Alcator C-Mod Run 951026 Information

Back to runs

Miniproposals
Miniproposal:120
Date Filed:10/18/1995
Title:Restart C-MOD 1995
First Author:Steve Wolfe
Session Leader:Steve Wolfe (shots 1-26)

Operators
Session leader(s):Steve Wolfe
Physics operator(s):Ian Hutchinson
Engineering operator(s): Unknown

Engineering Operator Run Comment
plasma start-up

Session Leader Plans

Physics Operators Plans

Session Leader Summaries
Entered: Jul 7 2004 03:49:13:930PM
Author: To Be Determined
Run Summary for 951026

MP#120 Restart C-MOD
SL: Wolfe
PO: Hutch

The plan was to try for plasma after completion of power tests. As it turned
out, the power tests didn't get completed.

We successfully tested, with the hybrid, OH1 commutation,EF1 commutation,
EF2U&L, EF3, EF4 double swing, and EFc. After a few difficulties, the fill and
gas puff were confirmed to be working. The crossover delay on EF4 is 100msec,
which can possibly be improved. EF2 gains can be increased to get smaller
errors in Segment 1. This was loaded for shot 26, but not tested.

The OH2 supplies, which were working fine yesterday, wouldn't turn on under
PLC or Hybrid command. At 4PM we turned things over to the power system guys
to figure out why. At around 4:30PM, they found a logic problem in the PLC
that had been made accidentally while fixing a problem on the bank charging
relay. After this was fixed, the OH2's worked ok under PLC control, but had
COM failures in the bypass circuit on the commutation test.

There is also a serious problem with the magnetics. The TRAQ controllers seem
to be incompatible with the DAQA SCSI highway. Various things were tried by
Bob and T&J without success. The problem is apparently NOT the crate
controller, which was swapped without improvement. Bob had been convinced it
was the old TRANSIAC controller, and was going to swap those units for a set
of DSPs from the xray rack, but then it was noted that the DSP's were also
screwing up, though less frequently than the TRANSIAC.

Among the problems that got fixed, the EF4 bus was connected backward,the EF4
shunts were installed backward, there was a problem on EF4 crossover, I had
not properly reset the factor in EF2L in Segment 1, causing the hybrid to ask
for twice the current, and the OH2 supplies not staying on.

Problems not fixed by 4PM included: the OH2 supplies not commutating
properly, the Engineering PAD won't read old shots (JAS is working on it),
and the magnetics TRAQ problem (TWF &/or JAS are working on software.)

No plasma attempts
No fizzles
No duds
No plasmas.

Physics Operator Summaries
Entered: Jul 7 2004 04:36:54:910PM
Author: To Be Determined
Physop Log 951026

MP120 Restart C-Mod
___________________

Summary: Day spent chasing power-supply, control, and computer bugs.
Details in session leader's log.


Power Tests to start with.
Various Magnetics diagnostic problems.

Load from user9:[wolfe.mds.shots.hybrid.pcs] 951025001,
the nominal new fiducial shot. Lower current to 0.6MA during flat top.
Reduced the FSP gain to zero. Raised the ZXL gain back to 3 (mid-shot).

PCS is painfully slow bringing up the wire windows.

Cold start done by Horne.

02 EF2 3.5kA, EF3 1.2kA. EF4 400A.

05 EF4 test. Negative Curent.

08 Hybrid only test. No current.
09 Chopper control only. 500A of current. OK.
11 EF4 only gives -1097A when asked for -1100. Fizzle detector at 80ms.
12 Fizzle detector and timed invert to 1 and 1.1s resp.

Ian out for a while. EF4 double swing fixed. EF2 fixed. OH1 commutates OK.
OH2 commutation problem.

OH2 proves to be more than just commutation. Does not work on PLC.
Tested EF1 commutation. Tested gas fill and pulse.

Handed to Engineering for OH2 fixing at 4pm.
Magnetics camac still in bad shape.


Session Leader Comments
Oct 26 1995 07:25:33:700AMSteve WolfeRun plan for 951026

MP#120 Restart C-MOD in 1995
SL: Wolfe
PO: Hutchinson

Discharge clean in D2.

Plasma operations will start following successful completion of power tests.
Power tests should include use of Hybrid to control PF supply currents,
including the chopper. Steve Horne should be present for hybrid tests.

Engineering setup from 950526005. Fill "A-side" plenum to 19psi D2, and enable
for fill and puff.

Oct 26 1995 07:26:11:190AMSteve WolfeStart from the prepared shot in USER9:[WOLFE.MDS.SHOTS.HYBRID.PCS]PCS_951025001.*
which has been tweaked to accomodate the new configuration. This was an 800kA,
5T shot. Reduce the programmed current to 600kA.

Tweak the startup to get a good null, as indicated by breakdown.

Evaluate current rise (if any). Tweak fields and fill to optimize current
rise. Assess state of machine conditioning. If plasmas are too resistive and
don't seem to be cleaning up, go back to ECDC; else continue to try for
flattop. Pay careful attention to EF2/EF4 currents and try to keep off limits.
The starting shot has FSP control turned on. We probably will want to change
this to pure R_xl and Z_xl control.

Raise current to 800kA. Continue to monitor EF2 and EF4 current levels. Follow
usual disruption policy: don't push current if disruptions are a problem.
Continue to observe clean-up.

Raise current to 1MA, if machine condition and disruptivity permit.

If time permits, begin development of pulse-gas only startup. Otherwise, this
program will be continued into Friday.

Oct 26 1995 08:39:15:320AM951026001Steve WolfeShot 951026001

EF2 test - PLC Control.
-3kA pulse, ran ok
Errors on Bob's BP signals. Not obviously broken from traces.
TWF says he had not put in the fix for the TRAQ modules/ does so before
shot 2.

Oct 26 1995 09:04:16:440AM951026002Steve WolfeShot 951026002

Re-do both EF2's and do EF4

PF breaker pulled out. Fault on EF4.
Magnetics still screwed up. There is a signal on BTOR and F01 is much
smaller than it should be. And it thinks there's 100kA of plasma
current!

Tom will put in extra waits and look at the code.

Oct 26 1995 09:16:27:800AM951026002Steve WolfeShot 951026002

TWF has re-stored the data from the suspect TRAQ_3; it now looks right
(or at least not obviously wrong) for shot 2. Will look at shot 3 if and
when.

Oct 26 1995 09:37:12:750AM951026003Steve WolfeShot 951026003

There is a problem with the HYBRID. It ACCVIO's on COLDSTART.
Shape will be rebooted

HYBRID tree for shot 3 (Shape is rebooting)
EF4 looks sort of ok, +2300A; still a lot of ripple

Magnetics look ok this time.

Oct 26 1995 10:27:28:280AM951026004Steve WolfeShot 951026004

Shape is up, BCS responds.
The RT-node is there (SEARCH) but can't find it's job!
Horne is going in to power cycle it.
Dispatched coldstart seems to work.
Horne says there still seems to be a problem on dispatch that doesn't
show up when the same commands are run manually from shape.

The EF4 supply current shunt is reversed; they expected negative current
in the coil, got positive on shot 3. Reversed EF4 bus for shot 4. Also
adjusted loop gain.

This was an intentional positive EF4 current test. Looks good
Magnetics are screwed up again.
Hybrid timing is wrong by one table entry, starting
halfway through shot.

Oct 26 1995 10:31:04:590AM951026005Steve WolfeShot 951026005

EF4 test at higher current, PLC control.
-4000A ok.

Magnetics still screwed up.

Oct 26 1995 10:39:24:270AM951026006Steve WolfeShot 951026006

-5ka EF4 ok. Magnetics still bad.

Oct 26 1995 10:47:05:590AM951026007Steve WolfeShot 951026007

Try positive EF4 current. looks ok

Oct 26 1995 11:00:14:810AM951026008Steve WolfeShot 951026008

Hybrid test only. Horne will poke at memory after INIT, before Check.
Looks ok after init.

Looks ok.

Oct 26 1995 11:03:25:850AM951026009Steve WolfeShot 951026009

Chopper control power only
chopper control ok.

Oct 26 1995 11:18:23:900AM951026010Steve WolfeShot 951026010

Low current chopper test (500V on bank)
TWF is trying to run the MAG_TRAQ3 over REMCAM from Barolo analysis
server.

Chopper looks ok
Magnetics still bad.

Oct 26 1995 12:02:48:800PM951026011Steve WolfeShot 951026011

EF4 hybrid test.
Use shot programming, PLC limits at +/- 2000A, start at -1.1,invert at
+0.1, hybrid enable. This should give the startup, but not test the
double swing yet.

Looks ok, except I'm not sure the fizzle detector took it out. It single
phases after the FD comes in . Also, there is a lot of hash on the FD
signal. SFH reports a 100mV offset on the Ip line out of the iso-amp.

Oct 26 1995 12:34:04:350PM951026012Steve WolfeShot 951026012

EF4 hybrid test.
Use shot programming, PLC limits at - 2000A,+3000A start at -1.1,invert at
+1.1, FD at 1.0, hybrid enable. This should give double swing yet.

Pulled the PF breaker. Crowbarred at crossover.
FD looks clean. EF4 being worked on.

Oct 26 1995 01:12:20:310PM951026013Steve WolfeShot 951026013

Changed fuses and tweaked blocking logic in EF4 supply. Removed
1/3 of SCRs and Fuses from circuit.

Repeat EF4 hybrid test.
Use shot programming, PLC limits at - 2000A,+3000A start at -1.1,invert at
+1.1, FD at 1.0, hybrid enable. This should give double swing yet.

Looks good, about 100msec of dead time at crossover. Gets to 2000A
at about 1sec, but it's not flat out on voltage. The EF2 by EF4
may be holding it back.

Magnetics didn't store correctly, but no error on ActMon

Oct 26 1995 01:34:07:150PM951026014Steve WolfeShot 951026014

EF2U&L hybrid test. FD @0.08, Invert at +.1. PLC current limit at 2000A

EF2U gets 1830, EF2L got 3000A!!!!
Something is bad.
Operator error! Seg 1 EF2L had not had its factor redone!

Try again.

Oct 26 1995 01:38:46:250PM951026015Steve WolfeShot 951026015

EF2U&L hybrid test. FD @0.08, Invert at +.1. PLC current limit at 2000A

Much better. 1824 U and 1773 L. we are still getting more than we ask
for, but we'lll see what it does when the alt is loaded.

Oct 26 1995 02:00:55:970PM951026016Steve WolfeShot 951026016

OH1 commutation test, under hybrid control.

Note that there is plenty of room on the EF2 PID, so we can put back the
gain we lost when the integrator was halved. If it's offset, we'll
do something else.

Looks good. On to the OH2's

Oct 26 1995 03:15:37:710PM951026017Steve WolfeShot 951026017

OH1 + OH2's commutation test, under hybrid control.

Charging relays didn't close. Vinnie will watch them during a swiches
open cycle. (No currents). Relay didn't do the right thing.
This Must be fixed.

Granetz will go down to the cell to change out the Crate controller.
Magnetics still broken

Fixed the charging relay problem by increasing the accept range in the PLC.

Try it for real now (its been like an hour)

Looks very bad. No current on OH2L at all, OH2U was driven to -8kA at
OH1 commutation. Probably a power supply or PLC problem. Hybrid
signals look there and correct.

Oct 26 1995 03:21:49:480PM951026018Steve WolfeShot 951026018

OH2's PLC test. See if that still works. It doesn't.

Go on to the EF1's

Oct 26 1995 03:30:57:550PM951026019Steve WolfeShot 951026019

Try an EF1 commutation test, under hybrid control.
Looks good, around 4700Amps, good commutation.

Oct 26 1995 03:39:04:720PM951026020Steve WolfeShot 951026020

Try EF3 under hybrid control.
Looks ok, gets to 2000A before the FD kills it.

Oct 26 1995 03:49:10:030PM951026021Steve WolfeShot 951026021

Test with breakers open.

Oct 26 1995 04:01:12:670PM951026022Steve WolfeShot 951026022

PLC OH2 test.
Also gas fill and puff from "A"-side/ Didn't fill, bypass

No OH2 current. PG pulse seemed to work.

Oct 26 1995 04:02:29:110PM951026023Steve WolfeShot 951026023

Test. The gas worked ok this time.

Oct 26 1995 04:26:07:420PM951026024Steve WolfeShot 951026024

Test.

Oct 26 1995 04:33:49:400PM951026025Steve WolfeShot 951026025

VB found something in the PLC logic,
try PLC OH2 test again, no commutation

It worked.

Oct 26 1995 04:54:29:390PM951026026Steve WolfeShot 951026026

Back to OH1 and OH2 commutation test with Hybrid
Load P=4 for EF2U&L seg 1 gains

Breaker pulled out. Looks like OH2's crowbarred.
COM fault in OH2 bypass.

Finish up for today, see what they find tomorrow.

Physics Operator Comments

Engineering Operator Comments
ShotTimeTypeStatusComment
108:33:18:980AMTestOk
208:41:48:510AMTestOk
309:32:29:530AMTestOk
410:13:56:750AMTestOk
510:27:27:860AMTestOk
610:34:26:760AMTestOk
710:40:16:850AMTestOk
810:52:44:670AMTestOk
911:01:04:890AMTestOk
1011:09:14:830AMTestOk
1111:34:54:360AMTestOk
1212:11:53:520PMTestOk
1301:05:42:060PMTestOk
1401:16:27:830PMTestOk
1501:36:34:850PMTestOk
1601:55:38:730PMTestOk
1702:05:18:160PMTestOk
1803:18:10:390PMTestOk
1903:27:58:260PMTestOk
2003:36:23:960PMTestOk
2103:43:41:010PMTestOk
2203:51:51:140PMTestOk
2303:58:36:310PMTestOk
2404:08:15:900PMTestOk
2504:31:57:870PMTestOk
2604:40:49:430PMTestOk