Alcator C-Mod Run 1060307 Information

Back to runs

Miniproposals
Miniproposal:418
Date Filed: 3/4/2005
Title:LH system commisioning
First Author:Stefano Bernabei
Session Leader:John Liptac (shots 1-24)

Operators
Session leader(s):John Liptac
Physics operator(s):Steve Wolfe
Engineering operator(s):Bill Cochran,Bill Parkin

Engineering Operator Run Comment
Run Plan 1060307 LH Commishioning Experiments MP 418

Session Leader Plans
Entered: Mar 6 2006 10:14:53:227AM
Author: Ron Parker
Run Plan 1060307 LH Commishioning Experiments MP 418

Session Leader: Liptac, Parker

Plasma Conditions:
Same type of plasmas as for the Ti coupler commishioning experiments, 1050414. Low density L-mode, lower single null, nl_04~0.6, B0=5.4 T, and Ip~ 700kA. No ICRF.

Goals:
�?� Measure LH coupling, ie reflection coefficients, for different antenna phasings and plasma densities at the LH grill.
�?� Characterize the plasma density at the coupler at various coupler positions and outer gap sizes.
�?� Examine 3rd harmonic arc detection.

LH setup:
Begin with the launcher at 1650 counts. Set mechanical phase shifters to 90 degree current drive phasing. 20 kW per klystron.

Operation:
Coupler position scan
Shot 1: single segment, 90 degree ideal CD phasing.
Setpoint generator settings: Shot Length=30 ms, Duty Cycle=33%, Pulse Period=30 ms.
Shot 2: two segments, 90, and 120 degree CD phasing.
Setpoint generator settings: Shot Length=60 ms, Duty Cycle=33%, Pulse Period=30 ms.
Shot 3: three segments*, 60, 90, and 120 degree CD phasing.
Setpoint generator settings: Shot Length=90 ms, Duty Cycle=33%, Pulse Period=30 ms.
Shot 4: four segments*, 60, 90, 120, and 150 degree CD phasing.
Setpoint generator settings: Shot Length=120 ms, Duty Cycle=33%, Pulse Period=30 ms.
Shot 5: five segments*, 60, 90, 120, 150, and 180 degree CD phasing.
Setpoint generator settings: Shot Length=150 ms, Duty Cycle=33%, Pulse Period=30 ms.
Shot 6: five segments*, 180, 150, 120, 90, and 60 degree CD phasing.
Setpoint generator settings: Shot Length=150 ms, Duty Cycle=33%, Pulse Period=30 ms.
Cell access to change coupler to intermediate position ~1625 counts.
Shot 7: five segments*, 60, 90, 120, 150, and 180 degree CD phasing.
Setpoint generator settings: Shot Length=150 ms, Duty Cycle=33%, Pulse Period=30 ms.
Cell access to move coupler position as close as we are comfortable ~1600 counts.
Shot 8: five segments*, 60, 90, 120, 150, and 180 degree CD phasing.
Setpoint generator settings: Shot Length=150 ms, Duty Cycle=33%, Pulse Period=30 ms.

Outer gap scan* running as many segments as we can.
Shot 9: five segments*, 90, 90, 90, 90, and 90 degree CD phasing. Keeping outer gap constant to establish baseline.
Shot 10: 90 degree ideal CD phasing.
Cell access to set phase shifters to 60 degrees.
Shot 11: 60 degree ideal CD phasing.
Cell access to set phase shifters to 120 degrees.
Shot 12: 120 degree ideal CD phasing.
Cell access to set phase shifters to 150 degrees.
Shot 13 150 degree ideal CD phasing.
Cell access to set phase shifters to 180 degrees.
Shot 14 180 degree ideal CD phasing.

*Can run up to 20 segments/shot, depending on how it goes- more segments the better

Physics Operators Plans
Entered: Mar 6 2006 08:20:52:733PM
Author: Steve Wolfe
Engineering setup for Tuesday 7-Mar-2006

MP 418 Low power LH Commissioning

Session leader: Liptac,Parker
Physics operator: Wolfe

Overnight ECDC in D2, and bake at 60 C.

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

Power systems as on 1050414014 (0.7 MA, 5.4 T, lower single null)
Fizzle delay on OH1 and OH2's=300msec, Fizzle delay on TF=600ms
A-coil configuration: normal [+Dtop -Dbot -Jtop +Jbot] and enabled.

Gas Setup:
fill B-Top with 6 psi D2 Hybrid enabled (PG4)
fill B-side lower with 1 psi Ar + 2psi Neon 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 Leave DISABLED

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

Torvac setup: Gate valves open

ICRF: All four transmitters in RUN-ON by 08:30. 
ICRF will be allowed to fire after 0.9 sec for at least the first several
shots. Thereafter it depends on how long the LH pulses can be extended. 


Run plan: Start from seg2 of 1050414014, with (D)PCS changes as follows:
Re-compute magnetics predictors because of BP16 coil change in 1051101
Add B21_proj wire and take away Acoil programming on V screen.

I also intend to run using the beta test version of PCS,
and to at least test out the new dpcs_xjump routine to ramp-down runaway
shots early. This is a safety feature for the LH launcher when it is 
moved to its forward position, which is planned for shot#8. 

The in-shot outer gap scan called for in shot 9 and after in the SL_PLAN
will need to have steps in position timed to coincide with the LH pulses 
segments. 

We'll continue to keep track of the C-right cooling. As of Monday it
is clearly slow to cool, but it looks like we should be able to keep up
with the shot cycle. 








Session Leader Summaries
Entered: Mar 8 2006 07:57:17:843AM
Author: John Liptac
The run was very successful! A tremendous amount of coupling data was collected, and unambiguous, non-destructive verification of the LH power affecting the plasma was observed on shots 12 and 16. It was particularly clear on shot 16 where the power was increased to 30kw/klystron. All of the coupling data was gathered at 20kW/klystron.

There was no evidence of arcing. A "mystery cloud" appeared on some of the early shots. It was particularly visible on shot 5, but at a position far from the LH grill on wide2. Playing with the video display settings also revealed a low intensity glow near the antenna. The cloud's intensity diminished on later shots.

Coupler position scan completed, with multiple segments. A maximum of 6 segments were used. The LH phase control was demonstrated and results seemed to track qualitatively with what was expected. There is a very large amount of data to analyze to get quantitative results. A position of 1650 counts produced too low a density to be useful. The antenna was moved toward the plasma gradually and ended at a position of 1615 counts. The level of nl04 density also impacted the density at the grill significantly.

6 segments were used, keeping the phase constant, while the outer gap was increased by 2 mm between each segment in order to lower the density. The position of the antenna was left at 1615 for these experiments. Ideal phase coupling was studied using 90,120,150, and 180 degree phasing. The data errors at the end of the run prevented 60 degree phasing from being tested.



Physics Operator Summaries
Entered: Mar 7 2006 05:49:58:820PM
Author: Steve Wolfe
A pretty successful run. John got almost all of his run plan completed.
Unforutunately, we ended up with a no TF shot and two data system 
malfunctions that prevented his getting the last point on the scan of
the mechanical phase shifter. 

I tried the "soft ramp-down" adaptive programming with the dpcs_xjump
routine on shot 4, which in fact had runaways and was ramped down 
successfully. However, the way this jump-ahead sequence was implemented
it had the effect of shifting the gate for the dpcs_fizzle routine, 
which meant that the oh's were not reverse swung after the plasma 
termination at 0.5sec. This resulted in high coax resistance on the 
subsequent shot. The dpcs_xjump routine was turned off for the remainder
of the run, and the technique will need to be modified before we try
it again. See my notes in the logbook for shot#4 for more details.

Many of the shots today had early runaways, but these tended to go away 
during the pulse. Nevertheless, the presence of a tail was apparently 
a factor in some of the LH results. In particular, on shots with
an early runaway component we often saw what may have been a synchrotron
(visible) beam on some of the visible light diagnostics, including the
video views. See comments in the logbook by Reinke, Marmar, and others.

In trying to get rid of the runaways, I ended up with pg4 puffs as long
as 47msec. Later in the day I backed this down to 40msec. The two fizzles
we had were probably caused by too much pre-puff.

The B-side main (PG3) Piezo was not opening reliably today. It seemed
to take about .8sec at high voltage before it actually started passing gas.
Since the target density was quite low this was not noticed until fairly
late in the day. We experienced something similar when we ran in 
February. Tommy is going to zap the valve with a power supply after
the run. 

The only operator action required for today's experiment, aside from 
trying to get rid of runaways and avoid fizzles, was programming a 
step-wise gap scan during the pulse. This took a while to draw, but
seemed to do the right thing. Examples (all alike) are shots #15,17,19,21.

The main worry today was a high joint resistance on TF turn #49 (E-left)
bottom-middle tap. This went as high as 10uOhm on the post shot scanner.
The camac instrumentation was moved to include this turn and we ran a 
couple of TF only shots, which showed values in the 3uOhm range, somewhat
higher than any of the other turns being digitized. We put a back porch
on the TF and continued to monitor the resistance through the run. There
was some trend toward lower scanner values after full length shots, 
higher (up to 10.3uOhm) following fizzles. This bears watching. It 
may improve at higher fields, which are planned for tomorrows run. 

Note that the fizzle detector programming may need to be adjusted in 
order to get a back porch after Bob and Val's programmed disruptions
tomorrow. The present programming causes the TF to ramp down after
about 600msec. 

Scorecard:
----------

          Tests            3
          Duds             1 (no TF)
          No power (DAQ)   2
          Fizzles          2
          Plasmas         16
          ----------      ---
          Shots           24

All plasmas went full length, except for shot#4 which ramped down
early under adaptive program control. No disruptions today. 

But wait until tomorrow!


Session Leader Comments
Mar 7 2006 09:06:05:783AMJohn LiptacLH test shots look ok. Phase on 10 looks a little off ~35 degrees, but should be ok for the tests today.

Begin with coupler at a position of about 5 mm behind limiter ~1650 counts. Actual position behind limiter is unkown as the stops failed. Position will be adjusted according to density at the grill. Three positions are planned ~1650, ~1625, ~1600. The calibration is ~10 counts/mm.
Mar 7 2006 09:26:53:313AM1060307001John LiptacNo power shot to test CAMAC. LH power into dummy loads to check timing. Can check LH camera as well.

Fizzle detector prevented power for LH, timing looks good though. LH camera worked frame at 0.702 s
Mar 7 2006 09:44:31:110AM1060307002John Liptac10 ms LH pulse at 0.7 s. Plasma right out of the gate! 246 kW from IQ detectors. Nothing on the LH camera.
Mar 7 2006 10:11:06:160AM1060307003John Liptac90, 120 degree phasing start time of 0.695 s to better capture on camera. 246 kW from IQ detectors. Nothing on the LH camera. phase looks just like test shots, pretty good! Need to move launcher in to get higher density, it's too low 2e17m-3.
Mar 13 2006 01:18:44:760PM1060307004John LiptacLauncher moved to 1629 counts, in a couple of mm to increase the density.

Shot terminated by Steve's x-ray protection.

At this position couplers should be safe ~ 3 mm in, so the probes should be behind the limiter.

Try three segments again.

Nothing on HXR camera of note for the last three shots.
Mar 7 2006 10:38:48:160AM1060307005John LiptacThree segments:
60, 90, 120, degree phasing.
Start time of 0.695 s to better capture on camera.

IQ power: ~237 kW
LH Camera: Nothing on LH camera. Made two flashes corresponding to 2nd and third segments on wide2 0.733 0.767.
Density: 1e18

Mar 7 2006 10:54:09:970AM1060307006John LiptacFour segments:
60, 90, 120, 150 degree phasing.
Start time of 0.695 s to better capture on camera.

IQ power: ~238 kW
LH Camera: Nothing on LH camera. Nothing on wide2
Density: 1e18

Mar 7 2006 11:11:59:440AM1060307007John LiptacFive segments:
60, 90, 120, 150, and 180 degree phasing.
Start time of 0.695 s to better capture on camera.

IQ power: ~238 kW
LH Camera: LH camera has low intensity glow if viewed using high contrast settings. Nothing on wide2
Density: 1e18

looks good.
Mar 7 2006 11:53:28:253AM1060307008John LiptacFive segments:
180, 150, 120, 90, and 60 degree phasing.
Start time of 0.695 s to better capture on camera.

IQ power: ~244 kW
Density: 1e18


Mar 7 2006 12:38:19:680PM1060307009John LiptacLauncher moved in to 1620, 1 mm, to increase density.

Five segments:
60, 90, 120, 150, and 180 degree phasing.
Start time of 0.695 s to better capture on camera.

IQ power: ~238 kW
Density: 2-2.5e18 m-3, a good density. Brambilla code shows best coupling around 3e18 m-3.

Complete phase scan.
Mar 7 2006 12:59:11:087PM1060307012John LiptacLauncher moved in to 1615, 0.5 mm, to increase density.

Six segments:
60, 90, 120, 150, 180, and 90 degree phasing.
Start time of 0.695 s.

IQ power: ~238 kW
Density: 1-2e18, as the nlo4 went down

Maybe evidence of LH on HXR and ECE


Mar 7 2006 01:18:17:600PM1060307013John LiptacTry to increase nl04

Six segments:
60, 90, 120, 150, 180, and 90 degree phasing.
Start time of 0.695 s.

IQ power: ~238 kW
Density: ~2e18
Maybe evidence of LH on HXR again


Mar 7 2006 01:21:57:413PM1060307014John LiptacTry to increase nl04 for coupling

Six segments:
60, 90, 120, 150, 180, and 90 degree phasing.
Start time of 0.695 s.

IQ power: ~238 kW
Density: ~2.5e18, as the nlo4 went down

Signal down on HXR


Mar 7 2006 01:42:15:757PM1060307015John LiptacGap scan, moved 2 mm out between each segment. Should decrease the density for each segment.

Six segments:
90 degree CD phasing.
Start time of 0.695 s.

IQ power: ~238 kW
Density: drops as expected
Signal down on HXR
great result


Mar 7 2006 03:08:39:740PM1060307016John LiptacReload 12 and increase LH power to 30kW/klystron all 90 degree CD phasing. Try to make some x-rays.

Six segments:
90 degree CD phasing.
Start time of 0.695 s.

IQ power: ~390 kW
Density: ~2.5e18 m-3 great for coupling!
HXR and ECE results!

HXR x-rays ~30-40 keV


Mar 7 2006 03:35:27:443PM1060307017John LiptacBack to gap scan, reload 15. Back down to 20kW/klystron to be consistent with previous shots.

Six segments:
180 degree phasing. Start with 180 since this phase should show the biggest change in going from approximate to ideal phasing.
Start time of 0.695 s.

IQ power: ~244 kW
Density: ~1-2e18 m-3 over gap scan
Phase on klystron 10 looks off ~ 47 degrees.

No signal on HXR as expected.
Mar 7 2006 03:51:35:757PM1060307019John LiptacGap scan-

Six segments:
150 degree ideal CD phasing.
Start time of 0.695 s.

Power: 20kW/klystron
IQ power: ~250 kW
Density: 1-2e18 m-3 over gap scan density goes up again for last segment.

No signal on HXR.
Mar 7 2006 04:19:50:883PM1060307021John LiptacGap scan-

Six segments:
120 degree ideal CD phasing.
Start time of 0.695 s.

Power: 20kW/klystron
IQ power: ~250 kW
Density: 1-2e18 m-3 over gap scan still increases for last segment.

Data slow coming back.

Physics Operator Comments
Mar 6 2006 06:00:09:723PMSteve Wolfe
Starting to set up for tomorrow's run. 
I'm going to blow away some of the dpcs setup stuff that 
I was working on today, but I'll put it back when it's ready. 

Bring up a fresh pcs from /usr/local/cmod/codes/pcs3-Beta
Load from 1060210017 (last shot). That's going to wipe out 
my dpcs_xjump in the model (when I load it), but that's ok. 

Leave the startup as is:
Br_0=+.0002 (I may want to change that)
Rcur offset=1600
ic_ef4 = -1350A
pg4=28msec

Import seg2 from 1050414014
  Need to recalc all the magnetics predictors
  It won't import the dpcs because the tree is too old. Ok. 
  All magnetics predictors recalculated (including all three RCUR instances 
   on wire 1)
  Copy B21_PROJ from wire 9 in segment 3 onto wire 9 of segment 2 (was
    bolofeed) Didn't work, got another bolofeed? Ah, I haven't loaded it. 
    I need to specify the scratch area /home/wolfe/pcs_scratch. That got it. 
  Change the title from "1.0 MA SNL fiducial" to "0.7MA SNL LH target"
Build all
   Got timestamp errors on p13, p15 seg2. Touch the points in all-p's
Build all (again)
   Almost forgot- Go to v01-16 screen in seg2 and zero the acoil current
      waveform
Build all (again)
Save as /home/wolfe/pcs_trees/ 1060307000
Load/Clean 
Filling in configuration / 1-NOV-2005 12:26:36.00/
Filling in DPCS configuration /22-DEC-2004 09:58:59.00/
Cleaning PCS tree in /home/wolfe/pcs_scratch
Cleaning DPCS tree in /home/wolfe/pcs_scratch
about to ask alcdata to copy /home/wolfe/pcs_scratch/pcs_model to /cmod/trees/models/pcs/pcs_model
about to ask alcdata to copy /home/wolfe/pcs_scratch/dpcs_model to /cmod/trees/models/dpcs/dpcs_model
Load at  6-Mar-2006 17:59:43.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done


Mar 6 2006 06:12:28:260PMSteve Wolfe
Now to set up the dpcs_xjump procedure again. 
In a fresh IDL
IDL> path = !path
IDL> !path = expand_path('+/usr/local/cmod/codes/dpcs')+':'+path
IDL> .r dpcs_load_procedure
% Compiled module: DPCS_ADD_PROCEDURE_NAME.
% Compiled module: DPCS_LOAD_PROCEDURE.
IDL> mdsopen,'hybrid',-1
IDL> dpcs_load_procedure,'async_pros','dpcs_xjump',/add,/off
% Compiled module: DPCS_SAVE_CTX.
% Compiled module: MDSPUT.
% DPCS_ADD_PROCEDURE_NAME: Added routine name dpcs_xjump to \DPCS::TOP.LOADABLES.PROCEDURES:ASYNC_PROS:PROCEDURE_02
% Compiled module: MDSSETDEFAULT.
% Compiled module: MDSTCL.
% Compiled module: DPCS_RESTORE_CTX.
IDL> parameters = {threshold:1e14,gate:[.12,.6],jump_to:1.6,tavg:.01}
IDL> dpcs_load_procedure,'async_pros','dpcs_xjump',{parameters:parameters}
% Compiled module: SERIALIZE_STRUCT.

So that loads the parameters with some reasonable values; the node is
off, so we'll start without it. 

Now, just to be sure I don't lose this
 Load From -1
 Save as /home/wolfe/pcs_trees 1060307999
So now I should have both of them saved. 


Mar 7 2006 08:02:35:253AMSteve Wolfe
Today we will be testing a new callable procedure in 
DPCS named dpcs_xjump. The purpose of this procedure 
is to provide an early rampdown on runaway shots. This 
is considered a safety measure for the lower hybrid 
launcher, which is thought to be vulnerable to damage 
from runaway electrons when it is in its extreme forward
position. 

Basically, the action of this routine is to test the signal
on the neuts_hards input to DPCS and if this exceeds a 
threshold value to jump ahead to a pre-set time in the
ramp-down part of the shot. 

The routine is presently loaded in the DPCS model tree,
but has not been turned on for shot#1. The node name is
\dpcs::top.loadables.procedures:async_pros:procedure_2
The routine is activated by setting this node ON.
Behavior of the routine is governed by values in the 
tree, which can be loaded using the routine 
/usr/local/cmod/codes/dpcs/dpcs_load_procedure.pro
The present PCS GUI does not provide an interface to this
routine (but it soon will). 

Parameters currently set are:
  threshold = 1e14 n/sec (signal level which triggers a jump) 
  jump_to= 1.6sec (time in sequence to which to jump)
  gate=[0.12,0.6] (time interval during which routine is active)
  tavg=0.01 sec (effective averaging time applied to signal, to avoid 
                 triggering on noise spikes). 

Mar 7 2006 08:38:59:237AMSteve Wolfe
Touch  V_21 in seg_01 and seg_02 to force the
new interface to load the new V names
Load/Noclean 
Filling in configuration / 1-NOV-2005 12:26:36.00/
Filling in DPCS configuration /22-DEC-2004 09:58:59.00/
about to ask alcdata to copy /home/wolfe/pcs_scratch/pcs_model to /cmod/trees/models/pcs/pcs_model
about to ask alcdata to copy /home/wolfe/pcs_scratch/dpcs_model to /cmod/trees/models/dpcs/dpcs_model
Load at  7-Mar-2006 08:37:17.00
Open tree /home/wolfe/pcs_scratch      -1
Open tree done

Ready for shot#1

Mar 7 2006 09:25:53:517AM1060307001Steve Wolfe
shot#1: No power cycle to test things out. Camac had a problem
this morning, and we saw some unexplained flashes and noise during
yesterday's test cycles that need to be checked. 

Prad foil array is going to be offline today due to a CAMAC problem. 
NESoX will not be getting triggered (same problem). 

Got gas. I don't see any gas coming from the short B-main puff,
but maybe it just needs exercizing? 

LH timing is from 0.7 to 0.71sec. TCI feedback signal still
has a 200mV offset, being corrected by baseline_sub. 

I think we're ready to go

Mar 7 2006 09:36:22:597AM1060307002Steve Wolfe
shot#2: Try for plasma. 
Plasma full length. 
nl04=5e19. 
EFIT job hung on cmodws23. 
Ran it manually. Outer gap is 7-10mm. 
Tiny runaway level early, < 3e12

Mar 7 2006 09:50:00:893AM1060307003Steve Wolfe
shot#3: No changes. 
C-right is down to -165 at 10 minutes, waiting for LH
LH will go at .65sec this time
plasma full length
Runaways bigger this time, 2e13 early. Maybe I should turn 
on the xjump routine?

Cell access for LH to move the coupler in. 

Mar 7 2006 10:14:30:597AM1060307004Steve Wolfe
shot#4:Turn on dpcs_xjump. Br looked positive on last shot, 
Bz a little negative, but no change yet. 
Joe has loaded the dpcs_ftae procedure into test_pros, but
has not turned it on yet. 
LH launcher moved in 2mm. 
Runaway shot, and I think xjump worked. 
Xray signal was over 2e14. 
dpcs_xjump turned the plasma off, ramped down by 0.5sec
Turn it back off for next shot, since the LH people can 
still work with runaway shots at this radius. Try to get rid
of them in startup. 

Mar 7 2006 12:17:13:490PM1060307004Steve Wolfe
On shot#4 the dpcs_xjump routine successfully ramped 
the plasma down in response to a high runaway (hard 
Xray) signal. However, the way the routine is implemented,
it resulted in the dpcs_fizzle routine not double-swinging
the oh's after the plasma current ramped down. This was a bad
thing, because the early ramp-down left all the oh currents at
negative (pre-charge direction) values at the end of the
shot. The oh coax resistance was therefore very high on shot#5.

The problem arises because the jump ahead sequence moves all
the synchronous time switches ahead, so that the effective fizzle 
detector gate is also shifted, in this case moving the off time
from 1.8sec to ~0.32sec. Since the plasma didn't get below the 
threshold current for the fizzle detector until after 0.4sec
the dpcs_fizzle routine did not act. 

The right behavior would be to leave the switch times for the
dpcs_fizzle sequence alone, so they stay locked to real time. 
However, this is difficult to accomplish with the present scheme. 

It would perhaps be possible to set things up so that the end 
of plasma came within the shifted fizzle gate, by extending 
the end of the real f.d. gate to 1.9, doing the jump to 1.5sec
and ramping down faster. However, this is a rather dicey proposition, 
in view of the bad consequence for the oh coax resistance if it fails.
The issue is that a "normal" ramp-down, lasting about 500msec after
the jump still needs to activate the fizzle sequence because the 
oh's have not double swung. On a full length shot we don't want 
to activate this sequence after a successful ramp-down because
by then the oh's have reversed naturally. 

Mar 7 2006 10:32:49:597AM1060307005Steve Wolfe
shot#5:Turn off dpcs_xjump. 
Br0 from 0.0002 to 0
pg4 from 28ms to 32ms
ic_ef4 from -1350 to -1340
Add dpcs_ftae to test_pros and turn it on for this shot. 
plasma full length. Still has bad runaways early, up
to 3e14. 

EFIT job hung again, ran it manually. 
Got a coax high resistance warning on this shot. The
dpcs_xjump routine pushed the dpcs_fizzle out too far,
so it didn't double swing. Need to do it a little differnt
when I turn it back on. 

Mar 7 2006 10:52:37:207AM1060307006Steve Wolfe
shot#6: Raise pg4 from 32 to 34msec
plasma full length. No runaways at all <5e11
Late breakdown, ~9msec.
Leave it for next shot, probably will come back 
(runaways will probably come back too).




Mar 7 2006 11:01:03:673AM1060307007Steve Wolfe
shot#7: No pcs changes this shot.
I think I probably turned off Joe's routine again. 
The problem with not doing the load proc in the scratch_dir 
is that whenever I load a new shot it overwrites the model
dpcs. 

For next shot: open the tree in pcs_scratch and add dpcs_ftae
Load it. Should work next shot



Mar 7 2006 11:03:56:800AM1060307007Steve Wolfe
shot#7: No pcs changes this shot.
Plasma full length but runaways are back. 
Breakdown is back at the usual time. 
next shot: increase pg4 from 34 to 36 msec



Mar 7 2006 11:10:51:707AM1060307008Steve Wolfe
shot#8: increase pg4 from 34 to 36 msec
Reload dpcs_ftae routine (again) for Joe.
plasma full length 
hard xrays still bad.
Dpcs_ftae ran that time.

Cell access for lh to move the grill in some more.
next shot: increase pg4 again, still trying to 
get rid of the runaways.



Mar 7 2006 11:34:02:770AM1060307009Steve Wolfe
shot#9: pg4 to 39ms
plasma full length
Runaways are down this time, ~1.5e12

Next shot: no pcs changes

Mar 7 2006 11:41:59:987AM1060307009Steve Wolfe
shot#9: We're getting a high resistance reading on
TF turn 49. We will put a back porch on the TF waveform
and extend the tf invert time to 3.5sec.
Need a cell access to move the tf tap instrumentation


Mar 7 2006 11:46:45:457AM1060307009Steve Wolfe
shot#10: TF only test shot at 50kA. PLC control




Mar 7 2006 12:09:08:503PM1060307010Steve Wolfe
shot#10: TF-only test shot at 50kA. PLC control
LH moved the launcher in for next plasma shot 
(which won't be until shot 11). 

Got a 50kA TF pulse ok. Waiting for tap data



Mar 7 2006 12:32:30:650PM1060307011Steve Wolfe
shot#11: Take another PLC TF-only test at 150kA
Resistance is still a bit higher than the neighbors
but only 10% or so. Scanner reading came down.

Next shot: go for a plasma.  


Mar 7 2006 12:43:08:210PM1060307012Steve Wolfe
shot#12: Try for plasma, only change from 9 is back porch 
on the TF, TF invert to 3.5sec, TF fizz del at 1000ms.
PG4 still at 39msec

plasma full length. No runaways (<5e11)
bd a little late (7ms), but pretty good current rise.
I don't see the back porch? I put it in Seg1 !*?

Mar 7 2006 12:57:37:570PM1060307013Steve Wolfe
shot#13: This time put the back porch in seg2 where it belongs. 
Raise nl04 from 6e19 to 7e19 (density has fallen a little;
probably should have given it a little maintenance gas instead)
plasma full length. Had an early runaway burst this time,
up to 2e14.
Got our back porch this time.
next shot: add 2ms to pg4 again

Mar 7 2006 01:10:41:710PM1060307014Steve Wolfe
shot#14: pg4 form 39 to 41ms
Add 30V pg3 pulse in seg2
plasma full length. Runaways are a little better,
still over 2e13. Several bounces in startup,
bd a little late, 7.5ms

next: raise pg4 again, hope it doesn't fizzle.

Mar 7 2006 01:41:04:240PM1060307015Steve Wolfe
shot#15: pg4 form 41 to 43ms
Program a series of steps in RCUR for a gap scan
RCUR = .661 (where it was), .660,.659,.658,.657,.656
at .695-.705,.72-.735,.75-.765,.78-.795,.81-.825,.84-855
Also raised pg3 to 40V during seg2

plasma full length, big bounce in startup. Lots of
runaways early
Need to run the fast efit to pick up the steps.

Next shot:reload shot12

Mar 7 2006 01:50:02:053PM1060307016Steve Wolfe
Shot 16: Import seg2 from shot 12
raise pg4 to 45msec!
Plasma full length. One bounce, runaways down 
but still there ~1e13.
Lost the back porch, put it back for next shot.



Mar 7 2006 02:39:44:680PM1060307017Steve Wolfe
shot#17:Raise pg4 again from 45->47ms
Cell access to move mechanical phase shifter
Back to gap scan: reload shot 15 seg2. That also 
puts back the back porch

plasma full length. No runaways. Breakdown around
7.5msec, nice current rise, no bounce. Got the gap
scan. 

Next shot: another mechanical phase shift change
Mar 7 2006 03:20:17:617PM1060307018Steve Wolfe
Shot#18: No pcs changes. 
Fizzle. Looks like gas may have finally been too high
bd time same as last shot. Fields also look the same.

next shot: drop pg4 by 1ms (just to be doing something)

Mar 7 2006 03:32:55:100PM1060307019Steve Wolfe
Shot#19: pg4 from 47ms
Plasma full length. No runaways. 
No bounces, still a bit late in bd. 

Next: another cell access to change mech. phase shifter.
I'm going to think about tweaking Br

Mar 7 2006 03:44:58:570PM1060307019Steve Wolfe
Shot#19: There is a density rise after the LH pulse 
starting at .83sec. Before that the gas valve is puffing 
full on and the density is not rising. It looks like some
kind of a detachment. Also corresponds to the end of the gap scan.


Mar 7 2006 03:50:48:367PM1060307019Steve Wolfe
Shot#19: Outer strike point is moving up onto the nose
at about the time of the density rise. I don't want to 
change the equilibrium out from under John's scan, but 
I may need to if we're losing density control.
I don't understand how it is taking all that gas and not
getting the density above 5e19/m2

Mar 7 2006 04:01:18:900PM1060307020Steve Wolfe
Shot#20: No pcs changes this time, see what happens.
What happens is a fizzle. 
Next shot: drop pg4, drop Bz 

Mar 7 2006 04:22:48:820PM1060307021Steve Wolfe
Shot#21: Pg4 from 46 to 44ms, ic_ef4 from -1340 to -1350
add a 100V pg3 puff from 100 to 150ms, drop maint level to 30V
plasma full length
But we've got data system problems. 
No runaways. Very slow current rise this time, 
I think it could stand lower pg4 again. 
Still getting the density rise after .83sec
I think the B-side main valve isn't actually opening until
.8sec. The plenum pressure seems not to change until
then. 

Mar 7 2006 04:42:10:350PM1060307022Steve Wolfe
Shot#22: pg4 from 44 to 40ms. 
Extend 100V on pg3 to .25sec to try to get it open.
No TF. 

Mar 7 2006 04:57:12:040PM1060307023Steve Wolfe
Shot#23: pg4 from 44 to 40ms. 
Extend 100V on pg3 to .25sec to try to get it open.
NO Power shot - RF DAQ hung in INIT

Mar 7 2006 05:12:15:273PM1060307024Steve Wolfe
Shot#24: pg4 from 44 to 40ms. 
Extend 100V on pg3 to .25sec to try to get it open.
Another data system problem- NO POWER SHOT, no data.

Engineering Operator Comments
ShotTimeTypeStatusComment
109:11:57:970AMTestOkCycle Test Shot
209:26:52:643AMPlasmaOk
309:44:32:050AMPlasmaOk
410:09:43:347AMPlasmaOk
510:22:57:360AMPlasmaOk
610:40:37:143AMPlasmaOk
710:53:22:813AMPlasmaOk
811:05:54:423AMPlasmaOk
911:26:44:487AMPlasmaOk
1012:03:44:520PMTestOkTF 50 KA Test For Mag turn 49
1112:24:06:673PMTestOk150KA TF Magnet Test
1212:37:43:850PMPlasmaOk
1312:53:30:663PMPlasmaOk
1401:06:18:880PMPlasmaOk
1501:30:05:553PMPlasmaOk
1601:45:27:367PMPlasmaOk
1702:34:33:133PMPlasmaOk
1803:15:47:663PMPlasmaOk
1903:28:24:117PMPlasmaOk
2003:58:35:507PMPlasmaOk
2104:13:03:630PMPlasmaOk
2204:39:46:773PMPlasmaBadTF did not fire
2304:51:06:837PMTestBad
2405:04:51:600PMTestBad