| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Steve Horne |
| Physics operator(s): | Joseph Snipes |
| Engineering operator(s): | Unknown |
| Engineering Operator Run Comment |
| Plasma developement |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:09:710PM |
| Author: To Be Determined |
| Session Leader Summary -- 940428 Horne First plasma attempt, 1994 -- Horne, Snipes. MP 41 Wolfe has slightly tweaked 931020003, primarily in response to the failure of F17. Control algorithms by Wolfe and Hutch have been modified to work around the bad loop. Fizzle detector is not yet operational. Initial settings -- gas 1.4e-5, pull the drawn pulse_gas3 (used to be "14") to zero at .25 sec. Inversion at .25 sec (crossover, roughly) EF3 is questionable. The TF crowbar problem has been worked on. Bus voltages are still not working properly. A test signal is in ef3 bus voltage. The EF1U PFN Rogowski integrator is not resetting. Finally, the EF1L commuation resistance as measured by the PLC is a factor of two high. We started late (about 11:00) due to alternator problems, and quit a few minutes early to fix power supplies. Many systems worked well -- TF crowbar problem has apparently been fixed -- and there were flashes on shots 6 and 8. The breakdowns were very late; no sign of current rise. We took only 9 shots. We spent a lot of time chasing power supply problems and searching for an apparent hybrid fault which was in fact due to a power supply imbalance. When we continue tomorrow, we'll begin by running just the OH supplies to verify that they have been fixed. Since we can't make a plasma till they are behaving properly, we'll have to spend whatever time is required to make them work. Shot list -- shot 1 problems: 1) OH2 current limits were too conservative. 2) ef2 U didn't turn on. 3) ef3 didn't turn on. The EF2 u problem MAY be due to a bad chip in the PID!!! It worked fine yesterday. Try once more. shots 2-5 -- ef2u still not working. check out PID, hybrid cables, digitizers. Hybrid checks out fine. 6 -- To diagnose the problem, removed EF2 pair from b_cur controller. Flash! Suspect the BR predictor, as recalculated, may be broken. 7 -- turned off the BR controller. No flash. By comparing shot 6 to 931020003, it's clear that the OH2 supplies are (a) unbalanced, and (b) have lower gain than they had last october. This caused large radial field before t=-.4 sec. The BR control in the hybrid tried to compensate by backing off the demand on ef2_u. But this caused the demand to ef2_u, before it's switch closed, to be insufficient to hold the SCR in conduction. So it never turned on. The OH2 imbalance is due in part to differences in power supply gain. Various knobs on the supplies have been twisted since last October. Joe D tried to adjust them for the next shot. 8 -- flash. There is still a strange OH2 problem. 9 -- OH2 U failure first, OH1 and OH2L also failed. Quit to allow engineers to fix the supplies. Try again tomorrow. |
| Physics Operator Summaries |
| Entered: Jul 7 2004 04:36:50:787PM |
| Author: To Be Determined |
| ------------ USERNAME: SNIPES ENTERED: 28-APR-1994 17:00:17.36 VOIDED: NULL RUN: 940428 SHOT: 940428009 TOPIC: PHYSICS_OPERATOR TEXT: On the bright side of things, the run was successful in that we succeeded getting two plasmas, however brief. Shots 6 and 8 had 10 - 15 ms breakdown plasmas visible on the H port H alpha. On the other hand, the power supplies did not perform very well. There was a 20 V imbalance in OH2U and L. Fairfax thinks this may be induced by a problem in OH1. In any case, in practice, it resulted in a a 20 mT radial error field that made it very difficult to get breakdown. Nonetheless, it did happen twice, but the plasma did not last long and the breakdown was around 18 ms, which is rather late. So, once again, I was running the machine when we successfully obtained first plasma for the year. |
| Session Leader Comments | |||
| Apr 28 1994 11:56:53:710AM | Steve Horne | First plasma attempt, 1994 -- Horne, Snipes. MP 41
Wolfe has slightly tweaked 931020003, primarily in response to the failure of F17. Control algorithms by Wolfe and Hutch have been modified to work around the bad loop. Fizzle detector is not yet operational. Initial settings -- gas 1.4e-5, pull the drawn pulse_gas3 (used to be "14") to zero at .25 sec. Inversion at .25 sec (crossover, roughly) Commutation resistors, PLC settings for 931020003 were COIL Resistor (Milliohms) PLC current limit oh1 44.7 -16/+20 oh2u 8.3 -15/+25 oh2l 8.5 -15/+25 ef1u 102.4 +/- 8 ef1l 100. +/- 8 ef2u 45.2 4 ef2l 47.7 4 We don't need the high + limits if we invert at t=.25 EF3 is questionable. The TF crowbar problem has been worked on. Bus voltages are still not working properly. A test signal is in ef3 bus voltage. The EF1U PFN Rogowski integrator is not resetting. Finally, the EF1L commuation resistance as measured by the PLC is a factor of two high. | |
| Apr 28 1994 12:33:24:950PM | Steve Horne | First plasma attempt, 1994 -- Horne, Snipes. MP 41
Wolfe has slightly tweaked 931020003, primarily in response to the failure of F17. Control algorithms by Wolfe and Hutch have been modified to work around the bad loop. Fizzle detector is not yet operational. Initial settings -- gas 1.4e-5, pull the drawn pulse_gas3 (used to be "14") to zero at .25 sec. Inversion at .25 sec (crossover, roughly) Commutation resistors, PLC settings for 931020003 were COIL Resistor (Milliohms) PLC current limit oh1 44.7 -16/+20 oh2u 8.3 -15/+25 oh2l 8.5 -15/+25 ef1u 102.4 +/- 8 ef1l 100. +/- 8 ef2u 45.2 4 ef2l 47.7 4 We don't need the high + limits if we invert at t=.25 EF3 is questionable. The TF crowbar problem has been worked on. Bus voltages are still not working properly. A test signal is in ef3 bus voltage. The EF1U PFN Rogowski integrator is not resetting. Finally, the EF1L commuation resistance as measured by the PLC is a factor of two high. | |
| Apr 28 1994 12:33:34:410PM | Steve Horne | First plasma attempt, 1994 -- Horne, Snipes. MP 41
Wolfe has slightly tweaked 931020003, primarily in response to the failure of F17. Control algorithms by Wolfe and Hutch have been modified to work around the bad loop. Fizzle detector is not yet operational. Initial settings -- gas 1.4e-5, pull the drawn pulse_gas3 (used to be "14") to zero at .25 sec. Inversion at .25 sec (crossover, roughly) Commutation resistors, PLC settings for 931020003 were COIL Resistor (Milliohms) PLC current limit oh1 44.7 -16/+20 oh2u 8.3 -15/+25 oh2l 8.5 -15/+25 ef1u 102.4 +/- 8 ef1l 100. +/- 8 ef2u 45.2 4 ef2l 47.7 4 We don't need the high + limits if we invert at t=.25 EF3 is questionable. The TF crowbar problem has been worked on. Bus voltages are still not working properly. A test signal is in ef3 bus voltage. The EF1U PFN Rogowski integrator is not resetting. Finally, the EF1L commuation resistance as measured by the PLC is a factor of two high. | |
| Apr 28 1994 04:46:44:820PM | Steve Horne | First plasma attempt, 1994 -- Horne, Snipes. MP 41
Wolfe has slightly tweaked 931020003, primarily in response to the failure of F17. Control algorithms by Wolfe and Hutch have been modified to work around the bad loop. Fizzle detector is not yet operational. Initial settings -- gas 1.4e-5, pull the drawn pulse_gas3 (used to be "14") to zero at .25 sec. Inversion at .25 sec (crossover, roughly) Commutation resistors, PLC settings for 931020003 were COIL Resistor (Milliohms) PLC current limit oh1 44.7 -16/+20 oh2u 8.3 -15/+25 oh2l 8.5 -15/+25 ef1u 102.4 +/- 8 ef1l 100. +/- 8 ef2u 45.2 4 ef2l 47.7 4 We don't need the high + limits if we invert at t=.25 EF3 is questionable. The TF crowbar problem has been worked on. Bus voltages are still not working properly. A test signal is in ef3 bus voltage. The EF1U PFN Rogowski integrator is not resetting. Finally, the EF1L commuation resistance as measured by the PLC is a factor of two high. | |
| Physics Operator Comments | |||
| Apr 28 1994 12:34:59:390PM | 940428001 | Joseph Snipes | Slow start due to alternator problems. Started with shot 931020003.
First shot failed due to OH2 current limits set wrong. Will be changed to 25 kA for shot 2. Also noticed that PID output for EF2L was not coming out. |
| Apr 28 1994 12:51:54:330PM | 940428002 | Joseph Snipes | DAQ1 camac failed for shot 2.
|
| Apr 28 1994 02:21:38:580PM | 940428004 | Joseph Snipes | Continued EF3 problems. In addition, the hybrid computer does not
put out a PID output for EF2U. |
| Apr 28 1994 02:53:03:880PM | 940428006 | Joseph Snipes | Shot 6
EF2 control of Br removed from Br controller. |
| Apr 28 1994 03:02:04:050PM | 940428006 | Joseph Snipes | Shot 6
First plasma! Flash observed on H port H alpha. Started at 18 ms until 30 ms. Still have problems with radial field. Still have EF3 problems |
| Apr 28 1994 03:23:24:720PM | 940428007 | Joseph Snipes | Shot 7
Removed Br feedback control by setting the proportional gain to 0. |
| Apr 28 1994 03:50:31:660PM | 940428008 | Joseph Snipes | Shot 8
Returned proportional gain to Br controller to 5 as before shot 7. There is an imbalance problem between the OH2U and L power supplies that requires adjusting in the power supplies. |
| Apr 28 1994 03:51:26:940PM | 940428008 | Joseph Snipes | Shot 8
Imbalance problem between the OH2U and L power supplies still exists after switch closure at -0.65 s. |
| Apr 28 1994 04:22:22:710PM | 940428009 | Joseph Snipes | Shot 9
Moved gas puff on A side 20 msec earlier. OH2U died at -0.85 s. |
| Apr 28 1994 05:00:17:360PM | 940428009 | Joseph Snipes | On the bright side of things, the run was successful in that we succeeded in
getting two plasmas, however brief. Shots 6 and 8 had 10 - 15 ms breakdown plasmas visible on the H port H alpha. On the other hand, the power supplies did not perform very well. There was a 20 V imbalance in OH2U and L. Fairfax thinks this may be induced by a problem in OH1. In any case, in practice, it resulted in a a 20 mT radial error field that made it very difficult to get breakdown. Nonetheless, it did happen twice, but the plasma did not last long and the breakdown was around 18 ms, which is rather late. So, once again I was running the machine when we successfully obtained first plasma for the year. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 11:50:26:910AM | Plasma | Bad | no ef3 |
| 2 | 12:38:01:640PM | Plasma | Bad | oh1 ioc and comm - ef3 forced invert |
| 3 | 12:49:43:850PM | Plasma | Bad | oh1 comm fault |
| 4 | 02:04:50:360PM | Plasma | Bad | Power supply problems. |
| 5 | 02:30:46:190PM | Test | Ok | data system only |
| 6 | 02:44:36:440PM | Plasma | Ok | no faults |
| 7 | 03:03:25:400PM | Plasma | Ok | no faults |
| 8 | 03:27:32:970PM | Plasma | Bad | Oh2 l dosen't follow prog. voltage |
| 9 | 04:16:19:280PM | Plasma | Bad | comm faults |