Miniproposals

Miniproposal: Date Filed: Title: First Author: Session Leader: John Rice (shots 1-25)

Engineering Operator Run Comment

Plasma Run: Wolfe at Helm.

Session Leader Plans

Physics Operators Plans

Session Leader Summaries

Entered: Jul 7 2004 03:49:08:007PM

Author: To Be Determined

Run summary for 6/29/93 Physics Operator: Wolfe Assistant Operator: Rice Engineering Operators: Gwinn/Silva The goal of this run was to investigate startup with reduced voltage. Actually the main point was to reduce voltage on the coils in preparation for re-configuring the system; reducing the loop voltage at breakdown and current rise was of secondary importance, but nonetheless interesting. The run was advertised (see the RUN PLAN) as likely having a high proportion of duds, fizzles, and runaways; this prediction came true. There were three objectives: 1) Reduce EF3 voltage to approximately -1100V, giving high confidence in pur ability to operate with the EF3 coils in series and perhaps improving discharge reliability and robustness. 2) Reduce OH1 voltage to of order -600V, permitting us to attempt startup with no commutation of OH1 (600V is the nominal inversion voltage available from the OH1 supply.) 3) Demonstrate startup with OH1 not commutated. This mode of operation will enable us to push the pre-bias on OH1 and OH2 with less fear of OH2 overcurrents arising from commutation failures. The first two objectives were successfully, and almost painlessly, achieved. Operation with V_EF3 as low as -1050V and OH1 bus voltage around -600V was demonstrated. The resulting plasmas were well-behaved, with currents near 300kA, and nearly flattopped, with durations to 380msec. Operation with OH1 commutation turned off was less successful. Only fizzles, and one horrible runaway shot that reached 100kA, were obtained; the OH1 supply was found to be only capable of about -450V in inversion. Vacuum/Wall Condition: ---------------------- ECDC in H2 for about an hour and a half before the run. No mass 28 event observed during cooldown, but there was a mass 28 & 4 event between shots 13 and 14, coincident with a failure on HEAT (see Problems). Detailed Summary: ----------------- The first five shots were power supply tests (EF2 and EF3). EF3 failed to fire at first (see Problems section below) but finally worked on Shot 5. The first plasma attempt was shot 6; the programming was taken from Earl's shot 930625009, except the argon pulse was turned off. This shot was chosen as a low elongation, non-challenging, vanilla plasma; the idea was not to have to worry too much about the late discharge evolution but concentrate on the first 30msec. Shots 7-13 were a series of lowering the voltage on EF3, and making corresponding changes on the other supplies. Net changes were: V_EF3 reduced from -1500V to -1050V V_OH1 inversion reduced by 150V (to 600V net on the Bus) V_OH2U&L inversion reduced by 25V (actually ended positive) Shots 7 and 8 were both high runaway shots, with Mo gettering. The rest were good shots with few or no disruptions, sawtoothing the whole shot, nearly flat current, lasting till 350msec or longer. The max current went from 320 ka on shot 9 to 280kA on shots 12 and 13. Current rise rates went from 3-4MA/sec on #9 to 2.5MA/sec on 13. At the end of the scan, breakdown voltages were between 4 and 5 Volts, with peak voltages around 6V. The OH1 supply did not run out of current until .29 sec. Shots 14-25 were attempts to startup with OH1 commutation off (CAP bank not charged, OPEN and CROWBAR times after supply inversion). These resulted in fizzles, and some out-and-out duds. Shot 20 was a runaway shot, not unlike #7. Several of the fizzles had reasonable looking field structure, but the evolution may not have been appropriate. Maximum loop voltages were between 4 and 5 Volts. EF3, OH2 and EF1 voltages were adjusted to try to get the current to take off. It is possible that the EF3 voltage was still too high although it was reduced to -900V on shot 17. Problems: --------- The Tuesday curse strikes again: EF3 failed to turn on (held in inversion) on test shots 2-4. THe problem was apparently due to fault detection relays not latching on properly, for reasons that remain mysterious; the problem went away when Joe located it, though he did not do anything to fix it. The gas system failed to operate properly before the first plasma attempt. This was traced (by M. Graf) to a logic error involving ECDC settings and order of operations. This will (?) be fixed tonight; it has apparently been present for several weeks, and may be part of the reason Silva has experienced problems setting up ECDC in the mornings. As a result of the preceding, the first plasma shot was not attempted until almost 11AM. The HEAT scanner died after shot 13, and had to be restarted. This caused the temperature to drop to -4C, and caused (?) a mass 4 and 28 event on the RGA. The EF3 timing jitter is still present, and seemed to get worse late in the day. For the last two shots I tried putting -50V on the V_EF3 programming from -15msec to -1msec. The timing on these two shots seemed consistent, and the front porch current level was about 100amps. It is unclear whether this constitutes an acceptable or reliable approach. The EF3 glitches (sawtooth patterns on the current, funny phasing on the supply voltage) are still present, despite a reported fix on Monday. Shot 15 is one of several examples. On a positive note, a 60 cycle problem on the EF2s, which had apparently been there "forever", was fixed. However, running OH1 in full inversion uncovered three problems. First, we only get about -450V, as opposed to a nominal -600V. Second, running at that level produces a substantial 60 Hz ripple. This may or may not have caused problems with the commutationless operation shots, but it certainly didn't help. Third, it was found necessary to ask for -1000V and move the inversiontime 2msec early in order to get a prompt inversion response. Simply asking for -600V at .001 (shot 14) caused the voltage to dribble down to its final level by +.008sec; this seems to be a feature. Scorecard: ---------- Power system tests 5 Good Shots 7 Fizzles 9 Runaway shots 2 No breakdown 2 --- Total 25

Physics Operator Summaries

Session Leader Comments
Jun 29 1993 08:09:20:210AM		Steve Wolfe	Run Plan for 6/29/93 Physics Operator: Wolfe Assistant Operator: Rice Engineering Operator: Silva/Gwinn Objective: Investigate lower voltage breakdown and current rise scenarios. 1. Try reducing EF3 voltage at breakdown and over the first 30 msec. There are two reasons this is attractive: a) reducing the EF3 voltage from the present 1500 V increases the margin for series operation on the EF3 coils, which is necessary for high current operation. A reasonable target is 1100V, comfortably under half the maximum obtained in April/May . b) A slower EF3 current rise may increase robustness of the startup against jitter in the EF3 turn-on timing.
Jun 29 1993 08:09:59:400AM		Steve Wolfe	2. Reduce the OH1 voltage (and OH2 voltage correspondingly) to a level consistent with eliminating OH1 commutation. This should be feasible if the OH1 bus voltage can be reduced to < 600 V. This step, in addition to simplifying the startup, has advantages for going to increased pre-bias current (more volt-sec), due to decreased risk of over-currenting the OH2 coils on a commutation failure. 3. If (2) is successful, turn off the OH1 commutation and demonstrate startup with OH21 supply inversion only.
Jun 29 1993 08:10:39:200AM		Steve Wolfe	Sequence: --------- Start from one of Earl's shots from last Friday, eg 930625009. Reduce EF3 voltage in 100-200V increments. Observe early discharge evolution, particularly radial position. Reduce OH1 and OH2 voltages as indicated to keep current rise rate consistent with dBz/dt. If jitter in EF3 turn-on seems to be a problem try using 10-20msec of forward bias on EF3 instead of holding the voltage positive as is done now. Using the V waveforms, reduce OH1 and OH2 voltages at t=(0 .. .030). Target a net maximum OH1 bus voltage of -600V. Adjust other voltages to keep null and current rise position satisfactory. Turn off commutation on OH1 and program V_OH1 waveform to correspond to total bus voltage obtained in previous step.
Jun 29 1993 08:11:08:020AM		Steve Wolfe	Disclaimer: ----------- This will be in the nature of a breakdown studies run. A high proportion of fizzles and duds is anticipated. Furthermore, a significant number of runaway shots may occur. It would not be advsable to move the RF antenna in for this run.
Jun 29 1993 08:49:58:100AM	930629001	John Rice	Torus pressure 5 e-6 w/GV3 closed. H2 pulsed gas from B-side upper. Vessel temp 18 C no 'cooling' leaks yet. EF2 test.
Jun 29 1993 09:06:13:070AM	930629002	John Rice	IGT 4 e-6. EF2L test. no EF3. +/- 250 A induced in parallel EF3s. ENA
Jun 29 1993 09:20:27:110AM	930629003	John Rice	IGT 3 e-6. same attempt as shot 2, EF2L and EF3s. no EF3s
Jun 29 1993 09:33:48:250AM	930629004	John Rice	IGT 3 e-6. same attempt as shot 3, EF2L and EF3s, this time with hybrid control, instead of PLC lack-of-control. got nothing, and liked it.
Jun 29 1993 10:25:07:210AM	930629005	John Rice	IGT 2 e-6. same attempt as shot 4, EF2L and EF3s, Objective 1a) calls for lower EF3 voltage, the 0 V tests have been a success. Turn-on jitter also absent. Some relays weren't behaving correctly? EF3 ok. Problem of funny fasing still present. Graf 95, Goetz 91
Jun 29 1993 10:59:25:940AM	930629006	John Rice	fill pressure 4 e-5. finally a shot attempt. good shot, current's a little low, sawtoothed til the end, moderate hards.
Jun 29 1993 11:34:31:430AM	930629007	John Rice	changes: fill pressure 3.7 e-5. Vef3 (1ms) from -1500 to -1300 (24ms) from -1533 to -1300 EF3U timing shot 5 -300 A @ 3.5ms 6 -300 A @ 4.11ms 7 -300 A @ 4.8ms 65 kA plasma, disrupted at 60 ms, quasi-fizzle, saturated hards. failed early. sputtered along.
Jun 29 1993 11:50:59:270AM	930629008	John Rice	changes: fill pressure 4.0 e-5. EF4 current between -.2 sec and +1 ms from 655 A to 605 A results: EF3U timing shot 8 -300 A @ 4.1ms more arcs and sparks, but it was a long one. saturated hards. low density. moly splatter on shots 7 and 8 gettering water like crazy.
Jun 29 1993 12:11:52:910PM	930629009	John Rice	changes: fill pressure 4.3 e-5. Voh1 @ .008 s from -190 V to -100 V .024 from -196 to -100 .026 from -300 to -200 .069 from -300 to -200 results: EF3U timing shot 9 -300 A @ 4.4 ms a few sparks at the end. nice current fall, no disruption, sawtoothing all the way.
Jun 29 1993 12:36:50:280PM	930629010	John Rice	changes: fill pressure 4.0 e-5. Voh1 @ .008 s from -100 V to -50 V .024 from -100 to -50 .026 from -200 to -150 Vef3 @ .001 from -1300 to -1200 .024 from -1300 to -1200 results: EF3U timing shot 9 -300 A @ 4.8ms good shot, no hards,no disruption. not convinced changes were loaded.
Jun 29 1993 12:48:30:920PM	930629011	John Rice	changes: fill pressure 4.0 e-5. Voh1 @ .008 s from -50 V to 0 V .024 from -50 to 0 .026 from -150 to -100 .069 from -200 to -100 Vef3 @ .001 from -1200 to -1100 .024 from -1200 to -1100 results: EF3U timing shot 10 -300 A @ 4.8ms 11 5.5ms good shot, no hards,no disruption, didn't go sparkle, sparkle
Jun 29 1993 01:10:37:310PM	930629012	John Rice	changes: Vef3 @ .001 s from -1100 V to -1050 V .024 from -1100 to -1050 results: EF3U timing shot 12 -300 A @ 6.0 ms good shot, no hards,no disruption, similar to shot 11
Jun 29 1993 01:41:41:990PM	930629013	John Rice	changes: Voh2l @ .008 s from 0 V to 25 V .010 -10 +15 .024 -10 +15 .026 -20 +5 Voh2u .008 0 +25 .010 -10 +15 .024 -10 +15 .026 -20 +5 results: EF3U timing shot 13 -300 A @ ms good shot, no hards,no disruption
Jun 29 1993 02:05:46:700PM	930629014	John Rice	changes: trun off commutation on oh1 voh1 -600V from .001s to .070 s if this one goes it will be amazing. results: EF3U timing shot 13 -300 A @ 5.5 ms 14 5.9 dud, no flash even, 6 V vessel temp 11 deg due to scanner failure between shots. mass 18 gone, masses 4 and 28 up. small double seal leak?
Jun 29 1993 02:22:29:010PM	930629015	John Rice	changes: turn off commutation on oh1 Voh1 -1000V from .001s to .070 s ICef4 from +600 to 652A from -.2 to +.001 s results: EF3U timing shot 15 -300 A @ 5.7 ms fizzle, late breakdown, not so much a milestone as a rock in the road vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 02:37:09:110PM	930629016	John Rice	changes: turn off commutation on oh1 Voh1 inversion demand earlier by 2 ms point at t=-.001 to -.003, +4.2V point at t=+.001 to -.001, -990.7V results: EF3U timing shot 16 -300 A @ 5.5 ms fizzle, vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 02:55:26:830PM	930629017	John Rice	changes: commutation on oh1 is off Vef3 @ .001s -1050V -900V .024 -1050 -900 .026 -1062 -900 .046 -1061 -900 results: EF3U timing shot 17 -300 A @ 7.0ms fizzle vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 03:08:26:350PM	930629018	John Rice	changes: fill pressure 3.6 e-5 commutation on oh1 is off Voh2u+l @ .001s 20V 40V .008 25 45 .010 15 35 .024 15 35 .026 5 25 results: EF3U timing shot 18 -300 A @ 6.0ms fizzle, vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 03:34:27:780PM	930629019	John Rice	changes: fill pressure 3.6 e-5 commutation on oh1 is off Vef1u+l @ .002s from -287 V to -150V .047 -257 -130 Voh2u+l -.001 12.8 40 results: EF3U timing shot 19 -300 A @ 6.1ms fizzle vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 03:47:35:400PM	930629020	John Rice	changes: fill pressure 3.6 e-5 commutation on oh1 is off Vef1u+l @ .002s from -150 V to 0V .047 -130 0 results: EF3U timing shot 20 -300 A @ 7.0ms 100 kA plasma! this one we may be able to diagnose. hards saturated. vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 03:58:35:990PM	930629021	John Rice	changes: fill pressure 4.0 e-5 commutation on oh1 is off ICef4 from 650A to 600A between -.200 to +.001s results: EF3U timing shot 21 -300 A @ 6.0ms another fizzle vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 04:28:20:360PM	930629022	John Rice	changes: fill pressure 4.0 e-5 commutation on oh1 is off ICef4 from 600A to 625A between -.200 to +.001s results: EF3U timing shot 22 -300 A @ 7.4ms fizzle, boy did that break down late. vessel temp 17 deg. masses 4 and 28 still up.
Jun 29 1993 04:46:03:920PM	930629023	John Rice	930629023 changes: fill pressure 4.0 e-5 commutation on oh1 is off We have been getting 60 cycle on OH1 inversion voltage. Dave claims it is due to running into the inversion endstop. For this shot, he pushed back the endstop, and we decreased the V_OH1 demand from -592 to -450V from .014 to .070 sec. results: EF3U timing shot 23 -300 A @ 8.2ms fizzle, broke down at 5.6msec, Bz looks high (positive) vessel temp 18 deg. masses 4 and 28 still up.
Jun 29 1993 04:55:26:670PM	930629024	John Rice	930629024 changes: fill pressure 3.6 e-5 Vef3 from -15ms to -1 ms -50V results: EF3U timing shot 24 -300 A @ 4.1ms dud vessel temp 18 deg. masses 4 and 28 still up.
Jun 29 1993 05:01:01:010PM	930629025	John Rice	930629025 changes: ICef4 from 625 to 725 A results: fizzle EF3U timing shot 25 -300 A @ 4.2ms vessel temp 18 deg. masses 4 and 28 still up.

Physics Operator Comments

Engineering Operator Comments
Shot	Time	Type	Status	Comment
1	08:45:08:540AM	Test	Ok	ef2 test
2	08:57:52:600AM	Test	Bad	no ef3
3	09:13:11:180AM	Test	Ok	ef3 test no go
4	09:21:58:120AM	Test	Bad	no ef3
5	10:20:48:290AM	Test	Ok	ef3 pulse
6	10:47:41:870AM	Plasma	Ok	.26 ma
7	11:24:25:920AM	Plasma	Ok	ok
8	11:40:04:470AM	Plasma	Ok	lots of spark
9	12:00:52:030PM	Plasma	Ok	ok shot
10	12:20:45:960PM	Plasma	Ok	on a roll
11	12:37:55:410PM	Plasma	Ok	on a big roll
12	12:53:50:160PM	Plasma	Ok	shot ok
13	01:15:36:300PM	Plasma	Ok	heater fault
14	01:44:38:980PM	Plasma	Ok	Plasma shot No faults.
15	02:13:30:840PM	Plasma	Ok	Plasma shot No faults.
16	02:24:21:630PM	Plasma	Ok	Plasma shot No faults.
17	02:38:49:560PM	Plasma	Ok	fizzle
18	02:58:23:680PM	Plasma	Ok	fizzle, bad ef3 waveform
19	03:24:08:540PM	Plasma	Ok	fizzle
20	03:36:47:720PM	Plasma	Ok	no faults --
21	03:52:55:180PM	Plasma	Ok	no faults
22	04:06:21:920PM	Plasma	Ok	fizzle
23	04:25:34:080PM	Plasma	Ok	Plasma shot No faults.
24	04:46:44:940PM	Plasma	Ok	No plasma No faults.
25	04:58:21:650PM	Test	Bad