Alcator C-Mod Run 1060517 Information

Back to runs

Miniproposals
Miniproposal:298a
Date Filed: 5/20/2002
Title:C-Mod Plasma Startup
First Author:Jim Irby
Session Leader:Steve Wolfe (shots 1-21)
Miniproposal:409
Date Filed: 1/19/2005
Title:Phase Space of the Scrape-off-Layer Plasma
First Author:Brian Labombard
Session Leader:Brian Labombard (shots 22-33)

Operators
Session leader(s):Steve Wolfe,Brian Labombard
Physics operator(s):Steve Wolfe
Engineering operator(s):Andy Pfeiffer,Sam Pierson,Bill Parkin

Engineering Operator Run Comment
MP#298a C-Mod Startup (reverse field)SL: PO: Wolfe

Session Leader Plans
Entered: May 17 2006 08:55:03:353PM
Author: Brian Labombard
Run Plan for 1060517

Prepare for MP#409 "SOL Phase space studies" - Reversed Field

Goal
Continue the SOL 'phase space' mapping studies at 0.8MA and 5.4Tesla, but with
reversed plasma current and field direction.

Run Preparation Tasks (for probe diagnostics)
- Configure ASP bias supplies: PA93 amplifiers ("DC supply") in standard sweep mode,
"Sweep A" and "Sweep B" set to -200 volts
- Configure divertor probes in standard "sweep" mode

Run Preparation Tasks (for optical diagnostics)
- Connect the following fiber views for the fast diodes
Outer SOL views 1-3,5,7,9-14,31-40 (total of 21)
Innerwall SOL views B and C
- Connect the following fiber views for the CHROMEX
Outer SOL views 18-22,24-29 (total of 11)
Innerwall SOL views A,D,E,F - otherwise known as 1,4-6
- Set timing on fastdiode digitizers to 0.1 s after the A-B limiter NINJA puff timing

Plan
Set up reversed field, LSN and USN discharges that are similar to
Shot#1060324005 (LSN, 0.8MA, 5.4T, running on segment#2) and
Shot#1060324002 (USN, 0.8MA, 5.4T, running on segment#3)

ASP, ISP programmed to scan at 0.65, 0.85, 1.05 seconds
NINJA gas puff for GPI and boron V spectroscopy entering plasma at 1.1s
Set NINJA for D2 puff at A-B limiter and B016-001. Plenum pressure of 10 PSI.

Shots
1,2,3 LSN, NL04 = 0.6, optimize probe targeting
4,5,6 USN, NL04 = 0.6, optimize probe targeting

7,8 USN, NL04 = 0.4
9,10 LSN, NL04 = 0.4

11,12 LSN, NL04 = 0.8
13,14 USN, NL04 = 0.8

15,16 USN, NL04 = 1.0
17,18 LSN, NL04 = 1.0

19,20 LSN, NL04 = 1.2
21,22 USN, NL04 = 1.2

23,24 USN, NL04 = 1.4
25,26 LSN, NL04 = 1.4

Allow for repeated attempts of these shots.

Shot total is ~30

Physics Operators Plans
Entered: May 16 2006 01:29:56:317PM
Author: Steve Wolfe
Physop Run Plan and Engineering Setup for Wed. 17-May-2006
Power System tests 
followed by
MP#298a C-Mod Startup (reverse field)
SL:
PO: Wolfe 

Overnight ECDC in D2, bake at 60C
Run starts at 09:00 (lockup at 8:30) ends at 17:00

Power system tests: (assuming alternator is ready)
 Operate supplies under PLC control to assure coils
 are properly configured for reverse field

For MP#298a (C-Mod startup) 
Power systems as on 1060324005 (0.8MA, 5.4T, LSN)
 Except: A-coils [-Dtop +Dbot +Jtop -Jbot] which is 
   reversed from the normal setup.

Gas setup:
 fill B-Top with 6 psi D2 Hybrid enabled (PG4)
 fill B-side lower with 1 psi Ar + 2 psi Ne Hybrid DISABLED (PG1)
 leave B-side upper as is Hybrid DISABLED (PG2)
 fill B-main (C-side) with 40 psi D2 Hybrid enabled (PG3)
 leave NINJA as is DISABLED

Enable the following gate valves and shutters, assuming no vacuum problems:
ECE, VUV, Z-bolo shutters, DNB

Torvac setup: Gate valves open

PCS setup:
  Load from 1060324005 (LSN in seg2, USN in seg3)
  Flip waveforms for reverse field (all segs)
  Call predictor for B21_PROJ (segs 2-4)
  Tune startup to taste
Confirm that dpcs_fizzle source_shot=1060516999

Plan: 
  Get startup to run. Get rid of the runaways on the original shot
  Make sure shape is reasonable (compare to 1060324005)
  Let RF check things out
  Call Brian to take over as SL and start MP#409 shots
    which will start from this shot setup. 
  

Session Leader Summaries
Entered: May 17 2006 09:58:12:103PM
Author: Brian Labombard
Run Summary for 1060517

Preparation for MP#409 "SOL Phase space studies" - reversed field

Steve Wolfe managed to get a reversed field plasma on the
first try today (shot#14). This quick startup allowed us to
start setting up for MP#409 on shot#22. After a little
scramble, we managed to get the inner wall probe and analysis
software to work in the reversed field. Also the
A-port probe continued to look ok, having 'recovered' from
some deep insertions (unintended) during a previous run
of MP#409.

By the end of the run, we got some great reversed field target
plasmas in LSN and USN with good probe targeting. We have already
recorded useful data at NL04=0.68. With continued luck, we hope
to have a productive go at MP#409 tomorrow.

Quick shot log:

ISP pulse duration set to 22 ms.

24- LSN, Ip=0.8MA, NL04=0.68, ASP(+6) could be deeper, FSP (+3) far out, ISP could be deeper
25- fizzle
26- LSN, Ip=0.8MA, NL04=0.68, ASP(+8), FSP (+7), ISP-good - next time try 1mm less gap.
27- USN attempt - fizzle
28- USN, Ip=0.8MA, NL04=0.66-0.63 no probes scanning.
29- fizzle
30- USN, Ip=0.8MA, NL04=0.68, ASP(+8), FSP (+7), ISP-good
31- USN, Ip=0.8MA, NL04=0.68, ASP(+9) excellent, FSP (+7)-ugly!, ISP-great
32- no power shot
33- USN, Ip=0.8MA, NL04=0.46, ASP(+9) excellent, ISP-good

Physics Operator Summaries
Entered: May 17 2006 06:38:51:023PM
Author: Steve Wolfe
Physics Operator Summary for 1060517 

Power System tests     (SL/EO: Cochran)
MP#298a C-Mod Startup  (SL/PO: Wolfe)
MP#409 SOL Phase Space (SL: LaBombard)

This was a pretty good start to the reverse field mini-campaign.
Power system testing was completed in 13 shots, finishing around noon. 
All the coil system connections were verified to have been correctly reversed,
and no engineering problems were encountered. 

During the power tests, the Physop was able to troubleshoot two PCS problems:
 For the first couple of shots the demands from DPCS were bad (high from
 -3.5sec) because I had accidentallty clicked Segment#1 OFF. This resulted
 in the gas coming on full from -3.5sec until the fizzle detector gate at 0.08
 sec on shot#1, which closed gate valves but was otherwise harmless. Could
 have been worse. This was corrected for shot#3 (gas had been turned back off
 for #2). 

 The dpcs_fizzle sequence that I had prepared in saved shot#1060516999 turned
 out to be faulty because I neglected to do a build_segment on it, and/or a
 build_all on the shot. This caused the Mout demands for the OH's to have the
 wrong sign. The problem was noted during the PLC testing, and
 corrected. Shots#6 and 7 had the properly built fizzle sequence turned on in
 segment 4, with
 \dpcs::top.loadables.procedures:async_pros:procedure_01:source_shot set to
 $shot, and subsequent shots have source_shot set to 1060517007. This setting
 should be maintained for all reverse field shots. 

NOTE: If the new Beta PCS GUI is used and a shot from earlier this year loaded
(rather than just importing segments) the value of source shot will need to be
changed back to this shot number. The GUI will pop up a warning/acknowledge
message with this information. At present, the Beta GUI does not provide an
interface for making this change, and it will be necessary to use the
traverser on the model file AFTER LOADING from PCS. I'm working on the
necessary interface and may have it up in the pcs3-Beta directory by the end
of the week. 

The first plasma attempt was shot#14, and it ran full length, although the
startup was not pretty. I was so surprised I forgot to load my logbook entry. 

The startup was adjusted over several shots, although I would not call the
final program robust it was doing better than on some recent normal field
runs. The major changes were to IC_EF4 (it was too positive) and changing the
sign of the RCUR offset (from +1600 to -1200; the latter value might still be
tweaked a bit to change the Bz ramp rate). 

Once the RF got on the air, the power came up quickly to the 3MW level, with
all antennas operating more or less cleanly. We had LSN H-modes on shots
#17-21. For shot#22 and after the RF was off again, so that Brian could start
taking data for MP#409. 

The B21_PROJ feedback was broken for the first plasma shots, driving the Acoil
full on while reporting increasing error signals. It turns out that although I
had recalculated all the predictors using the new tap configuration, I had not
updated the nodes in the analysis tree that the predictor code uses to get the 
coefficients to use a reverse-field plasma surface to calculate the m/n
spectra. It took me a while to figure out what the problem was and fix the
trees, so from shot#18-25 I either turned the acoil off or ran it under
waveform pre-programming.  The B21 scope is returning bogus values (although
the magnitudes are correct, the phases are out by 180deg) for shots before #24
when the trees were fixed, except for #22 which was re-done as noted
below. I'll put a note in the logbook when I update #14-21, and 23. 

The relevant tree nodes were updated first in the shot#22 pulse file and then
in the analysis and engineering models by running the routines 
~wolfe/idl/error_fields/load_nonaxi_nodes.pro and
load_nonaxi_applied_nodes.pro
with surf_shot=1060517022. The B21_PROJ predictors were then recalculated
again, resulting in sign changes on the coefficients of the acoil current and
180degree orientation change on the projected direction. Once this was done
the AOUT error signals on the B21_PROJ wires again made sense. Feedback was
turned back on in segment 3 for the first upper null shot (#27) and reinstated
in all segments by shot#28.  Any shot from #28 on should be ok for B21_PROJ
feedback in any of segments 2-4 (#4 may need to have the integrator C gate
moved up to 0.3sec). 

Brian officially started MP#409 on shot#22. We get to count about 3hours of
Research time today. We got a pretty good headstart on the reverse field part
of this experiment, completing one matched density case in lower and upper
null, and getting one shot at reduced density in the upper null
configuration. There were no difficulties getting the flipped target plasmas
to go as programmed, although Brian did want some 2mm tweaks to the gaps for
target optimization. 

Shot#32 had a data system problem and had to be bypassed, and there were some
residual dislocations on shot#33. Martin reports that several of his scopes
blew away trying to look at it, and Ian's run_summary routine hung up on
either 32 or 33. It was a good shot though, so hopefully the data Brian needs
are there. 

Manfred Bitter and company were here testing their new soft xray spectrometer
system, and from what I understand, they had success. See Manfred and/or John
Rice or Alex Ince-Cushman for details. 

Scorecard
---------

    Tests      13
    Sys err     1
    Duds        0
    Fizzles     5
    Plasmas    14
  =========   ===
    Shots      33

Session Leader Comments
May 17 2006 02:15:25:290PMBrian LabombardRun Plan for 1060518 (with potential to set up for reversed field shots on 1060517)

MP#409 "SOL Phase space studies" - Reversed Field

Goal
Continue the SOL 'phase space' mapping studies at 0.8MA and 5.4Tesla, but with
reversed plasma current and field direction.

Run Preparation Tasks (for probe diagnostics)
- Configure ASP bias supplies: PA93 amplifiers ("DC supply") in standard sweep mode,
"Sweep A" and "Sweep B" set to -200 volts
- Configure divertor probes in standard "sweep" mode

Run Preparation Tasks (for optical diagnostics)
- Connect the following fiber views for the fast diodes
Outer SOL views 1-3,5,7,9-14,31-40 (total of 21)
Innerwall SOL views B and C
- Connect the following fiber views for the CHROMEX
Outer SOL views 18-22,24-29 (total of 11)
Innerwall SOL views A,D,E,F - otherwise known as 1,4-6
- Set timing on fastdiode digitizers to 0.1 s after the A-B limiter NINJA puff timing

Plan
Set up reversed field, LSN and USN discharges that are similar to
Shot#1060324005 (LSN, 0.8MA, 5.4T, running on segment#2) and
Shot#1060324002 (USN, 0.8MA, 5.4T, running on segment#3)

ASP, ISP programmed to scan at 0.65, 0.85, 1.05 seconds
NINJA gas puff for GPI and boron V spectroscopy entering plasma at 1.1s
Set NINJA for D2 puff at A-B limiter and B016-001. Plenum pressure of 10 PSI.

Shots
1,2,3 LSN, NL04 = 0.6, optimize probe targeting
4,5,6 USN, NL04 = 0.6, optimize probe targeting

7,8 USN, NL04 = 0.4
9,10 LSN, NL04 = 0.4

11,12 LSN, NL04 = 0.8
13,14 USN, NL04 = 0.8

15,16 USN, NL04 = 1.0
17,18 LSN, NL04 = 1.0

19,20 LSN, NL04 = 1.2
21,22 USN, NL04 = 1.2

23,24 USN, NL04 = 1.4
25,26 LSN, NL04 = 1.4

Allow for repeated attempts of these shots.

Shot total is ~30
May 17 2006 02:39:53:023PM1060517022Brian LabombardStarting set-up for MP#409 - reversed field

Shot 22 NL04=0.65 Ip=0.8, Bt=5.4 REVERSED
ASP, FSP, ISP scanning at 0.65, 0.85, 1.05s
ASP - far in SOL
FSP - far SOL
ISP - Good targeting.

Next: Bring probes in

May 17 2006 03:05:47:570PM1060517024Brian LabombardShot 24 NL04=0.68 Ip=0.8, Bt=5.4 REVERSED
ASP, FSP, ISP scanning at 0.65, 0.85, 1.05s
ASP - at separatrix
FSP - far SOL
ISP - Good targeting.

Next: repeat with FSP deeper

May 17 2006 03:43:38:307PM1060517026Brian LabombardShot 26 LSN, NL04=0.68-0.65 Ip=0.8, Bt=5.4 REVERSED
ASP, FSP, ISP scanning at 0.65, 0.85, 1.05s
ASP - great target!
FSP - Not working very well
ISP - Good targeting - best so far.

Next: try upper null

May 17 2006 03:53:04:820PM1060517028Brian LabombardShot 28 USN, NL04=0.66-0.63 Ip=0.8, Bt=5.4 REVERSED
no probes scanning.

Next: reduce inner and outer gaps by 2mm and scan probes.

May 17 2006 04:22:26:133PM1060517030Brian LabombardShot 30 USN, NL04=0.68 Ip=0.8, Bt=5.4 REVERSED
ASP, FSP, ISP scanning at 0.65, 0.85, 1.05s
ASP - could be deeper
FSP - limping along...
ISP - OK, could be deeper

Next: Repeat, with ASP deeper and smaller inner gap

May 17 2006 04:30:43:947PM1060517031Brian LabombardShot 31 USN, NL04=0.68 Ip=0.8, Bt=5.4 REVERSED
ASP, FSP, ISP scanning at 0.65, 0.85, 1.05s
ASP - excellent!
FSP - limping along...and getting into trouble - time to turn it off.
ISP - good

Next: Go to NL04 = 0.4

Physics Operator Comments
May 17 2006 07:42:11:810AMSteve Wolfe
Setting up for plasma shots for 1060517,
although they still have to go through the power
test sequence. 

Using the old gray PCS for now, since the procedure-editing
pieces of the alpha version aren't quite ready for prime
time yet (but close). 

Load from 1060324005 (Brian's shot for MP#409, 800kA,5.4T,6e19)
Seg1: All_Vs :Flip wires 1-9,10,11 (all PF coils, TF)
      All_Ps :Flip wires 9-12 (all PF coils, Ip)
      Change title to "Reverse field startup"
Seg2: All-Ps :Flip wires 13,15,16 (EF2byEF4,EF4,Ip)
      All_Vs :Flip wires 1-9,10,11
      Change title "+0.8MA SNL"
      Can't redo B21_PROJ because EngOp has tap settings wrong
Seg3: All_Ps :Flip wires 13,15,16
      All_Vs :Flip wires 1-9,10,11
              Zero Acoil_cur (this was just wrong at 2500)
      Change title
      Try to remember to redo B21_PROJ later
Seg4: All_Ps :Flip wires 13,15,16
      All_Vs :Flip wires 1-9,10,11
      Change title
      Try to remember to redo B21_PROJ later
Load at 17-May-2006 07:41:45.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done



May 17 2006 07:46:15:843AMSteve Wolfe
Continuing to set up for run 1060517
Seg1: Br0 from .0002 to 0.000
PG4 from 22ms to 20ms
Note IC_EF4U=+1335A, leave it for now

Load at 17-May-2006 07:45:54.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done


May 17 2006 08:15:33:797AMSteve Wolfe
Andy has updated the C-Coil tap settings, so I can
go ahead and recalculate the B21_PROJ predictors
Do it in segs 2-4, check version return for setup=
-Dtop +Dbot +Jtop -Jbot . 
I hope this is all correct. It's possible I should have
just left them alone, since the helicity is the same?
Go with it this way, if I'm wrong we'll need a cell access 
to fix it, but if I think about it any more I'm sure to 
get it wrong!

Load at 17-May-2006 08:12:43.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

May 17 2006 09:10:50:907AM1060517001Steve Wolfe
Shot#1: EF2U PLC test. We will have gas on the first shot
Got a lot of gas on that shot? Fizz should have taken
it out at .08sec
Also lots of magnetics errors for low signal, but
that's expected. 

Looks like B-main puff was on from very early?
DPCS is doing something weird too, it exited early,
couldn't find the fizzle segment?

Turn off gas for now. Something is broken in the
PCS setup. 

May 17 2006 09:16:09:110AM1060517002Steve Wolfe
Shot#2: EF2L test, no gas this time. 
No PCS change yet, still trying to figure out what happened

For next shot I'll disable the dpcs_fizz and see if that
is all that's wrong. 

May 17 2006 09:57:16:453AM1060517002Steve Wolfe
Shots#1 and 2 had crazy saturating outputs on DPCS
because Seg 1 had been switched off (operator error). 
Seg2 was on/sync with a start_time=0.1sec
All the waveforms seem to have been full on at t=-3.5sec,
which is NOT what I would have expected, but maybe it will 
make sense if I look at what the code does in extrapolating
backward in this case. 

There were also errors in the log file associated with 
assembling the dpcs_fizzle waveform from segment 4 of
shot 1060516999. I think I had this set up correctly 
yesterday, but may not have. The dpcs_fizzle routine was 
disabled on shot#3, on which I turned seg1 back on (and ran ok)


May 17 2006 09:30:44:733AM1060517003Steve Wolfe
Shot#3: disable the dpcs_fizz routine and see if that
is all that's wrong. 
This will be EF3 test, PLC control
I see what's wrong, Seg1 is off!
Reload, leave dpcs_fizzle off for now, 
it was giving an error. 

Load at 17-May-2006 09:23:29.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

OK, those Mouts look more reasonable, except that
the dpcs_fizzle isn't there of course. 

Next shot: reenable dpcs_fizzle and see if
that was also because there was no seg1 of a 
different problem

May 17 2006 10:12:46:093AM1060517004Steve Wolfe
Shot#4: EF4 forward bridge 
dpcs_fizzle enabled in dpcs::top.loadables.procedures:async_pros:procedure_01
with source_shot=1060516999, segments=[4]; if I screwed 
up making those pulse files I may have to re-do this with 
a local seg4. 

Looks ok, positive current in EF4. Fizzle sequence (with
backporch) triggered ok. Positive Bz confimed. Note there
is reverse current in EF3, so the mflux isn't as straight
as I might expect.

I'm not positive the polarity of the oh demands after
fizzle trig makes sense?

Next shot: re-enable gas

May 17 2006 10:20:02:500AM1060517005Steve Wolfe
Shot#5: re-enable gas (B-top,B-main). EF1U only test.
Got some gas. EF1U positive current.
Gas looks good.
I'm not happy with the sign of the Mout demands on the
fizzle sequence. The aout signals seem ok, but the voltages
are the same sense as the pre-charge. Maybe I needed
to do a build before saving the pcs tree?



May 17 2006 10:31:22:763AM1060517006Steve Wolfe
Shot#6: EF1L PLC test.
Import seg4 from 1060516999
Interesting that it gave me the warning message
Do a build segment on seg4
Build_all, 
Click On/Async and do it again(build_seg, build_all)
Load into model
Load at 17-May-2006 10:24:21.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

Change source_shot from 1060516999 to $shot
EF1L positive current
Fizzle demands for oh's are in opposite direction this time. 
** Needed to do the build **


May 17 2006 10:47:00:013AM1060517007Steve Wolfe
Shot#7: Reload with seg4 title changed to "reverse field fizzle"
Change dpcs_config:config_id to ask for 1060517007
OH1 forward bridge test.
OH1 positive current, get negative Bz
Looks ok. 

Next is oh2u forward bridge
Change source_shot from $shot to 1060517007

May 17 2006 10:57:55:623AM1060517008Steve Wolfe
Shot#8: oh2u forward bridge
Change dpcs_fizzle source_shot from $shot to 1060517007
Turn off seg4 
Load at 17-May-2006 10:47:59.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

OH2U positive current, fizzle signals still look ok. 
Negative radial field, ok. 

Next: OH2L test. Re-import seg 4 from earlier shot

May 17 2006 11:09:39:280AM1060517009Steve Wolfe
Shot#9:  OH2L test. Re-import seg 4 from shot#3
(this is dn->un ssep control; I don't want it,but it's there)
Click sync (already off) and load
Load at 17-May-2006 10:59:54.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

OH2L positive current, positive radial field ok
next: TF test

May 17 2006 11:29:48:797AM1060517010Steve Wolfe
Shot#10: TF test, no PCS changes
Positive 50kA TF current
Bz=+3mT, mflux doesn't look like I remember it for TF only?
There's a null near R=1m, Z=0.
Btor is positive all right. OK.

May 17 2006 11:34:45:203AM1060517011Steve Wolfe
Shot#11: TF 100kA test
TF test ok

Next: Chopper test

May 17 2006 11:51:25:513AM1060517012Steve Wolfe
Shot#12: Chopper test (Hybrid control)
OK, got reverse current from normal

next: program acoil in seg1

May 17 2006 12:12:34:140PM1060517013Steve Wolfe
Shot#13: program acoil in seg1 [-.9 to 0] at 2000A
Acoil pulse ok (we'll see if the polarity is what I 
want, but it's undoubtedly what I asked for)
Engineering setting had the old tap settings. 
Change it manually, so I can stare at the result and 
still be confused.

Next: Zero acoil current waveform in seg1
Go for plasma


May 17 2006 05:42:08:620PM1060517014Steve Wolfe
Shot#14: plasma full length. 
Startup ugly, but it made it. 
No runaways apparent on the TV, neuts/hards and bolo
signals missing. 

H-alpha trace looks strange, slow wobulations. 
Current trace ok, density all over the place (but mostly
lower than program). Bz very positive. 


May 17 2006 12:36:30:860PM1060517015Steve Wolfe
Shot#15:  enable b-side lower (50ms). Ef4 from 1335 to 1320
Fizzle. Still hesitating, looks like multiple bounces,
loses it vertically. 

next: Change rcur offset from 1600 to 0.



May 17 2006 12:52:42:220PM1060517016Steve Wolfe
Shot#16: Change rcur offset from +1600 to 0.
If this helps, I may go to -1600 next.
RF is supposed tp fire on this shot (if it goes)
Plasma full length, but had runaways. 
See if neuts/hards trace is back after Matt fixed it; Yep.
Startup nicer, still a little positive on Bz, some
bounciness before 15msec.
Got RF, H-alpha looking better (still L-mode).
One big injection around 1.0sec. 

next: raise nl04 a little, drop ef4 another 10amps

May 17 2006 01:13:35:217PM1060517017Steve Wolfe
Shot#17:ef4 from 1320 to 1310. nl04 from 6e19 to 8e19. 
Plasma. Still runaways, but I think we got H-modes that time.
Injection at .71sec, H-modes look ELM-free. 

I think A-coil is doing the wrong thing though.
It seems to be coming on at .5sec, and going to the
rail. Must have something wrong. 

Well, the .5sec is because the integrator gate is set wrong.
Change that to .3sec. However, the feedback looks backward
based on the b21 scope. Also from the Aout signal.

Turn it off for a shot. 

May 17 2006 01:25:08:533PM1060517018Steve Wolfe
Shot#18:B21_PROJ gains from P=1.0 to 0. and I=4.0 to 0
RCUR offset from 0 to -1000.
Plasma, still with runaways and H-modes
That one looks like a bouncer in startup.
Another injection around 0.7sec 

Next: More negative again on ef4. Run Acoil pre-programmed 

May 17 2006 01:45:41:883PM1060517019Steve Wolfe
Shot#19: EF4 from 1310 to 1300. 
Program 2000A on Acoil from .6 to 1.5 (flattop interval)
Plasma, still getting runaways. Brian poked on this shot.
Bounce is worse this time. Am I going the wrong way on EF4?
According to the b21 scope, intrinsic field is at +21deg and
applied is -158, so they should be compensating. 
Don't understand the feedback driving it to the rail. Could
the controller gain be backward?
 
Leave it on manual for now, and tweak at the startup. 

May 17 2006 02:04:15:980PM1060517020Steve Wolfe
Shot#20: pg4 from 20 to 24msec (asking for a fizzle)
plasma, xrays not as bad, looked like an h-mode on the tv
Runaways look gone. Startup plasma moves out, almost
loses it; bz not rising (negatively) fast enough.

The B21_Proj error signal is going the wrong way. All the
predictor terms have reversed sign. I think somehow the 
Acoil one shouldn't. This may be a bug in the algorithm,
or in the values stored in the tree. I think possibly
this is because somewhere I've got a surface normal 
set for a negative current plasma! However, in that case,
I could also be fooling myself about whether I've 
got the configuration right, too.


Next: RCUR offset one way or the other. 

May 17 2006 02:10:59:023PM1060517021Steve Wolfe
Shot#21:  RCUR offset from -1000 to -1600
Plasma, injection early (.7sec), but survived.
RF, H-modes, probes.
Less of an outward swing on startup, no runaways

Next: repeat. RF off, Brian will poke deeper. 
I'm going to stare at some error field routines and
see if I can sort out what these traces mean. If we
stop getting H-modes, maybe I'll be able to test whether
the acoils should be in this direction or not by making
and/or suppressing a locked mode.

May 17 2006 02:21:09:807PM1060517022Steve Wolfe
Shot#22: Repeat
plasma, no runaways, no injection
startup ok, no bounce
No locked mode. According to the b21 scope,
B21 is below 2G, intrinsic up to 6G.

Next: Acoil off 

May 17 2006 02:33:21:307PM1060517023Steve Wolfe
Shot#23: Acoil off.
Fizzle. Big lump on the Halpha, looks inside.

Next: drop the gas. 

May 17 2006 03:00:12:167PM1060517024Steve Wolfe
Shot#24: PG4 from 24 to 22ms. Acoil off.
Plasma, no runaways. Probe, no rf (by design)
Bz now is beginning to look a little too negative,
but startup is nice.

No sign of a locked mode. I/n isn't high enough for an 
intrinsic I guess. Max amplitude of b21 is about 5G 
according to the B21 scope. The aout scope also says
B21_PROJ=+.0005 (whatever the sign means).



May 17 2006 03:09:12:213PM1060517025Steve Wolfe
Shot#25: Recalculate the B21_PROJ predictor in seg2. It
now says 
21-Jul-2005,total 2/1 non-axisymmetric field at phi=  18.67 degrees, Acoil=-DTOP+DBOT+JTOP-JBOT
(used to say -158 degrees)
Leave Acoil off, look at aout signal on this one. 

Another fizzle. Better work on the startup some more. 
Lumpy halpha again, bz too negative or too much gas?

Next: Brian wants to decrease left gap by 2mm

May 17 2006 03:24:08:163PM1060517026Steve Wolfe
Shot#26: PG4 back to 20msec from 22
CLEARIN from .0015 to -.0005
RCUR from .658 to .657, etc.
Put acoil back on at 2000A 
(note Acoil coeff in predictor changed sign)
plasma, no runaways.
startup ran outside a bit, but no bounce.
AOUT for B21_PROJ now agrees that the acoil is reducing
the error field. 

Next shot: Go to seg3 (upper null)

May 17 2006 03:33:46:917PM1060517027Steve Wolfe
Shot#27: Go to uper null (seg 3)
Recalculate B21_PROJ predictor (again), and leave it on.
Set integrator gate to 0.3sec in seg3.
Set nl04 from 6 to 8e19
Increase pg1 puff from 40 to 50ms

Fizzle. Still lumpy and moving inside. I need to either adjust
EF4 of Rcur offset

May 17 2006 03:49:37:837PM1060517028Steve Wolfe
Shot#28: EF4 from 1300 to 1310
Rcur offset from -1600 to -1200

Go ahead and recalculate the b21_proj predictor in seg4
Turn the B21_PROJ gains in seg2 (P=1.0, I=4.0) and zero
the acoil current on seg2 all_v screen
That should leave this set up for tomorrow, with all segments
ready. 

Plasma no runaways. Bz still looks a little high, 
lump on Halpha is earlier. 
B21_PROJ feedback looks ok now.  

Next: Brian wants 2mm less inner gap, and also outer gap less by 2mm
Figure out what to do with this startup. 

May 17 2006 04:06:00:057PM1060517029Steve Wolfe
Shot#29: Decrease CLEARIN from -.001 to -.003
Ef4 from 1310 to 1320. Pg4 from 20ms to 18ms
Fizzle. Bouncer, Got a magnetics error 
"f13 does not have finite signal". 

Put back the 2ms on pg4


May 17 2006 04:12:55:337PM1060517030Steve Wolfe
Shot#30:PG4 from 18 to 20msec 
CLEARIN still at -.003
Plasma, poky probe, some runaways. 
Still bouncing in startup. Injection at .13sec.

Next: put back 5amps on ef4 


May 17 2006 04:25:11:820PM1060517031Steve Wolfe
Shot#31: ef4 form 1320 to 1315 
Another 1mm on CLEARIN, -.003 to -.004
plasma, a few runaways, still bouncy.
gaps are smaller, as requested. 

Next: add 2ms on pg4

May 17 2006 04:41:49:650PM1060517032Steve Wolfe
Shot#32: pg4 from 20 to 22msec
Decrease nl04 from 8 to 5.5e19; decrease pg3 in seg3
No Power Shot, data system problem

May 17 2006 04:59:49:480PM1060517033Steve Wolfe
Shot#33: pg4 from 20 to 22msec
Decrease nl04 from 8 to 5.5e19; decrease pg3 in seg3
Plasma, no runaways, startup swinging outside but 
no bounce. Bill had to blow away a bunch of CAMAC inits.
Density looks good at 4.5e19. Brian's happy, I'm happy. 

Quit while we're ahead. 
END OF RUN.

Engineering Operator Comments
ShotTimeTypeStatusComment
109:02:47:580AMTestOkEF2 Upper 3KA 500V PLC Test
209:13:37:530AMTestOkEF2 Lower 500V 3KA
309:26:11:437AMTestOkEF3 3KV 3KA
409:58:37:467AMTestOkEF4 Forward Bridge Test 1KV 2KA
510:13:17:907AMTestOkEF1 Upper forward bridge test +500V 2KA
610:28:00:953AMTestOkEF1L Forward Bridge Test 500V 2KA
710:40:23:780AMTestOkOH1 Forward Bridge test 1KV 2KA
810:53:06:843AMTestOkOH2 Upper Forward Bridge Test 300V 2KA
911:06:46:013AMTestOkOH2 Lower Forward Bridge test 300V 2KA
1011:19:33:390AMTestOkTF 50KA Reverse field
1111:31:59:297AMTestOkTF 100 KA Reverse field
1211:44:37:343AMTestOkEFC 400V/1.0KA Reverse field
1311:55:17:373AMTestOkA-coil test
1412:13:17:060PMPlasmaOk
1512:26:34:220PMPlasmaOk
1612:47:38:000PMPlasmaOk
1701:01:14:140PMPlasmaOk
1801:18:38:843PMPlasmaOk
1901:35:31:500PMPlasmaOk
2001:48:13:680PMPlasmaOk
2102:00:49:353PMPlasmaOk
2202:15:28:307PMPlasmaOk
2302:28:39:980PMPlasmaOk
2402:52:59:620PMPlasmaOk
2503:05:45:667PMPlasmaOk
2603:18:29:120PMPlasmaOk
2703:31:13:493PMPlasmaOk
2803:43:50:900PMPlasmaOk
2903:56:34:023PMPlasmaOk
3004:09:03:010PMPlasmaOk
3104:21:32:883PMPlasmaOk
3204:35:28:870PMTestBad
3304:52:42:290PMPlasmaOk