SPE 110383

Cement Sheath Evaluation in Nonconventional EnvironmentCase History

J.L. De Paula, V.F. Rodrigues, and R. Vicente, Petrleo Brasileiro

Copyright 2007, Society of Petroleum Engineers

This paper was prepared for presentation at the 2007 SPE Annual Technical Conference and
Exhibition held in Anaheim, California, U.S.A., 1114 November 2007.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.

Abstract
Sonic and ultrasonic tools have been widely used for cement
sheath evaluation along the last decades. The evaluation main
goal is the definition of the presence of cement in the annulus
and its ability of promoting hydraulic isolation among
different permeable zones, regardless of its density,
compressive strength or even quality of the slurry. The
challenges interpreting those logs are enormous because the
influence of many variables in the acoustic signal. The
difficulties increase dramatically if the well configuration is
not conventional. In those cases, huge mistakes can occur
requiring from the interpreter experience, understanding of the
tools and acoustic principles, and detailed information about
the well, casing, cement, and so on. The paper describes in
detail three well documented cases histories where nonconventional conditions in the wellbore prevailed and required
extreme care and experience to obtain the correct log
interpretation. The first case reports the evaluation of an
annulus with cement between two concentric casings; the
disturbances caused by reflections from the second interface in
the CBL signal are discussed and recommendations presented
to bypass this obstacle. In the second case, the challenges to
evaluate the cement sheath in a fiber glass casing are
presented; usually in this case, the first signals arriving at the
receiver do not come from the casing. Finally, the last case
history describes the difficulties to come up with the right
evaluation logging a liner cemented with lightweight slurry.
For all cases, guidelines and recommendations are suggested
to avoid misinterpretations and assure correct decisions about
cement sheath evaluation.
Introduction
The cement sheaths play an important role in the integrity of
a well throughout its life. To guaranty cement sheaths integrity
one should consider the cement sheath properties, taking into
account the effects of well operations, the cement slurry
design and testing, the cement slurry placement, and the

1
cement sheath evaluation (CSE) . This paper focuses CSE in
nonconventional environment from the end users point of
view. After a section defining and making some
considerations about nonconventional environments, a section
on the CSE process, and a brief refreshment section about
sonic and ultrasonic logs interpretation, three case histories are
presented. The CSE in the Brazilian State Owned Operator
(BSOO) fields is based on thorough analysis of the drilling
and cement operations reports, including well conditioning,
and the interpretation of sonic and ultrasonic logs outputs.
Alternative techniques, including inter-zonal communication
tests, are rarely performed in Brazil.
All the case histories focus land wells and are related to
zonal isolation. The first case is about a casing in casing CSE.
It illustrates the CSE process in the BSOO land operations in
the year 2001 and an initial misunderstanding in the
interpretation of the log outputs. Nowadays there are strict
corporative standard procedures, including check-lists for
cement logging planning, logs quality control and logs
interpretation, for CSE in the BSOO. The second case focuses
a project just started in Argentine, that makes use of fiberglass
production casing to deal with water injection corrosion. This
case has a bit of innovation as the fiberglass CSE demands
modifications of the logs processing and interpretation
procedures. The third and last case reports the first CSE of a
lightweight cement slurry with hollow ceramic microspheres
in Brazil, in 1996. It illustrates the case of an important
cement evaluation of a nonconventional slurry, in a remote
area from the point of view of gas and oil exploration and
production, in the south of Brazil.
It should be mentioned that most of the wells drilled
offshore Brazil, in the last decade, are horizontal ones
completed with open hole gravel packs. Thus the horizontal
sections are covered with sand control screens. However, there
are a few cases of cemented horizontal sections for multiple
fracturing purposes, which CSE has not been easy. In all the
horizontal offshore wells the production casings land at the
tops of the reservoirs close to the horizontal, making their
cementation an important issue. Many times a thorough
analysis of the drilling and cement operations is sufficient for
the CSE of these production casings. When the drilling and
cement jobs reports show any significant abnormality, and in
geomechanical sensitive area, cement logs are run. As large
casings and lightweight slurries have been used, the CSE is
more complex. This paper doesnt present offshore wells case
history.

Nonconventional Environment
The cement sheath environment is comprised by the cement
sheath itself, the casing or concentric casings, the rocks and
contained fluids, surrounding the cement sheath, the fluid
inside the casing, and the well configuration. The well
trajectory and formation depth, with the associated pressure
and temperature, also characterize the environment. Let us
define nonconventional environment, for CSE purposes, as
those that require special care in cement logs interpretation. It
means that interpretation is not straightforward and can not be
made by simply using rules of thumbs, charts or computer
packages maps. The nonconventional character may come
from the fluid inside the casing, the casing geometry or its
material, the cement sheath characteristics, the fluid filled rock
properties, and the well configuration, including trajectory and
depth. The more nonconventional aspects the more difficult is
the logs interpretation and the CSE. Most of the mentioned
aspects have been found in the BSOO offshore and land fields
in Brazil and abroad.
Unconventional production casings are becoming common
in the new deepwater developments, where 10 in, 11 in
and even larger casing are being used. These casings achieve
the top of the formations close to the horizontal.
Regarding trajectories there are also a few cases of
horizontal production liners cemented for multiple fracturing
purposes and a couple of extended reach wells.
Unconventional casing materials, such as fiberglass epoxy,
are in field phase trials. The first case history presented is
about concentric casings and the second is about wells cased
with 5,65in fiberglass.
Regarding cement types, although conventional slurries are
by far the most common, several unconventional slurries have
been used. Cement slurries with hollow ceramic microspheres
are becoming common in the production casing cementing of
ultra deep waters wells. The third case history focuses a
lightweight slurry in an exploratory gas well.
Concerning rocks the hydraulic isolation of oil carbonate
reservoirs, stimulated with acid formulations, from the aquifer
has been a big issue. Inter-zonal communication after acid
stimulation has been studied in Brazil2 for a long time, and it
is still a recurrent problem.
Cement Sheath Evaluation PDCA
This section makes some considerations on the PDCA (plando-check-action) cycle in the CSE process.
The cycle starts with planning, which should include the
objectives of the CSE, the procedures to be used in the do
step, and the control items for the check and action steps.
Although the experts recognize the importance of a detailed
and careful planning for the success of the CSE, many times it
does not happen. It seems to be a matter of culture.
Stimulation programs, for instance, are usually much more
detailed than CSE programs, except in special cases. A
deficient program may reduce the scope of the objectives
and/or compromise the logging operation (do) and subsequent
steps, check and action. The CSE objectives must be clearly
defined. They may include structural support, protect useable
water zones, protect against casing corrosion, prevent interzonal communication or sustained casing pressure (SCP),
attend to regulatory requirements, and production zones

SPE 110383

isolation. As mentioned before standard procedures were

established in the BSOO to assure a good planning.
The do step requires a good program, suitable tools and
trained people. A few years ago, many times the engineers at
the rig site did not have a detailed program. Proper tools are
still a challenge in some environments, such as long horizontal
wells, where tool centralization is hard to achieve. Skilled
people are sometimes an issue in face of the current
experienced personnel shortage and the long time required to
train properly an engineer in CSE.
The check step comprises the logs quality control and their
interpretation.
The action may be approve or reject the logs and make the
decision on the cement sheath ability of satisfying the
requirements of the CSE objectives. When the cement sheath
is absent or is not good enough to fill the requirements, the
decision may be to perform a remediation squeeze. In this
case, the remediation viability must be analyzed. It should be
emphasized that the CSE must consider the well life span, and
that an important follow-up, many times missed, is the use of
the CSE results for cement operations design improvement.
Sonic and Ultrasonic Logs Refreshment
As many texts3-11 thoroughly discuss sonic and ultrasonic
logs interpretation, this paper will not discuss it in detail.
However this section presents refreshment of some principles
as a background for the case histories.
To interpret cement evaluation logs, besides knowing the
basic physics, and the capabilities and limitations of the
logging tools, one should be able to analyze all the aspects of
cement design and placement that can impair the cement
sheath quality.
Acoustic impedance (Z) of a material is the product of its
density and its acoustic velocity. The regular tools, signal
processing and interpretation procedures are proper for
environments where the fluid Zf is significantly smaller than
both the casing steel Zcsg and the set cement Zc, and Zcsg is
larger than Zc. Any notably variation, designed or involuntary,
in the densities of these three materials will affect the logs
outputs. That is why the lightweight slurry case history and the
fiberglass case history illustrate nonconventional CSE.
Experience reveals that from time to time cement setting
takes much more time than predicted by lab or even in loco
slurry tests. Recently, a sequence of three deepwater
exploratory wells offshore Brazil required long waiting times
on cement and several logs runs to allow a correct CSE. Later
on an incompatibility between the drilling fluid and the
cement slurries was found. Despite the careful fluids and
cement slurry design this could not be anticipated.
Refreshments on the use of CBL tools outputs3-9.
Although the typical CBL interpretation charts and many of
the guidelines for cement evaluation logs interpretation are
based on cement compressive strength, the acoustic
impedance is a better parameter to use for CSE.
One should consider that a free pipe signature may be
caused by many causes such as mud in the annulus, unset
cement in the annulus, a microannulus between the set cement
and the casing, and/or weak cement fully bonded to the casing.
A compressive strength of at least 250 psi and complete

SPE 110383

acoustic coupling is required for the transmission of the signal

through all of the paths.
When examining the amplitude of a cement sheath along a
casing remember that spurious variations may be caused by
deviations in casing thickness, cement sheath thickness,
cement density, cement compressive strength, and tool
excentralization. Cement properties variations with vertical
depth are not uncommon once they are affected by pressure
and temperature variation.
One should remember that the typical waveform expected
when the cement is entirely coupled with the casing but not
with the formation may come from thick wall cake, mud filled
hole enlargements, soft rocks, gas cut or foamed cement, or
concentric casing intervals. The user must analyze the fluid
data (properties and fluid loss), the openhole caliper (hole
enlargements) and the lithology logs available to assess the
cause of no bond between the casing and the surrounding rock.
When the composite waveform shows both formation and
casing responses the user should consider two causes: cement
channeling or a microannulus at the cement-casing interface.
To distinguish between channel or microannulus one may
check the cement report and run the logs with and without
pressure. In the cement report he or she may check the main
causes for channeling, such as poor casing centralization,
failure to rotate and/or reciprocate the casing while circulating
and cementing, solids settling from the mud or cement
(channel along the low side of the casing), free water
separation from the cement slurry after placement (channel
along the high side of the casing). The analysis of the pressure
history since the very beginning of the cement operation until
the time of logging may reveal the creation of a microannulus.
Finally, the run with the proper pressure applied down the
casing may close a microannulus or reveal a channel or a
big microannulus. Unfortunately, it is very difficult to run
the logs under pressure from a floating rig. In Brazil it is done
rarely and only when deploying the logging tools with drill
pipes, which is very expensive.
To avoid the formation of microannulus and/or stress cracks
in the cement sheath it is recommended to pressure testing the
casing immediately after bumping the top wiper plug, while
the cement is still liquid. This procedure only can be applied
when the plug has landed well and the collar can withstand the
test pressure.
It should be mentioned that to date, there is no field proved
logging tool that can identify a channel between the formation
and the cement sheath.
In front of fast formations the first positive casing arrival is
mixed with some portion of the formation signal. In these
cases, the amplitude has no physical meaning and the
interpretation should be based on the composite waveform.
With fast formations just a qualitative interpretation is
possible. A good cement is suggested by no pipe signals
Refreshments on the use of ultrasonic tools outputs10,11.
The introduction of ultrasonic tools constituted an evolution
of logging tools to better define the cement sheath quality,
specially in environments involving gas contaminated or ultralow strength cement. These tools can also provide information
on the casing geometry and integrity. As these tools can not
see the material outer of the cement sheath, they must be run
in combination with sonic tools.

The logging companies recognize that the CSE using the

first generations of ultrasonic tools were vague at best.
Inputting the correct parameters in the software algorithms,
setting the colors of the cement maps, variation of the cement
properties along the cemented interval, excessive optimism
about cement setting times, and gas cutting are among the
main reasons for the vagueness. On the other hand they claim
that new generation of ultrasonic tools and processing are
much more reliable.
For single transducer ultrasonic tools, Harness et al.
developed a method to discern foamed cement from fluid,
even when both have the equal impedance, for multitransducer
ultrasonic tools10. This method is based on the facts that solidfree liquids have a steady activity level on logs while
multiphase mixtures, like many lightweight cement, have an
irregular activity level. Thus the impedance of single phase
fluids shows small variances, while the impedance of cement
sheath shows significant variations, allowing discerning them,
even if they have the same impedance. An approach analyzes
the vertical and horizontal rate of impedance change11.
Another approach calculates the standard deviations of the
impedance in four directions and compares them with set
thresholds.
First Case History Concentric Casings
This case history focuses the CSE in the concentric casings
interval of an exploratory well in the Bahia state, northeastern
of Brazil, in 2001. This well was drilled to 2,750 m total depth
and cased with 7in casing from 1900 m to surface. It found
one oil zone (1,720-1,730 m) and two gas zones (1,645-1655
m and 1,165-1,200 m). The upper gas zone, cased with 9 5/8in
and 7in concentric casings, should be isolated from the lower
gas zone and from the surface (Figure 1a). The logging
operation was concluded late in night, so the field engineer
decided to rely on his own interpretation and perforated the
casings in the 1,218.5-1,218.8 m interval for a cement squeeze
remediation. After that he was able to discuss his decision
with an internal advisor who decided that a cement squeeze
would be useless.
A little bit of history reveals that the CBL log was considered
inadequate to evaluate cement sheath in concentric casings
because the resonance of the outer casing induces strong
signal disturbances in the measured amplitude. In 1987 it was
verified in a field application in Brazil, in the Esprito Santo
District, that CSE in concentric casings could be done by
looking for the reflections of both the outer and the inner
casings. If one can see these reflections on the transit time
and on the VDL curves, then one can suppose that there is a
good bond between the cement and the casings12. In 1988 an
experimental study validated with field applications, revealed
that high CBL amplitudes obtained in concentric casings were
often an artifact originated from a very good cement sheath
between both casings, and a long measuring window9. Thus
the solution was simply shortening the amplitude measuring
window to assure that only the first peak (Peak E1) was
recorded. In case of large amplitude measuring window,
where the second peak (Peak E3) is recorded, instead of peak
E1, the high amplitude should be ignored. When the effects of
the external casing is clear in the transit time (TT) curve and
on the VDL (apparent frequency increase of the first arrivals

of the waveform, and chevron effects from the outer casing)

one can say that there is good bond between the two casings. It
should be mentioned that the CSE in concentric casings with
an acoustic log is a qualitative one and is not able, for
instance, to detect a mud channel.
The CBL/VDL logs of the transition region from the single
casing to the concentric casings interval of the case history are
presented in Figure 1b. First of all one can observe that the 9
5/8 shoe was placed in fact at 1,223 m and not at 1,217 m as
initially planned. Also, the figure shows a sharp increasing in
the amplitude values as the logging tool moved up above the 9
5/8 shoe (from 12 mV to 42 mV) that could mislead someone
to interpret the concentric region as poorly cemented.
However, this conclusion would not be compatible with the
stretching observed in the transit time curve (from 274 s to
278 s). Since all the parameters that could affect the transit
time measurements such as tool centralization, casing ID, and
wellbore fluid did not change along the well one can affirm
that the stretching is due to a good bonding between the
cement in the concentric annulus and the casing. The bonding
is good enough to attenuate and flat the first positive peak of
energy that arrived at the receiver. The use of an appropriated
scale range of 100 s (250 s to 350 s) instead of the usual
200 s (200 s to 400 s) was fundamental to increase the
readability of changes in the transit time. The presence of
cement in the concentric annulus made possible to the energy
reflected from the outer casing to reach the receptor just after
the casing arrivals, while the window gate was still opened,
resulting in high values of amplitude (around 42 mV). There
was also a change in the waveform signature with a frequency
increase of the first arrivals of the waveform and chevron
effects due to the reflections from the outer casing. Thus the
CBL amplitude around 42 mV (free-pipe was 62 mV) should
be ignored and the cement sheath considered good enough for
the necessary hydraulic isolation. The cement squeeze was
cancelled. The well was completed in the oil zone and in the
lower gas zone in September 2001 and hydraulic fractured in
May 2002. The upper gas zone was completed and acid
stimulated in December 2003. The well is producing oil and
gas as expected without communication between zones, in
particular between the upper gas zone and the surface. Some
conclusions are:
The procedure of ignoring the high CBL amplitude and
relying on the evidences of outer casing reflections on the
transit time and VDL was confirmed correct by this severe
case history gas zone isolation.
The field engineer was not aware of the information of the
internal paper (1986) and of the scientific study (1988).
A detailed program could have avoided the field
interpretation mistake.
A transit time scale range of 100 s must be used to increase
the readability of changes in the transit time allowing the
detection of logging tool decentralization and stretching.
Second Case History Fiberglass Casings
This case history focuses a water injection project in a land
field located in Argentine, which wells are being cased with
fiberglass epoxy casing. Fiberglass are supposed to have,
amongst other advantages, excellent corrosion resistance, low
scale build-up, are not bacteria nutrient, and can present high

SPE 110383

temperature capability. The fiberglass pipes are proper for

production casings/liners as they can be cemented, perforated
with shaped charges and support conventional completion
packers.
The production casings installed in this project are 5.65 in
OD, 5.90 lb/ft, 4.85 in ID, 0.40 in wall thickness, 6.80 in box
OD, 9,0 m joint length. They are rated for 1,150 psi external
collapse pressure, 2,000 psi internal operating pressure, 30,000
lbs axial load, and temperature up to 82C. They are run inside
a 9.625in carbon steel surface casing, and cemented until
cement slurry returns at surface.
The fiberglass differs from carbon steel in many aspects.
The most relevant for CSE are sonic velocity (vc) and acoustic
impedance (Z). For the fiberglass casings used vc is 3,387 m/s
and Z is 6.6 Mrayl. For steel casing vc is 5,334 m/s and Z is
41.6 Mrayl. Thus, the fiberglass vc is lower than the steel vc,
and the fiberglass Z is very close of the regular set cement Z.
The cement evaluation logging tools and respective signal
processing and interpretation procedures in the market were
developed for carbon steel casing. A study in the late eighties
showed that ultrasonic tools could be modified to verify
cement bond to fiberglass casing using a suitable signal
analysis program13. A computation algorithm was validated in
a lab experiment and an adapted pulse echo tool was used to
run a log of a fiberglass test well in Oklahoma, U.S.A. A brief
literature review did not reveal any further study for sonic or
ultrasonic logging tools application in the fiberglass scenarios.
Thus the project team organized a workshop with logging
companies to define tools and procedures for CSE in the
fiberglass production casing. The use of sonic and ultrasonic
tools was discussed.
As before said the ultrasonic tools can be used for CSE in
fiberglass casings with minor modifications in the signal
analysis program and a few assumptions - no intrinsic
attenuation in casing, and Z of the fiberglass casing larger than
the Z of the set cement.
Regarding sonic tools the relatively low sound velocity in
the fiberglass casing doesnt allow the clear separation of the
signals coming from the casing, cement sheath and formation.
A free pipe run was carried out before the cement job in the
first well logged in the project (Figure 2a) for data acquisition.
It can be seen that the amplitudes are very high (around 80
mV), above the expected value for equivalent steel casing (72
mV). The transit time (TT) was around 320 s as figured out
for the fiberglass casing. After the cement job in the same
well, a log run (Figure 2b) showed low amplitude (around 3
mV), very high TT, and the VDL waveforms revealed perfect
correlation with the lithology (GR curve). Thus, the amplitude
is meaningless and should not be used for CSE in fiberglass
casings. The CBL log in the concentric casings, 5.650in
fiberglass inside 9.625in steel, interval showed TT around 275
s (lower than the TT of the free pipe run, 320 s) with
several stretches and cycle skipping (Figure 2c). Thus the first
signal comes from the cement/formation or from the outer
steel casing. Hence, once again, the amplitude is meaningless.
Concerning the ultrasonic tools two wells were logged with
regular ultrasonic tools and the signals were processed with
the before mentioned modifications. Figure 2d shows an
extract of the output of an ultrasonic tool run in combination
with a CBL tool, in 5.625in fiberglass casing inside 8.50in

SPE 110383

openhole, cemented interval. The cement slurry was a class G

cement one with 12.5-13.5 ppg density. The measured TT of
the fluid inside casing was 196-202 s and the acoustic
impedance close to 1.6 Mrayl. Tracks 1, 2, and 3 are,
respectively, GR, CBL, and VDL. The CBL presents intervals
with zero or negative values and the VDL correlates perfectly
with the lithology. Thus the CBL amplitude is meaningless.
Track 4 is the cement impedance and tracks 5 and 6 are the
VDL waveform from the ultrasonic tool, taken along two
perpendicular diameters. On the whole the log revealed
reasonable results with the well being well cemented.
Unbonded sections could be seen at the bottom and at the top
of the cemented casing.
Thus the first results show that the CSE in fiberglass epoxy
casings can be made with ultrasonic tools and that CBL
amplitudes are meaningless in this scenario. Further studies
are necessary to gain confidence in the CSE in fiberglass
epoxy casings.
Third Case History Lightweight Cement
There are some conditions that can result in a relatively high
amplitude (less attenuation), which can deceive the sonic log
end user. These are lower compressive strength, low cement
density, thin cement sheath, and thick-walled casing.
Lightweight cement (low density cement) evaluation has been
a concern in the industry5 and in Brazil, in particular,12 for
many years.
This case history focuses the first application of lightweight
cement with hollow ceramic microspheres in Brazil, in August
1996. The location was a remote area, from the point of view
of gas and oil exploration and production, in the south of
Brazil. Thus there were logistic restrictions including the
availability of only a CBL tool for the CSE. This area was also
an important gas growing market. Hence, when the well hit
the first gas zone in the south of Brazil, the well test, including
the CSE, became a very important subject.
The well was drilled with nitrogen foamed drilling fluid and
cased with a 9 5/8 in production casing to 2,580 m and a liner
from 2,470 m to 3,740 m (Figure 3a). The available
centralizers allowed centralization only from the bottom to
2,400 m (stand-off > 90%). The cement design adopted a one
stage lightweight slurry (11.08 ppg), instead of a two stage
conventional slurry (15.0 ppg), to deal with a low pressure
(7.8 ppg equivalent) sandstone reservoir. The hollow ceramic
microspheres allowed to achieve high compressive strength
cement with a low density slurry. UCA analysis showed, for
the top of the liner conditions, that the 11.08 ppg slurry
developed 1,895 psi compressive strength in 24 hours and
2,979 psi in 52 hours. It also showed sonic velocity of 2,235
s/m and acoustic impedance of 3.8 Mrayl.
The cement job run as planned, except by the fact that the
top of the cement was tagged 130 m above the landing collar.
The CSE was based on the drilling and cementing reports
analysis and on the CBL log outputs. There was no ultrasonic
tool available for this job. Lightweight cements affect the
amplitude measurements of the CBL as they present less
sound attenuation than the conventional cement slurries. Thus
high amplitudes, such as 15 to 40 mV, may be indicative of
good bond. The waveform display is also affected showing
waves amplitudes less than expected11. The CBL tool was

checked in a preliminary run in the 9 5/8 casing and then

adjusted to the 7 in liner, which had no free pipe interval. The
liner was logged in two runs, without pressure at the surface
and with 1,800 psi at surface. The CBL showed amplitudes
from 15 to 40 mV, with the highest values in front of the
permeable zones (Figure 3b). The VDL showed both casing
and formation signals with good correlation with the lithology
(GR curve). According to the previous considerations the
cement was considered good enough for zonal isolation. In
face of the great importance of this exploratory well the
completion engineer decided to perforate just above the main
target (gas sandstone in the 3,441-3,453 m interval) with 8 jets
and perform a dry test. After perforating the interval 3,431.03,431.6 m, a drill stem test string was tripped in hole and a dry
test showed no gas or liquid production. The well was then
perforated in the 3,441-3,453 m interval, stimulated and tested
showing no zonal communication. Thus, this case history
proved, in a severe condition (gas zone) that it is possible to
perform a good qualitative CSE in lightweight scenarios, even
with only sonic log outputs. It would be better to have the help
of an ultrasonic log.
Conclusions
1. Good logging planning, materialized in a detailed program,
and teamwork can avoid mistakes in the cement sheath
evaluation in nonconventional environments.
2. Hydraulic isolation of oil carbonate reservoirs, stimulated
with acid formulations, from the aquifer has been a
recurrent issue in Brazil.
3. The procedures for cement sheath evaluation in concentric
casings, developed in the late eighties, were proved correct
in the first case history.
4. The second case history showed the initial concerns and the
first results of cement sheath evaluation in fiberglass epoxy
casings in a land field water injection project. Further
studies are necessary.
5. The third case history proved that the simple application of
recognized procedures in the literature allowed the correct
cement sheath evaluation in a lightweight cement case, even
without the help of ultrasonic log output. A dry-test
confirmed the good zonal isolation.
Acknowledgements
The authors thank Petrleo Brasileiro S.A for the support and
permission to publish this paper.
Nomenclature
API = American Petroleum Institute
BSSO = Brazilian state owned operator, PETROBRAS
CBL= cement bond log
CSE = cement sheath evaluation
GR = gamma ray
ID = internal diameter
OD = outside diameter
TT = transit time
TVD = total vertical depth
UCA = ultrasonic cement analyzer
VDL = variable density log
Z = acoustic impedance

References
1. Ravi, K., McMdchan, D.E., Reddy, B.R., Crook, R.: A
Comparative Study of Mechanical Properties of DensityReduced Cement Compositions, paper SPE 90068, presented at
the 2004 Annual Technical Conference and Exhibition, held in
Houston, Texas, 26-29 September, 2004.
2. Silva, M.G.P., Miranda, C.R., Vicente, R.: Optimization of
Cementing and Acidizing Operations is The Key To Minimize
Acid Damage To The Cemented Annulus, paper SPE 36112,
presented at the SPE Latin America/Caribbean Petroleum
Engineering Conference, 23-26 April 1996, Port-of-Spain,
Trinidad
3. Grosmangin, M., Kokesh, F.P., Majani, P.: "A Sonic Method for
Analyzing the Quality of Cementation of Borehole Casings,
paper SPE 1512 presented in Denver, October 2-5 1960, JPT
(Feb. 1961) 165-71: Trans., AIME, 222
4. Pardue, G.H., Morris, R.L., Gollwitzer, L.H., Moran, J.H.:
"Cement Bond Log A Study of Cement and Casing Variables,"
JPT (May 1963) 545-555.
5. Bruckdorfer, R.A., Jacobs, W.R., Masson, Jean-Pierre. : CBL
Evaluation of Foam-Cemented and Synthetic-Cemented Casings,
paper SPE 11980 presented in San Francisco, October 5-8, 1983
6. Nayeh, T.H., Wheelis Jr., W.B., Leslie, H.D.: "The FluidCompensated Cement Bond Log," paper SPE 13044, presented in
Houston, Texas, 18th September 1984.
7. Jordan, M.E., Sheperd, R.A.: Cement Bond Log: Determining
Waiting-On-Cement Time, SPE paper 14200 presented at the
SPE Fall Meeting, held in Las Vegas, Nevada, September, 1985.
8. Juten, J.J., ., Parcevaux, P .A., Guillot, D. J.: Relationships
Between Cement Slurry Composition, Mechanical Properties and
Cement Bond Log Output, paper SPE 16652 presented at
the 62nd Annual Technical Conference and Exhibition, held in
Dallas, Texas, 27-30 September 1987.
9. Jutten, J.J. Corrigall, E.: "Studies With Narrow Cement
Thicknesses Lead to Improved CBL in Concentric Casings,"
paper SPE 18028, Journal of Petroleum Technology, Nov. 1989,
p.1158-1196.

SPE 110383

10. Harness, P.E., Sabins, F.L., Griffith, J.E.: New Technique

Provides Better Low-Density-Cement Evaluation, paper SPE
24050 presented at the 1992 SPE Western Regional Meeting,
Bakersfield, California, 30 March to 1 April.
11. Frisch, G.J., Graham, W.L., Griffith, J.: Assessment of Foamed
Cement Slurries Using Conventional Cement Evaluation Logs
and Improved Interpretation Methods, paper SPE 55649
presented at the 1999 SPE Rocky Mountain Regional Meeting
held in Gillette, Wyoming, 15-18 May 1999.
12. Rodrigues, V.F.: Considerations on Cement Sheath Evaluation
Using Acoustic Logs, paper presented at the IX Well
Completion Seminar, held in Salvador, Bahia, Brazil, 5-9
October 1987 (In Portuguese)
13. Maki Jr., V.E., Hamilton, L: Cement Bond Logs of Fiberglass
Casing, paper SPE 18143 presented at the Annual Technical
Conference and Exhibition, held in Houston, Texas, 2-5 October
1988.

SPE 110383

7 23 lf/ft, TL @ 1,702.6 m

Stage collar @ 453

Top of cement: 665
Top of cement: 1050
Gas zone: 1,165-1200
9 5/8 Shoe @ 1,217
Perforations for squeeze:
1,218.5-1,218.8

Gas zone: 1,645-1655

13 3 /8 casing shoe @ 403

13.5 ppg slurry
890 m
15.8 ppg slurry
Stage collar @ 1,253.9
Top of cement: 1,550

15.8 ppg

Oil zone: 1,720-1730

Collar @ 1,253.9

Shoe @ 1,900.0

Figure 1a Well schematic Concentric casings

Amplitude = 42 mV
Transit Time = 278 s

Transition from concentric

to non-concentric region
9 5/8 shoe

Amplitude = 12 mV
Transit Time = 274 s

Figure 1b CBL/VDL including the transition from concentric casings to single casing

SPE 110383

Figure 2a CBL/VDL free pipe run in a 5.650 in fiberglass casing (before the cement job)

SPE 110383

Figure 2b CBL/VDL in a cemented interval of the 5.650 in fiberglass casing

Figure 2c CBL/VDL in concentric casings (5.650in fiberglass inside 9.625in steel) interval

10

SPE 110383

Figure 2d Ultrasonic log in 5.625in fiberglass casing inside 8.50in openhole cemented interval

9.625 in

2470 m

7 in 26lb/ft
3431m: 8 jets for dry test
Gas - sandstone

3440-3453 m

3740 m

Figure 3a - Well schematic

Lightweight cement

SPE 110383

11

GR

CCL

CBL

TT

VDL

Figure 3b CBL/VDL in 7in liner cemented with 11.08 ppg hollow ceramic cement slurry