Miniproposals

Miniproposal: Date Filed: Title: First Author: Session Leader: Robert Granetz (shots 1-12)

Engineering Operator Run Comment

mile stone, startup.

Session Leader Plans

Physics Operators Plans

Session Leader Summaries

Entered: Jul 7 2004 03:49:09:977PM

Author: To Be Determined

Session leader summary of run 940517 Not much else to say beyond Steve Wolfe's physics operator's summary. We should continue to be suspicious of the electrical connections at the magnetics feedthoughs, and I'm still following through with representatives of Ceramaseal about new components for the Amphenol connectors. Unfortunately, replacement of all the connectors on the machine is a very large job (nearly 1200 pins, one-third in vacuum) which will require an extended vacuum break and considerable downtime. It's not yet clear whether this will be required. The one flux loop that is still flaky, F17, is the exposed one on the lower shelf, and it has obvious physical damage all along its length. I thought I had figured out the algorithm for choosing the optimum pre-fill pressure before each shot, but the fizzle on the last shot proved me wrong. As far as I know, random setting of the pre-fill is as good as any other method. R. Granetz

Physics Operator Summaries

Entered: Jul 7 2004 04:36:51:210PM

Author: To Be Determined

Physics Ops Summary for 940517 Miniproposal 041 SL: Granetz PO: wolfe EO: Bertolino,Fairfax Yesterday, there was a He backfill and vacuum break to repair the bad magnetics signals on the B-port bottom connectors. It was found that the connectors had not been properly screwed in, so the repair was limited to tightening them up. The main goal of today's run was to determine if the fix worked. The run started at about 13:00, following cooldown and ECDC in the morning. Base pressure before discharge cleaning was 3e-7 Torr, dominated by H2O. We started with a reload of 940513026. There was a breakdown at 7.5msec, but it fizzled. All the flux loops except F17 appeared to be good. All the BPxx_BC coils which had been found to be bad since April were also apparently fixed, but BP19_BC, which had been good, now appears to be bad. For shot#2, Steve Horne and Bob replaced the BP19_BC coil with BP19_DE in the hybrid inputs, and we recalculated all the magnetics predictors after updating the Hybrid.config nodes. This resulted in a 50msec plasma, which appeared to have position problems. We played with offsets and pressure for the next several shots, without much success, then decided to proceed to recalculate all the predictors to use the revived flux loops. This was done on shot #6, which fizzled. There seemed to be a fairly large Br_0 error (about 3mTesla). For shot 7 we adjusted the Br_0 offset from -4.5e-3 to -1.5e-3, and got a dirty .23 sec long plasma. Loop voltage was high for the whole shot, whcih was very disruptive. For shot 8, the pressure was raised to 1.3e-5 (from 1.1e-5) and we got a .17 sec long shot that almost got to current flattop, and looked much cleaner. This one had a vertical position problem. For shot 9, we reduced the ZCUR(2) gain at 50msec from 2.5 to 1.0; the shot was a fizzle that died before 50msec. Shot 10 repeated the same settings with a lower pressure of 1.1e-5; this was a good shot, reaching current flattop (600kA) and lasting till 450msec. This one died from bad ZCUR at the end, and also had one large excursion in Zposition at 60msec. The remaining two shots were fizzles. We had tried to turn off ZXL feedback and reduce the early derivative gain on ZCUR, but neither plasma lasted long enough to see the effect of these changes. Conclusions: ============ The Flux loops and BC Bp coils appear to be stable for now, with only F17 of the flux loops still unreliable. Predictors using these loops have been loaded; there may (probably are) still some offset problems to be tuned out. The fact that F17 and BP19_BC seemed to improve during the day is not the best sign for the continued reliablility of the remaining magnetics; the connectors may still be flakey. The machine is still dirty following yesterday's break. There were clear increases in H2O on the RGA following the "good" plasmas. This may be responsible for the fizzles on shots with no early changes from shots that went. Breakdowns were earlier than in the last two weeks, and we had no duds, even on a shot which apparently had up to 3mT of radial field error. This may be good, or it may be an indication of a dirty machine. It may be necessary to tweak up the Bz evolution to account for the early current breakdown. MFLUX seems to be working now that the F22 loop is back. The nulls on the shots that went, and on most of the fizzles, looked good. Scorecard: ========== Duds 0 Fizzles 7 Plasmas 5 (1 runaway) ------- -- Total 12

Session Leader Comments
May 17 1994 01:34:25:840PM	940517001	Robert Granetz	Fizzle. It looks like F17 and BP19_BC are (the only) bad magnetics on this shot. We'll have a short cell access to swap BP19_DE for BP19_BC in the hybrid. Next shot, repeat.
May 17 1994 01:39:44:140PM	940517002	Robert Granetz	Very short plasma pulse. Ipmax=100 kA, dies at 0.055 s. Plasma dives down. OH2L crowbarred, so it's hard to determine if any additional magnetics misbehaved. Nothing obvious, other than F17.
May 17 1994 02:05:09:240PM	940517002	Robert Granetz	Repeat for next shot.
May 17 1994 02:14:05:250PM	940517003	Robert Granetz	Fizzle. No new problems with magnetics (we think). Even F17 worked. We are puzzled. There is a small difference in the BP19 offset in the hybrid now, so we'll change this for the next shot. No obvious problems on the magnetics.
May 17 1994 02:19:20:160PM	940517004	Robert Granetz	Fizzle. Hybrid offset setting for RCUR will be changed to the opposite sign. Also, for the next shot, lower pre-fill from 1.2 to 1.0e-5.
May 17 1994 02:44:57:260PM	940517005	Robert Granetz	Runaways. OH2L crowbarred again, probably due to hybrid demand. For the next shot, but back F22, F23, F25 flux loops into the predictors, and put the offsets on RCUR and ZCUR back to 10000.
May 17 1994 02:48:26:360PM	940517005	Robert Granetz	Also, change pre-fill from 1.0 to 1.1e-5.
May 17 1994 03:04:11:940PM	940517006	Robert Granetz	Fizzle. With the inclusion of the repaired flux signals, we now find a 3 mT radial field. For the next shot, change BR demand by 3 mT using the minus_one_volt coeff.
May 17 1994 03:21:06:350PM	940517007	Robert Granetz	Plasma. Ipmax=0.32 MA, duration=0.23 s. Very disruptive. Loop voltage is up at 4 volts the whole time, indicating a dirty plasma. Not surprising considering yesterday's vacuum break. Next shot, raise pre-fill from 1.1 to 1.3e-5.
May 17 1994 03:39:46:830PM	940517008	Robert Granetz	Plasma, sort of. Ipmax=0.53 MA, duration=0.17 s. Less disruptive than the previous shot, and much more current. Vertical position control is pretty bad. For next shot, reduce the gain on ZCUR feedback for the early portion of the shot.
May 17 1994 03:56:31:320PM	940517009	Robert Granetz	Fizzle. Reason unknown, so let's blame it on pre-fill pressure. Next shot, decrease pre-fill from 1.3 to 1.1e-5
May 17 1994 04:22:58:160PM	940517010	Robert Granetz	Getting there. Ip=0.6 MA (flattop, sort of), duration=0.458 s. We lose vertical position control at the end. Next shot, turn off ZXL feedback, and delay step increase in derivative gain of ZCUR from 0.060 to 0.100 s.
May 17 1994 04:37:04:350PM	940517011	Robert Granetz	Fizzle. Should have raised the gas (I think). Next shot, raise pre-fill from 1.1 to 1.3e-5
May 17 1994 04:52:46:920PM	940517012	Robert Granetz	Fizzle. I guess I shouldn't have increased the pre-fill.

Physics Operator Comments
May 17 1994 12:51:16:460PM	940517001	Steve Wolfe	Shot #940517001: --------------- Loaded shot #940513026 P=1.2e-5 Result: fizzle - bd at 7.5msec F17 still bad, other flux loops look ok BP19_BC is now bad, BP18,24,26 look ok Null looks ok, Bz comes in a little late.
May 17 1994 01:42:23:730PM	940517002	Steve Wolfe	Shot #940517002: --------------- Change from BP19_BC to BP19_DE in hybrid inputs (and in tree) Recalculate all non-vacuum magnetics predictors to pick up the new BP loop Result: PLasma - short, PF breaker pulled out Plasma lasts 50msec, then seems to hit something.
May 17 1994 01:59:48:730PM	940517003	Steve Wolfe	Shot #940517003: --------------- Repeat Result: fizzle.
May 17 1994 02:13:59:680PM	940517004	Steve Wolfe	Shot #940517004: --------------- Change all offsets on RCUR and ZCUR's from 10000. to 12000. Result: fizzle - 6.5msec breakdown. RCUR and ZCUR both show larger error than previous.
May 17 1994 02:33:36:810PM	940517005	Steve Wolfe	Shot #940517005: --------------- Change all offsets on RCUR and ZCUR's from 12000. to 8000. Lower fill to 1.0e-5 Result: xrays, PF breaker pulled out again. Looks a lot like #2. Lots of bursts.
May 17 1994 02:41:56:620PM	940517005	Steve Wolfe	Shot #940517005: --------------- Change all offsets on RCUR and ZCUR's from 12000. to 8000. Lower fill to 1.0e-5 Result: xrays, PF breaker pulled out again. Looks a lot like #2. Lots of bursts.
May 17 1994 02:58:47:510PM	940517006	Steve Wolfe	Shot #940517006: --------------- Change RECON_VAC:STATIC_SHOT to 1006 Copy IAN2_MODEL to IAN_MODEL Turn on Hybrid inputs for F22,F23,F25 Recalculate all magnetics algorithms for F22,F23, and F25 ok P=1.1e-5 Change offsets on RCUR and ZCUR back to 10000. Put factor on BR_0 to 40. Result: fizzle - bd at 8msec. Br claims to have -3mT error. MFLUX agrees there is radial field.
May 17 1994 03:15:05:780PM	940517006	Steve Wolfe	Shot #940517006: --------------- Change Br_0 offset from -4.5e-3 to -1.5e-3 Result: Plasma. Lots of injections. Current only makes it to 300kA. Looks dirty to me. Yuichi claims something hits his lower split limiter.
May 17 1994 03:20:06:620PM	940517007	Steve Wolfe	Shot #940517007: --------------- Change Br_0 offset from -4.5e-3 to -1.5e-3 Result: Plasma. Lots of injections. Current only makes it to 300kA. Looks dirty to me. Yuichi claims something hits his lower split limiter.
May 17 1994 03:42:30:090PM	940517008	Steve Wolfe	Shot #940517008: --------------- Raise gas to 1.3e-5 Result: Short plasma. PF breaker pulled. Plasma dies at 170msec OH2L voltage goes to zero at 120msec Shot looks cleaner
May 17 1994 03:54:14:830PM	940517009	Steve Wolfe	Shot #940517009: --------------- Reduce ZCUR gain at 50msec from 2.5 to 1.0 Result: fizzle
May 17 1994 03:55:54:630PM	940517009	Steve Wolfe	Shot #940517009: --------------- Reduce ZCUR gain at 50msec from 2.5 to 1.0 Result: fizzle
May 17 1994 04:14:03:060PM	940517010	Steve Wolfe	Shot #940517010: --------------- Lower P to 1.1e-5 Result: Longer plasma looks cleaner We almost lose it down at 60msec dies around 450msec.
May 17 1994 04:35:33:180PM	940517011	Steve Wolfe	Shot #940517011: --------------- Turn off ZXL Push I gain on ZCUR2 to 75msec Delay step on D gain on ZCUR2 to 100msec. Result: fizzle - bd at 6.8msec
May 17 1994 04:51:04:850PM	940517012	Steve Wolfe	Shot #940517012: --------------- Raise gas to 1.3e-5 Result: fizzle (told you so Bob) breakdown at 9msec
May 17 1994 04:52:13:150PM	940517012	Steve Wolfe	Shot #940517012: --------------- Raise gas to 1.3e-5 Result: fizzle (told you so Bob) breakdown at 9msec

Engineering Operator Comments
Shot	Time	Type	Status	Comment
1	12:34:20:070PM	Plasma	Ok	no faults
2	01:28:00:690PM	Plasma	Ok	oh2l comm fault
3	01:47:03:170PM	Plasma	Ok	no faults
4	02:06:37:150PM	Plasma	Ok	no faults
5	02:25:48:470PM	Plasma	Ok	oh2l comm fault, oh1 over voltage
6	02:50:53:590PM	Plasma	Ok	no faults
7	03:06:20:290PM	Plasma	Ok	no faults
8	03:27:45:660PM	Plasma	Ok	oh2l comm fault at disruption
9	03:47:49:290PM	Plasma	Ok	no faults
10	04:06:10:580PM	Plasma	Ok	Plasma disrupted 600 kA; no faults
11	04:30:35:810PM	Plasma	Ok	Fizzle. No other faults. Plasma current reached 40 ka before dying at 3
12	04:47:23:560PM	Plasma	Ok	Fizzle. No Faults