL (~4 to T2
9 As shown in ~IG. 1, each measuremer.t m1, m2, m3 and m4,
~is stored in measurement storage &pparatus 22 for each increment
11 j' of depth dI, dI + 1~ - etc., each increment being on the order
12 !i of six inches or less.
-13 1! lf mcasurement storage capacity is limited, it is advan- ., .
14 tageous to combine some of the measurements to minimize the
16 '~ needed capacity. For example, measurements ml and m2 for the
16 j~ same depth increment (see FIG. 2A at Position I and dep~h in-
17 1~ crement dI) may be subtracted in measurement selection and
~8 combination apparatus 24 to 40rm a new measurement ~ - m2 ~ ml~,
19 which in turn may be stored, replacing both ml and m2 or, if
j sufficient storage capacity exists, as an additional measurement.
21 As the array is advanced through the borehole from dI to
22 ~t. dJ~ other measurements may be combined to form replacement or
23 ~ additional measurements. As the array is advanced to depth dL
24 (see Position L in FIG. 2), a complete set of measurements will
be produced. Those previously acquired are now available from
26 storage 22A and those acquired at depth dL now available as
~7 current measurements. ~hus, it-is then possible to combine
28 these produced measurements to provide compensated measurements
29 for the borehole interval shown in FIG. 2 below dI.
80 ' :
gl - 21 -
æ ~
.
1~17~7 20.l67
.;
, i
~, i
1 ~, For exa~ple, by subtracting ml from m3 produced at dL, the
2 ~ currenl depth at position L shown in FIG. 2A and combining this
8 result with measurements m2 and ml at dI previously produced at
4 i position I, or its previous combination m5 at dI, the borehole
6 j compensated measurement for the borehole interval illustrated in
6 !i FIG. 2 corresponding to the short T-R distance investigation is
7 ¦i provided.
8 ,~ The above combinations result,- for an acoustic logging
9 ¦ embodiment, in adding two ~t-measurements for~the same borehcle
lo !j interval, one corresponding to ~ two-receiver-measurement and the
11 ¦ other to a two-transmitter measurement, to prov-'de the desired
12 i borehole compensation. Depending upon the separation between
13 i! the like-transducer pairs, the result may need resca ing. If,
~ for example, tne separation is one-foot, the correct ~t, as
16 !1 indicated by output A, will be obtained by dividing the ~inal
~6 ,¦ combination by two.
17 ¦' In add~tion to combining these ,wo ~t measurements as
18 described above, various measurements at various depth levels
~9 ; could be compared to indicate borehole conditions requiring
1' compensation or combined in a manner to provide the average
21 'i measurements. For example, m2 at dI (see FlG. 2C at I) and
æ j, at dJ (see FIG. 2C at J) may be ædded or averaged to form m6.
23 ~hese average measurements could then be combined to provide ~t !
.
24 measurements or for other purposes.
2~ In FIGS. 3A and 3B, there are illustrated typical circuits
26 i ~or surface and downhole apparatus for performing transmitter
27 and receiver selection, ac~uisition and combination of indivi-
28 dual measurements in accordance with the present invention.
29 .
81 - 22 -
82 `
~ ~ 20.1671
, 1.
17~ ,
. .
1 1 While these circuits are illustrated for providing compensated
2 ll acoustic travel time measurements, similar circuits may be used
8 ; for compensated amplitude ratio measurements, for example, by
4 l1 modifying the circuits of FIGS. 3A and 3B to also acquire signal
6 l amplitude or gain setting information along with the time
6 ,' measurement information. The amplitude information may be pro- I
q 1¦ cessed in accordance with the teaching of the previously discussed
8 '! zill patent. I
9 ,l A general description o~ the operation of the apparatus of
~0 ,' FIGS. 3A and 3B will be given followed by a detailed description.l
11 ,1 In general, the measurement sequence commences with a
12 ,I depth pulse corresponding to a depth increment and then clocking
13 I the individual transmitter and receiver selection cycles to
14 j complete the sequence. Four cycles and corresponding logic modesl
15 ll are shown in Table II below to select either transmitter T3 or
16 , the more distant transmitter, T4j and receiver T2 or the more
~7 distant receiver, Tl.
18 ii These logic modes, denoted M and M or N and N, are used
19 l/ to select respectively the appropriate transmitter or receiver
'1 as will be described later. The travel path portions given in
21 1 Table II for each o~ the measurements are illustrated in FIG. 4A
22 (to be discussed later) and apply as well to either the contact
23 skid embodiment shown or to a mandrel embodiment.
24 T A B L E II
CYCLE NO. MODE TRANS. RECR. PATH l~AS.
26 1 M N T3 T2 A + B + D m
27 2 M N T3 Tl A + B + C + E m2
28 ; 3 M N T4 T2 F + G + H m
29 ~ M N T4 Tl F + G + I + J
~ ;:
81 - 23 -
æ
7 ~7
; ~ 20.1671
,
1 The selected transm'tter is fired and the propagated signal
2 ` received at the selected receiver, amplifled with a gain setting
8 ilappropriate for the particular T-R cycle and transmlt~ed uphole.
4 j A reference timing pulse is generated in a fixed time relation-
~ llship to the ti~ of firing the transm~tter and used as a basis for
6 jlcompensating signal losses in the cable and as a time reference
7 ¦¦point to gate the signal detection circuits used for detecting the
8 ~Itime of arrival of the received signal. Time measurements are
9 !Iperformed by gating clock pulses into a counter beginning with
~I the transmitter firing reference pulse and stopping with tha de-
tection. m e counter's ~ontents then become the individual meas-
~2 ,, urements m already discussed:in regard to the previcus:figures.
13 ', m ese measurements are stored or recorded according to their
.14 ,~ cycle position for later processing.
16 , For a detailed .description, refer now to FIG. 3A. m e se-
:16 ~! quence begins with a depth pulse rom depth pulse.generator 305
lq i start~ng a rate oscillator control 310 and clock 324. m e
1~ oscillator 310 and clock 324 cooperate to generate control sig-
19 " nals at a rate such that several complete sequences of four
,~ cycles each will be performed per second. Clock pulses are fed
. to cycle keyer 330, which generates four cycle control pulses
22 , utilized as binary states of M and Nj i.e., M, M, N and N by
23 , steering logic to produce mode signals representing the modes of j
24 , Table II. Cycle control pulses are used to synchronize steering .
26 . logic 331 and 332 to select a new T-R combination, and to syn-
26 chronize downhole gain system 334 to establish the gains appro-
27 : priate for the received signals of each cycle.
28 '
;
30 ~ ,,
31 :
æ - 2~ -
20.1671
7~9
" ~ ' .
1 Referring to FIG. 3C, a t~mJing d-iagram is shown to illus-
2 ,, trate time relationships between clock pulses (on line 1) from
8 clock 324, M and N binary mode signals (2 and 3) from steering
4 ,~ logic 331, cycle control pulses (4, 6, 8 a~d 10) and their delayed
6 I counterparts (5,7, 9 and 11, respectively) from cycle keyer 330
6 ~¦ and counter reset and up/down steering control pulse~ (12, 13 andl
7 i 14 for counters ~1, 2 and 3 respectively) from steering logic 332,
8 jl used in the surface circuits shown in FIG. 3A to synchronize down
.Q IJ. hole circuits shown in FIG. 3B.
0 ~ While the actual circuits, such as square wave~generators, i
or example, which may be used to generate these.signals.and
`12 ~ p~lses and their delayed counterparts are not shown for sim~licity
:13 1~ of the circuit diagrams, how to make-these circuit~ is w~ll known
14 ~; to those in this art. Similarly for clarity of the diagræms, notl
j all the connections between various circuit components using these
6 i~. sig~als and pulses to enable gates, select codes, reset counters
17 i. and the like are shown. The identity of each or alternative
18 1i signals and pulses is shown where appropriate.
19 ,, From FIG. 3C, it can be seen that a depth pulse starts a
.20 ~ series of clock pulses 1-9 (shown on line l) to define one com-
-al ! pl ete meaæurement sequence as previously described in regard to
.22 FIG. 2 and Tables I and II. On the initial clock pulse, mode
.23 signal M selects T3 (line 2) and N selects T2 (line 3) to begin
24 ;. the Cl cycle to produce ml~ Counter #1 (shown at 391 in FIG. 3A)
2$ - may be reset (RS) (as shown at line 12 in FIG. 3C) on the upward
26 edge of the steering pulse from steering logic 332. Thereafter,.
27 counter-~l will start counting ciock pulses from high frequency
28 clock 389 when gated to it via clock gate 390 during the initial
29 portion of cycle Cl (line 4).. Normally, counter ~1 is stopped
! by receiver signal detection before the end of this.portion
'~.
æ - 25 -
20.1671
7 ~
.1 ,
1 ~ at clock pulse 2, and if not, this pulse or a delayed pulse Cl'
2 '' may be used to stop counter ~1, but its contents would be in-
8 'i vælid in this case.
4 ', Up/down coun~ers #2 and ~3 are showr. at 394 and 394A in
~ I! circuits shown as 24A and 24B respectively in ~IG. 3A, and are
6 ; used in conjunction with memories #1 and #2 also shown there to
7 l, combine the particular measuremen~s for each counter in the
g j~ up (+) or down (-) mode as indicated on lines 13 and 14 respec- ¦
9 ,1 tively of ~IG. 3C. Similar counters ~4, ~2~, #3A and #3B a~e
~ shown in dashed lines in circuits ~4A and 24B of FIG. 3A.
~ or example, counter #2 of circuit 24A of FIG. 3A is
`12 ~1 shown (line 13) in a count down mode at clock pulse 1 and, as
13 ~, will be described in detail later, counts down during the
~ initia'l part of cycle 1 when ml, the ~ ind''cating the measure-
16 j ment was depth delay or memorized from a previous ml measure-
16 fl ment at a deeper depth, for example, is input from~memory
17 ~ then at clock pulse 2, while stiIl in the down mode, the
l$ , current ml is input to counter ~2 direct from courter #1; i.e., i
19 ~, without delay or memorization since counter ~1 contains the
cur-ent measurement after the initial part of each cycle. ~hus,
21 at clock pulse 3, counter ~2 has ~ccumuIated - ~ - ml and is
æ .~ then switched to a count up (+) mode. Then dur~ng the iritial
23 part of cycle 2, m2 is input from memory ~1 to add +m2* and,
a4 during the latter part of cycle 3 (at clocl. pulse 6~;the current
26 m3 counted during the initial part of cycle 3 is input from
26 counter ~1 to add ~m3 to the prior accumulation in counter #2.
27 At the end of cycle 3, counter #2 contains -ml - ml + m2
28 ' + m3, which have been input in that order. Subsequently, at a
29,
,
- 26 -
~2
7~
convenient time ~shown as during cycle 4 on line 13 in Figure
3C~, counter #2 may be output and reset ~RS) to begin at the
next depth pulse, as another sequence as described above at
clock pulse 1. As will be explained later, and as shown in
Figures 1 and 3A at A, this combination of measurements
corresponds to one of the borehole compensated measurements
provided by the techniques of this invention.
Up/down counter #3 shown as 394A in circuit 24B
of Figure 3A is similarly diagrammed in Figure 3C on line 14.
However, its sequence begins after clock pulse 3 rather than
clock pulse 1, as for counter #2 discussed above. At the start
of cycle 2 with clock pulse 3, counter #3 beglns, by switching
to the down mode. Then during the latter part of cycle 2, and
during the initial part of cycle 3, -m2 and -m*3 are input.
Then at clock pulse #7, counter #3 is switched to the up mode
and +m*4 + m4 input from memory #2 and direct from counter #1,
respectively during cycle 4. Thus, at the end of cycle 4,
counter #3 contains -m2 - m*3 ~ m*4 I m4. As shown in Figure
3C on line 14, contents of counter #3 may be output during
the following cycle 1 and counter #3 then reset ~RS) and
switched to the down mode to begin its sequence again at clock
pulse 3. As will be described later, and as shown in Figure
3A at B, this combination of measurements corresponds to an-
other of the borehole compensated measurements provided by the
techniques of this invention.
Cycles 1 through 4 shown in Figure 3C are summarized
in Table III and will be further discussed in its description
later. With the general use and timing of the control signals,
mode pulses, counters and memories now described, the
- 27 -
1~17~t7
particular circuits of FIGS. 3A and 3B will be described. The
M and N mode select signals generated for each cycle by
steering logic 331 will be used as the first two bits of a
code signal.
-27A-
7 ` 20.l67l I
, ,
.,
1 At the start of each cycle, a code signal is transmitted
2 ~ from code transmitter 336 in FIG. 3A downhole to code receiver
8 ~~ 340 in FIG. 3B. The code signal may contain as few as six bits
4 of information designating which of the two transmitters (one
8 , bit), which o~ the two receivers (one bit), and which of 16
~ i! ga~n settlngs (four bits) are to be used. Additional bits ~or
7 ,l additional gain or attenuation settings may be desirable to
8 ,l increase gain resolution when amplitude/attenuation measurements;
9 ~l are also being made.
0 Now concerning the operation of the downhole-circuits,
11 jl refer to FIG. 3B. In general, code transmission results in
12 connecting the selected transmitters and receivers to appropriate
13 downhole circuits and setting the downhole gain. Next, the
14 ~ selected transmitter ~s fired and the timing count begins. An
16 ~, automatic galn control system is used to standardize the signal
16 ! amplitudes by varying the gains for each different T-R cycle.
17 Now discussing the detailed operation of the downhole
18 circuits of FIG. 3B, the code signa~ from code transmitter 336
19 j are received by downhole code receiver 340 in circuit section
``I 20 , llB shown in FIG. 3B, and a code bit representing the M or M
2L mode is routed to the transmitter selector 344 which connects
22 either firing circuit 352 or 354 to T3 for M or T4 for M,
23 respectively. Similarly, the N or N bit is routed to the
24 receiver selector 350 and either receiver T2 ~or N or Tl for N
is connected through receiver selector 350 to variable gain
26 æmplifier 348.
27 Gain bits in the signal code (~our illustrated) are routed
28 ~ to downhole gain selector 346 which uses these bits to connect
29
, 80
31
æ - 28 -
; ` " 20.1671
~ ~1 7 ~
., .
1 I selected fixed attenuators and gain amplifiers to provide the
2 1 desired gain represented by the code. The resulting 16 possible
8 gain variations are shown for simplicity as represented by
4 ,I variable gain amplifier 348 controlled from gain selector 346.
6 jl The gain will be determined automatically by analysis of the
6 l~ received signals as will be explained later but for now it will
7 l, suffice to appreciate that longer T-R distances, as for example,
8 ~ T4 to Tl, are given relatively higher gains than the shorter
g T-R distances like T2 ~ T3.
~I Reception downhole of a given code by code receiver 340
11 also causes conditioning of fire pulse receiver gate 360 to
12 ! interpret the next downhole transmission as a fire pulse com-
l~ I mand. Through an appropriate delay provided by delay 341, recep-
14 tion of the code also enables previously disabled downhole out-
l put gate 342 to then allow uphole transmission of output from
16 ~¦ power amplifier 368, which might have previously interfered
17 with the code transmission. Thus, it can be seen that the
18 downhole circuitry of FIG. 3B uses the code to condition asso-
~9 ciated electronics to connect the appropria~e transmitters and
receivers and set the desired gain and gates in expectation of
21 ~ a subsequent fire pulse command.
22 , Returning to FIG. 3A, the uphole circuits are enabled
2S ~ synchronous with the operation of the downhole logic, to provide
24 the firing pulse and receive the associated reference pulse and
2~ subsequent receiver signal. Cycle keyer 330 shown in FIG. 3A
2S generates, for each cycle, a signal that is sent to steering
27 logic 332 that in turn generates signals to reset to zero a
28 first counter 391 and, depending upon the particular cycle, ~l
29 .
: ;
31
82 29
20.1671
7 ~'
, . ,
1 through C4, to provide various gating signals to gates, counters
2 1 and signal processing circuitry, most of which has already been
8 described in relation to FIG. 3C.
4 I Delayed control pulses from cycle keyer 330 divide each
E ~' cycle into subcycles for detection and automatic gain determina-
6 1I tion, and gating completed current measurements or previously
7 l! stored measurements into the signal processing apparatus 24 o~
8 '" FIG. 3B, as appropriate for the particular cycle. For example,
9 ,I when signal processing circuitry 24A and 24B is utilized, these
,; steering signals may be used to clear counter 394 and 394A and
11 ! condition them to process the next input by counting down or, if
12 , the input is a digital word transfer, to combine the word with a
13 j negative sign. Subsequent steering signals cause these counters
14 ii to accept further input by counting up or adding to the previous
1~ !, conten~s.
lB , Besides providing clock pulses which begin each cycle to
17 , cycle keyer 330, clock 324 also provides, after a suitable delay,
18 ~ a control pulse to fire pulse circuit 320. This delay, provided
19 by circuits internal to clock 324, is such as to allow time for
`il both the downhole and surface circuitry to be conditioned, as
;. .
21 ~ already described, to receive the fire pulse. With the downhole
22 and uphole circuits ready, this delayed clock pulse causes fire
23 pulse circuitry 320 to initiate a fire pulse command (FP) which
24 is transmitted downhole and properly interpreted by the previous-
ly conditioned fire pulse receiver gate 360 shown in FIG. 3B.
æ Returning to FIG. 3B, the fire pulse command is gated
27 through to fire pulse detector 362 and upon detection, causes
28 To generator 364 to initiate a downhole firing pulse. This
23 causes transmitter selector 344, previously connected to selected
30,
81
32 3
` 1~9~7~7 20-1671
'. I
transmitter firing circuits, 352 for T3 on mode M or 354 for T4
2 , on mode M (see Table II), to fire the selected transmitter, T3
~ ~ or T4. The To generator 364 also initiates a To pulse for use
4 il as a reference signal both by downhole and surface circuitry.
~ l~ For surface use, the To pulse is transmitted uphole via power
6 !l amplifier 368 and through the now enabled output gate 342 to
q I signal receiver 370 and system automatic gain control 372 located
8 ~l at the surface (see FIG. 3A).
9 ~! At the surface, the To pulse is used as an amplitude refer-
¦¦ enceO Since it is established downhole by To generator 364 witha¦
11 standard reference amplitude, the system automatic gain control ~
12 ! 372, To gate 374, peak reader circuit 376 and To gain set control
13 1 378 (shown in FIG.3A) are utilized to re-establish at the surface
14 lll this standard amplitude reference. Thus, this internal system
'1 control provides compensation for cable losses, phase distortion,
16 1' drift, etc.
17 ~ The To pulse is gated via To gate 374, which has been
18 ~ previously conditioned to allow To through to peak reader 376
19 ~ via line 374A. This conditioning is synchronized with the
operation of fire pulse circuit 320. The peak amplitude of
21 1~ surface received To pulse is read by peak reader 376 and compared
22 i' with a reference amplitude by To gain setting circuitry 378
23 , which ad~usts, if necessary, the automatic gain control circuit
24 372 to re-establish the amplitude for subsequent To signals to
26 the reference amplitude level. Additional signal conditioning
26 circuitry may be included for cable losses using this known To
27 , signal standard.
28 The peak To amplitude read from peak reader 376 is also
29 provided to an amplitude detector 380 as an amplitude reference
31.;
: æ - 31 -
~ 20.1671
., I
Il l
1 !~ for use in detecting the receiver signal which will follow To~ as
2 , w ill be explained.
8 m e To pulse provides an accurate time reference related
4 ll to the transmitter firing. This time reference is determined by
!j a zero crossing detector circuit 375 connected through Togate
il 374 to consistently detect the To zero crossing point. This To
~ ll detection pQint is used as a beginning for the time measur~ment
8 j¦ by providing a time-related To dete~tion si~nal as a start sig-
g !l ,nal clock gate control 388 of FIG. 3~ which enables a clock gate
Q l~ 390 to pass high frequency clock pulses from clock 389 to counte~
#l which, having been previously set to zero, begins counting
12 ~! the clock pulses. The clock pulses should have a high enough
13 ' frequency, for example, 2.5 megacycles, to provide the desired
14 ,~ time resolution. ~he To detection signal is used in turn to gate i
~ off the To gate 374 and gate on a received signal detector gate
lB j 379 such that subsequent signals will be interpreted as the next ¦
17 ,, expected receiver signal. With the uphole apparatus of FIG. 3A
18 , already beginning the time measurement at the start o~ the clock
19- ~ pulse counting, refer now to downhole circuits shown in FIG.
3B~
21 ~ The To signal generated by ~0 generator 364 is delayed by
Z2: delays Dl and D2 provided by delay circuit 365, as shown in FIG.
28 ` 3B, and used to respectively open and close a receiver gate 366
24~, for a time interval correspor.ding to the expected arrival time o~
25~ the received signal. These delays will understandably vary
26 with the design distance between the transmitter and receiver
27 and can be determined in a well known manner.
28 A~ter propagation in the ~ormation, the acoustic pulse
29 transmitted from the selected transmitter is propagated through
,
81
æ - 32 -
- ~ ~ 20.1671
1~17~7
.,
1 ~ the borehole and formation and received by the selected receiver, !
2 ` which has been previously connected through receiver selector
8 1 350 to an already set variable gain amplifier 348. The selec-
4 ~', tion and gain setting was discussed earlier in regard to the
~ ' operation of code receiver 340. The received signal is allowed
: 6 ~~ through receiver gate 366 now enabled as described above to the
7 ' previously described power amplifier 368 and still open output
8 l' gate 342 where it is transmitted to surface circuitry shown in
9 ~ FIG. 3A.
~I Returning again to FIG. 3A, the amplified receiver signal,
11 ~ here denoted as Rx, is received at the surface and reconditioned
12 ~1 at signal receiver 370 and amplified by system AGC 372,already
13 , described,to provide a cable compensated amplitude. It is then
14 gated through a receiver detector gate 379, previously condi-
15 i tioned by a To detection signal generated by zero ¦
16 `, crossing detector 375 to allow the signal to pass to detection
17 il circuits 380 through 384.
18 , As illustrated in FIG. 3A, the arrival of the received
19 . signal Rx is detected by simultaneously comparing Rx amplitude
i. in amplitude detector 380 with a To reference amplitude supplied¦
21 by peak reader 376 and examining Rx with a zero crossing detec-
22 tor 382 and a peak detector 384. A typical Rx signal is shown
23 in FIG. 9A and has positive and negative half-cycles which
24 ; increase in amplitude during the first few half-cycles. As
2~ illustrated in the circuits cf FIG. 3A, three conditions are
: 26 required for detection: 1) a zero crossing must be detected
2~ by zero crossing detector 382, this detection being delayed
,; .
28 internally by a delay corresponding approximately to one half-
29 : cycle; 2) the subsequent amplitude of Rx as compared by
; i
.
31
æ - 33 -
amplitude detector 380 must exceed a small fraction of the To
reference amplitude; and 31 an Rx amplitude peak must be
detected by peak detector 384 within the half-cycle delay
following the zero crossing detection. All three detection
indications are provided to AND gate 385, such that the first
occurrence of an amplitude exceeding a threshold amplitude
referenced to the To amplitude which is preceded by a zero
crossing in the appropriate polarity, and which is followed
by an amplitude peak of the same polarity within the half-cycle
delay, completes the detection.
An Rx detection indication is output from AND gate
385 and causes hold circuit 386 to hold the peak amplitude
detected by peak detector 384 for use in setting the downhole
gain for subsequent reception with the same transmitter-receiver
combination in cooperation with downhole gain setting circuit
334. Independent gain settings are made and stored in the
downhole gain setting circuit 334 for subsequent use with cor-
responding cycles. Further, these gains are determined on
the same part of the signal used for the measurement. A fur-
ther description of this automatic gain setting technique is
provided in United States Patent Nos. 4,040,001 and 4,172,250.
The Rx detection signal output from AND gate 385 is
used to reset the receiver signal gate 379, previously described,
and, more importantly, to cause clock gate control 388 to gate
off clock pulses coming from clock 389. These pulses had been
gated on earlier to counter 391 by means of the clock gate 390
- 34 -
20.1671
~7
.,
,' 1.
1 by the To detection as previously described. Thus, the To and
2 I Rx detection are used to cause the determination of a clock
8 pulse count corresponding to the time measurement ~or this
4 i! given cycle. In this manner, the counter #l now contains the
,
6 il number of 2.5 megacycle clock pulses corresponding to the travel
B 'l time relative to the To and received signal detections. The
q ! counts contained in counter #l may be in turn regarded as the
8 l individual time measurement corresponding to the particular
9 I measurement cycle such as ml for cycle 1, m2 for cycle 2, etc.
il The contents of counter #l at the completion of the counting
11 ! may then be transferred through various gates to utilizing
12 , devices at the times shown in FIG. 3C, as provided by appro-
13 1~ priately delayed control pulses C' from cycle keyer 330.
14 ~ For example, where the indiYidual measurements are to be
recorded for later processing, the delayed control pulses C
16 through C~ each cause the counter contents, corresponding to
17 ' ml through m4, respectively, to be gated through gate 392B
18 j to a suitable recording device connected at point C such as a
19 l digital tape recorder, not shown.
Alternatively, particular cycle control pulses are used
21 to selectively gate the counter contents into the memory and
22 counter circuits 24A to provide one compensated signal, and to
23 similar circuitry in 24B if two different spacing compensated
24 ,. .
26 '
~7 :
28 .
29 ..
81
82
20.1671
.
1~17~7
i
1 signals are desired. These circuits accomplish the relationships
2 , ~or combining individual measurements illustrated in Table III
8 below:
4 '
~ TABLE III -
.. . .
6 !1
7 CONTR GATE CONNECTION CONTENTS OF COUNTER;
PULSE MEAS. FROM TO C#l C#2 , C#3
8Cl i~ ~ ~ ~ ~ r~
12 Cl' mlC#1 M#l ml OUTPUT
,¦ . _. C~ 2 . . _ * RESET
14 ' C2 m2* M#l C#2 2
. -ml - ml *
16 li' ___ .
1~ , C2 ' m2 C#l M#l. m2
18 1! C#3 -m2
~9 ,! ,
C3 m3 * ~#2 C# 3 - m3* - m
.
22 i' .. ... _ .. .
C3~ m3C#~ M#2 m3 +m3 + m2* .
2~, " __ C~Z _ 1 1 ...
C4- m4~M#2 C#3 A +m4*
26' `' ~ -m3 - m2
2~' .. _ . .
28 C4' m4C#l M#2 m4 OUTPUT
29 C#3 RESET +m4* + m4
: -m3* - m -
31 * measurement from previous position
æ - 3~ ~
1~ 6 7~7 " 20.1671
1 Table III above illustrates the general cycle for each
2 measurement m. ~uring the primary portion of the cycle, here
8 ' denoted as subcycle C, clock pulses are being accumulated in
4 ` counter #1 (C#l) for the new measurement at the current depth
6 , dJ, for example, as illustrated in FIGS. 2A and 2B. The corre-
¢ l~, sponding measurement m* made at the previous position dI in the
7 !~ illustrative example, is transferred f~om memory M to a second
8 ' counter which has been previously conditioned for this cycle to
9 ~ count down or subtract; or count up or add; for example. This
0 makes room in memory for current measurement such that memory
~ need only have a capacity for the number of measurements acquired
12 ~i between dI and dJ because the current measurement m (at dJ) may
13 ' replace the measurement m* (stored at dI).
4 The next subcycle C' begins after a long enough delay has
~6 1 been provided to allcw the completion of the current measurement~
16 i.e., after the expected signal has been recei~ed from downhole
17 1 and detected and C#l has stopped counting. Then m is gated from
18 c#l to memory M, replacing the corresponding previous measure-
19 ment m*. During subcycle C', m is also gated to the particular j
second counter C#2 or C#3 for this cycle. As illustrated for
21 the two compensated measurements A and B, each m goes in turn
.,
22 to one memory M and one additional counter; e.g., ml goes to
23 , M#l and C#2; m2 to M#l and C#3; m3 to M#2 and C#2 and m4 to
24 M#2 and C~3. Thus, each M stores two different m's and the
26 counters C#2 and C#3 combine two current m's and two previously
26 stored m*'s.
~i .
27 From examination of FIGS. 3A and 3B and Table III above,
28 it is readily seen that both the measurements and apparatus
29 :
~1 '
æ - 37 -
20.167
7 ~7
.' . I
1 components serve multiple uses. The same control, ampli~iers,
2 cable compensation, automatic gain system, time reference,
8 - detection circuitry, high frequency clock and clock pulse counter
4 are employed for each individual measurement. m is not only
~ provides lower cost apparatus but provides compensation ~or
6 i systematic measurement errors, as will be explained in more
~ 7 -I detail later. For now, it need only be appreciated that i~ a
; 8 ~ component inaccuracy causes ml ~o be in error, m2, m3 and m4
9 1 will also be in error by the same amount in the same direction.
However, in accordance with the advantages of this invention,
lL l when these systematically erroneous measurements are combined
12 I as illustrated above, these errors will be compensated out ~ust
13 il as a systematic error induced by sonde tilt, for example, is
14 compensated.
16 As previou~ly discussed and as illustrated in Table III,
16: i various measurements are typically used twice, ~irst in referenc ,
17 to the receivers and then with the transmitters. Steering logic
18 ` 332 provides the control pulse mode signais allowing the con-
19 ¦ tents oi counter ~I to be transferred to the gated memories or
i counters- used to per~orm the measurement combinat~ons. For
21 j example, at the completion of the ~irst measurement ml cycle
22 I whic~ corresponds,as indicated-by Table II, to the T3 - T2
23` Il measurement~ as ilIustrated in FIG. 3A~ the clock control pulse
24 ' CI or, pre~e~ly a delayed version of it, Cll, as in Table
III above, may be utilized to gate the counter contents to
26 memory in circuit 24. Pre~erabl~ the earlier Cl pulse is used
2~ ' to gate out a previously stored measurement ~rom memory. The
28 i~ timing relationships ~or various M, N, C and c~, (delayed) state~
. i
29 and combinations for the associated measurements ml through m
have already been described in regard to FIG. 3C. These rela-
31 tionships will be detailed now in regard to the particular
32 circuit components.
., .
- 38 -
20.1671
7 ~ 7
.
. I .
1 Memory is utilized to delay measurements m* made at
2 ' an earlier position such as at dI illustrated in FIG. 2, so that
8 ~I they may be combined with current measurements which are in
counter #1. In the preferred arrangement, counter #l contents
6 i, are gated through gate 392 to memory #1, both at the completion
6 ~ of cycle #1 and cycle #2, to store measurements ml and ~ . After
q ll the number of complete cycles corresponding to the movement of
8 il the transducer array from the position illustrated as dI to the
9 '~ position illustrated as dL in FIG. 2A have been stored, these
m~ 1i measurements are available from the output o~ memory #1 such
11 1 that control pulses provided to gate 396 gate out these pre-
12 ,I vlously stored measurements for utilization in counter #2. In
13 i¦ this manner, Cl' would cause ml at dI and C2' similarly cause m2
14 ,l at dI to be gated through gate 392 into memory #1 in serial
lB~ i~ arrangement. This memorization process of ml and m2 in memory
16 -¦¦ #1 continues until, for example, at dL, the previously memorized
17 '~li measurements become available as output o~ the memory. At this
18 '~ time control pulse Cl' would continue to provide new ml measure-
19 ments to memory as well as to counter #2 through gate 393.
,, -As already explained, counter ~2 has been previously
2L ,~ conditioned prior to Gl to interpret subseauent input in a
æ 1I countdown or negative sense. T~hus, when Cl is supplied to
23 ~i memory #1 output gate 396, ml* corresponding to previously stored
24 ml* measurement at d~ is gated to counter ~2. Then, at a delayed
version of Cl denoted Cl', the current ml is also gated to
26 I counter #2 and to memory. In this manner, ml* at dI and ml at
27 dL are gated to and combined at counter #2 in the same sense;
28 ! i.e., either by continuing to count down ~or their combined
29 ; count or added with negative signs. Thus, in counter #2 at
31
39
32 ,
~ 20.1671
"
.' I
~ ' the end o~ the Cl cycle is -ml at dL ~ ml* at dI. The next clock
2 ' cycle C2 would add m2* at dI through gate 396 to counter #2, but
8 ! now conditioned to consider input in a positive or count up
4 'j sense. Then the current m2 at dL would be stored. Thus, at the
l~ end of a C2 cycle, counter ~2 would contæin m2* at dI ~ ml at
6 jl dL ~ ml* at dI. During the next clock cycle C3, measurement m3
7 ~ll at current depth dL would be gated through gate 393 to counter #2
8 ll still in its add mode such that the result becomes m3 at dL + m2*
9 , at dI ~ ~ at dL ~ m2* at dI. The~, at a subsequent convenient
!I clock ~ulse, illustrated here as C4~,the contents o~ counter ~2
~1 fl is gated out through gate 397 to point A as the compensated signal.
12 jl Where the transmitter-receiver selection has been in accordance
13 il with T~-ble II, compensated signal A corresponds to a short T-R
14 , distance investigation. Counter #2 is subse~uently reset and the
16 ,¦ processing ~or another compensated signal sample corresponding
16 ~' to the next sequential depth increment begun in the above
1'7` I described manner.
18 ~ For a long ~-~ distance investigation B, correspond-
19 ,j ing circuitry 24B, shown in FIG. 3A with separa~e memory #2 and
!i counter #3 and corresponding gates, may be utilized. In 243,
ZL ,I these comFonent$ have been designated with the same numbers
2~ ,~ used in 24A but now include an additional designation "A". Of
Z3 i~ course~ these 'tA" gates are controlled by dif~erent control
24 pulses as ind~cated therei~ since they involve di~erent measure-
ments obtained at di~eren~ times. Like the timing diagram of
26 FIG. 3C for the timing o~ the circuits shown in FIG. 3A, Table
27 , III summarizes the operations of both circuits 24A and 24B in
28 terms o~ the control pulses, senseS and contents of the various
29 counters and memories illustrated in FIG. 3A and used
to derive the two compensated signals, A and B.
8~ '
æ 40 -
ar~s7
It will be realized where both A ~ B are desired,
memories l and 2 may be readily combined since their input and out-
put functions occur at separate control pulses and measurements
ml through m4 may be stored in that order and retrieved in that same
order. One suitable memory is described in United States Patent
No. 4,040,002. Each time a new measurement, as for example, ml
is ready, the oldest corresponding measurement is retrieved from
storage such that the newest measurement may replace the oldest
measurement and the memory managed on a replacement basis, thereby
conserving the memory capacity.
It will be readily recognized how the additional
measurements m2 through m4 may be acquired and utilized from the
description of ml above, the control logic and definitions to
acquire these measurements being provided by Table II and the
processing logic being provided by Table III. It should be
realized that the invention may be practiced by providing a
single compensated measurement, here illustrated as either A or
B, therefore employing only a single memory or additional counter
other than counter #1. In this case, two measurements may be
combined as acquired and only the result stored. The two current
measurements would not need to be stored.
It also should be realized that the processing pro-
vided by circuits 24A and 24B may be provided by a digital micropro-
cessor with its normally associated memory replacing memories
395 and 395A and its arithmetic registers replacing counters 394
and 394A, its control program utilizing the control pulses to
,' .
- 41 -
~ 7 ` 20.1671
. . .
., ~
1 Iperform the indicated transfers to and from memory and the regis-
2 Iters. As previously stated, these processes may also be provided
8 ,by utilizing output C recorded on a digital tape recorder which
4 is subsequently produced as input to a general purpose digital
~ I'computer and processed with an equivalent control program.
6 ,~ Referring now to FIG. 4A, there is illustrated a transducer
7 skid support which is tilted from the desired wall contact posi-
8 ~tion parallel to the borehole wall. Such tilt may be due to a
9 , variety of mechanical problems associated with the s~id to mandrel~
linkage, inadequate sidewall pressure, etc. Unfortunately, when
~ this tilt problem occurs, it may not be reflected in linkage
12 caliper or pressure measurements. However, in accordance with
13 one feature of this invention, not only can the tilt be detected
14 I but its effect compensated.
16 ,i The transducer array shown in FIG. 4A is arranged as was
16 assumed for illustration in FIGS. 2A and 2B; i.e., the receiver
17 ,` pair Tl and T2 is on the top and the transmitter pair T3 and T4
18 is on the bottom of the skid.
19 ` As denoted in Table I already described, four measurements
are made between different combinations of these transducers.
21 ` Two binary modes M and N are used to code the transmitter and
22 receiver selection which control the signal paths. In accordance
23 with one advantage of the novel transducer array, compensation
24 for borehole path length differences due to either tilt or bore-
26 hole washout between the near and far transducers in the pair is
26 provided by reversing the sense of the near and far transducer
27 ~ measurements; i.e., the far transducer becomes the near trans- '
28 ,
29 ~
81
æ - 42 -
` 20.1671
7~7
~ ' i
1 . ducer and vice versa. ~his ability is provided by utilizing a
2 ! pair of transmitters in the same sense that a pair of receivers
8 , are used to provide one of two sets of the transducer
4 ,', measurements. FIGS. 4A and 4B illustrate how this compensation
~ 1 is obtained for the skid type and FIGS. 4C and 4D for the non-
6 i,¦ skid type arrays.
~ 1 Consider the paths shown in FIG. 4A and Table II. Signals
8 li leaving T3 travel path A through the borehole to the formation
9 ,¦ and.the~ towards the receivers along path B, reaching T2 via
lQ i borehole path D and Tl via additional formation path C and bore-
11 ,I hole path E. .If the borehole paths E & D are equal, the differ-
12 ences between the signals of T2 and Tl will be essentially a
lg 3! measurement of the travel through formation path C~ corresponding
14 . to the interval between T2 and Tl. If, however, path D is sub-
stantially dif~e.rent from path E, this d~torts the short T-RR
16 ii measurement thought to correspond to formation path C as ~n the 1.
17 ilIustrated case where D is larger than E. The short-spacing
~8~ i~ travel time measurement ms equals m2 ~ ml = C + (E - D) because
19 the common paths A and B subtract out. Ideally, E - D and
there would be no-error.. However, in the above illustrated
2I. case, the error equals their difference, E - D, which is nega-
22 ` tive,. indicating travel.time wilI be too short..
23 An error would also be present for the long T-RR
24 measurement me made relative to ~4, since the borehole path
26 lengths E and J are also unequal. Here, m~ = m4 - m3 = I +(J-E),
26 ~ since common paths F and G subtract out. 'As illustrated, ~ is
27 larger than J, making the error due to their difference also
28 negative, and indicating this travel time is also too short.
29
31 : - 43 -
32
. . ` 20.1671
7~7
1 Despite the separation in paths illustrated in FIG. 4A,
2 il formation paths C and I for the formation interval between T2
8 i and Tl and borehole paths D and H at T2 are respectively almostthe,
j
4 ~! same, as are E and J at Tl. Even formations regularly varying
in acoustic properties radially from the borehole wall can be
assumed to still have nearly identical receiver borehole paths
7 i¦ ~or signals received over either the long or short T-RR dis-
8 il tances. Consequently, both the short-distance ms using T3 and
9 l; the long-distance m~ using T4 can be expected to have the same
~0 l¦ error~
~ Reférring.to FIG. 4C~ consider the nature of the error
12 ¦~- when the transducer array is moved from position (a), when the
13 jI receiver pair is adjacent interval I, to position (b), when
1~ j the transmitter pair is adjacent interval I. The borehole paths
lB. ~. ~or positio~ (a) are denoted as in FIG. 4A and ~or position (b)
lfi. ii by the s~me Ietter but primed; e.g., A and Al at T3, respectivel~.
lq .¦ With.the interval o~ interest: I between T3 and T4, the short-
18 '1 distance mea~urement for position (b) is m.'s= m3 - ml using T2
19 ~! and the long-distance measurement m.~ = m~ - m~ using Tl. Re~er-
. ring to Table..I, it can.be seen that the error for both ms and
2L ~ ~i is: F'-A~. If Fl is greater than Al, the error is positive
22: ,~ and thus in the opposite sense ~rom the errors in position (a).
Z~ 1 As~ FIGS~ 4B and 4D will 3how,. the error is also of the same
24 ~, magnitude..
25: Consider FIG.. 4B for the skid case shown in FIG. 4A, and-
26 recalling that position (a) errors wers (E-~) or (J-H), it
27 . can be readily seèn that since the tilt angle y is the same,
28 the paths E or J at Tl ~or position (a) taken in ratio to path
29 ~ A' at T3 ~or position (b) is proportioned to paths D or Ll at
9~. .
æ - 44 -
20.1671
~ 7
"
1 at T2 for position (a)taken in rati~to F' at T4 for position (b),~
2 due to the geometrical similarity. Thus, (E - D) = -(F' - A')
8 fl and in fact, the tilt angle ~ , may be computed. If m~ (or m~) ¦
4 is less than ms (or mQ), the illustrated case of tilt is present
~ l where the upper pair of same-type transducers is closer to the
6 f wall than the lower pair. If ~s is greater than ms, the reverse¦
7 , case would be indicated. This will be more fully appreciated
8 ~f f~ m FIG. 4D.
9 l In FIG. 4D, paths are illustrated with the transducers
superimposed to show the differences in parallel paths A' and F'
11 and paths E (or J) and D (or H). It can be seen that each path i
12 ~ is related to the tilt angle ~ , the distance from the wall
13 contact point of the transducer array, and the refraction angle
14 ~ . Since ~ and ~ are constant and the distance se~arating like
1~ transducers is the same (here shown as I) it can be shown that
~ the difference between path lengths for like transducers is also¦
17 , the same, such that D - E (or H - J) = F' - A~.
18 From the above, it can be seen that measurements between
19 i first the receiver pair and then the transmitter pair reverses
' the sense of the tilt error introduced in these measurements.
21 The effects of borehole shape rather than tilt is illus-
22 trated in FIGS. 5A and 5B. In FIG. 5A there is shown in
23 horizontal section the ideal position of a transducer T, i.e.,
24 centered in a round hole. Path 1 from a transmitter and the
26 ~ path 2 to a receiver are of equal length as are all the paths
26 around the circumference of the transducer. This causes the
27 transmitted energies radiated in different directions to be
28 received essentially at the same time and thus reinforce one
29 another to provide the best signal amplitude and phase stabilit~.
80 ,
81
æ
- 45 -
,
l~P~ 7~t~7
20 . 167
., ,
1 FIG. 5B shows the same transducer T parallel to the bore-
2 ~hole wall as in FIG. 5A (not tilted) but now the borehole is non-
8 circular, with the shape resembling two intersecting cylinders
., I
4 'with different diameters and non-coincident centers. This shape
6 is typically found in directional holes. It can be readily seen
6 ll that the borehole paths 5 from a transmitter and 6 to a receiver
7 ,not only vary in length but frequently do not even intersect the
8 ~Itransducer. This results in a marked reduction of the transmitted
9 lenergy coupled to the formation and a destructive out-of-phase
jlrelationship for the signals arriving at the receiver, since a
l1 ,signal traveling via path 7 will arrive much sooner than via path
12 l 8, for example. Consequently, large reductions in amplitudes are
13 iexperienced under such conditions.
14 To a lesser degree, the above signal problem also occurs
in tilt cases, since, in those situations, it is impossible to
16 lhave all the transducers in the ideal position. For example, as
~7 ,illustrated in FIG. 4C, various degrees of eccentering, even in
18 a round hole, would be present for each of the four transducers.
19 I mus, measurements m2 and m3 would be equal under the ideal
iFIG. 5A conditions, but unequal under the out-of-round hole of
21 FIG. 5B conditions or the eccentering associated with sonde tilt.
22 In this manner, this comparison of different measurements at dif-
23 ferent depths may actually detect different transducer operating
24 environment conditions such as caused b~ sonde tilt.
26 In the prior art T-RR-T arrays, the transmitters are
26 located at extremes of the array. Thus, if tilt causes one end
27 to eccenter, the two very much spaced apart transmitters operate
28 under substantially different positions even in a round hole.
29 ;
~0 '
31 - 46 -
~2
~ 20.1071
. . .
1 ' By comparison, same transducers in the TT-RR array describe;,
2 ' herein are closely spaced and operate advantageously in much the
3 . same positions with respect to the borehole wall.
4 ,~l As previously mentior.ed, it is desirable, particularly in .
j¦ acoustic investigations, to have long T-R distances to overcome
8 ,~ the effects, for example, of shale alteration. The same desire
7 !' exists in skid devices and in other types of measurements such
8 ii as high frequency, electromagnetic investigations, etc.
9 ' FIG. 6A illustrates one prior art borehole compensating
lQ 1~ array. The T-R distance is shown as occurring twice-, first from
~1 ' T.~l anL second from T~ to the array midpoint between Rl and R2.
t2 i For comparison, FIG. 6B illustrates the compensating array of the
18 ~¦ present invention~ as applied to the sidewall skid. The same
14 l3 span or receiver nvestigation interval and skid length are used .
16^ , in both FIGS.. 6A and 6B.~ However, for the same transducer array
~B i¦ length, the novel array shown in FIG. 6B provides a substantial
17 ;l increase in the T-R distance even for the shortest T-R investi-
18 I gation~ For the longest T-R.~nvestigat~on, this distance is the
19 ,1 entire array len~th, less. only one-half the span. In contrast,
, ~h~ maximum T-R distance o~ the prior art array is only one-half
81 Ji the entire.array length.. By the ~oveI overlapping of both the short
22 . and }ong ~-R distances,. the array illustrated i~ FIG. 6B,constructe~
2~ ~i in accordance with thi~.invention,. provides not only longer T-R
24 distances. for the same~array length, but provides two different
2~ ~-R distances contained within this length. Typical span be-
2B tween Iike transducers for acoustic time measurements is one or
27 two feet while the shortest T-R distances are at least four ~eet
28 but preierably six or eight feet. Thus, the reduction in length
29
:
. - 4~ _
æ
~ 20.1671
1 j obtained with the TT-RR array of this invention is on the order o~
2 six or more feet as shown graphically between FIGS. 7A and 7B.
8 ` Referring to FIGS. 7A and 7B, additional ~eatures of the
4 'I invention are shown. In FIG. 7A, both circuit connections and
6 the use of directional transducers in the prior art type of
6 I compensation array are illustrated. In order to use directional
7 'l receivers, two separate sets of receiver pairs must be employed,
8 Rn and Rf for receiving signals from the direction of Tu, the
9 I upper transmitter, and Rn and Rf for signals from To~ the lower
transmitter. Added to the complexity necessitated by the two
11 1 extra receivers is the usual electronic noise problem associated
12 with connections between the uphole circuits above the transducer~
13 to the bottom transducers. For the lower transmitter, for example,
14 ~ these connections must be strung through or around the upper
l transducers. A high voltage generator is usually located near 1
lS , one of the transmitters, here not shown but above Tu. In any cas~e,
17 one high voltage lead, here F~ , must be run past the receivers
18 to the remote transmitter. High voltage pulses typically
19 employed to fire such transducers must be shielded in order to
1 prevent electrical crosstalk into the receivers or receiver
2i leads Rn, Rn~ Rf, and Rf, and even then crosstalk can become 11 ,
22 ~ severe.
28 By comparing FIG. 7B illustrating the array according to
24 the invention with the prior art array of FIG. 7A described
26 above, it is readily apparent how the advantages of the inventive
2~ array can be used to overcome this electrical connection and
27 crosstalk problem. Since both the transmitters are together
28 and can be advantageously located on the same side of the
29
- 48 -
81
æ
~ 1in~ 7 20.1671
;!
.
1 ireceiver pair, no high voltage leads need pass near the receivers
2 or the receiver electronics. The high voltage generator may be
~ I~located below the receivers and their associated electronics.
4 ` Thus, only a relatively low voltage DC supply needsto be connected
6 ll from above. This arrangement provides good electrical signal
6 ll isolation and freedom from cross talk into the much lower level
7 receiver signals.
8 ~~ Further, the novel transducer arrangement will allow the
9 i use of both directional transmitters and receivers without the
'~ necessity of adding an extra pair of transducers to provide the
11 ~ needed directivity. Since both receivers are on the same side of
12 ~ both transmitters, each receiver and transmitter has a unique
13 ~ directivity requirement, requiring no additional transducers as
14 !' in the prior art array. Still further, since the same pairs of
15 i¦ transducers are always used, differences in additional pairs of
16 otherwise needed transducers to obtain directivity in both direc-¦
17 jl tions will not occur to affect the measurement.
18 l~ An additional advantage of the TT-RR type array is its
19 1 ability to compensate for refraction effects. As apparent from
1 the prior art compensation array shown in FIG. 7A~ the signals
21 approach the receivers from different directions and incliniations.
22 il This inclination is due to the well-known refraction effect which
2~ gives the appearance that the borehole signal path intersects the
24 formation at an angle somewhat less than 90, the actual angle
26 ; depending upon the formation/borehole fluid velocity contrast.
26 Two pairs of receivers Rn and Rfand Rn and R~ respectively~
27 are illustrated in FIG. 7A to accommodate the refraction effect.
28 Each receiver is aimed with its most sensitive direction along a
29 particular inclined borehole path. Each pair is displaced to
more effectively match the position of the formation interval
81
æ - 49 -
~ 7~7 20.1671
1 simultaneously under investigation between the two pairs. This
2 displacement may be termed a refraction displacement and deter-
8 mines the small spacing between the two receivers illustrated
4 therein, which are used to take the place of the usual single
~ ' receiver for directed reception from above and below, i.e., be-
6 tween either of Rn and Rf or ~ and Rf. Unfortunately, the re-
7 fraction displacement varies not only with hole size but with
8 1I formation velocity, such that a fixed spacing between these two
9 !I receivers can 'be designed but for one displacement corresponding
at best to a nominal borehole size, formation velocity, etc.
11 , However, in accordance with this invention, variations in
12 the refraction displacement can be compensated by varying the
13 delay distance or number of depth increments between measurements
14 , made between the different same-type transducer pairs. As can be
16 ~ seen in regard to FIG. 7B, the lower pair of same-type transdu-
16 I cers sees the ~ormation interval slightly above the actual bore-
~7 '`' hole depths for these transducers while the upper pair sees the
18 intervals slightly below their actual depths. Thus, the refrac-
19 I tion displacement compensation may'be readily provided by simply
' adjusting the delay between measurements made between thesepairs
21 ,~ before their combination, as for example, decreasing the delay
22 for larger displacements between the actual position and the ef-
23 fective position of a transducer caused 'by a larger borehole
24 size, higher velocity formations, etc.
26 An additional feature of the invention m~y be seen bycom-
26 paring the formation intervals investigated as shown in FIGS. 7A
27 and 7B. In the prior art compensation arrays shcwn in FIG. 7A
28 only interval I centered about its midpoint is investigated.
ag Thus, this excludes any possibility of investigating the critical
~ interval between this point and the bottom of the hole. However,
81
æ
- 50 -
~ 17~7 20.1671 '
., .
1 i
1 ~'as can be seen in FIG. 7B, the ~.ower interval I~ on theTT-RR array iies
2 '.very near the bottom of the hole and can 'be investigated by meas-
8 'urements made between the bottom pair of transducers. While not
J ', .
,jcompensated, both short and long T-R inve~tigations can be made.
6 , Circuits to provide the ~t measurement from the upper and
iB ~jlower intervals, ~tu for Iu and ~t~, for I~ are shown by dashed
7 ¦llines in FIG 3A. For example, the ml and m2 measurements gated
8 ¦!to memory #1 via gate 392 may also be gated to up/down counter ~'~A
' 9 !sh~w~ at 398 in circuit 24A~ This counter, li~e up/down counter
0 1l#~ sh~wn at 394,. is directed'by steering pulses from steering logic
~ 332 to count down or load with a negative sign for ml during Cl'
12 l¦and up or add with a positive sign for m2 during C2'. mus, at
13 lithe end o~. C2', the contents of counter #2A are m2-ml for the in-
14 iiterval currently between T2 and Tl or ~tu. Since m2 and ml are i'
li j~both referenced to the ~hort-distance transmitter T3, this is a
16 iishort T-R distance ~tu as can.be seen in FIG. 2A at position I.
. The timing and steering may be seen in Fig. 3C.
18, A long ~-* distanoe ~tu iS similarly provided using the re-
19 i maining measurements in another up/down counter ~3A shown at 398A
i oi FIG. 3A. This counter,.'steered like up/down counter #3, with m3
i! and m4 inpu* from;gate 392A.i~ circuit 24B provides m4-m3 for the ;
22 i1 interval currently between T2 and Tl or ~tu as can be seen in
2~ FIG. 2B at position I'.
24~ For position L, and.the Iower interval IQ, the short T-R
distance ~t~ is provided'by-up/down counter #2B at 399, steered
a6 ! like counter #2 with.ml and m3 input from gate 393 to provide
~ m3-ml; and for the long T-R distance ~t~, by up/down counter ~3B
28 at 399A steered like counter #3 with m2 and m4 input from gate
29 393A in circuit 24B. Thus, long and short T-R distance ~t inves-
tigations are provided for both the upper and lower intervals
shown in FIG. 7B.
æ
- 51 - .
; l
` 20.1671 ll
., I
1 Although neither ~tu nor ~tQ is borehole-compensated, they
2 , are bbviously useful to log the borehole intervals respectively
~ present just below the casing and at the bottom of the borehole.
4 iWhen used together, they are useful as 'borehole compensation indi-
~~ 'Icators, since their difference indicates the degree of tool tilt;
ff I e.g., ~tu < ~t~ corresponds to the FIG. 4C illustration.
7 ll Referring to FIGS. 8A and 8B, there is shown alternative
8 I,circuitry for one section of signal compensation circuit 24, pre-
9 !~viously described in connection with FIG. 3A. As previously men-
.
~~tioned, it is sometimes advantageous to compare, as well as to
11 I combine, the measurements. By comparing different measurements
12 'Ithat should be substantially the same, for example, measurements
13 ~Ibetween different transmitter-receiver pairs over the sameinterval
14 ll in the 'borehole, certain borehole operating conditions that cause
16 the measurements to vary may 'be detected. If the measurements com-
16 l¦pare to a reasonable degree, their differences may be attributed
17 ~to statistical variations such that they may be combined to pro-
18 ' duce an improved or compensated measurement. However, if the com-
19 l parison disclosed an unreasonable difference, the operating con-
' dition causing the error may be indicated.
21 !' Accordingly, a circuitry illustrated in FIG. 8A allows,
22 upon the occurrence of a depth pulse, gating at 181 of delayed
28 ~ measurement m* corresponding to a previous position and transdu-
24 ; cer combination available at the output of memory, as shown in
2~ FIG. 3A, to be passed to comparator 182. Similarly, the current
26 measurement m directly comparable to the memorized measurement m*
27 ; is also input to comparator 182.
28
29 : ,
- ~2 -
81
æ
~ 20.1671
;,
1 If, for example, the delayed input corresponds to m2 at dII
2 and the direct input corresponds to measurement m3 at dJ, as
8 il illustrated in FIG. 2C, it may be expected that under normal
.
4 I conditions the measurements would be substantially equal. How- I
ever, if a detection error has occurred in one of the measurements,
6 il a substantial difference will be noted.
7 , As shown in FIG. 8A, an unreasonable difference provides
8 ' a no-comparison signal, which may be used to indicate a detection
9 problem, such as cycle skipping. However, if the comparison is
reasonable, that indication is used to gate measurements m2 and
11 , m3 to adder 183 for combination to Produce a compensated average
12 measurement from the measurements.
13 ~ Alternative circuitry illustrated in FIG. 8B is more
14 ~ appropriate for indicating the borehole compensation required
1~ to compensate either time or amplitude measurements. The memoryj
16 ~~ delayed and the direct (current) measurements are gated to
17 and compared at 182A- If the comparison is reasonable, the
18 two measurements may be then combined as described above. How- ¦
19 ever, if the comparison is unreasonable, this indication may
,, be used to gate, via gates 181C and 181D, the measurements to
21 difference amplifier 183A, whose output is su~med at 184 and
22 1 used to indicate the relative error in the two measurements.
23 The circuitry shown in FIGS. 8A and 8B may also be used
24 for other compensating purposes. As previously described in
26 regard to FIGS. 4C and 5B, the condition of tool tilt produces
26 different degrees of eccentricity for various transducers and
27 corresponding differences in the arrival times and amplitude
28 measurements, which will be indicated by relative measurement
29 '
81
- 53 -
20.1671
,
1 indicator 184A. If the tilt results in an upper transducer pair i
2 that is more eccentered than the lower pair, it would be expected
8 , that the upper or memory delayed measurement would be shorter in
4 time and less in amplitude relative to the direct measurement.
~ Thus, the difference between the delayed and direct measurements
6 1~ will pro~ide a negative indication. Conversely, if the lower
7 I transducer pair is more eccentered, the indication would be
8 ,I positive. This will be seen from the following example.
9 1 Consider measurements m2 and m3, defined as shown in
, Table II, taken when their known positions along the borehole
11 ~ correspond to the same formation interval. This takes place
12 ; when the transducer array is moved, as for example, from position~
13 ~I dI to dJ in FIG. 2C. In effect, transducer T2 replaces Tl and
~ T3 replaces T4. Formation paths B and C for m2 at dI (here ~2*~
1~ , are substantially equal to path G for m3 at dJ (here m3), and any
16 l~ errors between m2* and m3 will be due to differences in the com-
17 parable paths A and F in combination with E and H, as can be seen
18 I from FIGS. 4A or 4C. Thus, the difference m2* ~ ~3 equals A +
19 i (B + C)+E - F - G - H = (A - F) + (E - H), assuming B + C = G.
As illustrated in FIGS. 4A or 4C,~ is ess than F and
21 ~ is less than H, such that the differences (A - F) and (E - H)
22 ~, do not cancel, but are of like sign (both negative here) and
28 combine to indicate both the nature of the error between these
24 two measurements and its magnitude.
26 While the preceding discussion has generally been directed
26 to acoustic measurements, additional methods and apparatus
27 directed to other types of measurements, such as high frequency,
28 electromagnetic measurements, etc. are possible and will be
29 ,,
81
æ - 5~ -
i
I
` 20.1671
1~ 7~7 : -
1 described. First, some inherent differences in the measurement
2 techniques used in these additional applications will be reviewed.
8 FIG. 9A illustrates the type of detections used typically
4 lin acoustic travel time measurements or other measurements, where
6 the signal period or wavelength is long compared to the resolution
6 required. The signal is normally propagated as a pulse having
7 ,positive and negative oscillations beginning with its arrival and
8 ! relatively little signal prior to that time. Thus, as illustrated
, . . .
g j'at I and II, corresponding to the reception signals that might be
lo expected respectively at the near and far receivers, relatively
11 little signal is pres~nt prior to its arrival. By design, the
2 Ifirst and relatively weaker half-cycle will be provided a polarity
opposite to that used for detection. A detection threshold ampli-
14 tude different from zero to avoid noise and in the opposite polarity
16 ~,from the first half-cycle is employed. The detection corresponds
16 ' to the point Tx when the amplitude first swings beyond this
17, threshold.
18 Thus, for I in FIG. 9A, the detection at the first receiver
19 occurs as illustrated at TX1 and the corresponding detection for
II at TX2. These detection points are related in time either to
21 each other as for example, where TX1 would begin a time interval,
22 and TX2 stop the time interval for the case of differential
28 measurements, or in the case of individual sequential measurements,
24 ~ Tx may be made relative to some earlier time such as To~ In this
26 manner, the measurement ml at III, corresponding to signal
26 received at T2, would begin at To and stop at TX1 while, for T3
2~ and m2 at IV, the measurement would begin at the reference To
28 time and stop at TX2. In this manner, the difference m2 ~ m
29 provides the interval measurement ~t as illustrated at V.
81
æ - 5~ -
~` ` 20.1671
L~
1.
1 In electromagnetic measurements, the signals travel at sig-
2 ~ nificantly higher velocities and their periods are quite short
8 compared to the required time resolution. Consequently, phase
4 ~detection is usually employed rather than the zero crossing or
threshold methoa lllustrated in FIG. 9A. The phase relationship
~ Ilmay be measured between signals received from the near and far
7 receivers to obtain a differential measurement or if individual
8 measurements are preferred, to a known reference signal of the
g ' same frequency. As illustrated at I and II of FIG. 9B, the two
signals are displaced slightly as will be seen by comparing the
11 l zero crossing detection points at III for the signal on ~e line
12 ~1 I with IV for the signal on the line II. Thus, as illustrated
13 ~ on line V, the phase shift ~ between the illustrated zero cross-
14 ll ing points corresponds in much the same manner as the ~t measure-
1~ ment illustrated in FIG. 9A. Particular circuitry to perform
16 the above illustrated differential phase measurements will be
~7 ; found in the previously discussed Calvert 3,849,721 patent and
18 ` U.S. Patent No. 3,944,910 issued on March 16, 1976 to Rama Rau.
19 For an illustration of an application of the novel trans-
ducer array to an electromagnetic measurement, refer now to FIG.
21 10. The novel TT-RR compensation array takes the form of trans-
æ mitter T and receiver R antennae supported on sidewall skid 37.
2S As with the acoustic embodiment already described, two separations
a4 are identified between same-type transducer groups, here Iu be-
2~ tween receivers Tl and T2 and Ie between transmitters T3 and T4. For
26 electromagnetic measurements,Iu and I~ will be in the order of
27 a few centimeters. Two T-R distances respectively 2 and 4 times
28 I may be provided on reasonable length skids. The actual dis-
81
æ - 56-
~ ~7 20.1671
.
'1.
1.
1 tances vary, as indicated by the division between T3 and T2,
2 depending upon the frequenc~ used in the measurement. This
8 ~ frequency-distance relationship is described further in the
4 i above patents. Where phase detection is employed, care must be
6 11 taken that the distancesprovide the proper basis for phase com-
~ l¦ parison. For example, combinations of frequencies and distances
7 !¦ which result in crossing through zero phase differences should
8 1! be avoided.
g l Much of the circuitry illustrated in FIG. 10 is described
0 " in the above Calvert and Rau patents, and will not be detailed
~ here. Provisions have been added to allow making individual
12 ', T-R measurements rather than the usual R-R differential measure-
13 ~, ments. This is accomplished by providing a transmitter related
14 1' signal for use as a reference signal in place of a missing
~ receiver signal. The mode control signals M and N already
i 16 ~' described in con~unction with the acoustic embodiment are utilize
7 as well in FIG. 10, here to steer the transmitter and receiver
8 , signals and the processing circuits. These control signals may
19 'I be provided by conventionally designed square wave generators
i 60A and 60B.
21 1 As shown in FIG. 10, the transmitter signals are switched
22 ,I from the high frequency oscillator 45 by switch 47 as controlled
23 , by mode M to either leads 47A or 47B and transmitted respectively
24 ` at T3 or T4- Simultaneously, signals are also delayed and
26 ~ attenuated to simulate formation conditions for short and long
26 T-R distances by delays Ds at 40A and D~ at 40B and switched
27 ~ through switch 41 to serve as reference input 41A to mixer 50.
28 .~ ,,
U '
æ - 57 -
~9 17~7
. 20.1671
1 ~I Transmitted signals propagate thr~ugh the formation and are
2 .~received at both T2 and Tl, but only one of these signals is
8 switched 'o mixer 51 dependir.g upon switch 43 as controlled by N.
4 ,IThe phase difference measurement is made using mixer circuits
~ ¦j48 and 49, zero crossing detectors 71 and 72 and sign reversing
6 iflip-flop 77 wlth integrator 78 to produce at 78A the phase or
7 j~travel time measurement for the particular T-R combination.
8 'liFurther changes in modes M and N result in a sequence of such
9 !¦measurements~. eack made in the above m~nner by utilizing delays
0 'l~S and DQ to provide the preferred range of phase differences for
~ the corresponding T-R distance. m e four ~-R combinations have
12 !lalready been described in con~unction with M and N in relation
13 ~Ita Table II.
14~ 'J Rather than using the transmitter reference signal approach i
16` 7,las a phase comparison basis~ as ilIustrated by circuits 40, 41
lB !iand 48, alternate circuits 44 through 44E may be used. As shown
17 by dashed lines~in FIG. 10, a 100 kHz oscilla~or 44 may be used i~
18 ,~con~unction with the high frequency oscillator 52 to provide syn-
19 ,,chronous 100 k~z clock pulses 44A which are thendelayed by either~
. delay Ds to.provide signal 44B. or delay-DD to provide signal 44C. .
2I ~ These selectively del.ayed signals are then routed by switch 44D
22. .~as determined by control pulse M such that output 44E may be used
28 .to re~lace '~he similar pulses normalIy output at 71A from zero
24 crossing detec.tor 7I.
- As disclosed in the Calvert and Rau patents, it is beneficial
26 to also measure, along with the phase difference or travel time
27 meaaurements, the amplitude or attenuation of the electromagnetic
28 signals. mus~ a second set of measurements corresponding to
29 .
ao ':
81 . _ 58 _
æ
7~t~
.
20 . 1671
., ,
1 ,peak amplitudes are desired. These are obtained simultaneously
2 with the individual phase measurements by circuits 80 through 90
8 ..shown in FIG. 10 such that a continuous corresponding sequence o~
4 Iamplitude measurements for each ~-R combir.ation are provided at
jl
~ 9OA.
6 ',l Since the T-R combination measurements are acquired at dif-
~ jlferent depths, a memory and gate circuit similar to that shown in
8 !FIG. 3A may be employed~ Since the compensation provided by the
9 .use of the TT-RR array applies both to time or phase type measure-
~ ments and to ampIitude or attenuation type measurements~ it is
11 iIdesired that these different type measurements, with each type
12 ~,ha~ing two di~ferent T-R distances, be provided the compensation.
!I Slnce the signals provided at 78A and 90A may appear as
14 ilsequences of analog voltage levels, they may be converted from
,lanalog to digital measurement sequences by A/D converter 94 syn-
16 llchronized to multiplex the input sequence using multiplexor 93.
lq 1I m e depth synchronization is provided for memory delay purposes
18 I,by depth pulses 92, and the measurement sequence synchronization
19 ;~is c~trolled by control pulses M and ~. m e resulting digital
output is then gated. from the A/D Converter to individual gate,
Zl ''memory and cou~ter circuits 24C through.24~, each constructed as
22 i! sho~n in Figure 3A for circuits 24A and 24B. These compensation
23 ,,circui.ts respect~vely- output first and second investigations rep-
24 resenting dif.ferent transmltter-to-receiver distances correspond- .
~ling to output A and output B already described in regard to cir-
26 . cuits 24A and 24B. However, in this case the investigations rep- '
27 . resent separate phase and attenuation measurements as shown at
28. 96 through 99 of Figure 10.
29 '
80 '.
31
æ 59_
!L7~7
20.l6n
.. Ii
1 Referring now to FIG. 11, there is shown a.further embodi-
2 ment of the invention corresponding to the type of measurement
8 where a given transducer may be operated either as a transmitter
4 .or a receiver, such as an antenna capable of transmitting or re-
B celving elect~omagnetic~waves. Thus, in FIG. ll, the transducer
~ pairs are denoted as antenna A2 and A~ ~or one long- and short-dis-
7 "tance antennae in one pair, respectively, and As and A~ for the
8 ,other pair.
9 ~. The abll~ty to switch a given transducer of one.type to
another prov~de& the advantage of differential measurements and a
lI better dut~ cycle. Thus, a given transmission may be simultane-
12 ously received by both receivers and measured either as differen-
13 I'tial.measurement, or individually relative to the same reference
14 ,signal. Since in effect two measurements are made at the same
lE Itime, each measurement may be averaged over a longer period.
16 , Modifications iIlustrated in FIG. 11 to the circuits in
1~ already described FIG~ lO provide for switching the transmission
18 ''signal generated by osciIlator 45 to either 47A or 47B. Switch
19 ~41A, which is separately but synchronously controlled by steering
pulse N, applies the short-~istance transmitter signal to either Al
2~ .or As.and the long-distance transmi.tter signal to either A3 orai.¦
22- S~m~larly, switch.42A selects. two adjacent antennae for use as
Z5 .receiver pairs and routes- the detected signals to the eparate
2~ :mixer circuits 48 and 49 prevlously described.
In this manner, differential receiver investigations may be
26 .obtained alte~natively from the upper interval Iu using As and A~ .
27 as near and far receivers while at the same time alternating be-
28 tween As and AQ as short- and long-d~3tance transmitters. ~hen,
29 without movement of the tool, differential receiver investigations
~0 ,.
81. !
_ ~o _
æ
~ 17~7 20.1671
1 may be obtained from the lower interval I~ by using As and Ai as
2 the pair of receivers while alternating between As and A~ as the
~ transmitters. Thereafter, in accordance with this invention, the
4 array isrmoved such that A~ and As are adjacent the interval Iu
~ previously investigated by As and AQ. Processing circuitry 95
6 depth synchronizes the measurements and combines them to produce
7 the compensated first and second investigation phase and attenua- ¦
8 tion measurements already described and illustrated in FIG. 10.
9 There has been illustrated method and apparatus for maxi-
mizing the use of a four-transducer array and measurements taken
11 between different combinations of the transducers. By utilizing
12 in a novel arrangement the same four transducers normally employed
13 to provide borehole compensation measurements, these transducers
.;
14 can be used to provide measurements for determining not one, but
two borehole compensated measurements, each investigating the
16 l~same formation interval with a different transmitter-receiver dis-
17 ~ tance. Since both investigations are compensated in the same man-
18 ~ ner, this compensation adds meaningfulness to any differences
19 occurring between these different investigations and the inter- ,
pretation significance attributed thereto, such as, for example,
21 indicating the presence of gas in a subsurface formation.
æ j In general, the novel transducer array allows double use of
23 the measurements derived therefrom. The two-receiver measurements
24 are used twice at each depth increment, once each in relation to
26 the near and far transmitters. Then, in turn, the two-transmitter
æ measurements are used twice, once each relative to the near and
27 far receivers. Even the transmitter-receiver distance is in ef-
28 fect used twice by overlapping this distance, which allows desir-
29 able increases in the T-R distances without the undesirable in-
creases in the array length associated with prior art arrays.
31
æ - 61` -
7~
20.1671
1 ~; Further, since all tr~nsmitter-type transducers are located
: 2 ;on the same side of the receiver-type transducers, signal propa-
8 ~gation takes place in the same direction ~or all measurements,
4 which readily facilitates the use of directional transducers.
Further, since same-type transducers are grouped together, they
~ operate in substantially similar borehole environments, which
7 allows both the combination and the comparison of individual meas-
8 ~;urements made with different transducer combinations.
9 , While the illustrative embodiments comprised acoustic and
10~ .lelectromagnetic measurements, novel.features of the invention
~ apply as well to other types of measurements.. Furthermore, al-
12 though the receiver pair was generally illustrated as being the
1~ .upper pair of tr~nsducers and the transmitter pair as the lower
14 .'pair of transducers, it will be appreciated that ~eatures o~ the
16 i~vention wil~ be.provided by the reversed arrangement. Similarly,'
: ~6 .~ the acquisitic~ o~ measurements may be made as the transducer -
17 -, array is moved either upwardly, as illustrated, or downwardly.in ~. ^.
~ 18 the~borehole. Although the described embodiments provide ~or
19 comblni~g measurements as they are acquired at the well site, it
will be appreciated that the individual measurements may be re-
21 corded.and combined at a.di~erent time and place.
2~ The above-described. embodiments are,. therefore, intended
~ . ,
~u to.be merely exem~lary and. all such variations and modifications
24 are.intende~ to be.included within.the scope o~ the invention as
defined in the appended claims.
2B
~ -
28
29
30 .
_ ~2 -
31
æ
