Alcator C-Mod Run 1020619 Information

Back to runs

Miniproposals
Miniproposal:298a
Date Filed: 5/20/2002
Title:C-Mod Plasma Startup
First Author:Jim Irby
Session Leader:Jim Irby (shots 1-25)

Operators
Session leader(s):Jim Irby
Physics operator(s):Steve Wolfe
Engineering operator(s):Gary Dekow,Bill Byford

Engineering Operator Run Comment
Plasma Startup, SL: Irby, PO: Wolfe, EO:Dekow/Byford

Session Leader Plans

Physics Operators Plans

Session Leader Summaries
Entered: Jul 7 2004 03:49:28:803PM
Author: To Be Determined
Run Summary for 1020619 Plasma Startup MP298A

SL: Irby
PO: Wolfe
EO: Byford/Dekow


We continued to clean up the machine today. The density was
programmed to 7e19/m^2 to keep from over-gasing the plasma. During
the early discharges most of the particles making up the plasma
were coming from the walls. NL04 for example rose to nearly 1.5e20/m^2
on shot 05.

After shot 09 we stopped for approx one hour to change the EF4
connections so that all the current was properly measured by the
bus Rogowskis. We had been living with a 120 amp imbalance
in the magnets. On shot 10 the imbalance was back to the 10 to
20 amp range. However, following this change, it took several
duds and fizzles and field tweaks to get a very nice plasma on
shot 17. By this point the density was almost under control and
only rising to 8e19/m^2 before development of what looked like
a MARFE beyond 1.1 s.

On shot 21 we successfully started moving the plasma up towards
a more double null configuration. Unfortunately, this was our last
plasma. The remaining shots were either duds or fizzles. We
also began running the DNB on shot 21, which may or may not have
affected operation. The MARFE formation was also moving later and
later into the shot as we cleaned up the machine.

Generally a very good day in that we are now clearly cleaning up the
machine and getting full length discharges. H_to_D was also
improving during the day. It ranged from 20 down to 14% at 200 ms
and was below 50% at 1.5 s.

John Rice mentioned that the new tangential x-ray detectors are working
well, and Howard says that the MSE signal looks very promising.


The plasma shots:

shot duration cii
---- ------- ---
03 0.39 .004
05 1.29 .005
07 1.30 .008
08 0.25 .008
09 1.73 .007
14 0.14 .008
17 1.68 .010
18 1.71 .011
20 1.69 .008
21 1.73 .006

Physics Operator Summaries
Entered: Jul 7 2004 04:37:14:863PM
Author: To Be Determined
Physics Operator Summary for Wed 19-JUN-2002

SL: Irby
PO: Wolfe
EO: Dekow,Cochran,Byford
MP: 298A C-Mod Plasma Startup

Engineering setup
------------------
ECDC overnight in D2 at 2e-4 Torr, stopping by 8:00 am.
We may want ECDC in D2 between shots.

Run begins at 09:00 and ends at 17:00.

Power systems the same as for shot 1020618012.

Gas setup:
fill B-Top with 6 psi D2 Hybrid enabled (PG4)
fill B-side lower with 1 psi Ar Hybrid enabled (PG1)
fill B-side upper with 10 psi D2 Hybrid DISABLED (PG2) (for ECDC)
fill C-side with 30 psi D2 Hybrid enabled (PG3)
leave J-Bottom as is Hybrid DISABLED (PG5)

Enable the following gate valves, assuming no vacuum problems:
ECE, DNB

-------------------------------------------------------------------------------
RF Setup:

None
-------------------------------------------------------------------------------
Run Plan:

Continue startup struggles.
If we get more reproducible plasmas, we will try to track down
the EF4 imbalance problem. If we start getting plasmas through
rampdown we will start the day 1&2 program of MP#298A.

Start from 1020618012; a .6sec shot
We may want ECDC in D2 between shots.

We would like to have the TCI working. We will probably also puff some
argon for HIREX Ti (and Te) measurements.

Results:
-------

An ok day, certainly not great, but progress was made.
We ran ecdc for 5 minutes between shots all day. The machine does
seem to be cleaning up. We had several 800kA plasmas run into ramp-down.
H/D on these was le 0.5, and the density ran around 7e19 with no gas
demand during flattop. On the other hand, the startup was not robust,
losing us several shots at the end of the run after nearly full length pulses
seemed established.

The swap of the two J-bot connectors restored the apparently broken JK coils
and a few of the GH coils. The weights for BP_14 and BP15 were put back on in
\analysis::efit_efitin:fwtmp2 . Bob also put the original \BTOR loop back
in. Unfortunately, on shot #6 BP16_JK died for real, probably due to the
disruption on shot 5, and Bob had to change the processed BP_15 to use bp16_bc.
BP16_JK was showing baseline jumps for the rest of the run. It's possible that
tightening the connector might fix it again, but for now we can use bp16_bc.

THe EF4 imbalance problem was understood and corrected, as described below.

The Power Room UPS failed last night, and was placed in bypass for the run.
This failure (possibly) was responsible for some CAMAC and PLC errors on
shot#1 which resulted in a dud due to no EF4; the Chopper also misbehaved
on #1 and the tf rogowski trace had a huge baseline offset. Service was called
on the UPS, and when the tech arrived this afternoon the unit seemed to be
working again. We left it bypassed, however, since no cause for the original
failure was identified.

After power cycling some PLC's and taking a test shot on the bus rogowskis,
we ran an EF4 and EFC hybrid control test on shot #2, which was successful.

Shot #3 was a short plasma, followed by a fizzle on #4, and a long (1.29s) plasma
pn #5, another fizzle on 6, and plasmas on #7-9. #9 got to 1.63sec, well into
rampdown.

While this was going on, we reviewed the modification that had been made to
the EF4 varistor connection with Bill Cochran, and determined that a current
path had been introduced which allowed current in ef4L that would not show up
on the EF4 bus rogowski, and actually went through the EF4U shunt. Some of the
EF4L current also could go through the EF4U rogowski. This was the likely
reason for the apparent large imbalances in the EF4 rogowski signals, and
for the very different programming required on IC_EF4 to get good Bz.

After shot #9, Bill restored the old varistor configuration, eliminating the
unmeasured and mis-measured current paths. This resulted in very well balanced
signals on the EF4U and EF4L bus rogowskis. However, it took more shots than
expected to recover plasmas, and the startup was not as robust, although the
fields looked good.

Shots #10-12 were duds. I was adjusting IC_EF4, and shot #12 with IC_EF4=-1350
had nearly identical Bz to #9. Shot #13 was a fizzle, after a small Br tweak
from .002 to .0025 . #14 dropped Br0 to .00225 and was a short plasma, which
got into trouble around 100msec, and 15 was another dud. #16 was also a dud,
with Br0 back to .002.

For shot #17 Br0 dropped to .0015, and the dnb gate valve and zbolo-shutter
were disabled. This shot ran to 1.68 sec. shot #18 was a repeat and ran a
little longer. On #19 we re-enabled the bolo shutter, and got a dud. However,
#20 was a repeat, with the shutter still open, and we got another 1.7 sec
shot.

For Shot #21 we lowered the upper xpoint by nearly 2cm, and got a nice long
(1.73sec) plasma with elongation over 1.7 and nearly flat density. The dnb
gate valve was re-enabled for this shot, but the beam didn't run.

For shot #22 we programmed zcur to be 1cm higher (from -.015 to -.005), but
shots #22-25 were all duds or fizzles. It may have been coincidence that the
beam was firing on all of these, or it might not. Small changes in fields and
gas puff did not recover a current rise. We never got the higher plasma
position shot.

At around shot 12 I discovered that the magnetics predictors had not been
updated. Apparently when Bill or maybe Jim loaded from shot 1010801015 he
ignored the warning about f24a being changed to f24. I recalculated the Br0
predictor for shot 13, but didn't get around to doing all of the others until
shot #24 (seg 2 done for shot #22, the unused seg3 and 4 on shot #23). When I
did seg 1 I found that there were two input mismatches, f24a and
bp26_gh. After the run, Bob told me that on May 31, during power tests, he
replaced bp26_GH with bp26_BC on input 55. Bp26_GH was the only one of the
coils on the swapped J-bottom connectors that was being used. However, while
Ian recalculated the predictors, and Bob made a logbook entry under magnetics,
noone updated the hybrid config_id, so the config message did not reflect the
change.

Since BP26_GH is now back where it belongs, after the run Bob swapped these
back, so BP26_GH is now on input_55, and we updated both the input_55 and
config_id nodes. This means that all the predictors I changed will need to be
redone tomorrow, but they would have had to be anyway if a shot before 22 got
loaded.


Scorecard
-----------

Tests 1
Duds 9 (including 1 power supply dud)
fizzles 5
plasmas 10
======= ==
shots 25


Session Leader Comments
Jun 19 2002 09:43:56:733AM1020619001Jim IrbyTook a test shot (no power) to check bus Rogowskis. Everything seems
to be okay.
Jun 19 2002 10:51:13:523AM1020619005Jim Irbynear full length plasma (1.29 s)

Density rises through out shot w/o gas input

Carbon is lower (.006)

HTOD 20% at 200 ms and rises to abt 60% at end of shot

Jun 19 2002 11:22:31:037AM1020619007Jim Irbynear full length plasma (1.30 s)

Density rises through out shot w/o gas input

Carbon .008

HTOD 16% at 200 ms and rises to 55%

Jun 19 2002 11:48:21:140AM1020619008Jim Irbyplasma, but only .25 seconds.
Ratiomatic not turned on.

HTOD 20% at 200 ms

Jun 19 2002 01:50:46:250PM1020619014Jim IrbyLost plasma at 140 ms or so.
Injections starting at 100 ms...
Jun 19 2002 03:05:36:993PM1020619017Jim IrbyBeautiful shot: 1.68 s

Density almost under control

HTOD 12% at 200 ms, peaking at 44% or so ----- gettting better?



Jun 19 2002 03:10:00:280PM1020619018Jim Irbyanother very nice shot: 1.71 s

HTOD abt 19% at 200 ms. Now has begun to roll over and fall until the MARFE
developed. Peaking late at 50%.

CII at .0011 --- not dropping very quickly
Jun 19 2002 03:46:20:040PM1020619020Jim IrbyPlasma for 1.69 s

Carbon a little lower (.008)

HTOD 14% at 200 ms, max of 43% late in discharge

Next shot Steve will start heading for a double null discharge....
Jun 19 2002 04:05:55:273PM1020619021Jim IrbyPlasma for 1.73 s

Steve was able to move plasma up and begin approach to double null

Carbon II down again (.0065)

HTOD a little higher peaking at 50%

Next shot we will continue moving plasma up...

Physics Operator Comments
Jun 19 2002 07:14:21:973AMSteve WolfeSetting up for 1020619 run
We had a fault in the power room UPS last night; still not fixed, but we
can probably run without it.

I'm going to try doing everything from the cmodws07 screen today,
although if T&J don't get the updates on the Linux scopes working it
may be painful.

Doing coldstart on Spanna.
rtnode -- data loaded, A matrix responds
rtnode -- data loaded, M matrix responds
rtnode -- data loaded, PID matrix responds
RT_NODE Done
Looks ok.

Run PCS from VMSTERM on spanna (@cmod$Models:[hybrid]pcs)
load-from 1020618012
note IC_EF4U=-1200, BR0 0ffset=.002
load into model
Jun 19 2002 07:44:00:233AMSteve WolfeIrby requests that we maximize shot count by reducing between-shot ECDC
to five minutes.
Jun 19 2002 08:33:50:707AMSteve WolfeReload the PCS shot with seg2 title changed to indicate correct field.
Jun 19 2002 09:18:24:067AM1020619001Steve WolfeShot #1: dud - no ef4, chopper funny, oh2l goes up after fizzle, because of
long fizzle delay.
Jun 19 2002 09:35:16:900AM1020619001Steve WolfeShot #1: TF rogowski signal has a huge positive baseline shift +335000
Jun 19 2002 09:51:55:800AM1020619002Steve WolfeShot #2: Chopper and EF4 test only (hybrid control)

Looks ok, ef4 still has imbalance, but runs at program current
efc looks ok.
TF baseline ok.

next shot: full shot, with tf and oh fizzle delays removed

Jun 19 2002 10:07:40:100AM1020619003Steve WolfeShot #3: full shot attempt, with tf and oh fizzle delays removed
plasma to .4sec, some hards on tv (HXray detector not installed)
PF breaker pulled - oh1 after the disruption; self-power circuit fired

It goes resistive around the time it diverts, ~.24sec
density is low,
Jun 19 2002 10:16:18:950AM1020619003Steve WolfeShot #4: increase pg4 puff from 21 to 23 msec
Had to go to spanna screen, alt-shift isn't working on the linux screen!
Jun 19 2002 10:21:42:387AM1020619004Steve WolfeShot #4: increase pg4 puff from 21 to 23 msec
Had to go to spanna screen, alt-shift isn't working on the linux screen!

fizzle - breakjdown slightly late 6.8msec, CII at .018mW/cm2/sr (was ~.005 on #3)
Jun 19 2002 10:52:00:010AM1020619005Steve WolfeShot #5: ecdc for 5min, no pcs changes
Bob switches btor signal back to the original btor loop.
plasma to 1.29 sec, density rising from 0.7sec on, with no
gas puff after .4sec
H/D rises from .2 to .6 before the disruption.
outer strike going above the nose.
ef1U goes to the rail


Figured out how to keep alt-mouse from doing things to the window, and it now works in
PCS, so back to the linux screen.

next shot: 5min ecdc, no pcs changes.

Jun 19 2002 11:00:55:247AM1020619006Steve WolfeShot #6: no pcs changes
fizzle, looks like a bouncer? early bd, bz may be down a tiny bit.

next shot: 5 min ecdc, no pcs changes
Jun 19 2002 11:27:15:490AM1020619007Steve WolfeShot #7: 5 min ecdc, no pcs changes
plasma, to 1.30sec ; similar to 5,
bad startup, 1 bounce, almost lost it.
efit doesn't converge, bp15 (bp16jk) is broken; now it really is broken, doed during #6,
although probably the doiruption on 5 did it in.

Next shot: dnb permissive on for next shot.
5 min ecdc, no pcs changes.
Bob will change bp_15 to bp16_bc
Jun 19 2002 11:46:12:047AM1020619008Steve WolfeShot #8: dnb permissive on.
5 min ecdc, no pcs changes.
Bob changed processed bp_15 to bp16_bc

plasma - short and ugly 0.25 sec
late bd, 7.8 sec
traq 2 failed, so efit ran without it.
ratiomatic was off, can't tell how much the dnb raised p0

next shot: repeat
Jun 19 2002 12:04:13:693PM1020619009Steve WolfeShot #9: repeat
plasma to 1.63 sec, into rampdown
density rises from about 1sec (marfe?)
loop voltage goes from 1.2 to 1.6V

next shot: Gary and Bill will change the EF4 configuration
back to what it used to be, eliminating the extra cble that doesn't pass
through the bus rogowski; this means I'll have to adjust the startup.
I'll guess a value of -1290, and see what we get.
Zbolo shutter will be enabled for next shot.
Jun 19 2002 12:48:25:600PM1020619010Steve WolfeShot #10: ICEF4 from -1200 to -1290; EF4 Varistors back to "normal"
dud. Bz is a little positive, compared to shot 9
#9 #10
Bz[-.005] -.0048 -.0018
[0.] 0.0028 -.0012
[.010] .00009 .0036

next shot: IC_EF4 from -1290 to -1320
Jun 19 2002 12:57:15:307PM1020619011Steve Wolfeshot#11: IC_EF4U from -1290 to -1320
dud, must not have been enough of an increase
give it another 30Amps
Jun 19 2002 01:23:15:097PM1020619012Steve Wolfeshot#12 IC_EF4U from -1320 to -1350; increase pg4 to 26msec
dud. BZ is almost identical to shot 9; br looks about .5mT more negative

next shot: raise br0 from .002 to .0025; it's reporting invalid predictor.
still has f24a? How did it get this way? call the alg, and then set the offset to .0025
also, raise fill again, to 28msec
Jun 19 2002 01:35:45:273PM1020619013Steve Wolfeshot#13:raise br0 from .002 to .0025;
also, raise fill again, to 28msec

fizzle. bd at 6.3msec, bounces

next shot: br0 back down to .00225
Jun 19 2002 01:50:31:860PM1020619014Steve Wolfeshot#14: br0 back down to .00225
short plasma, .14 sec, gets in trouble at seg switch.
bd is late .0078

next shot: repeat
Jun 19 2002 02:10:41:133PM1020619015Steve Wolfeshot#15: repeat, see if it happens again.
dud! br0 is more positive again, bz looks about right

next shot: br0 back to .002,
Jun 19 2002 02:19:04:447PM1020619016Steve Wolfeshot#16: br0 back to .002,
dud, not unexpected.

next shot: br0 down to .0015
Jun 19 2002 02:34:53:953PM1020619017Steve Wolfeshot#17: br0 down to .0015
disable dnb and zbolo shutters.

plasma, to 1.68sec nice and quiet.
bd around 7msec, slightly late.
density rises at ~1.1sec, no injections

next shot: repeat
Jun 19 2002 03:12:50:630PM1020619018Steve Wolfeshot#18: repeat
plasma to 1.7sec. Almost identical.
bd earlier, 5.4sec

next shot: repeat, re-enable zbolo shutter.
Jun 19 2002 03:23:06:100PM1020619019Steve Wolfeshot#19: repeat
dud! Earl insists that the bolo shutter can't cause anything like this,
but that was the only difference.
fields look the same.

next shot: no changes; if we get another dud, close it again.
Jun 19 2002 03:48:10:433PM1020619020Steve Wolfeshot#20: repeat
plasma to 1.7 sec again.
beam on next shot.

next shot: drop zxu from .048 to .03
also, lower pg4 from 28 msec to 24msec
Jun 19 2002 04:01:02:780PM1020619021Steve Wolfeshot#21: drop zxu from .048 to .03
also, lower pg4 from 28 msec to 24msec

plasma to 1.73; bd at 6.3
density is flatter.

next shot: raise zcur from -.015 to -.005
Jun 19 2002 04:17:18:307PM1020619022Steve Wolfeshot#22: raise zcur from -.015 to -.005. While I'm at it
do the predictor recalculations in seg 2. The only offsets are
clearin 12700 and ip -23000.

fizzle. bouncer, got to 42kA. bd at 5.3msec

next shot: 10A more positive on EF4

Jun 19 2002 04:30:58:063PM1020619023Steve Wolfeshot#23: raise zcur from -.015 to -.005.
10A more positive on EF4
recalcualte mag predictors in seg3 and seg 4 (for future use)

fizzle, another bouncer.
next shot: raise pg4 by 2msec

Jun 19 2002 04:47:45:323PM1020619024Steve Wolfeshot#24: raise zcur from -.015 to -.005.
IC_ef4 from -1340 to -1330
raise pg4 by 2msec to 26msec.
recalculate seg1 mag predictors; note that rcur is still using bp26gh (offset 5500)

dud. must have gone the wrong way.

next shot: ic_ef4u to -1360
Jun 19 2002 05:05:21:303PM1020619025Steve Wolfeshot#25: raise zcur from -.015 to -.005.
ic_ef4u to -1360 (maybe I went the wrong way?)

dud . Fields look ok, small negative radial field.
I think it's a gas problem. Maybe something to do
with the dnb, but I don't know what. Note that the
dnb gate valve opened on 21, but beam dodn't fire.
All the subsequent tries the beam went normally, and we
either got duds or fizzles.

Engineering Operator Comments
ShotTimeTypeStatusComment
109:12:36:220AMPlasmaBadEF4 and Chopper didn't go
209:48:58:610AMTestOkEF4 and Chopper test...Okay
309:58:24:143AMPlasmaOk
410:17:25:543AMPlasmaOk
510:37:40:223AMPlasmaOk
610:57:11:700AMPlasmaOk
711:11:36:487AMPlasmaOk
811:35:57:370AMPlasmaOk
911:49:26:157AMPlasmaOk
1012:39:29:410PMPlasmaOk
1112:53:39:087PMPlasmaOk
1201:07:42:390PMPlasmaOk
1301:24:41:380PMPlasmaOk
1401:40:21:543PMPlasmaOk
1501:56:35:453PMPlasmaOk
1602:12:16:367PMPlasmaOk
1702:28:13:293PMPlasmaOk
1802:51:20:523PMPlasmaOk
1903:10:57:907PMPlasmaOk
2003:35:08:990PMPlasmaOk
2103:52:27:597PMPlasmaOk
2204:12:13:190PMPlasmaOk
2304:27:33:993PMPlasmaOk
2404:42:43:283PMPlasmaOk
2504:57:22:557PMPlasmaOk