Exp 402T

Icefield tool 2007 and 2052 were received at X402 BOX back from the manufacture after necessary re-calibration. With such a short transit to the first site, the time available was unrealistic to do the BOX rotational tests and directly deployed without testing. Instead all 3 Icefield tools were subjected rotational checks to verify retrospectively all of the tools' functionality at the 402T BOX.

In this case a magnetic compass was used to measure the "net-magnetic north" for the specific location of the magnetic sensors on the tools in the lab (see photo below for approximate location with the tool mounted in the pressure barrel). Since the net-magnetic north varies so much onboard the ship, often in under 10 cm scale, it is best to try and be as accurate as possible when measuring net-magnetic north for "absolute" rotational tests. Again the tests are absolute to the net-magnetic north, and as the tools had ben shipped all around the world and back, it seems best to perform absolute and not just relative rotational tests.

All 3 tools were tested within reasonable accuracy and shown below. (Note: tool 2007 is slightly-off because the net-magnetic north was measured for a slightly incorrect location on the tool (about 40 cm away). The net-magnetic north was already deviated by ~30° for this location and explains the consistent shift in the data and its is assumed to the tool is functioning correctly). 

All KESSs were checked for proper alignment with the jig and determined to still be consistently and accurately aligned.

One Palm Pilot had its battery discharged and needed to have the software re-installed. All there are charged and fully functions and will be charged again towards the EOX to avoid the necessary re-installations.

Exp 401

Icefield tool 2043 was checked for KESS alignment, data recording, and tracking relative directional data using the the Multi-Shot Protocol (MSP) and gave great results. The only difference in rotational testing was that a hand held compass was used in approximately the tools sensor position in order to use actual values for the local magnetic North for the tests. Between this and using MSP, the tool correctly recorded 0, 90, 270, 360/0 ° within positioning errors.

It was noticed (again), that batteries used in the MOTs can experience fast accelerations and decelerations and that this often causes physical damage to the battery and they no longer can maintain continual contact. Battery failure has commonly been reported with used MOT batteries used in other electronic devices and each time, they showed damage. This a much more realistic explanation for the tools dropping data and restarting memory during deployment, than the power harness coming off. For deployment this expedition, batteries were shrink wrapped as a unit and the battery spring was re-tensioned (stretched) to not allow them to slam around in the tool during core rejection (and/or partial-rejection).

Impact damaged batteries likely cutting out during tool runs. New battery unit is just 6 AAs shrink wrapped together.

Run for U1609A. 0-24H. pulled early. over pull of 55k kept doing fulls. could risk tool. Switched to half-cores next core anyway.

1609A Data used to create bore-hole survey for upper upper 211 m. Shots #s translated into core top mbsf. Need to account for tool offset. Until 3H pipe not in the formation and already showing an inclination of XXX. implies that the pipe is being deflected by currents. Same for casing? Once in the formation tool follows initial heading and formation influence. Strong bottom currents expected.

U1611A Survey 1 (logging in): Magnetometer data looked okay but pipe not moving, so good chance deflected by the pipe and casing but since they are not moving, they would cause a constant deflection. inclination data seems good but shows the hole is deviated.

PLOT

U1611A Survey2. Done up while retrieving core #XX. Pipe rotating around 17 RPM. Mag data is garbage and deflected by the moving pipe. Means that the pipe does deflect the mag data. Inclination data okay but noisy b/c pumping at 45959849 and survey was rushed and tool likely bouncing on wireline. Wait 30 seconds next try.

PLOT

U1661B Survey 1. Done up retrieving 64R. Inclination data shows that this hole is vertical. And first successful bore-hole survey with MOTs. Seems that bottom currents did not affect the coring process in hole but done after the hole was nearly completed and initially pipe could also have been influenced by bottom currents, though site is 1.5 km offset so could be different environment as well. Since not cased, seems like casing might affect the lock-in of the initial deflection.

Plot

Exp 400

All Icefield tools were checked for KESS alignment, data recording, and tracking relative directional data correctly. Tool 2007 showed bad directional values and it was determined that it had be dissembled on the previous expedition to fix a power issue. However, once an Icefield tool is opened, it needs to be recalibrated, and this can only be done in the Icefield lab. It was then also noticed that the factory seal on 2052 was replaced with electrical tape and 2 of the 3 screws were stripped, so clearly had been worked on by IODP technicians and would also need factory recalibration after the work (this was done on X395). Due to poor documentation, it was not possible to determine if that last calibration was done after the tool being opened up by us or prior. As a result only too 2043 was used for orientation. Going forwards, the MOTs should not be worked on in house, but instead sent back when issues are encountered.

Tool 2043 was deployed in single runs for both for 1603B and 1604B using the Multi-Shot Protocol. The MSP worked well and was drastically easier to process and no longer required reapplying the tool's calibration. Also the stand-pipe pressure files were no longer necessary in order to pick the data since the data is only shown for the actual orientation period. Details on the protocol, the comparisons between the old method and the MSP, as well as the BOX tests and the pre-deployment comparison tests between the two methods are given in the MOT confluence pages.

MOT_Multi-Shot_Protocol.pdf

Exp 397

The two Icefield tools were extensively used during Expedition 397. The FlexIT tools were not used.

• Core orientation procedure during Expedition 397: pressure applied before firing (500-1000 psi), heave compensator off, 5 min orientation time
  161 APC cores were oriented during Expedition 397 - 84.5% gave very good orientation with a standard deviation of ± 5°.

• Examples of orientation data 

U1586A-11H: Excellent orientation data

U1586D-7H: Bad orientation data - firing 3 m off bottom due to heave

• Plots shown below represent the Heave and Standpipe Pressure information from Rigwatch in regard to the magnetic tool face measurement at three sites. Green dot: standard deviation of M-tool face <5°, orange dot: M-tool face standard deviation between 5 and 10°, red dot: M-tool face standard deviation >10°

  U1586A (water depth: 4691.07 m)	U1587C (water depth: 3478.99 m)	U1385G (water depth: 2592.39 m)

  During the expedition, APC coring was conducted when the heave was below 3.5 m or so, which allowed a recovery of good quality APC cores. Similar comparative plots are available for all oriented holes at Sites U1586, U1587 and U1385. Rigwatch data for Site U1588 was available a couple of days before leaving ship (onshore work)
  

• At Hole U1587A, Icefield Tool 2043 stopped recording after Core 13H (APCT-3 core). The cause is unknown - nothing happened during core recovery and drilling operations, and the battery was fine. The tool was used again during the expedition and no other issue was reported.
  
• For our record, pictures of top of liners were taken to document the set screw rotation/migration. Pictures are available at IODP_Share>PMag>Core_Orientation>Set_Screw_Photos>X397

U1586A-8H: Ricochet feature (left) after firing

Rotation tests

• Icefield Tool 2043

KESS alignment OK (snubber #101). Rotation test performed in the downhole laboratory

• Icefield Tool 2052

KESS alignment OK (snubber #107). Rotation test performed in the downhole laboratory

• FlexIT Tools 0936 and 0937

FlexIT tools were tested in the downhole lab because they might be used during the expedition if the Icefield tools are not available. They both passed the KESS alignment and rotation tests.

Results of the Icefield and FlexIT rotation tests are available at IODP_Share>PMag>Core_Orientation>Experiments_and_Tests>Exp397 Tool testing.

Exp 397P Tie-up

• Discussion with Ops (Bill Rhinehart on X397P/T) regarding the procedure(s) used by Siem during core orientation. Crews are doing things differently and a standardized protocol to be used by both crews is thought to be necessary. Bill and I had interesting discussions regarding drilling actions during core orientation. Bill looked at different parameters recorded by Rigwatch for the last two expeditions which took place in the South Atlantic Ocean (Expeditions 390 and 393) and compared those to the relative quality assessment of the MOT data. Unfortunately, bad weather during Expedition 393 impacted the quality of core orientation. The Excel spreadsheet that Bill made and other relevant data are found at IODP_Share>Pmag>Core_Orientation>Core Orientation procedure. This common project between the two IODP pmag techs and Ops (in touch with Siem) is ongoing and will be continued throughout next expeditions. A first "protocol" will be hopefully tested during Expedition 397 which aims at orienting all drill holes.

• One Palm device (Ice T) had its battery dead. When a Palm device runs out of battery and dies, the Inclin application on the Palm needs to be reinstalled. For that, connect the Palm device to the computer and use the HotSync application on the desktop (information on how to do that is found in the Icefield User Guide on Confluence).

Uploader MUT2 for orientation data

• Orientation data are now uploaded to LORE by using MegaUploadaTron 2 (aka MUT2) . Information on how to upload data with MUT2 is provided on the Confluence User Guides of FlexIT and Icefield. Note that the last version (to be used) of the CreateInclinFile.exe Labview program is now on the desktop of the Downhole lab computer.

Screenshot of orientation software and MUT2 available on the desktop of the Downhole lab computer

• For FlexIT data, one can use either the pmag lab workstation or the downhole lab workstation - both are equipped with a TransIT adapter. The PP/Core description workstation should not be used (no adapter and no tool registration done during X397P).
• For Icefield data, only the downhole lab workstation can be used for practical reasons.

Rotation Test as part of KTLF after workstation change

• FlexIT Tool 0936

Test performed on August 23, 2022 on the dock (tool in pressure case) - SUCCESSFUL

Workstation: downhole lab

• FlexIT Tool 0937

Test performed on August 24, 2022 on the dock (tool in pressure case) - SUCCESSFUL

Workstation: pmag lab (Kappa)

• Icefield Tool 2043

Test performed on August 26, 2022 on the dock (tool only, no pressure case) - SUCCESSFUL

Workstation: downhole lab

Palm device: Ice cube

• Icefield Tool 2052

Test performed on August 27, 2022 on the dock (tool only, no pressure case) - SUCCESSFUL

Workstation: downhole lab

Palm device: Vanilla Ice

• Icefield Tool 2052 was used for an extra rotation test to verify that the last Palm device (i.e., Ice T) was working properly. Test (tool only, no pressure case) was successful (results not shown but available on IODP_Share).
• Because time allowed, Icefield Tool 2052 was used for a rotation test with its pressure case (Palm Device: Ice Cube). Test was successful.
• All test data are available in IODP_Share>Pmag>KTLF>X397P>Data>FlexIT and IODP_Share>Pmag>KTLF>X397P>Data>Icefield. 

Exp 393

Tool Rotation Issues

Significant heave was experienced throughout most of X393. This prevalent ship motion often resulted in tool rotation and difficulty in picking orientations for cores. This is fairly common and it would be useful to define some concepts for making the picks more objective from expedition to expedition (Note: The following discussion assumes that the palm pilot's clock, and thus tool's clock, are well synchronized with Rigwatch so that the firing times determined from the stand pipe pressure, relate directly to the tool data and are accurate the MOTs' 10 second measurement interval.)

Figure 1 shows an orientation period with basically zero tool rotation and the orientation pick is very straightforward. There is minimal noise and the tool rotates slightly right after the firing time (not included in the averaging period for the pick).

Figure 1: A straight forward pick easily automated with a script or program. In all following examples the red oval shows the period averaged and the arrow show the accurate firing time (to 10 seconds).

When the APCT-3 tool is run, the pick can be slightly more complicated as there is a wait time for the mudline, the MOT, as well as the APCT-3 tool. Figure 2 shows an example with little rotation during the MOT wait period, but the rotated significantly after firing and during the APCT-3 measurement time. Here it should also be noted that even though there is a wait time at the mudline, there still needs to be wait time for the MOT. We have had issues in past with the MOT wait time being ignored on APCT-3 runs and had to explicitly add it to the ACPT-3 data sheet to ensure it was observed.

Figure 2: An example of an APCT-3 tool run that is also a straightforward pick. Again with accurate firing times and clock synchronization automation of the pick is simple.

It is fairly common to see the tool creeping rotationally before this firing time (both for APCT-3 and non-APCT-3 cores). An example is shown in Figure 3 and in this case it does not necessarily make sense to include the entire waiting time in the average for the pick as the tool has clearly rotated. Even if it is not a not large amount of rotation it, the initial values while pull the pick off of its actual value. Instead it is like better to average the end of the interval is an or even some sort of weighted averaged, emphasizing the later data.

Figure 3: An example showing "rotational creep". In this case only the end of the interval was included in the average, but there are potentially many other options for how to handle this situation, with various levels of complexity.

Numerous times at U1583E, the tool was rotating more than 360° during the orientation period. In some cases (Figure 4) the tool was briefly stationary allowing for an attempt at a pick. In this case automation of this pick would be vary difficult and it takes a subjective decision to make the pick. In other situations there was never a stationary period prior to firing the core and no pick was possible. It seems it would be best to define a minimum stationary time, but then if we decide that "knowing where the gun was aimed for 30 seconds prior to firing, means that we know where the bullet was aimed", then why do we need to wait a full 5 minutes and significantly slow down the drilling operation?

Figure 4: Heavy rotation during the orientation period (approximately shown with the yellow highlight) with a brief stationary period allowing for a reasonable pick. As mention before, accurately knowing the firing time is a requirement for making a pick under these conditions.

Operations Issues on U1583E

The core barrel jammed at the end of APC coring in hole U1583. As a result, tool 2052 was deployed for nearly 57 hours while numerous techniques were employed to recover the core and barrel. When the tool was finally back in the lab, the data was successfully download for the entire interval and pass all of the BOX tests without any issues.  Furthermore, the battery voltage prior to deployment was 9.7 V (measured underload) and was still at 9.6 V after deployed (measured not underload). This have several implications:

1. The tools can be deployed much longer than previously believed. Obviously a tool failing and losing for data is still a potential issue with not swapping tools, but on U1560C, the same tool was deployed the entire time of APC coring through EXP-U1560C-11H. This greatly simplified the data processing chain and helped with operations but avoiding changing tools multiple times.
2. It has already been noted on several occasions that the battery life was underestimated and changed too often. This further supports that observation.
3. During the recovery attempts, the tool was subject to unfavorable condition, main from large accelerations, which have been thought to be a potential cause of tool failure after hitting hard layers. This seems to suggest that this may not be the case or maybe not as important as supposed.
4. The orientation data is available for this entire process of the core barrel recovery if Ops would like to review it in order t better understand the failed verses successful attempts. 

Exp 390

• Icefield Tools 2007 and 2052 came back from shore after repair. We performed rotation tests on both instruments. Each tool was rotated by 90 degrees four times to complete a full rotation. Both tools initially failed the rotation test for different reasons. We found out that Tool 2007 does not contain an instrument calibration (Figure 1). Icefield company was contacted to have the calibration file of Tool 2007. Tool 2052 used an old (out-of-date) instrument calibration registered in the Palm device. 

Figure 1: Failed rotation test of Tool 2007 performed after return from the vendor (return to the initial position P1 was not recorded)

• Icefield Tool 2043 passed the rotation test (Figure 2).

Figure 2: Rotation test results for Tool 2043

• The Labview program CreateInclinFile.exe had to be rewritten by Beth Novak to upload/embed the new instrument calibration of Tools 2007 and 2052. Additional rotation tests were conducted for Tools 2007 and 2052 before APC/XCB coring at Site U1556 (during preparation for logging). A new stand (Figure 3) to perform rotation test to check the tool orientation was built because we could not find a former one.

Figure 3: Stand to perform rotation test (90 degree rotation)

• Tool 2007 was tested on May 3 (Figure 4 below) and Tool 2052 on May 4 (Figure 5 below). Both tools passed successfully the rotation test (Figures 4 and 5).

Note: there is no flashing light for Tool 2052 when it is on.

Figure 4: Rotation test results for Tool 2007

Figure 5: Rotation test results for Tool 2052

• At Site U1556 (Hole U1556C), Tools 2007 and 2052 were deployed to orient APC cores 1H to 16H. Tool 2043 was not used because corresponding pressure case (#104) has a crack (noticed during X392). Icefield Tool 2007 was deployed to orient Cores 1H to 6H. The tool stopped recording after 8.5 hours in the hole (after Core 4H). No record exists for Cores 5H and 6H. The tool started again recording three hours later and Tool-M=0. Tool 2007 was again deployed to initially orient Cores 10 H to 13H, but it stayed in the hole longer and returned on the rig floor after Core 14H. Tool 2007 again stopped recording after about 2 hours and Tool-M=0. Tool 2007 stopped recording four times and restarted during its second run:

first record: tool recorded for 112.5 min and stopped for 209 min.
second record: tool recorded for 17 min and stopped for 135 min.
third record: tool recorded for 100 min and stopped for 14 min
fourth record: tool recorded for 40 min and stopped for 10 min.
last record before ending survey: 5 min

Notes: during the third record, tool stopped and restarted very quickly (10 to 20 seconds later) within a 2-3min time window. Tool stopped for the first time about an hour after shoot time for Core 10H.
Note bis: during the first survey, tool 2007 stopped an hour after shoot time for Core 4H. Cores 4H and 10H were both with APCT-3 run.

Why it happened and why Tool 2007 had this erratic behavior need to be further investigated. If there is a cause-effect relationship between APCT-3 run and orientation record for Tool 2007, why is Tool 2052 not affected? Could it be some issue with the Palm device used when running Tool 2007 (i.e., Ice Cube)?

No issue was reported for Tool 2052.

• It was noted that Tool 2007 stopped recording about an hour after the shoot time of the core used for APCT-3 measurement. Is there an issue on how the MOT is attached in the barrel when APCT-3 tool is inside the hole? A Blank test was performed with Tool 2007. Tool was turn on for about 21 hours in the downhole room (horizontal position). The tool did not stop and dip did not change (around 0 deg). A second test was performed - this time, the tool was set in a vertical position. Dip value was as expected (90 deg for the set position) and recording time was correct. These observations infer that something happened/destabilized the tool while in the hole. The hypothesis of tool disturbed by APCT-3 run is plausible and needs further investigation.
• Broken snubber (#104) of Tool 2043 was broken apart to check the inner thread after it cracked during X392 (Figure 6). Dismantled snubber is sent back to shore. Technical drawing is added in the Instrument Resources and can be found in IODP_Share>Pmag>Pmag_Documents>Orientation Tools>Icefield broken snubber 104.

Figure 6: Snubber before and after being apart to check the inner thread.

• At Site U1559, two sediment holes were cored. Only Hole U1559C was to be oriented initially, with APCT-3 measurements. I asked the permission to use Tool 2007 in Hole U1559C (no APCT-3 run) in order to test the tool. The permission was granted. In Hole U1559C, Tool 2052 was deployed to orient Cores 1H to 6H. No issue was reported during core orientation. The tool seemed to work properly. However, coring conditions were not ideal and the top part of Hole U1559C might not have been oriented. Therefore, as we were going to use Tool 2007 in Hole U1559D, we decided to deploy Tool 2052 for Cores 1H to 3H and Tool 2007 for Cores 4H to 6H (with a different Palm device, Vanilla Ice). Unfortunately, Tool 2007 stopped recording during the survey at several intervals for 30 sec-1 min each. These breaks in the record did not jeopardize the core orientation as they happened 45 minutes after firing the last core. There is no reason (except shock due to firing) that could explain why Tool 2007 stopped recording.
• We started writing the end time of the survey to check the clock consistency of the record.
• A long-run test was conducted with Tool 2007 on May 24, 2022. The tool was left in the upright vertical position for 97h15min (survey was stopped on May 28, 2022). The dip value and Tool-M were as expected. Slight variations in Tool-M are due to the fact that the tool was moving in its stand due to the ship motion (Figure 7). The tool did not stop recording during the survey. The Palm device used was Vanilla Ice. It is possible that the tool has a technical/mechanical issue and disconnects when the barrel is fired. Icefield company was contacted and their reply from May 31, 2022 was: "If the tool seems to be working above ground and reporting consistent results, then it's difficult to say what could be happening downhole. Is it possible that there is any magnetic interference that could be getting introduced downhole that is skewing the magnetometer results? i.e. magnetic ore, drill pipe. This could explain discrepancies in magnetometer azimuth results and large jumps. [...] If the tool is working fine above ground though and it passed our calibration tests here not too long ago, I'm not sure what else could be affecting the tool in any way. " [Victor Randell, Icefield]
  

Figure 7: 97-hour long survey of Tool 2007 showing no break in the record

• Snubber and battery connector for Tools 2007 and 2052 were measured with a caliper to check the status of the strings and whether there is a big difference in these mechanical parts between the two tools (Figure 8). The string for Tool 2052 connector is slightly tilted, and the snubber is shorter compared to Tool 2007.

Figure 8: Comparison of mechanical parts between Tools 2007 and 2052

• Shock-like tests were conducted on Tool 2007 to test our hypothesis of disconnection during firing. Tool was inserted in its pressure case. The tests conducted on June 1 consisted of (i) banging the pressure case+tool, set horizontally, with a hammer along the pressure case and (ii) banging it vertically on a piece of wood. This short survey did not show any interruption in the recording. On June 2, another test was conducted. This time, the tool was set horizontally and the top of the pressure case was banged several times with an hammer. The survey did not stop. However, as reproducing shock due to firing is not possible in the lab, this hypothesis remains the most plausible. Tool 2007 is shipped to shore at the end of X390.
• All tests conducted during Expedition 390, including those presented above are available in the folder "Exp390 Tool testing" at IODP_Share>Pmag>Core_Orientation>Experiments_and_Tests.

Exp 396

• tool 2052 failed rotation tests. Each time the tool was rotated 90 degrees, a different angle was recorded which varied from ~ 20 degrees to 210 degrees. Initially the tool was tested side by side with tool 2043 (which passed all tests) and it was believed that there was interference between the two tools. Tool 2052 was tested on its own and still displayed the same behavior. See image below for an example where the tool recorded rotations of ~50, 50, 160, and 100 degrees. It was decided to send tool 2052 back to icefield to be inspected. Tool 2007 was returned to icefield on the previous expedition Exp395P for faulty behavior. This leaves only tool 2043 onboard until the tools are returned.
• 4 cores were oriented at site U1567A and another 4 cores were oriented at site U1568A. The MOT corrected declination was compared to the core mean rotated declination and most cores were within 20 degrees between the core mean rotated declination and the MOT corrected declination. Core 5 at U1567A was approximately 180 degrees from the core mean rotated declination. The set screw hole on the top of the core liner was not observed for site U1567A but no evidence of rotation was observed at site U1568A.

Exp 390C

• All magnetic orientation tools (MOTs) appeared to be functioning properly (Icefield tools deployed – no Flex-IT tools). Both Icefield and Flex-IT Keyed End-Seal Snubbers (KESSes) were properly aligned and documented.

Exp 384

Relative Testing of Magnetic Orientation Tools Onboard the Joides Resolution

After leaving port and losing the ability to conduct diagnostic tests away from the magnetic influence of the ship, a reliable method was needed to continue testing all 5 MOTs. Because of the extremely variable magnetic environment onboard the ship, both spatially and temporally, it is not feasible to achieve absolute magnetic tool face (MTF) values. Several attempts have been made to do this in the past, and while creative they all have different flaws. On top of the derrick or on the zodiac are likely the best magnetic environments “onboard” the JR but both have major issues with safety, permission, and the physical motion from the sea itself. As others have tested in the past, we decide to conduct the tests from a relative rotation point of view. However it was apparent that the magnetic environment that each tool potentially experienced was different enough, that each tool needed to be independently normalized to its initial position.

An example is shown in Figure 1 from MOT Icefield 2043 showing 6 repetitions of clockwise rotation (viewed from above: increasing declination) in 4 steps. All data was normalized relative to the average value of Position 0 of Test #1 and shows the tools ability to properly record relative rotations versus time. All subsequent test data are normalized in the same manner.

Figure 1

All 5 MOTs were tested in the same manner, with 6 different tests of various relative rotations to themselves and each other. Each time, all of the tools correctly recorded the relative rotations versus time. Example test results showing all 3 Icefield tools plotted together after being normalized are shown in Figure 2 and the results for Flexit tools are shown in Figure 3. From these experiments one can see that the 5 MOTs are capable of recording relative rotation verses time.

The results of these relative rotation tests, coupled with dockside tests confirming the MOTs are correctly measuring absolute MTF, provide strong evidence that the MOTs are in fact performing satisfactorily.

Figure 2

Figure 3

The raw excel file used to conduct and record the relative rotational experiments  as well as a blank test log is located on the Vol1 server at V:\IODP_Share\PMag\Core_Orientation\Experiments_and_Tests\Exp384 tool testing\orientation testing\Test Logs. These are included only as a suggestions, but using them would help to keep the testing methods consistent and much easier to interpret from study to study. A significant period of MOT testing period for Exp384 was spent on “discovery and detective” work from previous studies.