| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Steve Horne |
| Physics operator(s): | Robert Granetz |
| Engineering operator(s): | Unknown |
| Engineering Operator Run Comment |
| power supply test, & M.P.41 |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Physics Operator Summaries |
| Entered: Jul 7 2004 04:36:50:990PM |
| Author: To Be Determined |
| Physics operator summary for Run 940429 Session leader: S. Horne Physics operator: R. Granetz Mini-proposal 041 (get machine going again) Well, it was not the most successful run. Most of the shots were spent trying to fix the OH2 supplies, which were not following the voltage demand very well until shot 8. (There was a late breakdown on shot 5, which was basically a fluke.) On shot 9 we stumbled upon a very weird bug in the hybrid software: if the value of +10 is entered for the PID proportional gain, it seems to be loaded into the hybrid as -10 ! No indication of this is reflected in the displayed waveform. Since we were programming BR, this caused very confusing behavior of the OH2's. Unfortunately it took us three shots to figure this out. Shots 12-14 were serious attempts at plasma. BR and BZ and dBz/dr were tweaked until we finally got a reasonable null on shot 14, with a flash starting at 0.007 s. There was no current rise however. Shot 15 was supposed to be an exact repeat of shot 14, but OH2U had a 2 kA difference between the two shots. The OH2's still do not seem to be completely debugged. Shot 16 was also supposed to be an exact duplicate of 14, but EF3 failed. Summary: 1 good null/breakdown (shot 14) 1 weak breakdown (shot 5) 1 bizarre hybrid software bug found (shot 9) And multiple power supply problems, especially with the OH2's |
| Session Leader Comments | |||
| Apr 29 1994 10:14:54:620AM | Steve Horne | Continuation of MP 41 -- Horne, Granetz
The OH supplies are suspect. Engineering wants a PLC controlled OH1 only shot; we will follow with a shot with OH only, hybrid controlled, invert at 40 msec to test the stack. | |
| Physics Operator Comments | |||
| Apr 29 1994 10:49:20:580AM | 940429001 | Robert Granetz | PLC test of OH1 only.
|
| Apr 29 1994 10:49:43:130AM | 940429002 | Robert Granetz | Another PLC test of OH1 (opposite polarity this time)
|
| Apr 29 1994 11:00:17:730AM | 940429003 | Robert Granetz | One more PLC test of OH1, time time at full voltage.
|
| Apr 29 1994 11:13:36:810AM | 940429004 | Robert Granetz | Test of OH1, OH2U, and OH2L under hybrid control. We are trying to determine
if yesterday's problems with OH2L have been fixed. (It wouldn't get to the requested voltage yesterday, resulting in a large imbalance between OH2U and OH2L.) |
| Apr 29 1994 11:24:47:330AM | 940429004 | Robert Granetz | Excellent. The OH2's were well-balanced, as demanded by the BR control.
Next shot will be a full plasma attempt. |
| Apr 29 1994 11:58:20:530AM | 940429005 | Robert Granetz | Attempt at full plasma...no programming changes from the end of yesterday's
run. Got a flash, but the problem with OH2 imbalance has resurfaced. Actually neither supply reached its demand, but OH2L was significantly worse. Presumably this problem must have something to do with a synergistic coupling to the TF. My guess is that the TF causes the primary voltage to the OH2's to droop too much. Next shot: a test of the OH1, OH2's, and TF |
| Apr 29 1994 12:27:52:230PM | 940429006 | Robert Granetz | This shot is a test of the OH stack with the TF on.
Well, the OH2's did better, but still not good enough. OH2 demand was 21 kA, and OH2U got to 19 kA, while OH2L got to 17 kA. We're taking a 20 minute break to let the engineers go in and tweak things. |
| Apr 29 1994 12:47:46:770PM | 940429007 | Robert Granetz | Not any better.
|
| Apr 29 1994 01:44:55:280PM | 940429008 | Robert Granetz | Another test of OH1, OH2's and TF. This time the adjustments the engineers
made were much more successful. OH2 demand was 21 kA, OH2U got to 21, OH2L got to 20. Next shot will be an attempt at plasma. |
| Apr 29 1994 02:25:01:080PM | 940429009 | Robert Granetz | For this shot, the proportional gain on the BR feedback was doubled from
5 to 10. Somehow this really screwed up the OH2's. In fact, at the exact time when BR feedback turns on, the OH2U suddenly turns around and quickly goes positive. We checked the sign of the M-matrix elements for the BR wire, and they are correct. We don't understand this at all. No hybrid errors reported in the error log. For the next shot, return the PID gain for BR back to 5. |
| Apr 29 1994 03:10:06:530PM | 940429010 | Robert Granetz | The SCR switches didn't close on either of the OH2's. They charged up
through the resistor and therefore didn't commutate. Next shot: repeat |
| Apr 29 1994 03:14:37:290PM | 940429011 | Robert Granetz | Very good balance between the OH2's. All the power supplies basically
worked fine. There was no breakdown. BR and BZ were off by 5-10 mT each. Next shot, up PID gain on BR from 5 to 7. |
| Apr 29 1994 03:25:18:090PM | 940429012 | Robert Granetz | All supplies worked. The extra gain on BR doesn't seem to have done
much. BR was +0.008 T at commutation, BZ was +0.003 T. No breakdown. For the next shot, put in the value -0.005 for the "minus 1 volt" offset term in the BR predictor. |
| Apr 29 1994 04:03:33:010PM | 940429013 | Robert Granetz | All supplies worked. No flash. Obviously we got the right amount and the
right sign for the BR offset, since BR=1-2 mT on this shot. Bz=0.000 mT. It didn't breakdown because dBz/dr=-.020 to -.030 mT/m. For the next shot, we will increase OH1 demand current by 500 amps, which should increase dBz/dr by +0.020 mT/m. |
| Apr 29 1994 04:13:51:160PM | 940429013 | Robert Granetz | We've got about a 45 minute delay due to a communication problem with
the TORVAC rack, which requires getting into the cell. In the meantime, we will increase the change to I(oh1) from 500 to 750 amps so that dBz/dr will be about 0.000 mT/m rather than -0.030 mT/m. This will also introduce a BZ of -0.008 T. Therefore EF4 will be increased by 125 amps. |
| Apr 29 1994 04:36:50:920PM | 940429014 | Robert Granetz | Got a flash starting at about 0.007 s. The null, as calculated by Ian's
MFLUX program, looks pretty good. No current rise however. No changes for next shot. |
| Apr 29 1994 04:51:59:570PM | 940429015 | Robert Granetz | No breakdown. OH1 faulted, but not until late in the pulse. For some
unknown reason, OH2U had 2 kA less than the previous shot, causing the null to be badly screwed up. For the next shot, no changes made to the hybrid programming, but change the pre-fill from 1.4 to 1.3e-5 torr. |
| Apr 29 1994 05:13:50:880PM | 940429016 | Robert Granetz | No breakdown, no null. EF3 was completely screwed up. End of run.
|
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 10:42:52:940AM | Test | Ok | oh1 only |
| 2 | 10:49:36:930AM | Test | Ok | oh1 only |
| 3 | 11:02:12:210AM | Test | Ok | oh1 only |
| 4 | 11:15:38:630AM | Test | Ok | oh's only |
| 5 | 11:40:08:130AM | Plasma | Ok | no faults |
| 6 | 11:57:01:640AM | Test | Ok | oh's and tf only |
| 7 | 12:36:55:390PM | Test | Ok | no faults |
| 8 | 01:28:05:020PM | Test | Ok | no faults |
| 9 | 01:45:11:070PM | Plasma | Ok | no faults |
| 10 | 02:23:19:020PM | Plasma | Bad | Commutation did not operate on OH2 supplies |
| 11 | 02:54:56:290PM | Plasma | Ok | no faults |
| 12 | 03:10:46:940PM | Plasma | Ok | No faults |
| 13 | 03:29:28:340PM | Plasma | Ok | no faults |
| 14 | 04:19:59:560PM | Plasma | Ok | no faults |
| 15 | 04:31:49:790PM | Plasma | Ok | oh2u comm fault after shot |
| 16 | 04:56:29:670PM | Plasma | Ok | ef3 fault |