Miniproposals

Miniproposal: Date Filed: Title: First Author: Session Leader: John Rice (shots 1-26)

Engineering Operator Run Comment

Plasma run.

Session Leader Plans

Physics Operators Plans

Session Leader Summaries

Entered: Jul 7 2004 03:49:07:493PM

Author: To Be Determined

Summary of Run 930602 Session Leader/Physics Operator: Granetz Assistant Physics Operator: Rice Engineering Operator: Daigle/Gwinn ------------------------------------------------------------------------------- Overview: The two principal purposes of today's run were intended to be: 1) Try new predictors for R_CUR and Z_CUR. These have been recalculated by Steve Wolfe using lower weightings on the vessel filaments relative to the plasma filaments, and larger sigmas on the Bp coils. 2) Investigate the tendency of the plasma to become disruptive when it is pulled off the inboard wall. (Examine 2-pi and array bolometers, H-alpha emission from inner wall, molybdenum content, x-rays (sawteeth/disruptions), hard x-rays, and especially density. Unfortunately the shot quality and reproducibility was dismally poor, and we did not get good enough plasmas to begin looking at item (2). However, we did get to test the new position predictors (item 1), and found that they worked significantly better than the old predictors. These new ones have now become the working predictors. ------------------------------------------------------------------------------- Prior to run: This is the first run after a holiday weekend (i.e. today is Wed, last run was last Thursday). ECDC was done on Sat and Sun, resulting in some cleanup according to the RGA (if it can be trusted). Base pressure is in the mid 10^-7 torr range. Another important change: the RF limiter has been moved in from R=0.919 to R=0.900 m. ------------------------------------------------------------------------------- Detailed summary: The power supply programming for the start of the run was copied from shot 930527032 from last week. This was a long (>0.3 s) discharge on the inboard wall. The run was held up for over an hour due to a problem with the bitbus communication to the hybrid. This problem was overcome, but the cause is still not understood. The first shot experienced a failure of hybrid control, and part of this problem was a dead power supply on the AFOL's from the bus rogowskis. After this was fixed, we began a disappointing day of mostly `fizzles' (short breakdowns) and `no breakdown' shots. Even though we started by copying a particular shot from last week, we never were able to come close to reproducing it. Most of the day was spent doing minor `tweaks' of the vacuum fields to slighly improve the vacuum null at breakdown. As usual, the machine repeatibility from shot to shot was not very good. In addition, using the new, greatly improved R_CUR and Z_CUR predictors, we experimented with feedback control of the plasma current centroid. I would say that our gain on Z_CUR was too high, resulting in some oscillation of the z-centroid as seen by the x-rays. Control of the R-centroid was not clearly demonstrated. We also played around a little with the pre-fill gas pressure, as well as the pulse gas programming. We had one shot (930602008) that obviously was dominated by runaways, most likely due to insufficient pulsed gas. This was presumably due to a large change in the threshold offset of the piezo valve which was effected sometime during the past weekend. Re-programming the pulse gas voltage control eliminated this problem. All-in-all, it was a pretty disappointing day, with the exception of the successful tests of the new position predictors. Note: all the power supplies worked flawlessly today. Kudos to our engineering and technical staff. ------------------------------------------------------------------------------- Shot statistics: Total shots: 26 No breakdowns: 8 (only 1 due to failure of hardware) Fizzles (i.e. < 0.15 MA): 9 Plasmas (i.e. > 0.15 MA): 9 (max Ip=0.34 MA, max disruption-free period=0.10 s) ------------------------------------------------------------------------------- Suggestions/comments for next run: Start from a shot last week that was not so marginal with respect to getting a breakdown. Reduce the gain on the Z_CUR feedback control.

Physics Operator Summaries

Session Leader Comments
Jun 2 1993 08:29:21:990AM		Robert Granetz	Session leader: Granetz Assistant: Rice Engineering: Gwinn/Daigle Purposes of run: 1) Investigate the tendency of the plasma to become disruptive when it is pulled off the inboard wall. (Examine 2-pi and array bolometers, H-alpha emission from inner wall, molybdenum content, x-rays (sawteeth/disruptions), hard x-rays, and especially density. 2) Try new predictors for R_CUR and Z_CUR. These have been recalculated by Steve Wolfe using lower weightings on the vessel filaments relative to the plasma filaments, and larger sigmas on the Bp coils. 3) If time permits, try playing with the R_CUR controller (maybe add EF4 into the soup).
Jun 2 1993 08:29:44:890AM		Robert Granetz	Important happenings since last run: This is the first run after a holiday weekend (i.e. today is Wed, last run was last Thursday). ECDC was done on Sat and Sun, resulting in some cleanup according to the RGA (if it can be trusted). Base pressure is in the mid 10^-7 torr range. Another important change: the RF limiter has been moved in from R=0.919 to R=0.900 m. Power Supply settings (on/off times, commutation resistors, PLC current limits, etc) will be the same as the last shot, 930527032. We will start with the last plasma shot, 930527032, which was a long (>0.3 s) discharge on the inboard wall.
Jun 2 1993 10:43:45:440AM	930602001	Robert Granetz	This shot has been delayed by about 1.5 hours due to a problem loading the hybrid over the bitbus. Steve Horne, Josh Stillerman, and Tom Fredian are working on it. Manual "init"ing has enabled us to try for the first shot.
Jun 2 1993 10:46:14:150AM	930602001	Robert Granetz	No OH1. All other supplies just ran up to their PLC limits and railed until commutation. Oh yes, no plasma. The real-time storage scopes also did not get any signals. There still seems to be a problem with communication to/from the hybrid...possibly due to some new AFOL's.
Jun 2 1993 10:47:23:840AM	930602001	John Rice	Bit bus errors on hybrid have delayed the first shot. No flash. No OH1
Jun 2 1993 11:03:04:180AM	930602001	Robert Granetz	The 5 volt power for the AFOL's on the bus rogowskis was dead. It has been replaced.
Jun 2 1993 11:10:30:730AM	930602002	Robert Granetz	Fizzle. Gas pre-fill was at 6e-5 torr. Change to 5e-5 for next shot.
Jun 2 1993 11:10:31:930AM	930602002	John Rice	Flash, no current rise. CIII filter now on TV picture Fill pressure 6 e-5 T
Jun 2 1993 11:27:22:620AM	930602003	Robert Granetz	Fizzle again. Bz & Br okay. dBz/dR slightly negative. Adjust using 200 amps more in OH1 for next shot. Pulsed gas programming also was lowered for the next shot to account for recently changed voltage offset of the piezo.
Jun 2 1993 11:38:52:030AM	930602004	Robert Granetz	Fizzled again. Adjust Bz using EF4 for next shot (increase EF4 by 60 amps)
Jun 2 1993 11:59:16:850AM	930602005	Robert Granetz	Another fizzle.
Jun 2 1993 11:59:33:760AM	930602006	Robert Granetz	Another fizzle.
Jun 2 1993 12:25:09:820PM	930602007	Robert Granetz	Another fizzle
Jun 2 1993 12:31:19:610PM	930602008	John Rice	330 kA Ip. No pulsed gas? Lots of hard x-rays, also on softs. Hard x-ray gain changed.
Jun 2 1993 12:56:54:560PM	930602009	John Rice	Pulsed gas increased by 30 V, fill pressure changed to 6 e-5 T. Flash only
Jun 2 1993 12:57:07:170PM	930602009	Robert Granetz	Fizzled. Pre-fill was raised from 5 to 6e-5 and pulsed gas valve programming was increased by about 15 volts.
Jun 2 1993 01:23:03:240PM	930602010	Robert Granetz	Ip=0.155 MA. single disruption at 0.078 s ends the shot. Plasma moved downward by 10 cm as the shot progressed. The new predictors seemed to do much better than the ones presently being used for position control. For the next shot
Jun 2 1993 01:42:44:110PM	930602010	Robert Granetz	For the next shot, load in the new Z_CUR predictor on the actual control wire.
Jun 2 1993 01:31:51:490PM	930602011	Robert Granetz	Similar, but not identical to the previous shot. Ipmax=0.2 MA. Shot ends with a single disruption at 0.073 s. Plasma starts moving downward, then is forced back up, then dives downward. Simultaneously it is moving inwards.
Jun 2 1993 01:41:40:690PM	930602011	Robert Granetz	For the next shot, put in the new R_CUR predictor also.
Jun 2 1993 01:51:35:630PM	930602012	Robert Granetz	Slightly shorter duration. Ipmax=0.2 MA. Single disruption at 0.064 s ends the shot. Vertical position is lost after about 0.050 s
Jun 2 1993 02:09:48:660PM	930602013	Robert Granetz	No breakdown! I don't think it's anything we did.
Jun 2 1993 02:31:36:520PM	930602014	Robert Granetz	Carbon copy of shot #012
Jun 2 1993 04:02:36:790PM	930602015	Robert Granetz	No breakdown
Jun 4 1993 04:21:41:130PM	930602015	Steve Golovato	shot 930604015. fizzled again.
Jun 2 1993 04:01:47:220PM	930602016	Robert Granetz	Ipmax=0.25 MA. Shot lasted to 0.10 s. Plasma position feedback is definitely trying to do something.
Jun 2 1993 03:10:41:400PM	930602017	Robert Granetz	No breakdown
Jun 2 1993 03:18:04:350PM	930602018	Robert Granetz	No breakdown
Jun 2 1993 03:35:12:650PM	930602019	Robert Granetz	Ip=0.2 MA, ends at 0.073 s.
Jun 2 1993 03:57:42:630PM	930602020	Robert Granetz	No breakdown
Jun 2 1993 03:57:48:200PM	930602021	Robert Granetz	No breakdown
Jun 2 1993 04:26:24:900PM	930602022	Robert Granetz	No breakdown again. Fix EF3 leading "porch" to get rid of extra vertical field at t=0 for the next shot. Also raise fill pressure to 8e-5 torr for the next shot.
Jun 2 1993 04:38:46:740PM	930602023	Robert Granetz	Fizzle. For next shot: tweak EF3 "front porch" amplitude less by another 50 amps, and decrease EF4 by 50 amps.
Jun 2 1993 04:45:18:030PM	930602024	Robert Granetz	Ip=150 kA, duration=0.077 s. Sawteeth.
Jun 2 1993 04:52:16:130PM	930602024	Robert Granetz	For the next shot, increase PID gain of R_CUR feedback from 0.1 to 0.25
Jun 2 1993 04:57:01:040PM	930602025	Robert Granetz	Fizzle
Jun 2 1993 05:16:35:940PM	930602026	Robert Granetz	Best shot of day, but that's not saying much. Ipmax=0.15 MA. Duration=0.117 s Briefly sawtooths. Becomes disruptive late in the shot. Vertical position is very good, but in/out position moves toward inside starting when the feedback turns on at 0.050 s.

Physics Operator Comments

Engineering Operator Comments
Shot	Time	Type	Status	Comment
1	09:32:30:930AM	Plasma	Bad	ef1/oh2 com fail - no hybrid control
2	11:02:36:400AM	Plasma	Ok	Plasma shot no faults
3	11:11:28:140AM	Plasma	Ok	Plasma shot no faults.
4	11:28:25:970AM	Plasma	Ok	Good plasma shot
5	11:42:13:060AM	Plasma	Ok	Plasma shot no faults.
6	11:54:05:250AM	Plasma	Ok	Good Plasma shot no faults.
7	12:09:51:990PM	Plasma	Ok	Plasma shot no faults.
8	12:21:14:590PM	Plasma	Ok	plasma current no faults.
9	12:39:13:450PM	Plasma	Ok	Good shot.
10	12:57:31:340PM	Plasma	Ok	good shot.
11	01:18:48:660PM	Plasma	Ok	Plasma OK
12	01:39:01:980PM	Plasma	Ok	oh2u and l com fail
13	02:00:49:090PM	Plasma	Ok	no plasma
14	02:11:29:270PM	Plasma	Ok	200 ka
15	02:32:43:270PM	Plasma	Ok	Plasma shot no breakdown no faults.
16	02:42:19:880PM	Plasma	Ok	Plasma shot OK OH2 U single phased.
17	02:56:55:210PM	Plasma	Ok	good shot
18	03:09:20:230PM	Plasma	Ok	Plasma shot No faults.
19	03:24:27:580PM	Plasma	Ok	Plasma shot no faults.
20	03:40:55:540PM	Plasma	Ok	Plasma shot no faults.
21	03:53:03:770PM	Plasma	Ok	Plasma shot No faults.
22	04:07:33:400PM	Plasma	Ok	Plasma shot no faults.
23	04:22:40:850PM	Plasma	Ok	Plasma shot No faults.
24	04:41:00:320PM	Plasma	Ok	Plasma shot no faults.
25	04:53:07:080PM	Plasma	Ok	Plasma shot no faults.
26	05:05:46:880PM	Plasma	Ok	Plasma shot fault OH2 L.