Alcator C-Mod Run 1031126 Information

Back to runs

Miniproposals
Miniproposal:362
Date Filed:11/12/2003
Title:Shaping and Control Development for high elongation, high I
First Author:Steve Wolfe
Session Leader:Steve Wolfe (shots 1-29)

Operators
Session leader(s):Steve Wolfe
Physics operator(s):Robert Granetz
Engineering operator(s):Bill Byford,Bill Parkin,Gary Dekow

Engineering Operator Run Comment
MP #362 ITER Single-null Shape; SL: Wolfe; PO: Granetz; EO: Dekow/Byford

Session Leader Plans

Physics Operators Plans

Session Leader Summaries
Entered: Jul 7 2004 03:49:34:960PM
Author: Steve Wolfe
Session Leader Summary for Wed Nov 26, 2003

MP#362 (SN ITER shape part )
SL: wolfe
PO: Granetz
EO: Dekow,Byford,Parkin

The goal is to develop a discharge in the ITER shape, kappa=1.85,delta_x=0.49,
with the ITER field of 5.3T and q_95=3, which corresponds to I=1.6MA and to an
I_n=I/aB=1.4.

Plan
------
Engineering setup - standard except EFC Voltage Limit should be raised from
800V to 900V. C-coils should be in "standard" configuration +Dtop -Dbot -Jtop
+Jbot and current programmed to 2500A (all-V screen). C-coil PLC setting
should be 600V, 3000A.

No RF until we get to our final shape.

We will need to monitor halo current signals as we raise the current.

Starting from a recent 1MA shot, 5.3T shot, e.g. 1031023019, modify the
equilibrium to get higher elongation:
ZCUR to 0.0
RXL to -.005
ZXL to -0.0175
RXU to -0.04
ZXU to +.030
CLEARIN from +.005 to +.008
Raise NL04 to 8e19.

Confirm that shape is what we want, i.e. kappa=1.85, delta_u~delta_l~0.49,
aout=.215m

If this disrupts with vertical instability, then back off on CLEARIN and begin
tweaking Gains on wire 2 and wire 3. Present nominal values are P2=0.7 D2=0.7
I2=4.5 P3=4.0 D3=6.0 I3=0 Back in 1995 we ran with P2 and D2 at least 1.5, and
never varied it down. We need to observe the oscillation frequency, if any,
to decide which gains need adjusting. Low frequency oscillations, around
20-50Hz, are associated with the OH2 response, and may indicate we need a
lower wire 2 gain; a fast mode, 200-300Hz, is associated with the chopper on
wire 3. If we decide we need chopper D gain above 10 then change fast ZCUR to
wire 10, which has a factor of 10 higher derivative time constant. Be sure
that the chopper bias is correct to avoid non-symmetric railing; this may
require adjusting the chopper voltage (ALL_V) or changing the resistor value.

If we can't get to the required elongation at 1MA without disruptions after
about 10 shots, then proceed to 1.2MA at whatever kappa we have stably and try
again there. Note change in n/n_crit as current is increased.

Raise current in 100-200kA increments to 1.6MA (which should be I_n=1.4, q~3)
or maximum controllable with the chopper. Watch out for locked modes as we
raise current and increase A-coil current as indicated. Look at my B_21 total
estimates for guidance in setting A-coil programming.

The RF E transmission line is open, so we cannot pulse RF today without a lot
of radiation, so we won't be able to do any coupling studies even if we finish
early. Therefore, in that unlikely eventuality (note that we are ending an
hour early), we will either proceed to the FIRE-shaped DN development or do
some development of the JET shape in preparation for Hender's experiment
MP#361.

Results
-----------

Despite lots of disruptions, and post-disruption fizzles, we made considerable
progress toward the goal. We established the shape and on shot 28 got to the
target current of 1.6MA, with I_N=1.4, kappa=1.84, deltax=0.5; actual q_95 was
2.9 rather than 3, so the shape wasn't exact. N/N_Crit on this discharge was
1.05. However, we did not get this discharge to run full length, though the
disruption in this case was not a VDE, but seemed to be related to running the
OH1 onto the current rail, which had been left at -27kA and actually limited
at 25kA. For the last shot we increased the PLC limits on OH1 and EF3 (to
-33kA and 13kA, respectively) but the disruption-induced fizzle got us.

The path up to the high current had disruptions, mostly VDE's at every
stage. There was only one full length shot, #18, which was at 1MA. This shot
and some of those at 1.2MA had ohmic H-modes, but we lost these as we went up
in current, for no clear reason. The n/ncrit of the H-mode phase was lower
than the L-mode, which probably helped on shot 18.

We made a number of gain tweaks on the fast ZCUR, and ended up using wire 10
and a D10 gain of 1.2 (equivalent to 12 on wire 3); this compares to the
starting value of D3=6. The values used were D3=6,9,7.5 and D10=1.2. Changes
to the fast P gain did not seem to have strong effects on the vertical
disruptions, though they did change the amplitude and frequency of the
oscillations well away from the final event. We ended up at P10=3, compared to
a starting value of 4. We ran this between 2 and 8; the higher values caused
larger swings of chopper current (rail to rail) but were not apparently
helpful. Note that the Df="12" also caused rail to rail chopper excursions,
but the plasma survived these. At shot 8 we changed the chopper pre-program
voltage to zero (nominal was +100V) to try to symmetrize the current and avoid
flooring. This seems to work, although it leads to a no-plasma current level
centered about 2000A rather than 1500. We should probably adjust the resistor.

We made no changes to the slow zcur gains, although the Tinios run had both P2
and D2 double what we run now. This is the next thing that should be tried. The
D10 gain may well be too high (it is banging the rails), but it seemed to do
somewhat better than D3=9 at 1.4MA.

More time, though probably not a full run, will be required to complete this
development.

Other Issues
------------

From the start of the run, we had a new behavior with the oh1 pre-charge
current. This has been running at -19.7 to -19.8kA (bus rog) forever; with no
change in the PCS program today it was running about 20.1kA. This value was
consistent with the observed change in Bz at breakdown. The hybrid Aout
signal also indicated that it was 300A off-target and had reduced the demand to
compensate, but the gain was apparently insufficient; looking at yesterday's
traces, the aout values showed a good match and the demand was ~100V,
consistent with the V program, as opposed to 60V today. We compensated by
changing the program waveform and lived with the 300V error value, but this
seems to be a real change either in the hybrid or in the supply. There seems
to be a difference right from supply turn-on at -1.1sec. The 300A differnce is
already present at -1sec, and the hybrid demand is already lower by that time.
The effect is consistent with a small reduction in the oh1 resistor value, or
with a difference on the induced voltage. The only other supplies on at that
time are te EF1's and EF4.




Physics Operator Summaries
Entered: Jul 7 2004 04:37:19:863PM
Author: Robert Granetz
Physics operator summary for run 1031126 (Wednesday):

MP #362 - ITER single-null shape

Session leader: S. Wolfe
Physics operator: R. Granetz
Engineers in charge:

Summary: We achieved a 1.6 MA discharge with the proper parameters (kappa=1.83,
delta=0.49-0.50, I_N=1.4, q95=2.9, nl_04=0.8-1.0e20 m-2) on shot 28.
This shot disrupted -- not due to vertical instability, but rather
to hitting the OH1 PLC current limit. We may want to pick up from
here, with the PLC limit set to -33 kA, when we use this equilibrium
for future ITER studies. In order to obtain this plasma, we played
around with P and D gains on the fast ZCUR, and the voltage on the
EFC chopper supply. In the end, we had to move the ZCUR control
over to wire_10, which has 10 times higher derivative gain capability,
and set the derivative gain to 1.2 (corresponding to 12.0 if it had
still been on wire_03). We had the A-coils on for the entire run, in
the normal configuration with 2200 amps. We did not see any sign
of a locked mode, so this bodes well for even higher plasma current.
There were many high current disruptions today, with several shots
having halo currents > 150 kA (up to 210 kA). We had several stress
limit warnings at the 50% level on EFCU. The numerous disruptions
also led to numerous fizzles, but no duds.

Scorecard: 17 plasmas
11 fizzles
0 duds
0 test or no-power shot cycle
1 aborted and/or screwed up by computer problems
----------
29 total

-------------------------------------------------------------------------------
Engineering setup for Wednesday, 26 Nov 2003

MP #362 - ITER single-null shape

Session leader: S. Wolfe
Physics operator: R. Granetz

ECDC overnight in D2 at 2e-4 Torr, stopping by 07:00.
Bake at 60 C overnight as well.

Run begins at 09:00 and ends at 16:00. (NOTE EARLY ENDING TIME)

Power systems the same as on 1031023019 (1.0 MA, 5.3 T, lower single null),
EXCEPT that the PLC voltage limit for EFC should be raised from 800 to 900 V.

A-coils should be hooked up as on 1031125007 (+Dtop -Dbot -Jtop +Jbot), and
the power supply should be enabled. C-coil PLC limit settings should be set
at 600 volts and 3 kA.

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

NO NINJA!

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

-------------------------------------------------------------------------------
RF Setup: No RF until we get to our final shape, which will probably be late
in the run

-------------------------------------------------------------------------------
Run Plan:

We are trying to get 1.6 MA, 5.3 T (q95=3) equilibria with the following
ITER shape:
- single lower null
- kappa = 1.85
- lower and upper triangularities approximately 0.49
- a=0.215 m
We will start at 1.0 MA and work our way up in increments. Vertical
stability is likely to be an issue, and we anticipate that the run will
involve some adjustment of the slow and fast ZCUR PID gains. Monitoring
and stabilization of locked modes will probably be required as the plasma
current is raised. Halo currents will also be monitored.

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

ECDC was done overnight, as well as a 60 C bake.

Hybrid coldstart done by the engineering group; no errors found

Loaded PCS segment 1 from yesterday's 1031125002 segment 1.
Loaded PCS segment 2 from 1031023019, with the following changes
to get higher elongation:

ZCUR to 0.0 m
RXL to -.005 m
ZXL to -0.0175 m
RXU to -0.04 m
ZXU to +.030 m
CLEARIN to +.008 m

Also raise NL04 to 8e19.

Copied segment 2 to segment 3 and set up wire 10 to be fast ZCUR with
one-tenth of the gain, and turned off wire 03, and set Ref 9-12 to IP.
This is for possible use if we need a stronger fast ZCUR feedback,
since the derivative time constant is 10 times faster on wire 10.

Also turn off argon permissive.

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

The start of the run will be delayed while a problem with the EF1 power
supply is worked on.

Shot 01 -- Plasma. Disrupted at t=0.55 s due to vertical instability. Kappa
got above 1.8; n/ncrit got to 1.2.

Next shot: reduce CLEARIN to +0.005 m, increase RXU to -0.03 m,
tweak Ip rampup.

Shot 02 -- Aborted. No power cycle due to hangup in init.

Shot 03 -- Fizzle. Bz a little high.

Next shot: add -10 amps to IC_EF4

Shot 04 -- Fizzle. Steve notes that the OH1 current is couple of hundred
amps more (negative) than normal.

Next shot: add +200 amps to IC_OH1 (-20000 --> -19800)

Shot 05 -- Plasma. Made it into rampdown before VDE disruption. Kappa is
1.85, delta_U and _L is 0.58. EFC current is not symmetric.
We may be dithering with an upper null.

Next shot: add +1 cm to both RXL and RXU to reduce the
triangularities.

Shot 06 -- Plasma. VDE disruption at t=0.55 s. The triangularities did
indeed come down close to the desired values.

Next shot: start playing with EFC chopper. Change its
pre-programmed voltage to +200 V (from +100 V). We're not really
sure of the direction, so this is a best guess.

Shot 07 -- Plasma. Disrupted at t=0.72 s, but it was not a VDE. I think it
was an injection. Kappa=1.85, delta_L and _U approx 0.49, so
this is the shape we want. But the EFC current is even more
asymmetric, so maybe we went the wrong direction.

Next shot: change chopper pre-programming to 0.0 V (from +200 V)

Shot 08 -- Plasma. Disrupted very early (t=0.35 s)

Next shot: no changes.

Shot 09 -- Plasma. Disrupted due to VDE.

Next shot: change fast ZCUR P gain to 2.0 (from 4.0)

Shot 10 -- Plasma. Disrupted earlier.

Next shot: tweak Ip rampup; delay CLEARIN increase until after
t=0.5 s (to avoid OH1 crossover)

Shot 11 -- Plasma. Disrupted at t=0.64 s. kappa=1.85, delta_L=0.50,
delta_U=0.46

Next shot: increase proportional gain on fast ZCUR to 6.0 (from
2.0). Also add -3 mm to RXU.

Shot 12 -- Plasma. Disrupted at t=0.64 s. The oscillations on EFC current
are back. The plasma shape is exactly what we want.

Next shot: reduce fast ZCUR P gain to 4.0 (from 6.0)

Shot 13 -- Fizzle. Fields and fill pressure look okay.

Next shot: no changes.

Shot 14 -- Plasma. Disrupted at t=0.67 s, VDE.

Next shot: raise fast ZCUR derivative gain to 9.0 (from 6.0)

Shot 15 -- Fizzle.

Next shot: no changes.

Shot 16 -- Fizzle.

Next shot: add +100 amps to IC_OH1

Shot 17 -- Fizzle. Fields look okay.

Next shot: try a BR_0 tweak.


Shot 18 -- Plasma. Full-length shot; no disruption! Elongation=1.85, upper
and lower triangularities=0.50-0.51. So this is a stable shot (just
barely though) with the ITER shape at Ip=1.0 MA (q95=4.7-4.8).
This shot also has an ohmic H-mode (small ELMs) starting at t=1.0 s.
Startup isn't the greatest.

Next shot: increase Ip to 1.2 MA

Shot 19 -- Plasma. Upward VDE, disrupting at t=0.84 s. Ohmic H-mode starts
at t=0.75 s.

Next shot: reduce fast ZCUR P gain to 2.0 (from 4.0)

Shot 20 -- Fizzle. Disruption recovery?

Next shot: no changes

Shot 21 -- Fizzle.

Next shot: additional BR_0 tweak

Shot 22 -- Plasma. Downward VDE disruption. I_halo_lower=155 kA (first time
over 100 kA in a long time). Shape is still very good.

Next shot: raise Ip to 1.4 MA; change fast ZCUR P gain to 3.0,
tweak pre-fill gas puff.

Shot 23 -- Plasma. 1.4 MA Upward VDE; disrupted at t=1.25 s.
I_halo_upper=135 kA. Got a 50% warning on the EFC_U coil stress.
Our locked mode compensation is spot on. kappa=1.83, delta=0.49.

Next shot: decrease fast ZCUR derivative gain to 7.5 (from 9.0)

Shot 24 -- Fizzle. Disruption recovery?

Next shot: tweak pre-fill.

Shot 25 -- Fizzle.

Next shot: no changes

Shot 26 -- Plasma. 1.35 MA. Disrupted at t=0.62 s. This was a VDE, but it
didn't move very far down before quenching.

Next shot: transfer the fast ZCUR over to wire 10, which has 10
times higher derivative gain capability. Set ZCUR derivative gain
to 1.2 (corresponds to 12 on the old wire 03). Also turn off
wire 03 (by hitting the "zero all" button)

Shot 27 -- Plasma. Disrupted at t=1.12 s. This was NOT a VDE! Looks like
an injection. Good shape, with kappa=1.83, delta=0.49-0.50. EFC
current is oscillating wildly between both rails.

Next shot: raise Ip to 1.6 MA, and program nl_04 to go to end of
flattop.

Shot 28 -- Plasma. 1.60 MA. Disrupted at t=1.17. NOT a VDE! I don't think
it was due to a big impurity injection. Sawteeth go away just
before the disruption, but it's not clear there was a locked mode.
Shape is very good: kappa=1.83, delta_L=0.50, delta_U=0.49.
OH1 rails at it's PLC limit, so this may have caused the disruption.
Plasma touches the inner midplane wall just before the disruption.
q95=2.9 !

Next shot: raise PLC limit on the OH1 to 33 kA (actually -33 kA).
Also raise PLC limit on EF3 by 1 kA.

Shot 29 -- Fizzle.

End of run.

Session Leader Comments
Nov 26 2003 09:06:56:263AMSteve WolfeThere will be a short delay while the engineers replace a relay
in the EF1 supply
Nov 26 2003 09:42:42:487AM1031126001Steve WolfeShot#1: 1MA, high elongation setup.
Early disruption .54sec.

Vertically unstable like mad. 150Hz oscillation on zcur
Next shot: back off on clearin to .005, move RXU out by 1cm,
speed up current rise.


Nov 26 2003 11:49:12:620AM1031126002Steve WolfeShot#2: back off on clearin to .005, move RXU out by 1cm,
speed up current rise.

Data hang in init.
No power shot.
Nov 26 2003 10:05:02:077AM1031126004Steve WolfeShot#3: back off on clearin to .005, move RXU out by 1cm,
speed up current rise.

fizzle, late breakdown. Previous one was late too.
OH1 current seems to be 300A high?
Nov 26 2003 10:16:59:410AM1031126004Steve WolfeShot#4: back off on clearin to .005, move RXU out by 1cm,
speed up current rise.

fizzle. Still late breakdown and too much bz.
Nov 26 2003 10:29:56:040AM1031126005Steve WolfeShot#5: still trying for the second shot: back off on clearin to .005, move RXU out by 1cm,
speed up current rise.

plasma to 1.5sec, chopper is flooring but not railing.
n/ncrit at 1.13. Kappa = 1.86, triang=0.6, close to DN

next shot: decrease upper and lower triangularity
Nov 26 2003 10:44:00:030AM1031126006Steve WolfeShot#6: decrease upper and lower triangularity
disrupt at .54, n/n_crit=1.18, elong=1.84
same 150Hz oscillation
Chopper is still asymmetric. Try changing the Vout to from +100 to +200.
Nov 26 2003 11:06:26:827AM1031126007Steve WolfeShot#7: Try changing the Vout to from +100 to +200.
plasma disrupts at .72, chopper still flooring; looks worse,
I think we went the wrong way on the voltage.
n/nc = 1.15, eout=1.86, triang about 0.5, ssep=-.45cm
Bob says this wasn't a vde. frequency is higher before it
goes.

Try 0 on the chopper voltage, see if that symmetrizes better.
Nov 26 2003 11:25:11:483AM1031126008Steve WolfeShot#8: Try 0 on the chopper voltage, see if that symmetrizes better.

disrupts at 0.35sec, chopper level seems higher, but
the behavior is different.

next shot: repeat and see how reproducible it is.
Nov 26 2003 12:02:02:957PM1031126009Steve WolfeShot#9: repeat and see how reproducible it is.

not very, disrupts at .61, ~125kHz oscillation.
n/nc = 1.17

next shot: drop P3 gain from 4 to 2
Nov 26 2003 12:02:17:910PM1031126010Steve WolfeShot#10: drop P3 gain from 4 to 2

disrupts at .35 again; this is near oh1 crossover.

Delay the clearin and increase the current rise to try to separate
these effects.

Nov 26 2003 12:01:40:190PM1031126011Steve WolfeShot#11: Delay the clearin and increase the current rise to try to separate
these effects.

disrupts at 0.64, n/nc=1.17.

try P3 up to 6

Nov 26 2003 12:15:05:290PM1031126012Steve WolfeShot#12:try P3 up to 6

disrupts at .65, big efc oscillation this time, still somewhat asym
toward 0.
oscillation is around 125Hz again. shot 11 was slower.

P3 back to 4
Nov 26 2003 12:26:40:970PM1031126013Steve WolfeShot#13:P3 back to 4
fizzle
Nov 26 2003 12:40:19:147PM1031126014Steve WolfeShot#15:P3 back to 4

disrupts at 0.67sec, oscillation ~125Hz

Raise D3 gain from 6 to 9
Nov 26 2003 01:06:31:720PM1031126016Steve WolfeShot#16: Raise D3 gain from 6 to 9
Another fizzle.
Nov 26 2003 01:19:55:037PM1031126017Steve WolfeShot#17: Raise D3 gain from 6 to 9, another tweak of oh1current.
Yet another fizzle.

Nov 26 2003 01:38:27:447PM1031126018Steve WolfeShot#18: Raise D3 gain from 6 to 9, tweak to BR

Plasma full length!
And it got an H-mode
Shape looks ok, somewhat increasing triangularity, kappa=1.85
oscillating at close to 200Hz, chopper has big oscillations, but doesn't
rail on either side.
Some early runaways.
n/nc~1.1 in L-mode, 0.9 in H-mode

next shot: 1.2MA

Nov 26 2003 01:49:15:150PM1031126019Steve WolfeShot#19: 1.2MA

plasma disrupts at .85 in H-mode.

Drop P3 from 4 to 2, leave D3=9
Nov 26 2003 01:59:03:430PM1031126020Steve WolfeShot#20: Drop P3 from 4 to 2, leave D3=9
fizzle, not surprising.
Nov 26 2003 02:11:40:573PM1031126021Steve WolfeShot#21: Drop P3 from 4 to 2, leave D3=9
another fizzle.
Nov 26 2003 02:30:46:620PM1031126022Steve WolfeShot#22: Drop P3 from 4 to 2, leave D3=9
disrupt at 0.87sec

next shot: P3 from 2 to 3, current to 1.4
Nov 26 2003 02:46:06:277PM1031126023Steve WolfeShot#23: P3 from 2 to 3, current to 1.4

plasma to 1.24 sec

drop D3 gain from 9 to 7.5
Nov 26 2003 02:54:46:260PM1031126024Steve WolfeShot#24: drop D3 gain from 9 to 7.5
fizzle.
Nov 26 2003 03:06:25:570PM1031126025Steve WolfeShot#25: drop D3 gain from 9 to 7.5
And another fizzle.
Nov 26 2003 03:27:11:587PM1031126026Steve WolfeShot#26: drop D3 gain from 9 to 7.5
disrupts at .635


next shot: Change to wire 10 and set D10 to 1.2 (==12 on wire 3)
Nov 26 2003 03:42:25:917PM1031126028Steve WolfeShot#27: Change to wire 10 and set D10 to 1.2 (==12 on wire 3)
plasma to 1.1 but not a VDE. Injection at 1.104.

next shot: up in current to 1.6MA
Nov 26 2003 03:54:58:900PM1031126028Steve WolfeShot#28: up in current to 1.6MA

plasma to 1.17sec, not a VDE
hits the oh1 current rail

next shot: raise current rails on oh1, ef3
Nov 26 2003 03:58:51:197PM1031126029Steve WolfeShot#29: raise current rails on oh1, ef3

fizzle. Out of time.
Happy Thanksgiving!

Physics Operator Comments
Nov 25 2003 02:25:46:503PMRobert GranetzEngineering setup for Wednesday, 26 Nov 2003

MP #362 - ITER single-null shape

Session leader: S. Wolfe
Physics operator: R. Granetz

ECDC overnight in D2 at 2e-4 Torr, stopping by 07:00.
Bake at 60 C overnight as well.

Run begins at 09:00 and ends at 16:00. (NOTE EARLY ENDING TIME)

Power systems the same as on 1031023019 (1.0 MA, 5.3 T, lower single null),
EXCEPT that the PLC voltage limit for EFC should be raised from 800 to 900 V.

A-coils should be hooked up as on 1031125007 (+Dtop -Dbot -Jtop +Jbot), and
the power supply should be enabled. C-coil PLC limit settings should be set
at 600 volts and 3 kA.

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

NO NINJA!

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

-------------------------------------------------------------------------------
RF Setup: No RF until we get to our final shape, which will probably be late
in the run

-------------------------------------------------------------------------------
Run Plan:

We are trying to get 1.6 MA, 5.3 T (q95=3) equilibria with the following
ITER shape:
- single lower null
- kappa = 1.85
- lower and upper triangularities approximately 0.49
- a=0.215 m
We will start at 1.0 MA and work our way up in increments. Vertical
stability is likely to be an issue, and we anticipate that the run will
involve some adjustment of the slow and fast ZCUR PID gains. Monitoring
and stabilization of locked modes will probably be required as the plasma
current is raised. Halo currents will also be monitored.
Nov 26 2003 08:44:09:310AMRobert Granetz ECDC was done overnight, as well as a 60 C bake.

Hybrid coldstart done by the engineering group; no errors found

Loaded PCS segment 1 from yesterday's 1031125002 segment 1.
Loaded PCS segment 2 from 1031023019, with the following changes
to get higher elongation:

ZCUR to 0.0 m
RXL to -.005 m
ZXL to -0.0175 m
RXU to -0.04 m
ZXU to +.030 m
CLEARIN to +.008 m

Also raise NL04 to 8e19.

Copied segment 2 to segment 3 and set up wire 10 to be fast ZCUR with
one-tenth of the gain, and turned off wire 03, and set Ref 9-12 to IP.
This is for possible use if we need a stronger fast ZCUR feedback,
since the derivative time constant is 10 times faster on wire 10.

Also turn off argon permissive.
Nov 26 2003 09:03:37:717AMRobert GranetzThe start of the run will be delayed while a problem with the EF1 power
supply is worked on.
Nov 26 2003 09:38:02:987AM1031126001Robert GranetzShot 01 -- Plasma. Disrupted at t=0.55 s due to vertical instability. Kappa
got above 1.8; n/ncrit got to 1.2.

Next shot: reduce CLEARIN to +0.005 m, increase RXU to -0.03 m,
tweak Ip rampup.
Nov 26 2003 09:57:51:160AM1031126002Robert GranetzShot 02 -- Aborted. No power cycle due to hangup in init.
Nov 26 2003 09:58:02:990AM1031126003Robert GranetzShot 03 -- Fizzle. Bz a little high.

Next shot: add -10 amps to IC_EF4
Nov 26 2003 10:12:58:367AM1031126004Robert GranetzShot 04 -- Fizzle. Steve notes that the OH1 current is couple of hundred
amps more (negative) than normal.

Next shot: add +200 amps to IC_OH1 (-20000 --> -19800)
Nov 26 2003 10:34:11:447AM1031126005Robert GranetzShot 05 -- Plasma. Made it into rampdown before VDE disruption. Kappa is
1.85, delta_U and _L is 0.58. EFC current is not symmetric.
We may be dithering with an upper null.

Next shot: add +1 cm to both RXL and RXU to reduce the
triangularities.
Nov 26 2003 10:45:58:343AM1031126006Robert GranetzShot 06 -- Plasma. VDE disruption at t=0.55 s. The triangularities did
indeed come down close to the desired values.

Next shot: start playing with EFC chopper. Change its
pre-programmed voltage to +200 V (from +100 V). We're not really
sure of the direction, so this is a best guess.
Nov 26 2003 11:08:49:450AM1031126007Robert GranetzShot 07 -- Plasma. Disrupted at t=0.72 s, but it was not a VDE. I think it
was an injection. Kappa=1.85, delta_L and _U approx 0.49, so
this is the shape we want. But the EFC current is even more
asymmetric, so maybe we went the wrong direction.

Next shot: change chopper pre-programming to 0.0 V (from +200 V)
Nov 26 2003 11:36:01:680AM1031126008Robert GranetzShot 08 -- Plasma. Disrupted very early (t=0.35 s)

Next shot: no changes.
Nov 26 2003 11:35:40:523AM1031126009Robert GranetzShot 09 -- Plasma. Disrupted due to VDE.

Next shot: change fast ZCUR P gain to 2.0 (from 4.0)
Nov 26 2003 12:03:09:407PM1031126010Robert GranetzShot 10 -- Plasma. Disrupted earlier.

Next shot: tweak Ip rampup; delay CLEARIN increase until after
t=0.5 s (to avoid OH1 crossover)
Nov 26 2003 12:02:33:237PM1031126011Robert GranetzShot 11 -- Plasma. Disrupted at t=0.64 s. kappa=1.85, delta_L=0.50,
delta_U=0.46

Next shot: increase proportional gain on fast ZCUR to 6.0 (from
2.0). Also add -3 mm to RXU.
Nov 26 2003 12:27:40:547PM1031126012Robert GranetzShot 12 -- Plasma. Disrupted at t=0.64 s. The oscillations on EFC current
are back. The plasma shape is exactly what we want.

Next shot: reduce fast ZCUR P gain to 4.0 (from 6.0)
Nov 26 2003 12:27:50:247PM1031126013Robert GranetzShot 13 -- Fizzle. Fields and fill pressure look okay.

Next shot: no changes.
Nov 26 2003 12:46:00:813PM1031126014Robert GranetzShot 14 -- Plasma. Disrupted at t=0.67 s, VDE.

Next shot: raise fast ZCUR derivative gain to 9.0 (from 6.0)
Nov 26 2003 12:58:14:223PM1031126015Robert GranetzShot 15 -- Fizzle.

Next shot: no changes.
Nov 26 2003 01:07:39:733PM1031126016Robert GranetzShot 16 -- Fizzle.

Next shot: add +100 amps to IC_OH1
Nov 26 2003 01:21:09:427PM1031126017Robert GranetzShot 17 -- Fizzle. Fields look okay.

Next shot: try a BR_0 tweak.
Nov 26 2003 01:38:48:650PM1031126018Robert GranetzShot 18 -- Plasma. Full-length shot; no disruption! Elongation=1.85, upper
and lower triangularities=0.50-0.51. So this is a stable shot (just
barely though) with the ITER shape at Ip=1.0 MA (q95=4.7-4.8).
This shot also has an ohmic H-mode (small ELMs) starting at t=1.0 s.
Startup isn't the greatest.

Next shot: increase Ip to 1.2 MA
Nov 26 2003 01:50:42:010PM1031126019Robert GranetzShot 19 -- Plasma. Upward VDE, disrupting at t=0.84 s. Ohmic H-mode starts
at t=0.75 s.

Next shot: reduce fast ZCUR P gain to 2.0 (from 4.0)
Nov 26 2003 01:59:51:137PM1031126020Robert GranetzShot 20 -- Fizzle. Disruption recovery?

Next shot: no changes
Nov 26 2003 02:16:19:823PM1031126021Robert GranetzShot 21 -- Fizzle.

Next shot: additional BR_0 tweak
Nov 26 2003 02:36:22:807PM1031126022Robert GranetzShot 22 -- Plasma. Downward VDE; disrupted at t=0.87 s. I_halo_lower=155 kA (first time
over 100 kA in a long time). Shape is still very good.

Next shot: raise Ip to 1.4 MA; change fast ZCUR P gain to 3.0,
tweak pre-fill gas puff.
Nov 26 2003 02:59:40:820PM1031126023Robert GranetzShot 23 -- Plasma. 1.4 MA Upward VDE; disrupted at t=1.25 s.
I_halo_upper=135 kA. Got a 50% warning on the EFC_U coil stress.
Our locked mode compensation is spot on. kappa=1.83, delta=0.49.

Next shot: decrease fast ZCUR derivative gain to 7.5 (from 9.0)
Nov 26 2003 02:57:36:743PM1031126024Robert GranetzShot 24 -- Fizzle. Disruption recovery?

Next shot: tweak pre-fill.
Nov 26 2003 03:10:02:493PM1031126025Robert GranetzShot 25 -- Fizzle.

Next shot: no changes
Nov 26 2003 03:28:38:667PM1031126026Robert GranetzShot 26 -- Plasma. 1.35 MA. Disrupted at t=0.62 s. This was a VDE, but it
didn't move very far down before quenching.

Next shot: transfer the fast ZCUR over to wire 10, which has 10
times higher derivative gain capability. Set ZCUR derivative gain
to 1.2 (corresponds to 12 on the old wire 03). Also turn off
wire 03 (by hitting the "zero all" button)
Nov 26 2003 03:42:11:103PM1031126027Robert GranetzShot 27 -- Plasma. Disrupted at t=1.12 s. This was NOT a VDE! Looks like
an injection. Good shape, with kappa=1.83, delta=0.49-0.50. EFC
current is oscillating wildly between both rails.

Next shot: raise Ip to 1.6 MA, and program nl_04 to go to end of
flattop.
Nov 26 2003 04:01:10:527PM1031126028Robert GranetzShot 28 -- Plasma. 1.60 MA. Disrupted at t=1.17. NOT a VDE! I don't think
it was due to a big impurity injection. Sawteeth go away just
before the disruption, but it's not clear there was a locked mode.
Shape is very good: kappa=1.83, delta_L=0.50, delta_U=0.49.
OH1 rails at it's PLC limit, so this may have caused the disruption.
Plasma touches the inner midplane wall just before the disruption.
q95=2.9 !

Next shot: raise PLC limit on the OH1 to 33 kA (actually -33 kA).
Also raise PLC limit on EF3 by 1 kA.
Nov 26 2003 04:17:24:273PM1031126029Robert GranetzShot 29 -- Fizzle.

End of run.

Engineering Operator Comments
ShotTimeTypeStatusComment
109:23:53:013AMPlasmaOk
309:51:46:320AMPlasmaOk
410:04:24:810AMPlasmaOk
510:17:05:520AMPlasmaOk
610:35:03:600AMPlasmaOk
710:57:25:757AMPlasmaOk
811:12:06:760AMPlasmaOk
911:25:39:703AMPlasmaOk
1011:40:15:957AMPlasmaOk
1111:54:06:273AMPlasmaOk
1212:08:39:687PMPlasmaOk
1312:23:29:127PMPlasmaOk
1412:36:20:117PMPlasmaOk
1512:49:36:153PMPlasmaOk
1601:02:41:270PMPlasmaOk
1701:15:21:603PMPlasmaOk
1801:27:58:980PMPlasmaOk
1901:41:19:057PMPlasmaOk
2001:55:53:040PMPlasmaOk
2102:09:01:027PMPlasmaOk
2202:21:40:963PMPlasmaOk
2302:34:55:493PMPlasmaOk
2402:48:36:773PMPlasmaOk
2503:01:31:383PMPlasmaOk
2603:14:12:120PMPlasmaOk
2703:29:18:790PMPlasmaOk
2803:42:04:853PMPlasmaOk
2903:56:42:277PMPlasmaOk