Woods Hole Oceanographic Institution. Overview Report of DESH 5 Hydro Winch Testing on R/V ATLANTIS. May Upgrade and Renovation

Transcription

1 Woods Hole Oceanographic Institution Overview Report of DESH 5 Hydro Winch Testing on R/V ATLANTIS May Upgrade and Renovation R/V Atlantis Upgraded Markey DESH 5 Hydro Winches May 2013 Page 1 of 9

2 Report of the DESH 5 Winch Trials, May 3 to May 6, 2013 In late 2012 and early 2013 the DESH 5 winches from Atlantis were overhauled and upgraded at Markey Machinery Co. in Seattle. The upgrade had two main components; an improved levelwind assembly to allow the use of larger diameter sheaves with proper grooving and a new winch drive system with improved control. The new levelwind assembly is composed of a flagging block, metering block, VFD driven motor and the support structure. This assembly is now located inboard of the winch which required that the winches be relocated closer to the rails. The new drive system is a more modern and flexible Allen Bradley unit that was installed into the existing AC cooled cabinets. Besides having replacement parts readily available, this new drive allows for more inputs and can better control the winch s DC motor. The DESH 5 winch trial trip out of Woods Hole was arranged to allow testing of both components of the upgrade. Atlantis sailed at 1800 on Friday, May 3 and returned to Woods Hole on the morning of May 6. The test area was approximately south of Woods Hole in over 2000 meters of water. The sea state was 5 (very rough, 8 12 ft.) on May 5 and was sea state 4 (rough, 5 8 ft.) on May 6. In addition to Atlantis crew, a number of people were aboard to support the startup of the winch and to provide support in testing the winch functions and documenting the test results. Two representatives from Markey worked closely with the crew, SSSG technicians and others as the initial winch function tests were made and were followed by tests of the many new drive features. R/V Atlantis Upgraded Markey DESH 5 Hydro Winches May 2013 Page 2 of 9

3 UPGRADED LEVELWIND AND WINCH OVERHAUL CHECKS After a short set up period, the new VFD driven levelwind worked very well. Several 2000 meter casts were made with no requirement to adjust the levelwind performance. The flagging sheave on the top of the assembly followed the lead to the overboarding block well and no problems were noted. The motor, drive and winch components were checked during the casts and all performed smoothly. The first cast confirmed winch and control function as well as allowing SSSG to set up and check monitoring of the CTD pressure, wire tension and attitude. The next two casts were fully instrumented and provided very detailed and time stamped information which allowed evaluation of the automatic features. Test instrumentation on the CTD included an LADCP which senses pitch and roll, a wire tension load cell and a pressure sensor. Test instrumentation included the installed PHINS to measure ship s heading, pitch and roll as well as the 3PS winch monitoring system. While preparing for the following instrumented tests, we chose to measure the velocity of the CTD using its pressure sensor. That speed compared to the heave speed of the ship and payout rate of the winch would provide the best evaluation of the winch heave compensation performance. UPGRADED WINCH DRIVE CHECKS The upgraded winch drive system has a number of modes: 1) Standard manual mode, similar to the way the winches have been run aboard Atlantis since delivery; 2) MOCOMP mode; 3) Render/recovery mode; and Cast Hold mode. Descriptions of these modes are as follows: 1. The manual mode can be used from all 3 control stations; the console local to the winches, the doghouse and the computer lab. All stations were tested and controlled the winch well. Initially some adjustment was needed to improve the timing of when the motor is powered up and the brake is released. The doghouse controls use a new type of joystick which the operators liked. Markey personnel worked with the WHOI winch operators to customize the drives so that each joystick behaved identically. The Markey Techs were able to customize the drive s reaction to joystick movement to increase the low speed control. R/V Atlantis Upgraded Markey DESH 5 Hydro Winches May 2013 Page 3 of 9

4 2. MOCOMP mode is a new feature that uses a Motion Reference Unit (MRU) to sense the ship s motion and actively control the winch drum speed to negate the ship s vertical motion. This can only be controlled at the doghouse control station. MOCOMP can be used with the winch (a) stopped or in (b) pay in or pay out. The Markey supplied Kongsberg SeaTex MRU is located in a solidly mounted SS box local to the starboard hydroboom. MOCOMP (a), stopped, was tested at 2000m and 100m depths during the second cast. After stopping the winch at depth, the motion of the CTD was recorded for 10 minutes with the MRU turned off. Then the MRU was activated and CTD motion was recorded. Following that, the MRU was again turned off and data recorded. LADCP data from the CTD showed a slight decrease in pitch variability between MRU off and on modes. The tilt sensors showed a slight mean decrease while the winch MRU was turned on. That can be partly attributed to a decrease in wave activity (as evidenced by the ship heave velocity decrease in standard deviation from 0.43 m/s to0.40 m/s) during the MRU off time than the MRU on time. The CTD velocity and CTD load cell data were better evaluators of winch heave compensation performance. Figure cast2_test2_fig2 shows the change in CTD velocity and load cell during each test period. The CTD velocity standard deviation decreased from 0.60 m/s to 0.45 m/s. The load cell standard deviation decreased from 221 to 141 pounds while maintaining nearly equivalent mean loads (517 vs.507 pounds). Although the ships velocity and winch velocity were still sometimes out of phase, an improvement was seen in the winch heave compensation algorithm. Page 4 of 9

5 The next static test was performed at 100m depth. Sea state was nearly identical for both parts of this test. The CTD velocity and load cell standard deviations decreased from 0.43 to 0.34 m/s and 135 to 90 pounds. Mean load cell force was 496 and 492 pounds, respectively. This improvement was a similar percentage improvement in heave compensation as the 2000m static test. We conclude that we saw similar improvement with the heave compensation for the 2000 m and 100 m static tests. Following this cast the Markey engineers made further adjustments to the heave compensation algorithm based on these results Cast 3 was taken on May 6 down to ~100 m depth. A 30 minute static test was performed consisting of MRU off for 10 min, MRU on for 10 min, followed by MRU off for 10 minutes. Figure cast3_test1_fig3. Note that at the end of the first 10 minute period, the MRU was turned on with incorrect settings resulting in very erratic winch behavior. After 2 minutes, the MRU was turned off to correct that issue and then turned back on for the 10 minute MRU on test. Also note that in Panel 1 the pressure on the CTD decreases while the MRU is on. That is a result of the Markey MOCOMP requiring adjustment; the in and out compensation tended to ratchet the package up. A dramatic improvement in the heave compensation performance was evident during this test. The standard deviation of the CTD velocity decreased from 0.31 m/s to 0.11 m/s. Likewise, the load cell standard deviation decreased from 88 to 22 pounds. The standard deviation of the ship heave velocity stayed at 0.31 m/s showing that there was not any change in sea state over that time period. The final 10 minute test with the MRU off showed a decrease in ship heave velocity indicative of lower sea state. We noted a proportional decrease in CTD velocity standard deviation from the first 10 minute MRU off period. Page 5 of 9

6 MOCOMP (b), Pay in/pay out, was tested during Cast 2. The first test was a comparison of the winch heave compensation performance during a descent. The first part of the descent from 0 900m depth had the MRU turned off. When the MRU is turned off, the winch behaves like a normal winch without any heave compensation. Thus we expect the CTD vertical motion to be coupled to the ship's heave. From 900m to 2000m the MRU was turned on. The descent rate for the first 130 m was about 30 m/min. Then the speed was increased to 60 m/min until just beyond 800 meters depth where the winch was again slowed down to 30m/min. At 900 m depth, the MRU was turned on for the remainder of the descent to 2000 m at a rate of 30 m/min. The figures for this test are described below. There was an increase in the tilt variability while the MRU was on. Figure cast2_test1_fig2 has 4 panels. Panel 1 shows CTD pressure during the downcast. Panel 2 shows an increase in CTD acceleration while MRU was on. Panel 4 compares the load cell behavior between MRU off and on modes. We saw during the MRU off time that the load cell went to zero much more often than when the MRU was on. Normally the load cell going to zero might be indicative of the CTD cable going slack due to ship heave. However, the load cell was calibrated on deck and was found to have a rather large minimum load (force needed to register a non zero reading). We found the mean load cell force increase from 296 lbs.to 430 lbs. between MRU off and MRU on modes. The standard deviation of the CTD velocity increased from 0.37 m/s to 0.5 m/s. We also noted similar peak velocities (~ 106 m/min) despite the difference in descent rates. A closer look at heave velocities verified that the MRU winch velocity was out of phase with the CTD velocity. Panel 3 shows a comparison between the CTD package velocity, winch velocity, and ship heave velocity (from PHINS system). Page 6 of 9

7 Panel 3 was expanded in figure cast2_test1_fig3. We noticed with the MRU off, the CTD vertical motion was usually coupled to the vertical motion of the ship, but not always. Other components of ship motion may account for this phase shift. The ship's pitch and roll may have a significant influence on the CTD motion which may or may not be in phase with the ship's heave. A geometric adjustment of the ship's measured motion projected from the location of the PHINS to the starboard location of the boom over the CTD should improve the correlation between ship heave and CTD heave. With the winch MRU on, we found a significant increase in CTD heave due to winch heave compensation efforts that were out of phase. In normal CTD operations, this would not be acceptable and would result in not being able to use the winch with heave compensation on at all. After this cast the control system was again tuned and adjusted by the Markey engineers. Due to time restrictions to catch the slack tide in Woods Hole, the adjustments were not able to be tested. It is Markey s opinion that we will see a marked improvement during the next test. WHOI plans on continuing the testing process on Atlantis out of Astoria during the Jason voyages, when time permits, and we will plan to do further testing if necessary in conjunction with the Alvin test dives in September. Page 7 of 9

8 3. Render/recovery mode is a function where the Markey provided A B drive monitors the ship s 3PS tension monitoring system. The ship is able to define a maximum tension value in the Markey system and the drive will automatically control the winch speed to prevent overtensioning the wire. As the Markey and WHOI electronics engineers worked together to make this installation it was found that there was a conflict between the Markey and 3PS systems which prevented testing this function. Since the trials, these engineers have worked together and with 3PS to allow these systems to interact so that this function will run. The Markey and WHOI engineers will meet Atlantis on arrival in Astoria to complete this installation 4. Cast Hold is a function that allows the operator to run the winch in hands off mode. This function, which is run only for the doghouse, allows the operator to manually enter a meters out value into the touch screen operator interface and the winch will accelerate to a predetermined speed and pay out to the requested value. This function will also work on the pay in direction and, if an incorrect value is entered, automatically stop at a predetermined default location below the water, maybe 100m out. In addition to the function described above, this mode can be enhanced by engaging MOCOMP to actively negate the ship s motion or it can be enhanced by engaging the render/recovery mode which will prevent overtensioning of the wire. This function was not tested as it requires the input from the ship s wire metering system. A conflict between the Markey and 3PS systems has been resolved and we expect to install the solution in Astoria in late June. R/V Atlantis Upgraded Markey DESH 5 Hydro Winches May 2013 Thank you to the crew of Atlantis, the SSSG technicians, the Markey engineers and to Dan Torres who produced much of this material and graphics. Page 8 of 9

9 UPGRADED DESH 5 General Arrangement Drawing And R/V Atlantis Foundation Modification Drawing Page 9 of 9

10 D SHEAVE EXTREME POSITION D LINE LEAD ( 44) ( 18) 90 FLAGGING ANGLE (38) SHEAVE EXTREME POSITION 114 1/8 C C FAIRLEAD DRIVE ELECTRIC MOTOR 114 1/32 FLAGGING SHEAVE FAIRLEAD B 85 UPPER LIMIT LINE LEAD ISO VIEW B 20 LOWER LIMIT ESTIMATED* OLD FAIRLEAD LOCATION EXISTING DESH 5 FAIRLEAD ELECTRIC MOTOR AND GEARBOX 86 5/8 CERTIFIED FOR INSTALLATION BY: P Petrov FOR: DATE: 01/04/2013 NEW FAIRLEAD RELEASED FOR JOB: QUANTITY REQ D: RELEASE DATE: A * LOWER AGLE LIMIT (15 SHOWN) WILL DEPEND ON LINE LEAD (SIDE VIEW) AND FLAGGING ANGLE LIMIT (TOP VIEW) AS WELL AS FAIRLEAD HEAD LOCATION UPPER LIMIT IS 85 A CERTIFIED PPP 01/04/ REV DESCRIPTION BY DATE JOB REVISION HISTORY THIS DRAWING IS THE PROPERTY OF MARKEY MACHINERY COMPANY, INC. AND ALL PROPRIETARY RIGHTS ARE RESERVED BY MARKEY MACHINERY COMPANY, INC. THIS DRAWING AND INFORMATION CONTAINED HEREIN SHALL NOT BE USED OR DISCLOSED TO ANY THIRD PARTY EXCEPT AS AUTHORIZED IN WRITING BY MARKEY MACHINERY COMPANY, INC. PARTIES RECEIVING COPIES OF THIS DRAWING AGREE NOT TO USE THIS DRAWING IN ANY WAY DETRIMENTAL TO THE INTERESTS OF MARKEY MACHINERY COMPANY, INC. ORIGINAL JOB NO. ORIGINAL NEXT ASSY NO. UNLESS OTHERWISE SPECIFIED DIMENSIONS ARE IN INCHES (METRIC) TOLERANCES.X= ±.1.XX= ±.02.XXX= ±.005 FRACT. ±1/ " ±1/16 12" & UP ANGULAR ± 1 THREADS CLASS 2 DO NOT SCALE DRAWING 2 DATE: DRAWN BY: CHECKED BY: ENGINEER: REF DWG: weight: 11/28/2012 PPP N/A SEATTLE WASHINGTON DRAWING NO. SCALE: DESH 5 TOP ASSEMBLY D "=1 0" 1 U.S.A REV A SHEET 1 OF 2 A

11 /32 D 48 1/ /2 DRUM C (13 13/32) 18 1 D (10) 5 1/4 12 1/2 14 1/ C C 100 (86) (66 3/4) 95 3/ / (57 1/2) B (1 1/2) (29) (1 1/2) (16 3/8) (32 3/4) (49 1/8) (65 1/2) DRUM C B A SILL PLAN ORIGINAL JOB NO. ORIGINAL NEXT ASSY NO. UNLESS OTHERWISE SPECIFIED DIMENSIONS ARE IN INCHES (METRIC) TOLERANCES.X= ±.1.XX= ±.02.XXX= ±.005 FRACT. ±1/ " ±1/16 12" & UP ANGULAR ± 1 THREADS CLASS 2 DO NOT SCALE DRAWING 10 DATE: DRAWN BY: 11/28/2012 PPP CHECKED BY: ENGINEER: REF DWG: weight: N/A SEATTLE WASHINGTON DRAWING NO. SCALE: DESH 5 TOP ASSEMBLY D "=1 0" 9 U.S.A REV A SHEET 2 OF 2 A

12 SIGNED ORIGINAL ON FILE AT GLOSTEN

13 SIGNED ORIGINAL ON FILE AT GLOSTEN