A

SPE 49116
Optimization of Cuttings Removal
Measurements While Drilling

by Use of Real-Time Annulus Pressure

B.R. Pedersen, Baker Hughes INTEQ, J. Zajaczkowski, Statoil

Cepyflgh!

1S98, Scciety of Petroleum

Engineers,

Inc.

TMs paper was prepared for presentation at the 1S3S SPE Annual Tcwhnioal
ExMblion held in New Orfeans, Louisiana, 27FS0 September 1s98.

Conferen@

Service Description
The annular pressure reading is performed as a part of a
Modular Dynamic Pressure service (MDP) which comprises
downhole weight on bit, torque, bending, and internal and
borehole pressure readings. The sensors are fitted in a stainless
steel housing with an integral stainless steel barrier diaphragm.
The two sensors are combined in a sealed double ended
c@dge.
The sensor features a four arm active Wheatstone
Bridge strain gauge dift%sed directlly into a silicon diaphragm.
The system is powered from the MWD turbine driven power
supply. This limits measurements to be made while flow is
above a critical rate. The service is provided with 6.75 and
8.25 tool sizes. In addition an integrated annular pressure
sensor is mounted optionally in 4,75 MWD assemblies. A key
requirement for a downhole pressure measurement is a high
level of confidence in the accuracy of the pressure readings.
This is maintained through highly specified calibration and
verification routines. The dynamic range for the pressure
sensors is O-15000 psi within O-125 deg C operating range.
The sampling frequency for all sensors is 150 cycleskec.
Buckets of 5 seconds sampling is acquired. This gives 750
Mean and variance values are
samples in each bucket.
calculated for each channel. Average values for 2 buckets are
then stored in memory every 10 sees, and sent to surface apx.
every 2 minutes. The bit resolution is 7.5 psi in real time and
The analog voltage signal is digitized
0.28 psi for memory.
and corrected for non- linear drift and temperature drift, using
two sets of a third order polynomial. The six calibration
coefficients are calculated based on a sensor level response.
After the sensor is installed in the sub, the response is verified
on a system level. The response is specified with an accuracy
of +/- 1%. Response curve is verified each time the sub is
passing through the local maintenance shop.

and

Thk paper was selected for presentation by en SPE Pregram timmittee

following review ef
informstien crmained
in an abatracf submitted by the author(s). Centents of the paper, as
presented. have not been reviewed by the Smiety of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, dws not neceaeerily reflecl any
peeitien of the Seciety of Petroleum Engineers, its officers. or members. Papra presented at
SPE meetings ere subject to pubficetion review by Editorial Committees of the Seciety of
Petroleum Engineers. Electronic repreductiem, distribution, or storage of any part ef this pepr
for commercial purpeaes without the vnitten cenaent of the Smiety of Petroleum Engineers is
prohibited. Permission to reproduce in print ia restricted to an abstract of not more than 300
words
illuatratlona
may
not be mpled.
The
abstract
must
mntain
cefwpicuous
acknowledgment
of where and by whom the paper was presented. Write Librarian, SPE, PO.
Sox SS3S3S, Riihardeen, TX 750S3-SS3S. U.S.A.. fes 01 -972 -952 -94S5.

Abstract
This paper demonstrates
how real-time
monitoring
of
downhole annulus pressure can be used to reduce risks and
optimize the drilling process especially in the extended reach
applications. Field data shows how real-time pressure data are
used interactively to optimize cuttings removal and adjusting
mud weight. The ability to monitoring and subsequently
controlling
downhole
circulating
pressure
enables
tight
acceptance windows between estimated pore pressure and
fracture gradient.
Introduction
New directional
drilling technology
has encouraged
the
industry to drill longer horizontal wells, with increased rates of
penetration. This has created new challenges for efficient
cuttings removal, demanding
higher flow rates. Cuttings
buildup creates higher pressure drops, and increased chance of
fracturing the formation, or packing off. The desire to push
casing shoes further down has left tight margins between
fracture gradient and pore pressure. When these tight limits are
combined
with high pressure loss along the borehole,
surveillance and control of downhole fluid pressure becomes
critical.
This paper describes the use use of a downhole annulus
pressure service and its application for the Gullfaks field in
the Norwegian North Sea. The acquisition system is described,
and example logs are shown with interpretations.
The
experience is condensed to a Best Practice document intended
to be followed for the drilling operations.

The Gullfaks Field

Real-time monitoring of downhole annulus pressure was
introduced on Gullfaks field in fall 1996. Since then this part
of MWD service has become a standard on all wells drilled on
the field. The service was originally performed only in 8 1/2
reservoir section. After gaining experience from reservoir
sections, the service was also introduced in 12 1/4 and 17

565

B.R. PEDERSEN,

J. ZAJACZKOWSKI

SPE 49116

pressure loss for drilling with no rotation, and underestimating

the pressure loss for rotations above 100 rpm. On well C-36
the MDP sub measured an ECD of 1.72 S.G. with no rotation,
with mud weight of 1.62. At 130 rpm string rotation the ECD
increased to 1.77 S.G. That corresponds to a frictional m-essure
increase of 5070 with rotation.

1/2 hole sections, Statoil uses real time annular pressure data
for ECD controll and for monitoring of hole condition. Real
time annular pressure readings are considered
the best
available tool for hole cleaning surveillance.
Example well C-36
Well 34/10-C-36 was drilled by Statoil on Gullfaks field in
1997. The objective of the well was to penetrate 4 targets in
the Shetland group horizontally. The 24 section reached down
to 1475m with 72 degrees inclination at TC of section. 13 3/8
casing was set at 3205 m MD and 10 3/4 liner at 4585 m MD.
The well was further drilled with 9 1/2 hole assembly with
planned TD at 7367 m. The formation below 10 3/4 liner
consisted of sandstone, siltstone,claystone
and coals beds.
Pore pressure and formation strength varied a lot throughout
drilling interval giving a very tight window for minimum
mud weight and maximum ECD. Due to a very long 13 3/8
casing and high angle, hole cleaning
problems
were
encountered. Large amounts of cuttings were accumulated in
the 13 3/8 casing section above the 10 3/4 liner as the result
of almost 50% reduction of mud velocity in annulus. Mud
losses occured in the beginning of the 9 1/2 hole-section,
most probably in a coal bed. The ECD value measured by
was
MWD MDP tool before occurance of mud losses
considered as leak off test for futher operation. This was
confirmed later during drilling operation. Exceeding leak off
test value caused lost circulation. This was critical in order
to maximize pumprate and string rotation to improve hole
cleaning. Acquiring real time ECD values enabled adjusting
flowrate and string rotation to optimize hole cleaning and at
the same time avoid lost circulation.
One can see from fig. 1 how ECD increases while drilling in
rotational mode, as the hole is being loaded down with
cuttings. In steering mode, one can see a drop in ECD of
2/100 SG (Specific Gravity). This is in agreement with
expectations. An unexplainable steady drop of yet 2/100 the
next 90 minutes occurs. As the backreaming starts at 16:30
hrs, one can see the ECD increase with 2/100 SG as the hole is
being loaded down by cuttings.
The ECD values obtained while drilling were also useful
for designing of cementing program during completion of this
well. A precise value of maximum acceptable ECD without
loss could be entered into the model, hence reducing the
uncertainty.

Prl

,m

1 Dmth

E
,

Drilling while
rotating

Fast drilling,

7$1,91

increase in ECD

1;

..>

.. .
.:

>
lteering
5}

),.,?

,,,0.,

=~.
w-

,,.-.
--~
1

Increase in ECD

f - ~

00,

p=====+;

.:
f

l=..--+-.,;

-l-w

Ig 1 -Example of time based plot from well C-36.

Best Practices
Based on the use and experience of the annular pressure
measurements while drilling, Statoil has made a Best Bractice
document. The intent of this document is to establish best
practice for efficient hole cleaning during drilling operations.
The Best Practice document for use of annular pressure
measurements was originally written for the Gullfaks field but
is now being introduced
in general for Statoil on the
Norwegian Continental Shelf.
Experience shows the necessity to optimize hole cleaning.
Complex well designs and tight margins between mud weight
and formation strength, in addition to uncertainty in pressure

Simulation and Measurements

For ECD simulations on the Gullfaks field the Baker Hughes
INTEQ
program
EAP
Hydraulics
has been used.
Comparisons with empirical data shows that the Power Law
model shows best consistency with downhole recorded data.
The EAP power law model does not take into consideration the
effect of temperature, pressure, gelstrength, eccentricity of the
drillpipe, and rotation of drillpipe. Despite these limitations the
model gives a fairly good correlation with recorded values for
drilling with 100 rpm. The model is overestimating
the

566

SPE 491160PTIMIZATION
WHILE DRILLING

OF CUTTINGS

REMOVAL

AND MUD PROpE~TIES

ANNULUS

PRESSURE

MEASUREMENTS

Data Improve Reservoir Drilling Performance,

SPE
Drilling & Completion, March 1998.
2. Easton, M. D.J., Nichols, J., Riley, G.J., Optimizing
Hole Cleaning by Application of a Pressure While
Drilling TooI, SPE 37612.
3. Zajaczkowski, J., Retningslinjer for Gullfaksfeltet for
under
Hullrenskning
Effektivisering
Boreoperasjoner, St;oil document GF/980004R 1998.

prognosis,
limits the acceptable
limits for equivalent
circulating density (ECD). A lack of effective hole cleaning
has in several instances induced the ECD to exceed formation
strength, giving 10SSof circulation.
Through the use of ECD measurements the importance of
pipe rotation for cuttings removal is demonstrated. Rotation of
the string gives higher ECD. In highly deviated wells this
effect is amplified. In the best practices document a pipe
rotation of minimum 150 rpm is adviced. Wherever this is not
achievable,
the ineffective transport of cuttings must be
compensated by other means. After a steering section a rapid
start of string rotation can give a slug effect and an
undesirable increase of ECD. In these cases rotation must be
increased gradually over some minutes.
When tripping in and out of the hole there is no way of
pressure surveillance in real time. The effect of swab and surge
can only be estimated. Pump rate and ECD readings prior to
pulling out gives some indication of acceptable pulling speeds.
The restart of drilling after a connection can be critical in
wells with poor hole cleaning.
It is therefore adviced
establishing guidelines on how to resume to normal pipe
rotation and flow without exceeding the fracture limit of the
formation.
Conclusions
1. Measurement of ECD while drilling is a far mor reliable
and accurate tool than hydraulic simulations combined with
surface parameters. Continuous measurements is the best way
for hole cleaning surveillance and minimizing problems related
to loss of circulation. Downhole annulus pressure is recorded
downhole, transmitted to surface, transformed to an ECD
reading, and displayed at the displays of the driller, data
operator, MWD operator and company representative.
2. Hydraulic models can predict ECD fairly close (within
2/1 00 of SC). A better understanding and refining of the
models is achievable through comparisons of models and
recorded values.
3. Problems related to loss of circulation was previously
often blamed on formation properties. Some of them can now
be eliminated through the use of ECD measurements in real
time combined with best practice documents.
4. Having a verified and accurate limit for formation
fracture gradient limits the uncertainties when modeling a
cement job. The risk of fracturing and losses while displacing
cement is then minimized.
5. The effect of pipe rotation on hole cleaning is verified
through the use of ECD.
Acknowledgments
We thank Mahendra Gohil for producing the time based log.
References
1. Ward, C and Andreassen,

BY USE OF REAL-TIME

E Pressure-While-Drilling

567