Alcator C-Mod Run 930525 Information

Back to runs

Miniproposals
Miniproposal:
Date Filed:
Title:
First Author:
Session Leader:Robert Granetz (shots 1-26)

Operators
Session leader(s):Robert Granetz
Physics operator(s):Steve Golovato
Engineering operator(s):Vinny Bertolino

Engineering Operator Run Comment
Plasma shots all day

Session Leader Plans

Physics Operators Plans

Session Leader Summaries
Entered: Jul 7 2004 03:49:07:290PM
Author: To Be Determined
Summary of Run 930520
Session Leader/Physics Operator: Granetz
Assistant Physics Operator: Golovato
Engineering Operator: Bertolini

-------------------------------------------------------------------------------

Overview:

The main goal of this week's operation is to establish a quiescent,
steady state plasma, with a flattop current of 0.20-0.25 MA lasting for
at least 0.10 sec. Such conditions are desirable in order to study
feedback control of the plasma current and shape in the coming weeks.
Today's run made some progress towards this goal, with flattops of 0.03 s
at Ip=0.24 MA (shot #014) and disruption-free periods of >0.11 s, before
running into plasma position control problems.

-------------------------------------------------------------------------------

Prior to run:

ECDC was run for several hours yesterday. Changes have been made to the
hybrid computer and the PCS model file which affect the scaling of signals,
as well as the residual offsets. The bad SCR and fuse in the TF supply
has been fixed.

-------------------------------------------------------------------------------

Detailed summary:

The power supply programming for the start of the run was copied from
shot 930520031 from last week. The first 5 shots had problems due to a
very large ripple on OH2L, which was eventually found to be due to a bad
solder connection on a gate control board. Shot #006 was the first shot
on which all supplies worked properly, and #007 reached 0.22 MA with a
disruption at 0.075 s. This is about where we left off from last week.
The following four shots all fizzled, with two of them at slightly
higher fill pressure (8e-5 rather than 6e-5 torr). No programming
changes were made. It was subsequently noticed that the radial field, Br,
was being driven to -5 mT by the hybrid feedback. This appeared to be
due to a baseline offset in the hybrid, and was overcome by demanding
+5 mT more for Br (prior to shot #012). That did the trick, and all but
two of the following 15 shots surpassed 0.2 MA.

Shots #012, 013, & 014 had increasingly extended quiescent periods
before the occurrence of a disruption (up to t=0.113 s), at currents of
0.24 MA and flattops of up to 0.03 s. The disruptions did not have
any obvious causes. X-ray imaging clearly showed slow, continual
outward and downward movement of the plasmas. At times, the motion
as determined by magnetics may not have agreed with the x-rays.
The plasma motion was not responding to changes made in the hybrid
demand for R_CUR, possibly due to internal overflow errors. On shot #016,
the hybrid scaling for R_CUR and Z_CUR feedback was changed, which reduced
the outward motion, but didn't entirely eliminate it. It was noticed that
EF3 (which was being used to control R_CUR) was approaching the PLC
current limit, so this was raised for shot #017 from 12 kA to 15 kA.
From this point on, the plasma position control was quite poor, with
inward and very large downward motion (up to 20 cm). Plasma current
continued to ramp up instead of flattopping, with shot #022 reaching
0.36 MA. Changing gains for R_CUR and Z_CUR feedback did not help. It
is possible that the imbalance in the decay of the OH1 and OH2 currents
was causing the plasma to be pulled toward OH2L (i.e. the vertical
instability). Attempts to better match the relative decays in those
supplies did not help either, although by the last shot (#026) the
disruption-free period had again reached 0.112 s, even though the plasma
was moving in and very far downward. It was also sawtoothing the entire
time.

-------------------------------------------------------------------------------

Shot statistics:

Total shots: 26

No breakdowns: 2 (both due to failure of multiple power supplies)

Fizzles (i.e. < 0.2 MA): 10 (4 due to power supply faults, 6 to bad program-
ming)
Plasmas (i.e. > 0.2 MA): 14 (max Ip=0.36 MA, max disruption-free
period=0.113s)

-------------------------------------------------------------------------------

Suggestions/comments for next run:

Work on problems of position control. One might be tempted to re-start with
shot #014, but that would require reducing the EF3 PLC current limit, which
is an unsatisfactory way to operate.

The $64 question: What causes the disruption at 0.110-0.115 sec?

Physics Operator Summaries

Session Leader Comments
May 25 1993 12:58:33:480PMRobert GranetzNote: ECDC was done for 4 or 5 hours yesterday

May 25 1993 09:37:35:430AM930525001Robert GranetzCommutation resistors:

OH1 0.100 ohms (0.050 each)
OH2U&L 0.025 ohms
EF1U&L 0.180 ohms

Crowbar closure times:

OH1 0.040 s
OH2U&L 0.050 s
EF1U&L 0.050 s

Power supply invert times:

All PF supplies invert at 0.250 s.

May 25 1993 09:45:16:590AM930525001Robert GranetzThis run will begin by trying to repeat shot 930520031 from last Thursday.
It was a very long (>0.15 s), stable (no disruptions until 0.15 s),
sawtoothing discharge. Once we establish it, the goal is to get a flattop
for >0.1 s at Ip=0.2-0.25 MA. If successful, we can begin feedback operation
on Ip (plasma L/R time is of order 0.1 s).

May 25 1993 09:45:48:780AM930525001Robert GranetzFizzle. OH2L has a very large oscillation on it (1500 amps @ 55 Hz). It
must be fixed or we will waste the day trying to get repeatable operation.

May 25 1993 10:57:22:700AM930525002Robert GranetzFizzle. OH2L still has oscillation problem. OH1 faulted during commutation.

May 25 1993 10:59:56:400AM930525003Robert GranetzFizzle (short pulse reached 100 kA). OH2 still has oscillation problem.

Pause run for 15 minutes to investigate probably gate pulse problem.

May 25 1993 11:00:55:560AM930525003Robert GranetzCold solder joint on gate pulse control board of OH2L !

May 25 1993 11:03:01:030AM930525004Robert GranetzNo breakdown. No OH2's

May 25 1993 11:14:40:660AM930525005Robert GranetzNo breakdown. Multiple PF power supplies faulted early.

May 25 1993 11:35:21:030AM930525006Robert GranetzBefore shot #006, lowered the early voltage pre-programming on the OH2's in
order to avoid the brief railing of the supplies against the PLC limits at
about t=-1.5 s.

May 25 1993 12:09:15:110PM930525006Robert GranetzFizzle. All supplies worked. OH2L's ripple has been fixed.

May 25 1993 12:35:18:550PM930525007Robert GranetzGood shot. Ip=0.22 MA Shot lasted until t=0.155 s, but disruptions started
at 0.075 s. Plasma position was about z=-1 cm, as requested.

May 25 1993 12:37:26:740PM930525008Robert GranetzGas prefill raised from 6e-5 to 8e-5 torr for this shot.

Fizzle. Try a repeat

May 25 1993 12:43:10:290PM930525009Robert GranetzFizzle.

May 25 1993 12:56:40:850PM930525010Robert GranetzFizzle again. Gas pre-fill was lowered for this shot from 8e-5 back to
the nominal 6e-5 torr.

May 25 1993 01:25:39:500PM930525011Robert GranetzFizzle. There might be an offset problem with the hybrid's control of Br.

May 25 1993 01:30:13:540PM930525012Robert GranetzOffset the Br demand by +0.005 Tesla from the previous shots.

Good shot. Maybe what we did had something to with the good plasma, maybe
it didn't.

May 25 1993 01:41:35:080PM930525012Robert GranetzThere is no indication that the plasma hit any material surface to cause
the disruption. Try a repeat.

May 25 1993 01:46:26:180PM930525013Robert GranetzAnother good one. Disruption at 0.095 s.

May 25 1993 01:56:25:000PM930525013Robert GranetzChange hybrid programming of R_CUR from 0.665 to 0.650 m. The PID gain is
turned on at 0.040 s. (for the next shot)

May 25 1993 02:15:48:280PM930525014Robert GranetzBetter. Ip=0.24 MA. First disruption occurs at 0.113 s. Plasma is still
moving outward throughout the shot. The magnetics reconstructions do not
show the plasma moving outward. That's why the feedback doesn't seem to
be working correctly? X-rays definitely show plasma moving outward at
about 7 to 10 mm per 10 msec.

For the next shot, change programmed R_CUR from 0.650 to 0.640 m.

May 25 1993 02:20:47:950PM930525015Robert GranetzAlmost identical to the previous shot

May 25 1993 02:41:33:850PM930525015Robert GranetzHybrid may have an internal saturation. Change scale bit for R_CUR. Change
it for Z_CUR also, just to be on the safe side.

May 25 1993 02:46:23:980PM930525016Robert GranetzSimilar to previous 2 shots. Plasma outward motion is definitely reduced
compared to the preceding shots.

Next shot: raise EF3 PLC current limit.

May 25 1993 03:04:52:860PM930525017Robert GranetzIp=0.32 MA. Plasma went in and very far down. We don't know why it went down.
Apparently changing the EF3 PLC current limit caused this change. For the
next shot, change R_CUR programming back to 0.665 m.

May 25 1993 03:21:06:970PM930525018Robert GranetzIp=0.30 MA, but OH1 faulted during commutation. Plasma was not very good...
no sawteeth. Repeat for next shot.

May 25 1993 03:29:26:110PM930525019Robert GranetzOkay shot. Ip=.35 MA Disrupts at 0.110 s

May 25 1993 03:37:14:620PM930525019Robert GranetzOkay shot. Ip=.35 MA Disrupts at 0.110 s. Plasma shifts inward by 4-5 cm
at t=0.045, and dives downward after 0.085 s

May 25 1993 03:38:20:230PM930525019Robert GranetzFor next shot, increase PID gain on R_CUR by factor of 2. This will max out
the feedback for radial plasma position.

May 25 1993 03:43:30:320PM930525020Robert GranetzFizzle

May 25 1993 04:10:19:020PM930525021Robert GranetzFizzle. OH1 faulted

May 25 1993 04:11:53:090PM930525022Robert GranetzIp=0.36 MA. Better shot. Radial position okay, R(x-rays)=0.61 m
Plasma is still falling downward after about 0.080 s.

May 25 1993 04:17:13:320PM930525022Robert GranetzHybrid demands for Z_CUR feedback look proper, so let's just double the
PID gain on Z_CUR feedback for the next shot

May 25 1993 04:44:25:830PM930525023Robert GranetzIp=0.30 MA. Disrupted at 0.095 sec. May be being sucked into the lower
x-point. Try to adjust OH1/OH2 ratio to remain constant. Do this by playing
with OH1. Adjust the voltage preprogramming for the next shot.

May 25 1993 04:47:13:010PM930525023Robert GranetzOops, I think I programmed the voltage on OH1 in the wrong direction. Oh well,
it wasn't a large change anyway. Too late to change it now.

May 25 1993 04:48:04:830PM930525024Robert GranetzNearly identical to the previous shot. For the next shot, adjust OH1 voltage
upward by +250 volts after t=0.050 sec.

May 25 1993 05:04:56:960PM930525025Robert GranetzBad shot. Ip=0.2 MA. Plasma disrupted and ended at 0.060 s. It was quite
low.

May 25 1993 05:05:50:720PM930525025Robert GranetzFor the next shot back off on the OH1 voltage change after 0.040 s

May 25 1993 05:10:09:120PM930525026Robert GranetzIp=0.3 MA Sawtoothing. Disrupted at 0.112 s.

Physics Operator Comments

Engineering Operator Comments
ShotTimeTypeStatusComment
109:32:36:960AMPlasmaOkoh2l current ripple of 2ka, 20ka plasma
210:08:50:280AMPlasmaOkoh2l gate missing probable nothing found, will try another shot then che
310:32:33:820AMPlasmaOkBad Gate OH2 L, fixed
411:00:03:960AMPlasmaOkoh2l not synchronized with state pc, resync'ed
511:09:45:580AMPlasmaOkoh2l comm fault
611:59:25:440AMPlasmaOk60ka plasma, oh1
712:10:09:990PMPlasmaOk220ka-good plasma again!!
812:32:39:810PMPlasmaOk20KA
912:40:06:510PMPlasmaOk30KA PLASMA
1012:52:49:800PMPlasmaOk90KA
1101:10:36:370PMPlasmaOk110KA PLASMA
1201:27:14:870PMPlasmaOk240KA PLASMA
1301:42:16:830PMPlasmaOk230KA PLASMA
1401:59:27:500PMPlasmaOk240KA
1502:17:10:230PMPlasmaOk240KA PLASMA
1602:35:48:170PMPlasmaOk280KA PLASMA
1702:52:57:030PMPlasmaOk320KA PLASMA
1803:07:47:230PMPlasmaOk300KA PLASMA-- FAULTS IN EF1,OH1
1903:20:47:210PMPlasmaOk350KA PLASMA
2003:39:06:760PMPlasmaOk80KA PLASMA
2103:48:53:930PMPlasmaOk70KA OH1 COMM FAULT
2204:03:20:620PMPlasmaOk360KA PLASMA
2304:18:09:960PMPlasmaOk300KA PLASMA
2404:41:17:840PMPlasmaOk320KA PLASMA
2504:50:42:240PMPlasmaOk220KA PLASMA
2605:05:17:330PMPlasmaOk310KA PLASMA