Aade 17 Ntce 076
Aade 17 Ntce 076
This paper was prepared for presentation at the 2017 AADE National Technical Conference and Exhibition held at the Hilton Houston North Hotel, Houston, Texas, April 11-12, 2017. This conference is
sponsored by the American Association of Drilling Engineers. The information presented in this paper does not reflect any position, claim or endorsement made or implied by the American Association of
Drilling Engineers, their officers or members. Questions concerning the content of this paper should be directed to the individual(s) listed as author(s) of this work.
An increase in torque at the PDM will result in a corresponding set-up and memory dump at the drilling tool service facility.
increase in differential pressure along the drill string. This will Once the sensors are set-up (e.g. at the repair and maintenance
cause the SSRT to shorten length (or close), compressing a facility), they autonomously start recording while tripping in
stack of belleville springs until the closing force and spring and while drilling. No interaction with the sensors is necessary
force are equal. The amount the tool closes depends on the at a well-site, minimizing the cost of sensor deployment and
pressure increase in the drill string. making them transparent to rig crews and on-site engineers.
A decrease in torque at the PDM will result in a corresponding The downhole datasets gathered with compact dynamics
reduction in drill string pressure. This will cause the SSRT to recorders are “small data” which are well-structured and go
increase length (or open), uncompressing the stack of belleville through well-established physics-based equations to be
springs until the open force and spring force are equal. The converted to informative processed data, along with surface
amount the tool opens depends on the pressure decrease in the data or electronic-drilling-recorder (EDR) data [18].
drill string.
Proprietary software is used to merge downhole and surface
Because the pressure drop across the bit is held near constant, data and provides special visualization tools for data analysis.
it does not cause any change in the SSRT length. Only changes The software also applies data analytics algorithms to convert
in differential pressure across the motor will cause the SSRT to “small data” to actionable information as soon as surface and
extend or retract. The SSRT has 18 inches of travel on the downhole data are loaded into the software. This software and
closing stroke and 11 inches of travel on the opening stroke. workflow shorten the standard delivery time (several weeks) of
processed and actionable information within hours of tools
The belleville springs are rated at 4000 lbs. per inch of travel. being returned to service base. Figure 8 shows the basic drilling
Hydrostatic pressure is utilized in a unique way to adjust the dynamics measurements recorded at the SSRT and PDM.
operation of the tool for different drilling applications. The
response of the SSRT is opposite to what is seen from most Example #1
stroking tools such as jars, shock subs and bumper tools. These Example #1 is from two offset wells drilled with the same bit
tools will extend from the drill string internal pressure. and BHA. Both wells were low angle and drilled in 9 7/8” hole
size. The first well (#1a) utilized a carrier sub for embedded
Embedded BHA Instrumentation sensors positioned above the MWD, and offset well (#1b)
The embedded BHA sensors are designed to be compact utilized SSRT with embedded sensors. Both wells used 7-blade
enough to fit into existing downhole equipment such as SSRT bit with 16mm cutters, 7/8 6.4 PDM with 1.5-degree bent
and PDM [17,18]. The design allows for modification of housing.
existing assets to accept the sensors and eliminates the need for
additional subs and connections in the BHA. Well #1a (no SSRT) was drilled in 3 bit runs, drilling a total of
2003m in 110 drilling hours with an average rate of penetration
Figure 4 shows the embedded sensor installed under a hatch (ROP) of 18.20 m/hr. Well #1b (with SSRT) was drilled in 2 bit
cover in the SSRT. Figure 5 shows the embedded drilling runs, drilling a total of 1996m in 82.25 drilling hours with an
dynamics unit. It is contained inside a pressure barrel that is average ROP of 24.28 m/hr. Well #1b drilled the interval in one
rated for 15,000 PSI. less bit trip and at a faster overall ROP. The following
discussion highlights sections of data to compare the downhole
The embedded sensors are also designed to fit into the bit box drilling dynamics response with and without the SSRT. The
of a PDM to provide at-bit measurements. Figure 6 shows the snapshots of data are from identical depths while drilling the
embedded sensor installed in a motor bit box. Figure 7 shows section.
the embedded “puck” shaped unit.
Figure 9 shows a snapshot of data from Well #1a (no SSRT)
Both sensor designs contain the same electronics, solid-state and Table 1 gives a description of the drilling dynamics. There
sensors and batteries. The shape of the package is the only are sections where the maximum downhole string rotation
difference between the two sensors. speed was as high as 320 revolutions per minute (RPM). These
peak downhole rotation speeds correlate with an increase in
The sensor packages include onboard 3-axis inclinometers surface torque and differential pressure.
(±16G), 3-axis shock sensors (±200G), 3-axis gyros and two
temperature sensors. The sensor records burst data to memory The axial frequency spectrum data shows that there was axial
every 5, 10, 20, 30, or 60 seconds. The sampling frequency (and vibration activity and the lateral frequency spectrum shows that
anti-aliasing filters) is programmable between 25Hz and there was stick-slip present with intensity increasing when the
100Hz. maximum rotation speeds were at their peak.
The embedded sensor package has a communication port for These intervals of very high-peak RPM are the most damaging
to bit cutters. The top-drive rotary speed was set to 70 RPM;
AADE-17-NTCE-076 Stick-Slip and Torsional Oscillation Control in Challenging Formations 3
however, the downhole components were subjected to intervals surface rotary speed was set to 20-50 RPM and 130 RPM output
where the string was accelerating up to 320 RPM peak. To from mud motor. These peak downhole rotation speeds
preserve bit life these swings in downhole RPM must be correlate with an increase in surface torque and differential
managed to acceptable levels, while still maintaining a fast pressure. They also correlate with higher ROP and high
ROP. lateral/axial shocks at the bit.
A photo of the dull bit following this run is shown in Figure 10. The axial frequency spectrum data from the carrier sub does not
Damage to the cutters is typical for high stick-slip exposure, have any clear definition of axial vibration activity. The lateral
with the outer most rows (towards the shoulder) receiving the frequency spectrum shows that there was stick-slip present with
most damage. intensity increasing when the maximum rotation speeds were at
their peak. The bit from Well #2a was pulled for penetration
Figure 11 shows a snapshot of data from Well #1b (with SSRT rate and graded 1-2-CT-S-X-0-FC-PR.
over the same measured depth interval) and Table 2 gives a
description of the drilling dynamics. It is clear from the data These intervals of very high peak RPM are the most damaging
that the magnitude of downhole peak RPM was reduced. With to bit cutters. The PDM output was 130 RPM and the surface
the rotary speed still set at 70 RPM, the downhole peak rotation rotary speed was set to 20-50 RPM; however, the downhole
speeds were reduced to 175 RPM. This was a 55% reduction in components were subjected to intervals where the bit was
peak RPM experienced at the string. accelerating up to 500 RPM peak and the string was
accelerating to 420 RPM peak. To preserve bit life, the peak
The surface torque was smoother, and differential pressure was downhole RPM must be managed to acceptable levels, while
sustained at higher levels without the subsequent high peak still maintaining a fast ROP.
RPM measured downhole. The ROP was also higher than the
offset without SSRT. Figure 14 shows a snapshot of data from Well #2b (with SSRT
over the same measured depth interval) and Table 4 gives a
Figure 12 shows a photo of the dull bit following this run. The description of the drilling dynamics.
bit was still in good condition and better than that experienced
on the first well without SSRT (Figure 10). The data shows a reduction in peak RPM measured at the SSRT
and bit. The maximum downhole bit RPM was reduced to 350
Overall the data verified that the downhole response of the RPM and SSRT (string) was reduced to 120 RPM. The surface
BHA to stick-slip and torsional oscillations was reduced with rotary speed was set to 40 RPM and 180 RPM from the mud
SSRT in the BHA. Stick-slip was not eliminated; nevertheless, motor. These peak downhole rotation speeds correlate with an
it was “managed” to acceptable levels to improve ROP and increase in surface torque and differential pressure. They also
preserve bit life. correlate with higher ROP; nevertheless, lateral/axial shocks at
the bit were reduced.
Example #2
Two offset wells were drilled with the same bit and BHA in It is clear from the data that the magnitude of peak downhole
Example #2. Both wells were low angle and drilled in 8 3/4” RPM was reduced at both the bit and the string. There was a
hole size. The first well (#2a) used a carrier sub for embedded 64% reduction in peak bit RPM and 28% reduction in peak
sensors positioned above the MWD, and the offset well (#2b) string (SSRT) RPM compared to the offset well without SSRT.
utilized SSRT with embedded sensors. On this comparison test,
embedded sensors were also installed in the PDM bit box to get The axial frequency spectrum data from the carrier sub does not
an accurate data set at the bit itself. Both wells used 6-blade bit have any clear definition of axial vibration activity. The lateral
with 13mm cutters, 7/8 5.0 PDM with 1.5-degree bent housing. frequency spectrum shows stick-slip levels were lower using
the SSRT with lower intensity when the maximum rotation
Well #2a (no SSRT) drilled 1601m at an average ROP of 37.02 speeds were at their peak. The bit from Well #2b was pulled for
m/hr. Well #2b (with SSRT) drilled 2400m with an average penetration rate and graded 2-3-CT-G-X-0-LT-PR (note this bit
ROP of 25.26 m/hr. Well #2b drilled 799m further than Well drilled 799m further than offset without SSRT).
#2a eliminating one bit trip from the section. The following
discussion highlights sections of data to compare the downhole Overall the data verified that the downhole response of the
drilling dynamics response with and without the SSRT. The BHA to stick-slip and torsional oscillations was reduced with
snapshots of data are from identical depths while drilling the SSRT in the BHA. Stick-slip was not eliminated; however, it
section was successfully “managed” to an acceptable level to improve
ROP and increase run length by preserving bit life.
Figure 13 shows a snapshot of data from Well #2a (no SSRT)
and Table 3 gives a description of the drilling dynamics. There Conclusions
were sections where the maximum downhole bit rotation speed The SSRT proved effective at increasing ROP and run length
was as high as 500 RPM and SSRT (string) was 420 RPM. The by “managing” the peak RPM seen at the bit and BHA.
4 Steve Jones and Junichi Sugiura AADE-17-NTCE-076
Region Discussion
Downhole RPM burst data shows peak maximum values of 240 to 320 RPM. These peak RPM events occur
1
frequently with the longest duration for 10 minutes. Surface rotary speed set to 70 RPM.
2 Spikes in surface torque correlate with peak maximum RPM from burst data downhole.
3 Averaging 17 KdaN (kilodekanewton) WOB.
4 Spikes in differential pressure correlate with peak maximum RPM from burst data.
5 Averaging 18 m/hr. though this interval.
6 Low axial and lateral shocks (less than 5g). Peak shocks correlate with peak maximum RPM from burst data.
7 1.5Hz axial vibration visible on frequency spectrum. Indicates axial vibration possibly resulting from stick-slip.
Lateral vibration visible on frequency spectrum. Stick-slip at higher intensity during peak maximum RPM burst
8
intervals.
Region Discussion
Downhole RPM burst data shows peak maximum values of 150 to 175 RPM. This is significantly lower than
1
the 320 RPM experienced on the offset without stick-slip mitigation tool. Surface rotary speed set to 70 RPM.
Surface torque is much smoother compared to the offset without stick-slip reduction tool. This is a good
2
indication that the stick-slip reduction tool is functioning.
3 Averaging 13 KdaN WOB.
Differential pressure increases/spikes do not correlate with peak maximum RPM from burst data. This is a good
4
indication that the stick-slip reduction tool is functioning.
Averaging 30 m/hr. though this interval. Penetration rate is faster with less WOB than seen on offset without
5
stick-slip reduction tool.
6 Low axial and lateral shocks (less than 2g). Shocks are lower than offset without stick-slip reduction tool.
7 No axial frequency visible on vibration spectrum. Indication that stick-slip reduction tool is functioning.
8 Lateral frequency spectrum highlights lower levels of stick-slip which is consistent throughout the interval.
Figure 12: Example #1b – Photo of Bit Post Run (with SSRT)
AADE-17-NTCE-076 Stick-Slip and Torsional Oscillation Control in Challenging Formations 9
Figure 13: Example #2a – No SSRT (Embedded Sensors in Carrier Sub and PDM)
Region Discussion
Downhole mud motor bit box RPM burst data shows peak maximum values of 330 to 500 RPM. These peak
1 RPM events occur frequently with the longest duration for 2 hours. Surface rotary speed set to 20-50 RPM plus
130 RPM from mud motor.
Downhole string RPM burst data shows peak maximum values of 160 to 420 RPM. These peak RPM events
2 occur frequently with the longest duration for 2 hours. Surface rotary speed set to 20-50 RPM. String and bit
RPM trends correlate well.
3 Spikes in surface torque correlate with peak maximum RPM from burst data downhole.
4 10-16 KdaN WOB.
5 Several motor stalls visible on differential pressure and subsequent pick-up off bottom.
6 Averaging 6-8 m/hr. though this interval. Occasional increases to 24 m/hr.
Axial and lateral shocks at bit sustaining up to 30g with peaks up to 70g that correlates with peak maximum
7
RPM from burst data.
8 No distinct features on axial frequency spectrum.
Torsional vibration visible on frequency spectrum. Stick-slip at higher intensity during peak maximum RPM
9
burst intervals.
10 Steve Jones and Junichi Sugiura AADE-17-NTCE-076
Figure 14: Example #2b – SSRT with Embedded Sensors and PDM with Embedded Sensors
Region Discussion
Downhole mud motor bit box RPM burst data shows peak maximum values of 280 to 350 RPM. This is
1 significantly lower than the 500 RPM experienced on the offset without stick-slip reduction tool. Only a few data
point spikes above 400 RPM. Surface rotary speed set to 40 RPM plus 180 RPM from mud motor.
Downhole string RPM burst data shows peak maximum values of 100-120 RPM. This is significantly lower than
2 the 420 RPM experienced on the offset without stick-slip reduction tool. Surface rotary speed set to 40 RPM.
String and bit RPM trends correlate well.
Spikes in surface torque correlate with peak maximum RPM from burst data downhole. Downhole RPM remains
3
much more stable and consistent.
4 10-16 KdaN WOB.
5 No motor stalls indicated on differential pressure.
Averaging 10-18 m/hr. though this interval. Occasional increases 40-80 m/hr. This is faster ROP than the offset
6
without stick-slip reduction tool.
Axial and lateral shocks at bit sustaining up to 12g with peaks up to 40g that correlates with peak maximum
7
RPM from burst data. This is lower than the offset without stick-slip reduction tool.
8 No distinct features on axial frequency spectrum.
Torsional vibration visible on frequency spectrum. Stick-slip lower intensity than offset without stick-slip
9
reduction tool.