| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Matt Reinke |
| Physics operator(s): | Steve Wolfe |
| Engineering operator(s): | Sam Pierson,Bill Parkin,Ed Fitzgerald |
| Engineering Operator Run Comment |
| MP 567: Spatial Distribution and Dynamics of Disruption Mitigation Radia |
| Session Leader Plans |
| Entered: Sep 23 2009 12:47:23:697PM |
| Author: Matt Reinke |
| MP 567: Spatial Distribution and Dynamics of Disruption Mitigation Radiation
SL: Reinke PO: Wolfe/Irby This run plan will cover 1090924 (Wolfe) and the 1st half of 1090925 (Irby). The purpose of this experiment is to investigate what determines the length of time between when the gas jet first interacts with the plasma and when the prompt thermal quench (TQ) occurs. We will do this with a q95 scan (vary Bt) and a thermal stored energy scan. (DAY 1) We will also look at the intensity of the gas-jet radiation by varying the pressure of the gas jet and the gas jet species. (end of DAY 1 + DAY 2) By observing a variety of mitigated disruptions we will also look at the toroidal symmetry of the energy loss during these events as well as neutral and ion transport of the gas jet materials. MACHINE: Have ICRF (2-6 MW) available but not needed for the first few shots. B-SIDE-LOW filled with 2.0 psi Ne VUV gatevalve closed, ECE gatevalve closed NINJA can barter for a puff in the H-mode plasmas, but stay out of my Ohmics! DIAGNOSTICS: AXUV diodes with low gain settings Thomson scattering with control of pulse phasing (TBD during run day) FRCECE and GPC1 (if I can convince Amanda) TCI with burst mode enabled (window TBD) CXRS run in kinetics mode with 6 fibers/camera (3 OUT_PUFF, 3 IW_POL_SIG, 6 POL_BACK) spaced evenly over field of view. IR camera LBO can run in flat top as long as it's gone before the gas jet fires and doesn't disrupt the target (that's MY job) Limiter and Surface TC's HIREXJR for core rotation (Ar provided by the disruption mitigation) Fast magnetics GAS JET: Start with 15:85 Ar:He mixture at 70 bar with 1.5 ms pulse. Have other bottles on hand for species scan (DAY 2). Get GAS JET attached to the fizzle detector. Gas jet will be triggered at 1.2 seconds for all plasmas. RUN PLAN: A: q95 scan at fixed Ip and x2 density points 1) Establish 5.4 T, 1.0 MA, nl04=0.4 target plasma (NO DISRUPTION) and verify proper machine/diagnostic operations. Load basic equilibrium from 1090922007 and increase Ip to 1.0 MA. If locked modes/runaways a problem, increase density to nl04=0.5 (or higher) 2) x2 shots Bt=5.4 T 3) x2 shots Bt=4.1 T 4) x2 shots Bt=6.7 T 5) Increase density to nl04=0.8 [or x2 the density in (A1)] and x2 shots at 6.7 T 6) x2 shots Bt=4.1 T 7) x2 shots Bt=5.4 T B: W_TH scan at fixed Ip, Bt and nl04 by varying ICRF input power 1) Using target equilibrium from (A) increase nl04 to 1.0 and apply 2 MW ICRF power from 0.6-1.3 (ie on past disruption trigger) x2 shots 2) x2 shots w/ ICRF power to 3 MW 3) apply 2.0 psi Ne puff from 0.4-0.6 seconds ICRF from 2.5-4.0 MW from 0.6-0.75 and hold at 4 MW from 0.75-1.3 for x2 shots 4) apply 2.0 psi Ne puff from 0.4-0.6 seconds ICRF from 2.5-MAX MW from 0.6-0.9 and hold at MAX MW from 0.9-1.3 for x2 shots C: Gas Jet Parameter Scan 1) Decrease gas jet pressure shot-to-shot by pumping on the plenum. Choose stable, reproducible target from (A) or (B). Plenum pressure at 55,40,25,10 (bar) with only 1 shot at each pressure. 2) Pump out gas jet plenum and change gas jet species. In order of priority move through the following: (85:15 He/Ne, 85:15 He/Kr, 85:15 D2/Ar, pure He, pure Ar). All at 70 bar and 1.5 ms pulse length. Choose Choose stable, reproducible target from (A) or (B). Judging on remaining time do 1 or more shots for each bottle. 2*) Between changes in gas jet there should be time to run non gas-jet plasmas. Try and finish up MP 564 shots. a) load 1090922030 w/ 1.0 psi Ne puff b) lower/raise Ne pressure and repeat c) load 1090922010 w/ 4.0 psi Ne puff b) lower/raise Ne pressure and repeat |
| Physics Operators Plans |
| Entered: Sep 23 2009 01:25:15:447PM |
| Author: Steve Wolfe |
| PO plan for Thursday, Sep 24, 2009
MP#567 "Spatial distribution and dynamics of disruption mitigation radiation" SL: Reinke PO: wolfe Gas Jet operator: Granetz This run will employ the Gas Jet disruption mitigation system, which means the torus pressure will go high on every shot. Diagnostic gate valves will probably close and need to be reopened every shot. Torus pump-out after each gas jet pulse is expected to increase the inter-shot intervals. ----------------- Engineering Setup ----------------- Overnight ecdc: gas and pressure: D2 at 2e-4 Torr sweep: 44/45/103 cm scan: 20/120 s Run begins at 08:00 and ends at 18:00 **NOTE- Non-standard times** Power systems as on: 1090922007 **Except: TF current limit to 190kA ** Acoil: +Dtop -Dbot -Jtop +Jbot (standard) Hybrid Enabled Gas setup: Fill B-Top with 6 psi D2 Hybrid enabled (PG4) fill B-side lower with 2 psi 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) fill NINJA with 8 psi D2 DISABLED Enable gatevalves and shutters: Z-bolo and HIREX_SR ECE for first shot ONLY! Torvac gatevalve toggle (yes/no): no Boronization(yes/no): no Overnight ECDC (yes/no): YES ICRF(yes/no): yes, but not needed for first few shots LH(yes/no): no DNB(yes/no): NO Cryopump (yes/no): no Vessel temperature: 35/35/35 Gas Jet: start with 15:85 Ar:He mix at 70bar, 1.5ms pulse trigger at 1.2sec every shot except first SL requests interlocking gas jet with fizzle detector (so does the PO) -------------------------------------- Physop plan: ------------- With luck, no startup dpcs changes will be required for recovery from "mitigated" disruptions. Without luck, it's going to be a long day. Following the SL_PLAN sequence: A. q95 (Bt) scan 1) 5.5T, 1MA, nl04=4e19; load from 1090922007 and increase Ip from .8->1MA, decrease nl04 from .8 to .4e20. No gas jet on first shot. If there is a locked mode problem raise nl04 to 0.5e20. Use ECE to evaluate, then DISABLE ECE GATE VALVE BEFORE STEP#2! 2) 2 gas jet disruptions at 5.5T 3) 2 gas jet disruptions at 4.1T (tf=114ka) use normal startup, ramp to field. this looks like q95=3.05, may need to increase B slightly 4) 2 gas jet disruptions at 6.7T (186kA) 5) increase nl04 to 0.8e20 (or twice value in A1), 2 gas jet disruptions @6.7T 6) 2 gas jet disruptions at 4.1T (or whatever was used for #3) 7) 2 gas jet disruptions at 5.5T (could swap 5 and 7 if SL prefers) B. Stored energy scan at fixed B=5.5,I=1.0,nl04=1.0e20 by varying ICRF power No further physop changes on these beyond startup tweaks, hopefully: 1) 2 gas jet disruptions with Prf=2MW (0.6 to 1.3sec) 2) 2 gas jet disruptions Prf=3MW 3) puff 2psi Ne 0.4 to 0.6sec, rf at 2.5-4.0MW from 0.6->0.75sec, hold at 4MW 2 gas jet disruptions 4) repeat with neon and rf ramp from 2.5->MaxRF by 0.9sec, 2 gas jet disruptions C. Gas Jet parameter scan 1) Use target from A or B to be decided. Then no physop changes except startup. Decrease gas jet plenum pressure to 55, 40, 25, 10 bar (one shot each) 2) Gas species scan (Most likely won't get to this on Thursday) These require a cell access for each, and some additional time for pump and purge. If there is enough time may intersperse with 2* to clean up MP#564 (see below) a) 85:15 He/Ne b) 85:15 He/Kr c) 85:15 D2/Ar d) pure He e) pure Ar 2*) Between species changes, sneak in some shots to finish MP#564 For these we will want to re-enable ECE, then disable before next gas jet pulse! a) load 1090922030 (ELMy shape,800kA) reduce B-side lower to 1.0psi Ne, repeat rf waveform from 1090922030 b) lower or raise b-side lower Ne pressure, based on result of (a) and repeat c) load 1090922010 (EDA shape, 800kA) raise b-side lower to 4psi Ne d) raise (or lower) Ne pressure in b-side lower and repeat. 4) |
| Session Leader Summaries |
| Entered: Sep 24 2009 04:32:32:137PM |
| Author: Matt Reinke |
| The day started well but we ran into severe problems after shot 007 and didn't get any more disruption data. Steve did a phenomenal job trying to adjust for the loss of F15 on the fly and I nominate him for PhysOp of the year.
We were able to get shots right after the gas jet fired within 10-12 minute shot cycles which is much better than I had hoped. It took a few shots to get all the diagnostics to view the disruption properly as well as tweak the low-field shape. SHOT 003: q95=3.3, nl04=0.48 SHOT 004: q95=3.3, nl04=0.40 got TCI burst aligned, got CXRS aligned SHOT 006: q95=2.8, nl04=0.43 w/ runaways, got TS frames aligned, got IR camera on SHOT 007: q95=2.7, nl04=0.39 We'll try for more tomorrow! |
| Physics Operator Summaries |
| Entered: Sep 24 2009 06:42:01:200PM |
| Author: Steve Wolfe |
Physics Operator Summary for Thursday, Sept 24, 2009
MP#567 Spatial distribution and Dynamics of Disruption Mitigation
SL: Reinke
PO: wolfe
EO: Pierson, Parkin, Fitzgerald
Gas Jet operator and Flux Loop Zapper (GJOFLZ): Granetz
Only a little bit of the run plan was accomplished today.
After a promising start, we ran into a series of fizzles
and then the F15 partial flux loop got so bad that it was
failing every shot, during the shot if it wasn't bad at the
start, and zapping wasn't fixing it, although the resistance
did go back down after each zap.
I decided we shouldn't keep
running with this loop in the control system, and tried to
implement the software patches on the fly. Unfortunately,
the fix for "f15dead" that I had coded into fluxihh.pro back in
February of 2005, but never used, turned out to have a bug.
After several hours of trying to fix it, I decided to call the run
at about 16:00.
I finally found and fixed the bug, completed software testing based
on input data from today's shots, just before 18:00. I have now put
up the fixed code, and loaded PCS model with recalculated predictors
that can be used for the continuation of this run tomorrow. I expect
that the startup should be ok, although it may be helpful to increase
the RCUR offset in segment 1 from 2000 to 2300 (or not, shouldn't do
much either way).
A summary of what we did get done follows:
The run started almost on (early) time, despite a computer lossage
that caused all prior shots to be invisible, and hence unavailable
for loading. Josh got a workaround on that installed and shot#1
got taken right at 08:00. For unknown reasons, the gas system
malfunctioned, so #1 was a dud, but #2 was a good baseline (no jet
case at 4e19, 1MA), step A.1
We completed the two 5.5T shots (step A.2) on shots#3 and 4, and seemed
to be on our way. However, when we went to the 4.1T case it turned out
that the q95 was too low (~2.6) and it disrupted. The elongation of
the target, originally taken from 1090922007 with increased Ip, was only
1.50. Shot#6 had better elongation but also killer runaways. There
was clear synchrotrom emission on wide2, so #6 may be useful for
Dennis' runaway seeding paper; there was no indication the runaways
survived the MGI. ON shot#7 I increased the pg4 and intended to move
rcur in at 20msec, as I had success with getting rid of runaways
that way on Tuesday, but somehow that change didn't get loaded. Nevertheless
#7 was a nice thermal target at q95=2.7 to 2.8, just what we wanted for A.3,
aside from the fact that f15 was broken during the shot, and I had to re-run
the EFIT without it to get convergence.
Unfortunately, after #7 the run went sour. Shot#8 was a fizzle, So was #9,
and f15 needed another zap, appearing to go bad at commutation. The fact
that the gas jet is not interlocked to the fizzle detector was not helping
things, but may not have been the problem.
After #10 Bob and I tried
to trace down what the cable labeled "TORVAC to Gas JET" was, since
Bob thought that was the fizzle detector, but it went to the TORVAC
PLC near the Lab-side wall. Shot#11 also fizzled. We did 20 minutes
of ECDC and turned off the gas jet for shot#12, which was another fizzle.
Finally, after lowering the pre-puff down to 20msec, we got a full length
plasma on #13, so we tried to startup again (after another zap on f15).
Shot#14 disrupted early, possibly because of loss of control due to
f15 again going bad during the shot. Interestingly, to add to the
troubles, the gas jet, which was supposed to be enabled again, failed
to fire on #14. We never found out why, although some of the gka5
camac highway also failed around this time.
Following #14, I decided it was not a good idea to continue operation
with f15 failing on every shot, despite being zapped. I made an attempt
to implement the fix (see notes under "shot#15" in PHYSICS_OPERATOR and
MHD topics), but gave up when testing
with the test_dpcs2_software routine indicated there was a major
discrepancy with the new fluxihh predictors. I called the run
at around 16:00, and in fact did not demonstrate a satisfactory
fix until about 18:00. It should be ready to run tomorrow, I hope.
Old shots will need to have all the magnetics predictors recalculated
upon loading or importing.
Scorecard:
Tests/nopower 0
Duds 1 (gas fault)
Fizzles 5
Plasmas 8
=========== ==
Shots 14
|
| Session Leader Comments | |||
| Sep 24 2009 08:23:14:510AM | 1090924002 | Matt Reinke | Good target shot, things appear to be working properly. Let's start killing these things.
NEXT: Enable gas jet to fire at 1.2 sec. |
| Sep 24 2009 08:40:13:747AM | 1090924003 | Matt Reinke | Bt=5.5 T, nl04=0.475
DM system fired pre-TQ dt~1.00 ms CXRS kinetics mode missed the plasma, as did the TCI window. |
| Sep 24 2009 10:54:35:103AM | 1090924004 | Matt Reinke | Repeating to try and get the diagnostic windows for TCI, TS and CXRS tuned up. Shot cycle was 30 minutes so I think I'll have to abandon the x2 shots at each point to get coverage.
NO FIZZLE!!! q95=3.3, nl04=0.41 pre-TQ dt=0.99 msec TCI window good, CXRS seeing HUGE signal, TS timing seems a bit off. |
| Sep 24 2009 09:32:42:403AM | 1090924005 | Matt Reinke | Early disruption on this shot. Steve notes that qpsi is very low on this shot, probably caused the disruption. He'll increase elongation on this shot so it's the low edge q and then we'll have to go back to Bt=5.5 and repeat for this shape.
Bob fired even though it was an early disruption. Public flogging will commence after the run. |
| Sep 24 2009 10:52:45:310AM | 1090924006 | Matt Reinke | Lot of runaways but gas jet fired.
q95 at 2.8, nl04=0.42 It looks like there were two TQ-like spikes on the radiation, but that might be do to runaways. NEXT: repeat trying to remove the runaways. |
| Sep 24 2009 10:00:34:400AM | 1090924007 | Matt Reinke | q95=2.7,nl04=3.9
Gas jet fired, but flaky flux loop made the shape a little different than 006. NEXT: Repeat |
| Sep 24 2009 12:08:20:340PM | 1090924013 | Matt Reinke | We got the plasma back! Target looks good.
NEXT: Fire the DM system. |
| Physics Operator Comments | |||
| Sep 23 2009 04:26:46:993PM | Steve Wolfe | Setting up for 1090924, in case I get in later than usual for this early run. MP#567 Disruption mitigation radiation SL: reinke load from 1090922007 Seg2: change Ip from 800ka to 1MA nl04 from 8 to 4e19; leave point at .26,6.5e19, let it drop down to target Seg1: leave alone for now, note pg4=24msec. If we start getting runaways I may try the rcur tweak from 1090922 instead of raising pg4. Build All Save as /home/wolfe/pcs_scratch 1090924000 looking for.../home/wolfe/pcs_scratch/pcs_model NUM LONG = 3 copying /home/wolfe/pcs_scratch/pcs_model.tree to /home/wolfe/pcs_scratch/pcs_1090924000.tree % Compiled module: MDSTCL. copying /home/wolfe/pcs_scratch/pcs_model.characteristics to /home/wolfe/pcs_scratch/pcs_1090924000.characteristics copying /home/wolfe/pcs_scratch/pcs_model.datafile to /home/wolfe/pcs_scratch/pcs_1090924000.datafile looking for.../home/wolfe/pcs_scratch/dpcs_model NUM LONG = 3 copying /home/wolfe/pcs_scratch/dpcs_model.tree to /home/wolfe/pcs_scratch/dpcs_1090924000.tree copying /home/wolfe/pcs_scratch/dpcs_model.characteristics to /home/wolfe/pcs_scratch/dpcs_1090924000.characteristics copying /home/wolfe/pcs_scratch/dpcs_model.datafile to /home/wolfe/pcs_scratch/dpcs_1090924000.datafile Open tree /home/wolfe/pcs_scratch -1 Open tree done Can load from here tomorrow. Probably want to import seg3 and seg4 from 1090922030 and 1090922007 for part 2*, but leave that until tomorrow. | |
| Sep 24 2009 07:09:29:433AM | Steve Wolfe | As of 07:00 There is a computer problem this morning. Cannot see data from yesterday, last week, last year, whenever, which means we can't load settings for engineering or bring back shots we want to load. I could presumably load the shot I prepared last night which is in my pcs_scratch directory, but don't know where it would go. Apparently I can see the models, since PCS comes up. Josh has been called and is working on the problem. | |
| Sep 24 2009 07:28:30:760AM | Steve Wolfe | Begin set up based on shot stored in ~/pcs_scratch/ shot#1090924000 Load from /home/wolfe/pcs_scratch/*1090924000 Bob is investigating putting the trigger back through the interface box that has the dpcs drive and the f.d. Question: could I draw a 1.5msec pulse on that line? Answer: Not with the draw widget, because of the 1msec granularity! I would have to go into the node in the traverser, and the interface might muck it up if I brought it back up. Not a good option. No load yet, in case this would cause a problem in the models trees. Waiting for Josh | |
| Sep 24 2009 07:48:09:247AM | Steve Wolfe | Josh set up a backup and told us to log out and back in. Seemed to allow access through IDL mdsopen, but scopes still aren't finding shot from 1090922007, for instance. Matt has same problem Sam still having problem setting up cycle, restarting servers. Josh back on it. | |
| Sep 24 2009 07:58:48:620AM | Steve Wolfe | Again, since I didn't load it before logging out last time, bring up pcs and load from pcs_scratch 1090924000 Hold off doing a load_into_model Data from 1090922007 is now visible in a scope Try import seg2 from 1090922030 into seg3 Seems to have worked. Import seg2 from 1090922007 into seg4 Load into Model | |
| Sep 24 2009 07:59:02:713AM | Steve Wolfe | Again, since I didn't load it before logging out last time, bring up pcs and load from pcs_scratch 1090924000 Hold off doing a load_into_model Data from 1090922007 is now visible in a scope Try import seg2 from 1090922030 into seg3 Seems to have worked. Import seg2 from 1090922007 into seg4 Load into Model Load at 24-Sep-2009 07:55:39.00 Open tree /home/wolfe/pcs_scratch -1 Open tree done | |
| Sep 24 2009 08:08:18:933AM | 1090924001 | Steve Wolfe | Shot#1: Attempt at plasma Bob's valves are enabled with no gas in the plenum, trigger is connected to "state box" so fizzle detector should block on fizzle, allow on good plasma; his trigger time is 1.2sec No gas, no breakdown. Power supplies all went. Unfortunately Bob did not have his camac enabled, so wouldn't have fired anyway; null result on semi-test of f.d. interlock Something dumped the gas plena very early, b-main was down to 2.3psi, btop to 1.9psi; b-side lower is still at 1.75psi. |
| Sep 24 2009 08:18:51:353AM | 1090924002 | Steve Wolfe | Shot#2: Attempt at plasma again Bob's valves are enabled with no gas in the plenum, trigger is connected to "state box" so fizzle detector should block on fizzle, allow on good plasma; his trigger time is 1.2sec Try again, watch what happens on gas system. plasma full length, no locked mode, nl04=4.5e19, puffing a little gas until .7sec. Bob reports his valve signal came through, so the state box is not blocking. We'll see if the f.d works when the time comes. Next: repeat this shot with gas jet at 1.2sec (step A.2) |
| Sep 24 2009 08:31:11:120AM | 1090924003 | Steve Wolfe | Shot#3: repeat with gas jet at 1.2sec (step A.2) plasma, jet, boom. Here we go. This one almost lost it in startup though, turns around at 18msec and almost hits inside. Looks like a possible flake? I would up the gas, but maybe not after the disruption. |
| Sep 24 2009 09:02:00:730AM | 1090924004 | Steve Wolfe | Shot#4: increase pg4 from 24 to 26msec (despite MGI disruption) repeat with gas jet at 1.2sec (step A.2) IGT=2.5e-6 at INIT, 28minutes in recool Plasma, jet , again. startup hesitation but ran.Spike at 16msec on zmeter, later on CIII Torus gate valves didn't close this time, IGT went off. Next: 4.1 tesla point (A.3) |
| Sep 24 2009 09:21:58:137AM | 1090924005 | Steve Wolfe | Shot#5: Step A.3: 4.1 tesla, 114ka (seg2 only) leave pg4 alone, test startup robustness (if I were fiddling I'd drop it) IGT=2.4e-6 at INIT, 10+ minutes in recool The difference in pumpout is that the main torus valves were not closed after the gas jet pulse (PLC interlock level changed?) plasma disrupted early, gas jet fired anyway. So much for the f.d. interlock. Gate valves closed again. q95 down to 2.65 at B=4.2T? Probably need to raise the Bt target. F15 needs zapping; Bob will need a cell access. These elongations are considerably less than normal discharges, right around 1.5, so the 5.5T cases were already at q95=3.2. Next: try increasing the elongation to increase volume, q95 |
| Sep 24 2009 09:38:58:650AM | 1090924006 | Steve Wolfe | Shot#6: Step A.3: 4.1 tesla, 114ka (seg2 only) with more elongation Raise zcur from -.015 to 0. ZXU from (+.0.05 dropping to .047 to .0434) to +.035 flat; note ZXL is +.025, so should still be solid LSN Leave Bt at 4.2T, try to pick up this point, then have to re-do 5.5T IGT=2.0e-6 at INIT, 20minutes in recool plasma with saturating runaways, lasted till the gas jet. Elongation only came up to 1.58, q95 is still 2.8. Could stand a little more drop on zxu. Need to snuff the runaways. Sync rad on wide2 at the disruption. Next: raise pg4, and drop rcut at 20msec point. |
| Sep 24 2009 09:55:27:730AM | 1090924007 | Steve Wolfe | Shot#7: Step A.3: 4.1 tesla, 114ka (seg2 only) q95=2.8,B=4.2T raise pg4 from 26 to 28msec, and drop rcur at 20msec point from .63 to .61 Gate valves stayed open this time, so pumpdown is ok Until Irby's gate valve reopened, coming down again now. IGT=2.1e-6 in init, about 12 min cycle plasma to gas jet, runaways almost gone, possible h-mode dithering after 0.9sec F15 has big offset after disruption, Bob back in zap mode. Actually, f15 was bad during the shot, efit not converged. Startup has some hesitation, not terrible. Re-ran efit without flux_14, gets q=2.7 Probably ok, but I'd say repeat to at least get a second shot at this q95 |
| Sep 24 2009 10:05:43:010AM | 1090924008 | Steve Wolfe | Shot#8: Step A.3: 4.1 tesla, 114ka (seg2 only) q95=2.8,B=4.2T No dpcs changes. 1.9e-6, around 12 minutes (pump was ready, we were thinking) fizzle. Jet fired anyway. probably should have dropped pg4 |
| Sep 24 2009 10:20:46:917AM | 1090924009 | Steve Wolfe | Shot#9: Step A.3: tf=114ka (seg2 only) q95=2.8,B=4.2T Pg4 from 28 to 26msec Still trying for the second shot on A.3 IGT=1.1e-6 at INIT, 12:30 in recool (cooling delay) Another fizzle. F15 is broken again. No current at all. pressure still looks high |
| Sep 24 2009 10:45:07:480AM | 1090924010 | Steve Wolfe | Shot#10: Step A.3: tf=114ka (seg2 only) q95=2.8,B=4.2T Pg4 from 26 to 24msec Still trying for the second shot on A.3 IGT=6.2e-7 at INIT, 18:30 in recool while Bob and I were in cell The other input to the gas jet box seems to be the torvac state transition that controls the valves, not an f.d. line; I don't think it's got an interlock. Bob zapped f15 again, it's back to 3.7 ohms, from 12, but this is getting really flakey. I'm still not sure if it was bad at startup on #9, but it would have been during the shot so it's just as well it was a fizz. See what happens this time. We may need to take it out of use. Another fizzle. F15 worked on this one, according to Bob. pressure is down, fields look ok if the magnetics are working. I don't see why it's not going. Next: Put the rcur point at 20msec back to .63, it's the only other change aside from the fact that the jet keeps pulsing. |
| Sep 24 2009 11:03:20:980AM | 1090924011 | Steve Wolfe | Shot#11: Step A.3: tf=114ka (seg2 only) q95=2.8,B=4.2T I was going to re-do the rcur tweak I made back on shot#7, but it's already not there? I thought I had changed the value at .02 from .63 to .61, but it's still at .63. Must not have actually changed it? Opening the rga, doesn't look like there is a leak. Import seg1 from 1090924007, see what I had or hadn't done. It does have rcur at .020 at .63, so I must not have changed it Drop pg4 to 25msec from 28 and go again, close imp inj gate valve Another fizzle Something's wrong. Set up for 20minutes of ecdc when pressure comes down, turn off gas jet until we get a plasma back. |
| Sep 24 2009 11:50:37:090AM | 1090924012 | Steve Wolfe | Shot#12: After 20 minutes of D2 ecdc pg4 from 25 to 22msec. No gas jet this time. RGA trend scope is showing a burst of 28 and 32, levels an order higher than when we started ecdc. 32 is probably diborane, 14/28 ratio is .25 IGT 1.7e-6 at init, over 30 minutes in recool. fizzle again! hardly any plasma current, still looks like too much fill, very slow burnout, moving in (to the extent rcur means anything) F15 still ok. Next: drop pg4 to 20msec, probably get a dud! |
| Sep 24 2009 12:08:10:900PM | 1090924013 | Steve Wolfe | Shot#13: drop pg4 to 20msec, probably get a dud! plasma full length. F15 is bad again, 3e-3 offset. Zap it again. Elevated chisq in efit, so it happens during the shot. This loop looks like it's on its way out. still looks like there is more gas than we like. Leave pg4 where it is and hope. Gas jet back on for next shot. |
| Sep 24 2009 12:25:38:527PM | 1090924014 | Steve Wolfe | Shot#14: Gas jet back on for this shot. no dpcs changes plasma to .6 sec, disrupted early, no gas jet (Bob didn't enable gas jet apparently.) F15 has offset of 1.5e-2. has bad values during the shot. I think we can't run with this loop in the control. Next: Cell access, will try to reconfigure to run without F15. At least one hour delay, maybe closer to 18 hours. |
| Sep 24 2009 02:14:52:213PM | 1090924014 | Steve Wolfe | Working on dead f15 partial workaround. In PCS, load from -1 : It does bring up the config warning. Ack it. Start recalculating. I'll leave all the offsets as they were (RCUR=2000, BR_0=.0005) since I don't know how to do better. Call mag predictors in all four segments. Test by build_segment in each (not sure if this is sufficient). Test by build_all; no warnings or errors. Examine the predictors be sure each class has the right weights zeroed Looks ok, the main ones have f23,f25,f15 off, f14 on. Vacuum also has f27 and some other symmetical ones off. Load it! Load at 24-Sep-2009 14:14:41.00 Open tree /home/wolfe/pcs_scratch -1 Open tree done |
| Sep 24 2009 02:32:27:167PM | 1090924015 | Steve Wolfe | Working on the workaround for dead f15 partial flux loop. I have the model loaded with new coefficients. While waiting for people to get out of cell and power room, take test shot#201 in test_dpcs2_software environment, with source from 1090924014 tp check on change in offsets, etc. And it looks like there is a problem, the rcur comes out crazy, error at t=0 is over 40000. Something is not right. Zcur looks ok, BR_0 looks ok, but the f15dead switch doesn't seem to be working in fluxihh. |
| Sep 24 2009 03:58:33:980PM | 1090924015 | Steve Wolfe | There is no way I'm going to figure out how to fix the fluxihh software and get it tested by 18:00. The run is over. I'll try to get something working by tomorrow morning, but I don't guarantee it. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 08:00:23:323AM | Plasma | Ok | |
| 2 | 08:12:55:480AM | Plasma | Ok | |
| 3 | 08:25:46:137AM | Plasma | Ok | |
| 4 | 08:56:21:450AM | Plasma | Ok | |
| 5 | 09:09:14:730AM | Plasma | Ok | |
| 6 | 09:32:06:370AM | Plasma | Ok | |
| 7 | 09:45:32:247AM | Plasma | Ok | |
| 8 | 10:00:11:747AM | Plasma | Ok | |
| 9 | 10:14:57:933AM | Plasma | Ok | |
| 10 | 10:38:34:213AM | Plasma | Ok | |
| 11 | 10:58:54:010AM | Plasma | Ok | |
| 12 | 11:44:00:840AM | Plasma | Ok | |
| 13 | 12:00:53:683PM | Plasma | Ok | |
| 14 | 12:15:16:040PM | Plasma | Ok | |
| System Availability | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Sep 24 2009 07:59:54:540AM | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Sep 24 2009 09:31:46:497AM | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||