| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Jerry Hughes |
| Physics operator(s): | Steve Wolfe |
| Engineering operator(s): | Bill Cochran,Sam Pierson,Gary Dekow |
| Engineering Operator Run Comment |
| MP#614 Pedestal modifications with lower hybrid in EDA H-mode |
| Session Leader Plans |
| Entered: Dec 2 2010 09:23:58:620AM |
| Author: Jerry Hughes |
| Run plan for 1101202
MP 614 Overnight boronization requested, followed by He ECDC. Use recipe from the 29 Nov 2010 boronization. Special gas request: Fill B-side lower with 1 psi Ar Hybrid DISABLED (PG1) Fill H-bottom with 20 psi N2 Hybrid DISABLED (PG5) ICRF: all transmitters available LH: make available, but not on first shot Cryopump: yes, available on first shot. DNB: yes Requested diagnostics: TS, ECE, TCI, HXR camera, div probes, bolos, Ly-a, reflectometry, fast magnetics, Edge CXRS, SOL reflectomery. Special request for LH group: spectrum analyzer connected to probe on LH grill Starting shot: Load from 1100804027. 5.4T, 600kA, target nl04=6x1019m-2. Pumped. This shot is near DN with a ZCUR ramp programmed in around 0.7--0.8s, in order to shift from LSN to USN. Set H-bottom to chop and program a full-height PG5 pulse from 0.4 to 1.4s. Pulse PG1 from 0.30 to 0.35s for diagnostic Ar. (1) Begin boronization recovery. Allow ICRF to exercise antennas from 0.6 to 1.5s in each shot until >2MW is cleanly obtained from some combination of J+E. Exercise D also. Lower N2 pulse steadily shot to shot and confirm steady EDA H-mode can be maintained. Attempt to proceed with zero N2 puffing, although if heat loading or injections begin to look problematic, seed some more. Also, if the switch to USN is problematic, remove the ZCUR ramp for this step so that it stays slightly LSN. NO LOWER HYBRID YET! (2) Attempt to reproduce 1100804027, reloading if necessary. Prf=1.7MW from 0.6 to 1.5s. Plh=0.6MW at n_||=1.9 from 0.9 to 1.4s. If there is trouble keeping a steady H-mode, raise the RF power. (3) Run case obtained in (2) with cryopump WARM. Compare results. Evaluate whether to make the pump cold again. (4) Taking a good case from earlier, remove the ZCUR ramp in flattop. That is, repeat shot in a slightly LSN configuration. If results are different, then go back to base case and institute a ZCUR ramp back down in order to see whether there is a clear minimum SSEP for this to occur. If the results merit it, get a comparison with and without cryopumping. (5) Launch LH power from only the central sets of columns, at maximum level feasible, using the same n_||. This configuration should allow phase change on demand. Take comparison shot with counter-Ip phasing, and then one with heating phasing. (6) ICRF power scan. Take a well-established case from earlier steps and increase power. At 1.1s, step ICRF power up by 0.5MW, and at 1.25s, step up again by 0.5MW (or to max available). It is okay to try D-port out. Next, run a case in which ICRF is stepped down by 0.5MW at 1.1s, and then again at 1.25s. Then, to establish a control, rerun these shots with Plh=0. (7) Taking a favorable case from earlier steps, increase Ip by 0.1MA shot by shot until either pedestal modification is no longer observed, or we can no longer get steady H-modes. If plasmas go ELM-free, try putting in LH before ICRF, or attempt to ramp Ip up during an already established H-mode. |
| Physics Operators Plans |
| Entered: Dec 1 2010 05:22:13:547PM |
| Author: Steve Wolfe |
| Physics Operator plan and engineering setup for
Thursday, 2-Dec-2010 MP#614 Pedestal modifications with lower hybrid in EDA H-mode SL: Hughes PO: wolfe Overnight boronization ------------------------------ Boronization Parameters as for 1101130 ------------------------------ dP=120 psi ** optimized sweep parameters ** rmin/rdwell/rmax: 44/75/80 cm tdwell/tsweep: 60/120 s Zero EF4 current Pulser 0.7 on 0.3 off ---------------------------------- Post-boronization ECDC parameters: ---------------------------------- * Note: using He only * ---------------------- 1st Gas Setup Gas: He sweep: 44/45/103 cm scan: 20/120 s Pressure: 5e-5 Torr Duration: 3 hrs ********************************************************* Note: If there are problems with helium ecdc staying on with these parameters, I suggest the following steps: 1) Try either purging and refilling the B-top plenum with Helium or running directly off the Helium bottle 2) Try raising the pressure set point to 1e-4 Torr 3) Try changing sweep to avoid going out on inner wall, such as 45/46/103 cm, or to avoid going out at outer limit, such as changing max value from 103 to 100cm ********************************************************* ----------------- Engineering Setup ----------------- Run begins at 09:00 and ends at 17:00 Power systems as on: 1100804027 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 1 psi Ar Hybrid Enabled (PG1) leave B-side upper as is Hybrid DISABLED (PG2) fill B-main (C-side) with 40 psi D2 Hybrid enabled (PG3) fill H-bottom with 20psi N2 Hybrid Enabled (PG5) fill NINJA with 10 psi D2 DISABLED Enable gate-valves and shutters: ECE, VUV, HiREX Sr, Xeus Leave z-bolo shutter as is (should be open) Torvac gatevalve toggle (yes/no): no Boronization(yes/no): YES Overnight ECDC (yes/no): YES ICRF(yes/no): YES LH(yes/no): YES Cryopump (yes/no): YES DNB (yes/no): YES Vessel temperature: 45/45/45 **************************************************** Physics Operator Plan Load a recent successful seg#1, e.g. from 1101201021 Load seg3 from 1100804027 (600kA, nl04=6e19, ZCUR sweep from -.01 to +.015, clearin switch at 1.15sec) Per SL request, add "h-bot" to dpcs_piezo_chopper outputs and program 100V pulse from .4 to 1.4sec; I expect this to be too much N2, bur we'll reduce it later. Also, recalculate NL04 predictor and factor, possibly change hybrid_input_80:p_to_v_expr depending on Irby's examination of the fringe counter output. Probably also modify the NL04 gains, which were set at 8 or 9 and need to be much lower, and have some D gain, with the cryopump. Note that there is a feedforward pg3 puff until 0.6sec. The reference shot disrupted in rampdown. PO's notes indicate that a change in RXU programming after 1.38sec (shot 1100804030) helped get into rampdown, so I'll do that; actually the notes say ZXU, but it seems to mean RXU. Follow SL's plan until he comes to his senses or we run out of time: 1) Re-condition ICRF and otherwise recover from boronization. Once we're getting Prf>2MW, reduce N2 puff. Possibly remove ZCUR ramp, staying in LSN for the whole shot. The idea of this step is to get through boronization recovery, and maybe do some exploration and scoping for later while we do so. 2) Turn on LH and attempt to reproduce 1100804027: Prf 1.7MW from 0.6 to 1.5 Plh=0.6MW, N||=1.9; tune ICRF power to maintain steady H-mode. This means we reload from the earlier target, with no N2 puff and replacing the ZCUR sweep if it got removed during step 1. 3) Turn off cryopump and compare results. Change pg4 puff for non-cryopump startup, change NL04 gains, etc. Decide if this was a good idea, and if not, go back to cryopump 4) Now remove the ZCUR ramp to stay LSN, with LH and everything. Compare USN and LSN cases, with and without cryopumping. 5) Change LH to use only single klystron (central) columns. Run co- and counter-CD phasing, and heating phase. No PCS changes unless required to maintain equilibrium. 6) Scan ICRF power using staircase waveform (up and down). Compare with and without LH. No PCS changes unless required to maintain equilibrium. 7) Increase Ip shot to shot in small increments (100kA). Play with RF and/or LH timing. |
| Session Leader Summaries |
| Entered: Dec 3 2010 04:21:24:893PM |
| Author: Jerry Hughes |
| We made some progress today on MP614, which addresses the modification of
the EDA H-mode pedestal with LH waves. Mostly, though, we had to deal with the challenge of operating LH into an ICRF-heated H-mode. Many of our most promising discharges were ruined by injections which either diminished the density pump-out effect or killed the H-modes altogether. Mo injections were common with ICRF, an W injections were induced by LH, fairly regularly. Also, a large (>=2cm) outer gap was required in order to prevent overheating of the LH grill. (Perhaps this would be ameliorated by running at higher Ip, but we'll have to find that out on another day.) All discharges were at 5.4T, 600kA. LH n_|| was 1.9 in all cases, with P_LH ranging from 400kW to 600kW. We were able to check for density pumpout (and Te increase) with LH in the following four configurations: USN, pumped: yes, strong. (019) USN, unpumped: yes, not as strong. (018) LSN, pumped: yes, apparently just as good as in USN, pumped (030) LSN, unpumped: non-existent or very weak. (015) N2 seeding was used early, in the hope that boronization recovery would go as smoothly as it did on Tuesday's run. Recovery was satisfactory, and N2 residence on shots after we turned off the puff, while finite, did not negatively impact the experiment. Still, it might be better to use Ne seeding for this purpose in the future. |
| Physics Operator Summaries |
| Entered: Dec 2 2010 05:54:19:797PM |
| Author: Steve Wolfe |
Physics Operator Summary for Thursday, Dec 2, 2010
MP#614 LH effect on H-mode pedestal
SL: Hughes
PO: wolfe
EO: Cochran, Pierson, Dekow
The machine ran pretty well today, but the plasmas were
not fully cooperative with the goals of the run.
The start of the run was delayed due to a computer meltdown
(alchome was offline), and the first shot was a no-power test
to verify that all systems were back online. After that, the
data system seemed to work without fault for the rest of the day.
Boronization recovery using pumped marginal USN discharges
seeded with N2 from H-bottom went fairly well, although we
seemed to be getting more N2 in the discharge than expected
and steadily reduced the puff from shot#2 through 6, and turned it
off for good (or ill) at shot#7.
LH started operating on shot#7, and had some difficulties with
getting a satisfactory gap. On shot#9 we successfully
reproduced the density pumpout effect previously observed in USN, with
the cryopump operating. However, we were not succssful in getting
a case with cryopump off. The first attempts had problems with
W injections and we moved on to LSN only cases (cryo still off) at
shot#12, which was a beautiful steady EDA but had no LH. Shots@13-15
were unsuccessful attempts to get the LH into this target without
injections.
Shot#16-18 were back to Upper Null (after a Zcur sweep) with cryopump
still off. We continued to have difficulty with LH matching and
injections because of unsteady outer gap. These difficulties continued
with the cryopump back on for shot#19-23. Part of the difficulty seemed
to be associated with the zcur sweep up to the USN, so SL made the
decision (not in the plan) to start with USN and raise the ICRF power
to get into H-mode without a LSN phase. These shots (@24-26) were not
much more successful in terms of the MP objectives.
The last 3 shots were back to LSN for the whole discharge, keeping
the cryopump on. These actually produced some interesting relevant
results, with ICRF and LH staying on and reasonable length Hmodes,
though these became more ELM-freeish as the density came down.
We had a few early disruptions, including one runaway shutdown following
an injection in during startup.
Some late flattop and early rampdown disruptions were due to vertical
instability. Power systems behaved well, except that Bill is still
trying to track down the EF2U voltage ripple problem, which is present
on most shots today. The OH1 self-power circuit pulled breakers on
a couple of the disruptions, but no fuse loss was found.
Startup was pretty reliable, with only a couple of shots with early
runaways. Aside from changing PG4 between 50 and 22msec for cryo/non-cryo
conditions, I left it pretty much alone. Midday I added a little
to the -20msec PG3 peak. Late in the day I put
10A on EF4 to counter a late breakdown, then put it back when the
next shot bounced. The base pressure and N2 level on the RGA went pretty
high after we turned off the cryopump, but the shots kept on coming.
Scorecard:
No power/tests 1
Duds 0
Fizzles 0
Plasmas 29 including on xjump kill and 3 flattop disruptions
======== ==
Shots 30
|
| Session Leader Comments | |||
| Dec 2 2010 10:27:36:407AM | 1101202001 | Jerry Hughes | Shot 1: No power test. |
| Dec 2 2010 10:28:47:287AM | 1101202001 | Jerry Hughes | Shot 1: No power test. |
| Dec 2 2010 10:29:36:617AM | 1101202002 | Jerry Hughes | Shot 2:
Attempt a real plasma, with N2 puff, ICRF. No LH. Result: Disruption at 0.5s. |
| Dec 2 2010 10:31:10:143AM | 1101202003 | Jerry Hughes | Shot 3
Attempt: Repeat with delayed (to 0.6s) and reduced (to 50%) N2 puff. Result: Injections during initial ICRF turn-on. N2 probably doesn't reach plasma until sometime after 0.7s. Likely there's too much N2 for this level of input power. Also, it should probably come in earlier. Next: Reduce N2 chop to 40% and change start time of to 0.5s. |
| Dec 2 2010 10:31:32:950AM | 1101202004 | Jerry Hughes | Shot 4:
Attempt: Retry with 40% N2 chop, beginning at 0.5s. Result: H-mode at 0.666s, but short-lived. RF is ratty. There are some injections, but nothing really horrible. Next: Move start time of N2 to 0.45s. |
| Dec 2 2010 10:43:23:947AM | 1101202005 | Jerry Hughes | Shot 5:
Attempt: Retry with 40% N2 chop, beginning at 0.45s. Result: RF startup is clean. Judging by the edge Te trace, the N2 is making it in earlier. H-mode until D and E fault after 0.8s. H-mode is never recovered, perhaps because the plasma has gone USN at the end. Next: Yijun will try for 2MW flat from E+J, with a little extra from D later in flattop. Also, reduce N2 chop to 30%. |
| Dec 2 2010 10:58:34:890AM | 1101202006 | Jerry Hughes | Shot 6:
Attempt: 2MW demand from E+J. N2 chop is now very low. Result: Goes into H-mode easily, but isn't maintained through RF faults. H-mode doesn't come back in the USN phase, either. Next: Repeat with N2 turned off. Have LH pulse from 0.9 to 1.4. |
| Dec 2 2010 11:13:24:143AM | 1101202007 | Jerry Hughes | Shot 7.
Attempt: Establish a base case EDA with LH, as on 1100804027. Result: Peak ICRF and LH are reasonable. There are some faults. LH is perhaps reducing the H-mode density, but because of all the faults and multiple H-mode cycles, it is slightly ambiguous. Next: Retry |
| Dec 2 2010 11:38:21:847AM | 1101202008 | Jerry Hughes | Shot 8:
Retry. PO is increasing right gap slightly at LH group's request. Result: Aborted runaway shot. |
| Dec 2 2010 12:01:06:703PM | 1101202009 | Jerry Hughes | Shot 9:
Retry shot 7. Result: This is more like it. Good EDA H-mode from ~0.9 to 1.5s. LH does appear to lower the density, although not necessarily to the level that was reached on 1100804027. Prad is higher than on target shot, so maybe that is the reason. There is probably some lingering nitrogen. [Note that I am using 2.0MW instead of 1.7MW of ICRF here, as planned. I suspect that if I lowered the RF demand, however, I might have problems sustaining H-mode, due to the higher Prad. ] There is a nice QCM signature on BP4T_ABK. Lab frame freq drops from 90kHz to 50kHz. Next: Turn off cryopump and see what happens. |
| Dec 2 2010 12:11:18:013PM | 1101202010 | Jerry Hughes | Shot 10:
Attempt: Try Shot 9 without cryopump. Result: Big W injection right after LH comes on. Prior to that, things were moving along fairly well. Next: Delay onset of LH to 1.1s. |
| Dec 2 2010 12:30:02:020PM | 1101202011 | Jerry Hughes | Shot 11:
Attempt: Retry shot 9 without cryopump, and with LH start time delayed to 1.1s. Result: Unpleasant injections prevent the H-mode from persisting until the LH start time. The later inections may not have happened without the SSEP sweep. For the next shot, let's leave it lower null and leave the pump warm. Basically skipping ahead to step (4). |
| Dec 2 2010 12:40:59:170PM | 1101202012 | Jerry Hughes | Shot 12:
Attempt: LSN whole time. No pumping. Asking for LH from 0.9 to 1.4s. Result: The most beautiful EDA H-mode so far. Perhaps because there was no LH. Upside: The lower null, unpumped control shot has been obtained! Next: put LH into this. |
| Dec 2 2010 01:02:39:283PM | 1101202013 | Jerry Hughes | Shot 13.
Attempt: Retry shot 12 with LH. Result: Big W injection shortly after LH turn-on. Plasma disrupts at 1s. Next: Try to move RXL in and see if it helps. |
| Dec 2 2010 01:18:05:250PM | 1101202014 | Jerry Hughes | Shot 14.
Attempt: Retry with RXL moved inboard slightly. Result: Injection at 0.8s, leading to higher Prad which we never recover from. LH is firing into a fairly dirty H-mode. The event at 0.8s might be associated with the strike point being on the floor now. More W coming in during LH pulse, according to MLR. Next: PO will re-adjust RXL to try to keep the strike off the floor. Also asking LH to reduce demand to 400kW. |
| Dec 2 2010 01:31:47:023PM | 1101202015 | Jerry Hughes | Shot 15.
Retry with adjusted srike point and less LH power. Result: More nasties, during LH pulse. First injection is after 0.97s, 70ms after LH turn-on. No evidence of density reduction in the period of 0.9--0.97s. This period is probably long enough to have observed some pump-out, if it was going to happen. Next: Try USN again. |
| Dec 2 2010 01:46:24:437PM | 1101202016 | Jerry Hughes | Shot 16:
Attempt: Put ZCUR ramp in to get the plasma USN at 0.8s. Remain unpumped. LH demand same as on previous shot. Reverting to step (3) Result: Shot looks good up to 1.04s. Achieve steady EDA with SSEP=+5mm by 0.9s. LH on steady from 0.9 to 1.04s, then faulting. Before the faults, density reduction is there, and even persists a bit until LH is turned off at 1.4s. Next: Retry. Steve will try to get the right gap under control. |
| Dec 2 2010 02:01:19:327PM | 1101202017 | Jerry Hughes | Shot 17.
Attempt: Retry 016 with fix to right gap, at Wallace's request. Result: Gap still too small for the grill's liking. There's Fe coming in and radiating according to MLR. It's not clear the density has settled down before the LH turns on initially, so on next shot we'll delay the LH turn-on. Next: larger right gap and LH starts at 1.1s. |
| Dec 2 2010 02:15:04:230PM | 1101202018 | Jerry Hughes | Shot 18:
Attempt: Get a clean density pump-out in USN, unpumped. Result: Good shot. Density reduction is there. Counter-Ip velocity reduction from Hirex! Next: Retry with pump on, see if there's a difference in the density. |
| Dec 2 2010 02:32:36:150PM | 1101202019 | Jerry Hughes | Shot 19:
Attempt: Try Shot 18 with the pump on. Result: RF trips before LH turn-on, resulting in H-L transition. There is a slight uptick in Prad at 1.025s. Starting H-mode density is lower than in 18, presumably because pump is doing something. Next: Try for a repeat with cleaner RF. |
| Dec 2 2010 02:49:56:187PM | 1101202020 | Jerry Hughes | Shot 20
Attempt: Retry previous shot Result: Getting down to pretty low density now. |
| Dec 2 2010 03:01:50:953PM | 1101202021 | Jerry Hughes | Shot 21:
Attempt: Put in power steps in ICRF and two 150ms of LH. Results: Disrupts soon after H-mode formation. |
| Dec 2 2010 03:15:45:330PM | 1101202022 | Jerry Hughes | Shot 22:
Attempt: Retry shot 21 Result: Rf fault causes loss of H-L. And faults prevent reformation of another steady H-mode. Next: Retry |
| Dec 2 2010 03:32:11:700PM | 1101202023 | Jerry Hughes | Shot 23:
Attempt: Retry with improved gap for RF coupling. Result: Prad not well under control in the early H-mode, leading to some back-transitions. But, got in LH late in flattop with 2.5MW ICRF. There seems to be some pump-out. Next: Since LH seems content with 2.5MW in this condition, let's just set the ICRF power there to begin with and make the H-mode in USN. |
| Dec 2 2010 04:02:28:293PM | 1101202024 | Jerry Hughes | Shot 24:
Attempt: Stay USN the whole flattop. Picrf=2.5MW, stepping to 3.0MW at 1.2s Two LH pulses 150ms in width. Result: ICRF performs fairly well, but there is some radiation picking up, which kills the H-mode, and the LH doesn't have a good target to go into. 2nd H-mode has some combination of ICRF and LH, and nebar gets down to the minimum 1.6e20. |
| Dec 2 2010 04:05:36:427PM | 1101202025 | Jerry Hughes | Shot 25:
Attempt: Earlier, longer LH pulse. Result: For some reason, 2.5 MW of ICRF wasn't enought o get us into H-mode this time. Next: Try at higher power. |
| Dec 2 2010 04:14:08:550PM | 1101202026 | Jerry Hughes | Shot 26.
Attempt: 3MW ICRF. Result: Still stays out of H-mode until 1.2s. GH limiter appears to be taking a lot of heat. Next: Raise target density to 8e19. |
| Dec 2 2010 04:32:18:620PM | 1101202027 | Jerry Hughes | Shot 27
Attempt: Retry with higher target density Result: Still not so good. Disruption during H-mode, due to big impurity influx. |
| Dec 2 2010 04:43:29:683PM | 1101202028 | Jerry Hughes |
Shot 28. Back to the lower null case of 015, but this time pumped. Result: Not too bad. First H-mode was killed by a big injection. Recovered H-mode is at lower density. Small ELMs appear to be present. Next: Try to get continuous H-mode and unambiguous pedestal modification. Also raise RF demand at 1.1s |
| Dec 2 2010 04:52:13:067PM | 1101202029 | Jerry Hughes | Shot 29:
Result: Shot started out well, with steady EDA H-mode. LH turn-on at 0.9s briefly results in Te rise, SMALL nl04 drop, before Prad starts to rise, and we end up with a fairly ELM-free plasma. |
| Dec 2 2010 05:19:25:147PM | 1101202030 | Jerry Hughes | Shot 30.
Attempt: Retry shot 29 with 400kW LH power. Result: Cut off the plot at 1.04s and this is an awesome shot. Prf steady at 2MW. LH turns on at 0.9s and is at 400kW by 0.95s. Prompt drop in density, Prad observed, with an accompanying buildup in Te,ped. nebar reaches final value of 1.6e20, just as in the USN case. Pedestal reaches several hundred eV before Prad starts to rebuild. After 1.1s, the Prad rises and the density suppression goes partly away. Magnetics (BP4T_ABK) show a strong increase in QCM mode strength when lower hybrid turns on! |
| Physics Operator Comments | |||
| Dec 1 2010 03:13:27:597PM | Steve Wolfe |
Setting up for Thursday's run 1101202
Load from 1101201021
Import seg2 from 1100804027,Seg3
Note seg 1 settings:
RCUR offset 3500
BR_0 offset 0.000
IC_EF4U = -1295
PG4 50msec
PF3 peak 70V, backporch 30V
In seg2
Program H-bot (PG5) 100V from 0.4 to 1.4sec
zero the AMHD_FREQ waveform
Reset NL04 factor from 1.96850e-20 to 1.8e-20
I think this will rebuild the P waveform as well as the gain.
Note that the suggested value is 1.85185e-20, based on the current
calibration value in the tree, which is 1.35e20/Volt.
Check after build and load
Change NL04 gain to P=2,D=4,I=1
Set RXU flat at -.015 through 1.8sec (was ramping at 1.38sec)
Build_all
Save as 1101202000 in ~/pcs_scratch
Save as 1101202000 in alcdata-test::/cmod/trees/test/pcs and dpcs
Check scaling of NL04 waveform and gain in an IDL session:
IDL> mdsopen,'hybrid,xxx',-1
IDL> mdstcl,'create pulse 1101202000'
IDL> mdsclose
IDL> mdsopen,'hybrid',1101202000
IDL> p14 = mdsvalue('HYBRID::TOP.DPCS.LOADABLES.WAVEFORMS.P:POUT_14')
IDL> bita = mdsvalue('HYBRID::TOP.DPCS.LOADABLES.BIT_A:MATRICES')
IDL> print,reform(bita[13,79,*])
0.00000 -0.00000 -0.00000 -0.00000 -2.43000 -2.43000 -2.43000 -2.43000 -0.00000
IDL> print,p14/1.8e-20
1.95000e+18 1.95000e+18 1.95003e+18 -1.77500e+20 -1.77500e+20 -1.22317e+20 -1.66667e+20 -1.65063e+20 3.06519e+19 4.00000e+19
6.00000e+19 6.00000e+19 6.00000e+19 6.00000e+19 -1.00000e+20 -1.00000e+20 8.10000e+17
This should work to the extent that 1.35e20 is the correct TCI cal.
Evidence from yesterday and today indicates that the current value
is closer to 1.45e20, but I'll leave it for now. The old setting corresponded
to 1.27e20.
| |
| Dec 2 2010 08:34:53:610AM | Steve Wolfe | The Linux systems are down because (at least) Alchome is not communicating. Computer pros have been notified. Search and lockup will proceed as usual (without benefit of announcements). The run will start as soon as the computer systems are available. | |
| Dec 2 2010 08:59:42:270AM | Steve Wolfe | After computers came back to life, had to reboot cmodws50 to get logged in. PCS seems to come up ok. Load from 1101202000 Change to seg1: IC_EF4U from -1295 to -1290 Load at 2-Dec-2010 08:58:55.00 Open tree /home/wolfe/pcs_scratch -1 Open tree done Josh recommends that we/he check all diskless systems and run a no power shot to start. | |
| Dec 2 2010 09:04:01:617AM | Steve Wolfe | pcdaqdpcs3 and pcdaqmhd1 respond (allowed ssh) | |
| Dec 2 2010 09:17:37:223AM | 1101202001 | Steve Wolfe | Shot1 will be a no-power test with gas. H-bottom (N2) DISABLED for this test, so we don't fill the machine with N2 just in case the fizzle detector doesn't shut off. There will be a short delay while the EO reboots the EO workstation which was still frozen. |
| Dec 2 2010 09:34:50:067AM | 1101202001 | Steve Wolfe | Shot#1: No-power test shot with gas Got gas, data coming through. Looks like most things are working. Bill needs to reset a breaker in the power room, then we'll go for plasma, I think |
| Dec 2 2010 09:56:44:997AM | 1101202002 | Steve Wolfe | Shot#2: First plasma attempt, based on 1100804027 with mods as indicated above. Boronization recovery (step 1 of SL run plan). Result: Early disruption, .5sec Looks like a couple of small injections, folllowed by a monster. The N2 is coming in before the disruption, but it isn't clear that was the problem. Next: Delay the N2 to 0.6sec and reduce amplitude to 50V |
| Dec 2 2010 10:11:33:210AM | 1101202003 | Steve Wolfe | Shot#3: Delay the N2 to 0.6sec and reduce amplitude to 50V No startup changes. Result: plasma to 1.2sec, steady rise on bolometer Made it to upper null, loop voltage rises after it goes USN Ratty RF from J, D and E not bad; max 1.5MW or so. Density rises, main gas off after .6sec (feedforward), maybe some fueling from N2 after that. Zave rising a little from 2.2 to 2.5 during N2 puff. No ratiomatic this shot, reset for next. Next: reduce N2 to 40V, start from .5sec |
| Dec 2 2010 10:26:22:070AM | 1101202004 | Steve Wolfe | Shot#4: reduce N2 to 40%, start from .5sec Result: plasma to 1.5sec, ugliness on dalpha after 1sec Possible H-mode at .65sec, H-L at .73, no obvious neutron enhancement RF coming along, choppy, maximum 3MW Some small injections, looks like lower prad in the early rampdown Disruption is vertical instability in rampdown, n/nc goes high. Next: Adjust rcur and clearin in rampdown |
| Dec 2 2010 10:37:57:340AM | 1101202005 | Steve Wolfe | Shot#5: Start N2 puff at 0.45sec Adjust rcur to .656 and clearin to .005 in rampdown Also, hold ZXL at 0 in rampdown Result: plasma to 1.56sec, disrupts at almost exactly the same time. RF getting better, choppy between 3 and 4MW still getting an early H-mode in LSN, dies at USN Temperature falls fast after RF turns off, but it still looks vertically unstable. Next: more work on rampdown shape. |
| Dec 2 2010 10:53:51:740AM | 1101202006 | Steve Wolfe | Shot#6: Decrease N2 to 30% more work on rampdown shape RCUR in rampdown from .656 to .650 @2sec, CLEARIN from .005 to -.005 Slow down drop in zcur, flat at .015 to 1.5sec, then dropping to 0 at 2 Copied zcur into clearin by accident, redid it Load at 2-Dec-2010 10:47:06.00 Open tree /home/wolfe/pcs_scratch -1 Open tree done Result: Disrupts at exactly the same time but this one isn't as vertically unstable. RF is pretty steady at 2MW. H-L .7sec Next: N2 puff to zero, see how it runs without seeding |
| Dec 2 2010 11:12:19:640AM | 1101202007 | Steve Wolfe | Shot#7: N2 puff to zero, see how it runs without seeding LH will fire on this shot Result: Plasma full length Three H-modes, and an early marfe; also some early runaways, looks like an injection in late current rise ~.13sec RF has some trips, some injections during flattop as well LH pretty good at 600kW Next: repeat, no PCS changes, same RF request, trying for steady Hmode |
| Dec 2 2010 11:35:18:160AM | 1101202008 | Steve Wolfe | Shot#8: repeat, no PCS changes, same RF request, trying for steady Hmode Actually, LH is requesting 3mm larger outer gap Rcur from .6565 to .655 Result: Runaway startup, injection at .03sec, dpcs_xjump kills it, disrupts at .218sec Looks like it's a good thing, K-lim has some yellow-green Next: try again |
| Dec 2 2010 11:41:55:470AM | 1101202009 | Steve Wolfe | Shot#9: repeat Program same as 8, including increased right gap by 3mm in seg2 Result: plasma full length Steady single H-mode, Prad under 1MW, A few RF trips, 2MW; LH steady at 600kW Looks like a keeper to me. Next: Cryopump off, change startup, otherwise same |
| Dec 2 2010 12:11:51:460PM | 1101202010 | Steve Wolfe | Shot#10: Cryopump off, PG4 from 50msec to 24msec Base pressure is hanging around above 5e-6, M28 and 14 are high on the RGA. M28 is at 6e-8 and steady. Allow a cell access while it pumps down as far as it's going to. Decrease pg4 to 22msec IGT at 3.4e-6 after 25minutes wait, go ahead and try a shot. Result: plasma to 1sec, some early runaways, injection at .94 Matt says tungsten. It happens during LH pulse, around the E-port trip, but I think the injection comes first. Next: Delay LH pulse to 1.1sec, no pcs changes unless I decide to massage the startup |
| Dec 2 2010 12:26:47:720PM | 1101202011 | Steve Wolfe | Shot#11: Repeat with cryopump off, Delayed LH Raise PG3 at -.020 from 70 to 90V INIT with IGT at 5.5e-6 M28 still at 7e-8 on RGA Result: plasma to 1.4sec, 2 Hmodes, ending at 1sec, before LH Not a useful shot for the MP No early runaways this time, fairly fast burnout, no bounce Next: Go to LSN for whole shot (moving on to step 4) |
| Dec 2 2010 12:39:00:197PM | 1101202012 | Steve Wolfe | Shot#12: Remove ZCUR ramp, leave at -.009 through EOF delay CLEARIN reference change to 2.5sec No startup change Result: plasma full length LH wasn't really ready, was still into dummy load Great EDA H-mode, steady, Prad lt 700kW, Prf steady at 2MW Next: repeat with LH (we hope) |
| Dec 2 2010 12:59:18:553PM | 1101202013 | Steve Wolfe | Shot#13: repeat with LH (we hope) Wait till the pressure comes down after DNB conditioning into the machine. IGT to 5e-6 at 15minutes Result: plasma to 1sec,W injection during LH bounce or injection seen on zmeter at .013sec, not on dalpha; no runaways. Next: move RXL in a cm, move strikepoint down plate |
| Dec 2 2010 01:13:03:550PM | 1101202014 | Steve Wolfe | Shot#14: move RXL in to -.01 to move strikepoint down plate change mapping from LH launcher to plate No startup changes yet Result: plasma full length 2 Hmodes, LH faulting after 1sec; ICRF good at 2MW Injection at .81sec, Prad over 1MW rest of shot Next: adjust RXL before .9sec so strike stays off floor |
| Dec 2 2010 01:26:25:767PM | 1101202015 | Steve Wolfe | Shot#15: adjust RXL before 1sec so strike stays off floor This may make the second injection worse, if it's due to LH Result: plasma full length Still injections during LH pulse, strike still on floor early, grazing plate. Still 2 hmodes, first trip during LH. LH has some faulting, not as bad as 14. Next: put back zcur sweep to upper null |
| Dec 2 2010 01:47:25:660PM | 1101202016 | Steve Wolfe | Shot#16: Load seg2 of shot#9 into seg3 Paste zcur traces into seg2, move zcur sweep earlier by .1sec (to .6,.881sec). Move clearin ref switch to 1.0sec Leave lower xpoint programming, etc. as on #15 Result: plasma full length, good steady H-mode, RF starts faulting around 1.02sec, Gap decreases about that time. I suspect the clearin reference switch is contributing Also note that clearin is programmed to increase from .003 to .007 between 1 and 1.5sec Next: move clearin switch time to 0.8sec, decrease rcur after 1sec |
| Dec 2 2010 01:57:19:370PM | 1101202017 | Steve Wolfe | Shot#17: move clearin switch time to 0.8sec, decrease rcur after 1sec from .655 to .653 Result: plasma to 1.55sec, vertical problem? injection at 1sec, small faults on LH and ICRF outer gap still not steady. Next: Increase outer gap some more, decrease inner gap. |
| Dec 2 2010 02:13:19:053PM | 1101202018 | Steve Wolfe | Shot#18: Increase outer gap some more, decrease inner gap. Flatten CLEARIN to .003 (was ramping to .007) Decrease end point of RCUR from .653 to .651-.001=.650 Flatten rest of RCUR after .9sec to .650; note there was still a point at .6557 at 1.2 that hadn't gotten moved to .653 last time; that may have caused the problem, in which case this is overkill Result: plasma full length, made it down alive! looks like a steady EDA, decreasing density, LH is on from 1.1sec Right gap still has a notch between .7 and 1.0, fairly steady at 22mm after that. This looks more or less like it's programmed with RCUR moving in between .9 and 1.03sec. Next: repeat with cryopump on. I could adjust the RCUR again, but leave it alone to compare with cryopump. |
| Dec 2 2010 02:32:19:617PM | 1101202019 | Steve Wolfe | Shot#19: Repeat with cryopump, increase pg4 from 22 to 50msec No seg2 changes this time Result: plasma to 1.45sec, didn't quite make it to rampdown H-L at 1.1sec, RF trip just before start of LH pulse. J4 match changing, looks like density falling more with cryopump. This is all before the LH comes on, so it's not the effect we're looking for. Next: repeat, try for better rf match |
| Dec 2 2010 02:48:54:500PM | 1101202020 | Steve Wolfe | Shot#20: repeat, try for better rf match to keep it in H-mode No PCS changes Result: plasma to 1.65sec, looks like vert. instab. in rampdown Long sort of steady hmode, drop in ICRF power aroiund 1.08sec, near start of LH, and LH start is flakey. It looks like the change in plasma precedes the RF trip, resulting in increase in nl4 and change in match. May be associated with change in RCUR. Next:Adjust rcur program to try to get rid of swings near 1sec |
| Dec 2 2010 03:06:39:107PM | 1101202021 | Steve Wolfe | Shot#21:Adjust rcur program to try to get rid of swings near 1sec Make rcur program=.650 from .7sec on Result: early disruption at .75sec, so much for making it flat. Got the vertical instability early. I need to reduce this elongation Power room access - self=power circuit fired on oh1 after disruption There was an injection at .6sec, before my changes and right at ICRF turn-on. Prad rises to 1MW Next: Delay rcur=.650 to .85sec from .7sec, let it get to USN before increasing the gap to maximum. |
| Dec 2 2010 03:20:17:857PM | 1101202022 | Steve Wolfe | Shot#22: Delay rcur=.650 to .85sec from .7sec, let it get to USN before increasing the gap to maximum. Result: plasma to 1.52sec, loses the H-mode at .9sec Lots of rf faulting, Two LH pulses, neither very good Yijun says match is changing too much Maybe smaller gap will help. Next: try decreasing outer gap by 5mm (on average) |
| Dec 2 2010 03:28:56:670PM | 1101202023 | Steve Wolfe | Shot#23: Increase RCUR from .650 to .6525 Result: plasma full length, one long h-mode, not completely steady RF faulting better, still a couple of drops on E Two almost clean LH pulses Next: Start with USN and leave it there, raise ICRF to 2.5MW |
| Dec 2 2010 03:50:25:383PM | 1101202024 | Steve Wolfe | Shot#24:Start with USN and leave it there, raise ICRF to 2.5, 3 MW Put ZCUR to +.015 from 0.3sec Clearin to upper null reference from 0.1sec Result: plasma full length, multiple H-modes with some injections RF ok until 1.145sec; more faulting during 3MW phase at the end LH still faulting. Gaps are pretty steady, upper xpoint location is a bit bouncy. K and GH limiters are now yellow-green Small injection at .117sec, not sure if it's due to the prog change or just random Next: Repeat with different LH timing |
| Dec 2 2010 04:00:31:797PM | 1101202025 | Steve Wolfe | Shot#25:Repeat with different LH timing No PCS changes Result: plasma full length, H-mode doesn't come in until 1.2sec Early injection is gone this time. LH comes in at 0.9sec, some faulting after 1.1 Klim is now tan Next: Increase the ICRF, try to get the H-mode back earlier |
| Dec 2 2010 04:10:00:817PM | 1101202026 | Steve Wolfe | Shot#26: Increase the ICRF tp 3MW steady, try to get the H-mode back earlier No PCS changes Result: plasma full length, H-mode is still late Seeing inner wall reflection of glowing GH limiter, we don't want to keep doing this! LH is still faulting Breakdown is getting a bit late, maybe time to drop EF4 Next: raise target density from 0.6 to 0.8 |
| Dec 2 2010 04:25:41:400PM | 1101202027 | Steve Wolfe | Shot#27: : raise target density from 0.6 to 0.8 IC_EF4U from -1290 to -1300 Result: Disrupts at .7sec, early runaways, looks like a bouncy startup. Maybe I shouldn't have put those extra 10A on EF4 Prad goes screaming up after the RF turns on. Marfe just earlier. Next: Back to lower null, take back the 10A on EF4 |
| Dec 2 2010 04:41:04:167PM | 1101202028 | Steve Wolfe | Shot#28:Back to lower null, take back the 10A on EF4 Reload seg2 from shot#12 Note that we tried moving RXL in on a later shot, but this was the "control shot". Copy RXL from #15 Result: plasma full length 3 hmodes, one longish, one injection near .8sec ELM-like signals on Dalpha, clean LH, clean ICRF with one small fault at 1.35sec (E) Best shot in hours. Next: Repeat with a second ICRF power step. |
| Dec 2 2010 04:56:43:603PM | 1101202029 | Steve Wolfe | Shot#29: Repeat with a second ICRF power step. No PCS changes Result: plasma full length, not steady H-modes seem to be going ELM-freeish. Early injection around .11sec isn't helping either RF is clean at 2MW before step up at 1.1sec, then some faults LH is clean at 600kW Prad increases when LH turns on, density doesn't decrease, there is a change in QC frequency Next: ? |
| Dec 2 2010 05:07:10:917PM | 1101202030 | Steve Wolfe | Shot#30: LH down to 400kW, ICRF to 2MW flat No PCS changes Result: plasma full length, 2 hmodes, with some evolution in density and prad during the first one, LH and RF both clean at requested powers. That's as good as it's going to get (today) End of Run |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 09:30:00:007AM | Test | Ok | |
| 2 | 09:48:36:003AM | Plasma | Ok | |
| 3 | 10:01:14:203AM | Plasma | Ok | |
| 4 | 10:15:06:747AM | Plasma | Ok | |
| 5 | 10:31:04:527AM | Plasma | Ok | |
| 6 | 10:48:10:740AM | Plasma | Ok | |
| 7 | 11:04:01:633AM | Plasma | Ok | |
| 8 | 11:22:27:923AM | Plasma | Ok | |
| 9 | 11:34:55:707AM | Plasma | Ok | |
| 10 | 12:02:30:737PM | Plasma | Ok | |
| 11 | 12:16:32:550PM | Plasma | Ok | |
| 12 | 12:33:04:730PM | Plasma | Ok | |
| 13 | 12:50:32:747PM | Plasma | Ok | |
| 14 | 01:03:49:780PM | Plasma | Ok | |
| 15 | 01:17:58:270PM | Plasma | Ok | |
| 16 | 01:35:34:550PM | Plasma | Ok | |
| 17 | 01:50:08:837PM | Plasma | Ok | |
| 18 | 02:05:22:980PM | Plasma | Ok | |
| 19 | 02:23:15:407PM | Plasma | Ok | |
| 20 | 02:38:55:557PM | Plasma | Ok | |
| 21 | 02:53:24:103PM | Plasma | Ok | |
| 22 | 03:08:34:950PM | Plasma | Ok | |
| 23 | 03:22:32:187PM | Plasma | Ok | |
| 24 | 03:37:47:710PM | Plasma | Ok | |
| 25 | 03:50:23:040PM | Plasma | Ok | |
| 26 | 04:03:38:100PM | Plasma | Ok | |
| 27 | 04:18:43:497PM | Plasma | Ok | |
| 28 | 04:33:46:460PM | Plasma | Ok | |
| 29 | 04:46:19:927PM | Plasma | Ok | |
| 30 | 05:01:37:460PM | Plasma | Ok | |
| System Availability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Dec 2 2010 09:29:42:780AM | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||