OTC 17652

Receiver Balancing in Off-Centered Acoustic Logging Data

J.O. Blanch, K. Araya, G. Varsamis, and A.C.H. Cheng, SensorWise, Inc., and J. Market, C. Kessler, and G. Althoff,
Halliburton Energy Services

Copyright 2005, Offshore Technology Conference

orthogonal directions (X or Y), and subtracting the
This paper was prepared for presentation at the 2005 Offshore Technology Conference held in observed waveforms from the receivers on opposite side
Houston, TX, U.S.A., 25 May 2005.
of the tool (e.g. A-C, B-D). A crossed dipole
This paper was selected for presentation by an OTC Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as
measurement is performed by firing both X and Y dipole
presented, have not been reviewed by the Offshore Technology Conference and are subject to sources in succession, and subtracting the observed
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Offshore Technology Conference, its officers, or members. Papers presented at waveforms. Higher order modes such as the quadrupole
OTC are subject to publication review by Sponsor Society Committees of the Offshore
Technology Conference. Electronic reproduction, distribution, or storage of any part of this
mode can be generated in a similar manner.
paper for commercial purposes without the written consent of the Offshore Technology
Conference is prohibited. Permission to reproduce in print is restricted to a proposal of not
more than 300 words; illustrations may not be copied. The proposal must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract

Modern multi-pole acoustic logging tools require addition

and subtraction of signals from different combinations of
receivers. To enable the correct subtraction and
addition of acquired signals it is essential that the
receivers are well balanced. However, correctly
balanced receivers may change their properties as the
acoustic tool is lowered into a borehole and experiences
vastly different pressures and temperatures. By using
the properties of the Stoneley wave, it is possible to
create a balancing scheme such that subtraction and
addition of waveform data can be performed as if the
receivers were perfectly balanced.

Introduction

Modern multi-pole acoustic logging tools have the ability

to generate and detect different acoustic wave modes
propagating along the borehole and in the formation. A
requirement to accomplish this is the summation and
subtraction of the acoustic signals detected at receivers
distributed around the circumference of the tool. For
example, in the WaveSonic tool, there are four
receivers located at the four orthogonal positions around
the circumference of the tool, at 8 different distances
from the source position, resulting in a receiver array
consisting of 32 receivers. A schematic layout of the tool
is shown in Figure 1.
A monopole (or compressional or P) wave measurement
is performed by firing the monopole wave transmitter,
and summing the received waveforms at the four
receiver positions around the tool (A, B, C, D). A dipole
(or flexural) wave measurement is performed by firing
the dipole (or displacement) source in one of the two
Figure 1. Schematic layout of a wireline acoustic tool
Furthermore, because of borehole rugosity, source
radiation pattern, and other factors, a mixture of
monopole, dipole, and higher order modes are
commonly generated in the borehole. We rely on the
addition and subtraction of the waveforms measured at
the various receivers to isolate the different wave modes
for proper analysis. For example, improper subtraction
of the receivers may result in the (monopole) Stoneley
wave, being mixed with the (dipole) flexural mode, thus
contaminating the shear wave measurement from the
dipole. Thus the balancing of the receivers is an
important issue in the processing of modern multi-pole
2 OTC-17652

acoustic tools.

In this paper we describe a procedure that can correct

for the receiver imbalance of the measured waveforms.
In order to better understand the effect of receiver
imbalance on full waveform acoustic logging data, we
simulated a number of examples of receiver miss-match
using the algorithm of Byun and Toksz (2003).

Synthetic Examples of Imbalanced Receivers

Figure 2, 3, 4, and 5 show the slowness-time

semblances and the high-resolution dispersion estimates
(Araya et al., 2003) for synthetic monopole and dipole
data in a slow formation. The compressional wave
slowness was 152 s/ft and the shear wave slowness
was 305 s/ft. In this situation, the Stoneley wave
slowness is around 330-340 s/ft, (varying by Figure 3. Dispersion of a pure Stoneley (monopole) wave
frequency). It is very clear from the figure that the dipole
dispersion, though similar to the Stoneley dispersion,
has a different shape, and more importantly, approaches
the formation shear wave slowness at around 1-2 kHz.
It is worth commenting that because of the broad
bandwidth of the source used in the synthetic
calculations, the slowness-time semblance peak for the
dipole is slightly slower than the formation shear wave
slowness. This is because the slower higher frequency
part of the dipole is often included in the semblance
window.

Figure 4. Semblance of a pure Flexural (dipole) source

Figure 2. Semblance of a pure monopole source, showing the

compressional and Stoneley waves

In order to study receiver imbalance we created

synthetic data, which contained monopole, dipole and
also higher order interface waves. We created this
mixed data by adding pure dipole data with pure
monopole data or by creating a slightly (30 percent)
Figure 5. Dispersion of pure Flexural (dipole) wave
imbalanced dipole source. To create the imbalanced
dipole source, we used two monopole sources in a
unreasonable to expect the appearance of other (non-
dipole configuration and allowed one to be slightly more
dipole) modes in acoustic wireline data. Still, a slightly
powerful. Due to tool off-centering and the inherent
imbalanced source pattern is not a problem as long as
difficulty of producing perfect dipole sources it is not
the receivers are perfectly matched or balanced. The
OTC-17652
source patterns we show here are a mix of pure dipole
and half strength monopole, and a dipole where one
transmitter is 30 percent stronger than the other.
Specifically for the latter case, many standard tests
would deduce a very good dipole pattern, as the two
monopole sources are only 1.14 dB different.
To assess the ability to retrieve the dipole flexural wave
we perturbed the receivers in two different manners. In
the first type of perturbation, all C receiver waveforms
were multiplied by a single factor and in the second, all
of the receiver waveforms were multiplied by random
numbers. By using standard semblance and dispersion
analysis, the nature of the waveforms could be
determined. Before performing the standard processing,
the waveforms from receiver arrays A and C have been
subtracted, in the standard manners to enhance the
flexural (dipole) interface wave (XX).
Figure 6. Semblance produced from XX processing of a mixed
dipole and monopole source, with C array 50% (1.8 dB)
imbalanced.
Figure 7. Dispersion produced from XX processing of a mixed
dipole and monopole source, with C array 50% (1.8 dB)
imbalanced.
Figures 6 and 7 show the resulting semblance and
3
dispersion when one receiver array has been perturbed
50 percent. The source pattern is a mix of pure
monopole and pure dipole. As evident in the figures, the
standard subtraction (XA-XC) has not been sufficient to
manifest the flexural wave. The interface wave can
clearly be identified as a Stoneley (monopole) wave.
Figures 8 and 9 show the same receiver imbalance but
with a source pattern of a slightly imbalanced dipole
source. Again, it is obvious that it is not possible to
identify the flexural wave. There may be a small
difference from a pure monopole source in the
dispersion curve, but the monopole/Stoneley wave
clearly dominates the data. Figures 10 and 11 show the
semblance and dispersion from the slightly imbalanced
source. Here, the receivers for both the A and C array
have been randomly perturbed from -33 percent to +50
percent. The random perturbation is probably a more
realistic example, but it is clear from the figures that
even this perturbation (a total of 3.5 dB) is enough to
completely bias the Stoneley wave to the point where
there is no trace of the flexural wave, of any significance.
Figure 8. Semblance produced from XX processing of an
imbalanced dipole source, with C array 50% (1.8 dB) imbalanced.
Figure 9. Dispersion produced from XX processing of an
imbalanced dipole source, with C array 50% (1.8 dB) imbalanced.
4 OTC-17652
Figure 10. Semblance produced from XX processing of an
imbalanced dipole source, with randomly imbalanced A and C
arrays (-33% to 50%, i.e., 3.5 dB).
Figure 11. Dispersion produced from XX processing of an
imbalanced dipole source, with randomly imbalanced A and C
arrays (-33% to 50%, i.e., 3.5 dB).
Hence, a very small source imbalance would put
tremendous demands on the receivers to be very well
balanced. However, if either the source or the receivers
are perfectly balanced the other does not need to be.
Thus, with a perfect source the receivers do not need to
be perfectly balanced, and with perfect receivers the
source does not need to be perfectly balanced.
Receiver Rebalancing
The defining property of the monopole (Stoneley) wave
is its independence of azimuth. Hence, in a perfect
measurement the receivers surrounding the tool at a
certain distance from the source should record the exact
same waveform. Thus, if the receivers in a ring are
unbalanced, i.e., do not record identical signals, it is
possible to use the monopole Stoneley wave response
and calibrate the receivers to one another. The simplest
balancing scheme is to adjust the amplitude; however
more involved schemes can be constructed by utilizing
the response to create corresponding transfer functions
between the different receivers. When the transfer
function between the receivers has been determined it is
possible to apply it for all other recordings at a particular
depth or series of depths to obtain measurements, as if
they were made with perfect receivers. Subsequent
subtractions and additions of data would thus be correct.
Data example
Figure 12. The derived relative amplitudes from a receiver ring.
The A receiver Raa (thin purple line exactly at an amplitude of 1)
is used as reference for the other receivers. The relative
amplitude for receiver B is Rab, receiver C Rac, and receiver D
Rad. As can be observed from the data, the maximum deviation
is around 25%.
For commercial Array Dipole tools, receiver matching is
generally within 1 dB over the same ring and within 2
dB over the complete array. We should note that
matching of the receivers needs to be done in their
final embodiment and not at the raw piezoelectric
element level. Individually pressure and temperature
compensated receivers would offer a distinct advantage,
in that they can be matched in the environment in which
they would operate downhole. Furthermore, from our
discussion it is apparent that blind summation and
subtraction downhole of signals from imbalanced
sources or receivers will result in identification of
incorrect formation velocities. The capability to acquire
and transmit the original waveforms from the complete
OTC-17652
array is paramount to delivering a log with correct
velocities.
Figure 13. VDL of raw dipole data
Figure 12 shows an example of amplitude imbalance of
the receivers. This is from one of the rings of the
WaveSonic tool. The imbalance has been calculated
using the monopole source and corresponding
waveforms. The amplitudes are plotted relative to
receiver A. It is clear that in the top section of the well
the relative amplitudes of the four receivers are
5
significantly different from each other. This is probably
due to borehole rugosity. In the lower section of the
Figure 14. VDL of receiver re-balanced dipole data. Receivers
were balanced using monopole data.
well, the receiver amplitudes are closer but still distinctly
different from each other. More over, the difference is
fairly constant over a large depth range, thus we
interpret this as the result of receiver imbalance. The
differences are less than 25 percent and it should be
possible to overcome such minor source imbalances.
6 OTC-17652

Figure 13 and 14 show the processed dipole logs before

and after receiver re-balancing has been applied. As the
source and receiver imbalance is small, there is only a
slight increase in the quality of the log after re-balancing.

Conclusions

Very small combined imbalances of the source and the

receivers will make it difficult to produce a true flexural
wave log, and thus a correct shear wave velocity log. If
there are imbalances in the source or receivers, the
measured waveforms can be highly contaminated by the
monopole Stoneley wave, which would eventually result
in a Stoneley wave log (which will not yield the shear
wave velocity without considerable dispersion
correction). If either the source or the receivers (but not
both) are imbalanced, it is still possible to obtain a good
flexural wave log. If both the source and receivers are
imbalanced, they cannot be more than 1-3 dB out of
balance in order to produce an accurate flexural wave
based log.

We have presented a receiver re-balancing scheme

based on the properties of the Stoneley wave. The
Stoneley wave is extremely dominant when initiatedby a
monopole source, and thus the method will be robust
under almost all circumstances. Field tests show small
enhancements when the receivers and source are
initially well balanced.

Acknowledgement

We would like to thank SensorWise Inc. and Halliburton

Energy Services for permissions to publish data and
results. WaveSonic is a trademark of Halliburton
Energy Services.

References

Araya, K., Blanch, J., Cheng, A. and Varsamis, G., 2003, Evaluation
of dispersion estimation methods for borehole acoustic data,
73rd Ann. Internat. Mtg.: Soc. of Expl. Geophys., 305-308.

Byun, J. and Toksz, M. N., 2003, Numerical modeling of effects of

tool eccentricity on multicomponent monopole and dipole
logging, 73rd Ann. Internat. Mtg.: Soc. of Expl. Geophys., 289-
292.