Miniproposals

Miniproposal: Date Filed: Title: First Author: Session Leader: John Goetz (shots 1-18)

Engineering Operator Run Comment

plasma attempts

Session Leader Plans

Physics Operators Plans

Session Leader Summaries

Entered: Jul 7 2004 03:49:06:490PM

Author: To Be Determined

From: LEADER::HORNE Date: 7-MAY-1993 20:44:09 Description: run report, 930507 Shot leader -- Horne Assistant -- Goetz Engineering operator -- Fairfax Basic plan -- Load a precalculate sequence of voltages and currents into the hybrid, and bang away. All power supplies except EF2 and EFC were on. Vacuum was poor. We took 15 shots. Only three of these were judged "ok" by Fairfax. Most bad shots were caused by power supply woes of one sort or another, including a blown fuse and SCR in the EF1_L Commutation circuit.The hybrid controlled both the EF1 and OH2 supplies successfully using current feedback, though it took some tweaking. The TF crowbar fired routinely. We saw the first hint of "interesting" behavior due to feedback --For shot 11 we turned on current feedback for the OH2 pair with a 100 msec time constant. This was a good physics shot, with a lousy but recognizable null. (EF1 feedback was turned on for shot 10)For shot 12 the gain was raised to a 20 msec time constant, and the gain was switched in well after the commutation switch was closed. Some ringing appeared on the OH2 demand (perhaps EF1 also) when the gain became nonzero. For shot 13, I tried bringing up the gain from zero to about 5 in 4 steps of 100 msec or so each. There were still problems, so for shot 14 the OH2 staircase was scaled back in amplitude to go from zero to 1 in 4 steps. This worked, and we got another near null. To correct this we dropped the OH1 PLC current limit and shot 15, our last, had quite a good null, though not enough loop voltage. See the logbook for details. Conclusions: 1) We're getting close. 2) The power supplies are improving but there's still a way to go. 3) We need to address the "voltage across the SCR when it fires" problem. 4) Feedback controlled current seems to work well, once tweaked. 5) Wolfe's precalculated startup scenario performed well, though the changes to the programming we made to address 3) messed up the current evolution. We should develop a "pre-scenario" sequence which keeps the SCR's happy and tack it onto the beginning of every shot. Steve

Physics Operator Summaries

Session Leader Comments
May 7 1993 08:59:47:520AM		John Goetz	Run 930507 is being conducted with a base pressure of 2.1E-07. Evidence of a leak is shown on the RGA, with 28 being the dominant peak.
May 7 1993 09:58:19:880AM	930507001	John Goetz	Shot 930507001 All power supplies but EF2 on. Fill pressure 4E-05 TF crowbar OH1, OH2 switches not closed. Power supply voltages approximately as asked for by hybrid. EF4 voltage and currents approximately as asked for by hybrid. EF3 voltage and current approximately as asked for until 50 msec when loss of phase causes trip.
May 7 1993 10:30:14:820AM	930507002	John Goetz	Shot 930507002 All power supplies but EF2 on. Fill pressure 5E-05 TF timing advanced by 350 msec. Start at -.75 sec, inversion at .45 sec. TF crowbar. EF3 and EF4 followed voltage programming. Other supplies followed programming until commutation time. All commutation switches not closed. (forward bias ( 50V ) added at -1.6 sec for next shot)
May 7 1993 10:37:53:850AM	930507003	John Goetz	Shot 930507003 All power supplies but EF2 on. Fill pressure 4E-05 IGOR failure. No trigger sent. Forward bias ~50V added at -1.6 sec for all commutation switches
May 7 1993 10:52:56:990AM	930507004	John Goetz	Shot 930507004 All power supplies but EF2 on. Fill pressure 4E-05 State machine failure. No trigger sent. Permissive timeout problem. Forward bias ~50V added at -1.6 sec for all commutation switches
May 7 1993 11:14:36:230AM	930507005	John Goetz	Shot 930507005 All power supplies but EF2 on. Fill pressure 4E-05 Power supply timers are shutting off OH2 (perhaps others). Engineers fixing. EF1's aren't balanced well and feedback control is probably required. Current or voltage feedback not determined yet. TF crowbar. TF to start at -1.5 sec next shot in order to reach 100kA before crowbar. OH2 Jorway not triggered.
May 7 1993 11:39:41:610AM	930507006	John Goetz	Shot 930507006 All power supplies but EF2 on. Fill pressure 4E-05 OH1 commutation switch didn't close. OH2 at current limit TF crowbar. TF started at -1.5 sec. EF3 power supply didn't fire
May 7 1993 12:17:40:790PM	930507007	John Goetz	Shot 930507007 All power supplies but EF2 on. Fill pressure 4E-05 Forward bias on OH1, OH2, and EF1 increased to try and solve commutation troubles. Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID's are zeroed. TF started at -1.2 sec. TF crowbar. Pecularity in EF4 behavior. Current starts significantly later than the voltage. EF1 commutation failed.
May 7 1993 12:50:23:520PM	930507008	John Goetz	Shot 930507008 All power supplies but EF2 on. Fill pressure 4E-05 Forward bias on OH2's increased to 200V and moved in time to move away from the OH1 switch. Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID's are zeroed. TF started at -0.2 sec. One TF cabinet only. TF crowbar. Pecularity in EF4 behavior. Current starts significantly later than the voltage. Shots 6, 7, and 8 have same demand but different outputs. EF1U commutation failed. OH1 current on the rail. Programming done early in time to ensure switch closure has caused the OH1 and OH2's to depart significantly from scenario. Current limit will be set in PLC for correct currents at commutation. EF1 U and L current feedback switched on, 100 msec time constant for next shot.
May 7 1993 01:35:41:900PM	930507009	John Goetz	Shot 930507009 All power supplies but EF2 on. Fill pressure 4E-05 EF1 U and L current feedback switched on with 100 msec time constant. Feedback on EF1 is working. Raise the PID gain on both EF1U and EF1L from one to five. Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID's are zeroed. TF started at -1.2 sec. Two TF cabinets. TF crowbar. EF1U commutation failed. OH1 front porch dropped to ~350V and raised EF4 front porch to ~ 200V to try and turn on EF4 faster and solve the pecularity. PLC current limits on OH1 and OH2's have fixed the departure from the scenario. Both OH1 and OH2's commutated.
May 7 1993 02:38:15:060PM	930507010	John Goetz	Shot 930507010 All power supplies but EF2 on. Fill pressure 4E-05 Raised the PID gain on both EF1U and EF1L from one to five. EF1U feedback working fine. Current loss in EF1L. EF1U commutation failed. EF1L commutation too early. Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID's are zeroed. TF started at -1.2 sec. Two TF cabinets. TF crowbar. EF4 behavior changed with the front porch change but current still not consistent with voltage demand. Current feedback turned for OH2U and OH1L turned on with 100 msec time constant for next shot in attempt to stop oscillation at current limit.
May 7 1993 03:52:57:160PM	930507011	John Goetz	Shot 930507011 All power supplies but EF2 on. Fill pressure 4E-05 Fuse replaced in EF1L commutation circuit and SCR replaced in EF1U commutation circuit between shots 10 and 11. EF1L and EF1U commutated for shot 11. Feedback on Bz (before +.002 with EF4, after -.002 with EF3). TF started at -1.2 sec. Two TF cabinets. TF crowbar. Current feedback worked for OH2U and OH1L Comment: Best shot of the day so far. Fair null, but no breakdown. Bz about 10mT too negative. Commutation resistor changes: OH2U & OH2L from .032 to .046 Ohms for next shot. PID gain timing has been changed to 0 from -2sec to -1.4 for OH2's and EF1's. Also, OH2 PID gains set to 5 at -1.4sec. These changes for shot 12. PLC current limit on OH2's raised by 1000 amps to 16000 amps for shot 12.
May 7 1993 04:13:36:150PM	930507012	John Goetz	Shot 930507012 All power supplies but EF2 on. Fill pressure 4E-05 Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID zeroed. TF crowbar. Commutation resistor changes: OH2U & OH2L from .032 to .046 Ohms. PLC current limit on OH2's raised by 1000 amps to 16000 amps. PID gain timing has been changed to 0 from -2sec to -1.4 for OH2's and EF1's. Also, OH2 PID gains set to 5 at -1.4sec. *When PID matrix switched in at -1.4 sec, errors occured on EF1's and OH2's For shot 13, the PID gains for the OH2's and the EF1's are now 4 stairsteps starting at 1 and ending at 4.5 over a range of -1.4sec to -1.0 sec.
May 7 1993 04:43:33:080PM	930507012	John Goetz	Shot 930507013 All power supplies but EF2 on. Fill pressure 4E-05 Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID zeroed. TF crowbar. The PID gains for the OH2's and the EF1's are now 4 stairsteps starting at 0 and ending at 4.5 over a range of -1.4sec to -1.0 sec. Trouble with all supplies starting at ~-1.3 sec. The staircase on OH2's PID scaled down from a max of 5 to 1 for shot 14. Voltage programmed SCR closing pulse on OH2's will be from -1.7 sec to -1.53 sec for shot 14. Current programming changed also so that 10kA should appear at -1.53 sec.
May 7 1993 04:59:41:910PM	930507014	John Goetz	Shot 930507014 All power supplies but EF2 on. Fill pressure 4E-05 Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID zeroed. TF crowbar. The staircase on OH2's PID scaled down from a max of 5 to 1. Voltage programmed SCR closing pulse on OH2's will be from -1.7 sec to -1.53 sec. Current programming changed also so that 10kA should appear at -1.53 sec. All supplies commutated. Hardly any null at all. Bz 30mT too negative. OH1 PLC current limit dropped to 15kA for shot 15.
May 7 1993 05:35:16:630PM	930507015	John Goetz	Shot 930507015 All power supplies but EF2 on. Fill pressure 4E-05 Feedback on Bz (before +.002 with EF4, after -.002 with EF3). PID zeroed. Feedback on OH2's and EF1's with same stairstep rogowski algorithm. OH1 PLC current limit dropped to 15kA. TF crowbar. All supplies commutated. Slight dip on OH1 volts just before commutation. Comments: Null from 20msec to 28msec. No breakdown. dBz/dR too large. Bz needs to be flattened near 0 time.

Physics Operator Comments

Engineering Operator Comments
Shot	Time	Type	Status	Comment
1	09:42:04:700AM	Plasma	Bad	TF crowbar at -.35 sec; looks like spurious IOC on cabinet 4
2	10:09:56:330AM	Plasma	Bad	PF SCR switches did not close; hybrid programming was not forward biasin
3	10:31:48:760AM	Plasma	Bad	bad; no CAMAC trigger
4	10:39:34:360AM	Plasma	Bad	PLC/netowrk delays are causingACK failures and disabling pulse event; by
5	10:54:06:610AM	Plasma	Bad	OH2 Decoder did not fire; all PF supplies tripped off at -0.030 due to
6	11:15:53:440AM	Plasma	Bad	TF crowbar at -.3 sec
7	11:49:12:040AM	Plasma	Bad	TF crowbar
8	12:19:19:270PM	Test	Ok	TF tested at 30 kA; once phase group missing. EF1U SCR switch did not c
9	01:10:56:020PM	Plasma	Bad	EF1L did not commutate properly
10	01:38:29:650PM	Plasma	Bad	Fuse blown in EF1L switch; SCR short in EF1U switch
11	03:13:07:890PM	Plasma	Ok	good shot but no flash
12	03:53:33:700PM	Plasma	Bad	Onset of hybrid gain upset power supplies and caused commutation faults
13	04:15:48:400PM	Plasma	Bad	Both OH2's show 60Hz on feedback and command signals
14	04:45:40:220PM	Plasma	Ok	Good null but no flash
16	05:47:35:960PM	Test	Ok	Testing Data system (josh)
17	06:00:35:200PM	Test	Ok	josh testing data system
18	06:34:28:780PM	Test	Ok	Summary seems to have worked