| Miniproposals | ||||||||||
| ||||||||||
|
| Operators | |
| Session leader(s): | Jerry Hughes,Matt Reinke |
| Physics operator(s): | Steve Wolfe |
| Engineering operator(s): | Bill Cochran,Andy Pfeiffer,Gary Dekow |
| Engineering Operator Run Comment |
| automatically entered by signon - please replace with real comment |
| Session Leader Plans |
| Entered: Mar 1 2010 01:01:58:620PM |
| Author: Matt Reinke |
| At the end of the day we'll attempt to finish the work started on 1100226 for MP502 - HIREXSR Calibrations.
During the MP593 run, piggyback on getting the full spatial profile of He-like argon centered on the x3 detectors. Use possible/requested cell accesses if necessary. Test H-like/He-like vignetting =============================== 1) Load 1100226015 (USN, 1 MA, nl04~0.72, w/ cryopump and steady 0.5 psi Ar puff from 0.4 < t < 1.4) 2) run shots w/ and w/o He-like crystal mask, checking for vignetting after each pair of shots and, if necessary, adjusting crystal/enclosure. 3) run shots w/ and w/o H-like crystal mask like (2) at the same time as (2). |
| Entered: Mar 1 2010 11:07:50:710PM |
| Author: Jerry Hughes |
| Run plan for 1100302
MP 593: Vertical equilibrium shifts for triggering ELMs SL: Jerry Hughes PO: Steve Wolfe This run will attempt to induce and perhaps synchronize useful ELMs in normally ELM-free H-modes, and in I-modes. We will plan to take shots until 15:00, and then turn over the run to Reinke et al. and MP502. Prep: -- Working Gas Species: Deuterium -- ICRF at 80 MHz at 4--5 MW level. -- Special gas puffing: 1psi Ar puff enabled, puff at 0.3s, 50ms in length 10psi D2 NINJA puff enabled for GPI, CXRS --- please consult SL before triggering NINJA puff. -- A-Coils: Standard configuration -- LBO may piggyback near end of flattop --- please request SL permission before actually firing. -- Cryopump should be available for beginning of run Required Diagnostics and their set up: 1.) Edge TS, no position changes. 2.) ECE. Optimize for pedestal top 3.) 0-D Fast D-alpha for ELM timing 4.) Fast magnetics. Digitize for full length of shot. 5.) Fast EFIT set up from 0.8 to 1.1s Desired diagnostics: 1.) Passive and GP-assisted CXRS. 2.) PCI. Unmasked configuration, digitizing from 0.4 to 1.6s. 3.) Reflectometer. Set 0.5s digitization window to begin at 0.8s. 4.) Fast D-alpha GPI. 5.) Bolometry. 6.) Lyman alpha. 7.) HIREX Sr. Run plan: The initial part of the run will concentrate on trying to make ELM-free H-modes less transient by inducing ELMs. Cryopump should be cold. 1. Reload 1100209028, an Ip=1MA, Bt=5.4T, nl04=1e20 shot with an EDA H-mode and ~200Hz ZCUR modulation from 0.9 to 1.4s. Repeat this shot, keeping ICRF demand the same (on at 0.6s, ramping to 3.3MW at 0.8s. This will provide a comparison shot in a poorly boronized machine. Next, increase ICRF demand until max available power is put into the flattop (3--4 shots) 2. Keeping modulation the same, delay RF onset so that the gap between H-mode formation and vertical jogging decreases. Exact amounts to delay should be chosen based on the results of (1). The goal is to get the modulation started earlier in the H-mode formation, before the impurity buildup cools the plasmas. We may take want to extend the delay so that the modulation begins _before_ the L-H transition. (2--3 shots) 3. Try to further weaken any QCM oscillation that might exist by lowering density. Decrease nl04 to 0.8e20, then 0.6e20. (2--3 shots) 4. If (1--3) have managed to produce a condition where the modulation appears to be regulating the H-mode in an interesting way, run some control shots without modulation. (0--4 shots) The second part of the run will look at I-mode. 5. Load shot 1091203015 (1.2MA, 5.6T). This was an I-mode shot with sawtooth- triggered ELMs at ~50Hz. Attempt to reproduce this using ~5MW of ICRF. Next insert ZCUR modulation from 0.9 to 1.4s. Choose a drive frequency close to 100Hz, but not right on double the sawtooth frequency. Parameters: Period= 9, 10, or 11ms ModMax= -.004m, ModMin=0.0m , trampup/down=.002 s Adjust mod parameters as needed to maintain vertical stability and look for ELM triggering/pacing. (2--5 shots) Note: The shot we are loading from had a large Ne pulse, which we are not going to use. 6. Load shot 1100204014 (1.35MA, 5.8T). This was an I-mode shot with ELMs, some of which were not induced by sawteeth. Attempt to reproduce this using 4--5MW of ICRF. Next insert ZCUR modulation from 0.9 to 1.4s. Choose a drive frequency close to 100Hz, but not a harmonic of the sawtooth frequency. Parameters: Period= 9, 10, or 11ms, ModMax= -.004m, ModMin=0.0m , trampup/down=.002 s Adjust mod parameters as needed to maintain vertical stability and look for ELM triggering/pacing. (2--5 shots) |
| Physics Operators Plans |
| Entered: Mar 1 2010 01:29:52:010PM |
| Author: Steve Wolfe |
| Physop plan and Engineering Setup for Tuesday, Mar 2, 2010
MP#593 "Vertical equilibrium shifts triggering ELMs" and MP#502 "HIREX_SR calibration" Overnight ECDC in D2 gas and pressure: D2 at 2e-4 Torr sweep: 44/45/103 cm scan: 20/120 s ----------------- Engineering Setup ----------------- Run begins at 09:00 and ends at 17:00 Cryopump required for all shots Power systems as on: 1100209028 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 NINJA with 10 psi D2 DISABLED Enable gatevalves and shutters: ECE, VUV, HiREX Sr, B-spectrometer, Z-bolo Torvac gatevalve toggle (yes/no): no Boronization(yes/no): no Overnight ECDC (yes/no): yes ICRF(yes/no): Yes LH(yes/no): no Cryopump (yes/no): Yes Vessel temperature: 35/35/35 Physop plan: (See also SL Plan for MP#593) H-mode section: 1. Start from 1100209028 (Ip=1MA, Bt=5.4T, nl04=1e20, 200Hz zcur mod) Begin with ICRF as on 1100209028 (on at 0.6s, ramping to 3.3MW at 0.8) then increase power. Presumably the boronization will be mostly gone, so we won't get EDA. The transients at the H-L transitions may make the modulation unstable, so we may need to back off further, or go on to step 2. [3-4 shots] 2. Delay RF waveform relative to modulation, trying to have mod start earlier in H-mode and maybe before the H-mode, hopefully providing some ELM effect on impurities. [2-3 shots] 3. Reduce density to make the target even less EDA-ish [2-3 shots] 4. If the mod is making ELMs that seem to be doing something, take some unmodulated control shots [0-4 shots] I-mode section: 5. Load from 1091203015(1.2MA, 5.6T, ELMy I-mode). First, try to reproduce the original behavior (no modulation). Then add modulation, as indicated in Alcasim tests: Period= 9, 10, or 11ms ModMax= -.004m, ModMin=0.0m , trampup/down=.002 s Tune to taste (or stability) avoiding harmonics of the sawtooth freq, and look for ELM effects [2-5 shots] 6. Load from 1100204014(1.35MA, 5.8T, ELMy I-mode) and attempt to reproduce it. Again, add modulation: Period= 9, 10, or 11ms, ModMax= -.004m, ModMin=0.0m , trampup/down=.002 s and tweak as needed [2-5shots] Sometime around 15:00 (or earlier if things are really uninteresting), switch over to MP#502 (M. Reinke, "HIREX_SR calibration") Load from 1100226015 (ohmic, 5.4T, 1MA, nl04-7e19). Give Matt a cell access. Repeat shots, with cell accesses, until we run out of time or lose another breaker. |
| Session Leader Summaries |
| Entered: Mar 2 2010 03:31:12:310PM |
| Author: Jerry Hughes |
| Session Leader summary for 1100302
MP593 "Vertical equilibrium shifts triggering ELMs" SL: Jerry Hughes PO: Steve Wolfe Today we attempted to apply fast vertical jogs to ELM-free H-modes and I-modes in an attempt to trigger ELMs. The technique used is largely the same as that used on previous runs 1100202 and 1100209, but today we went forward without a boronization, reasoning that stationary H-modes would not be essential as targets for the vertical modulation. 200Hz Vertical displacements were used in 1MA, 5.4T H-modes, as was done on shot 1100209028. In that shot, numerous discrete spikes in D-alpha were observed, in apparent synchronization with the vertical jogs. Qualitatively similar behavior was observed today, though the D-alpha spikes were not as highly contrasting with the backgorund, and from comparison of D-alpha views, one might conclude that the bursts in D-alpha were due primarily to changes in the location of recycling light, as the equilibrium becomes closer or farther away from double null. Like on 1100209, these events did not correlate with any fast modification to the plasma. On the other hand, when we began to bring down the programmed density, there were signs that we were doing something with the jogs. In particular, on shot 012, from about 1 to 1.1s, there were several D-alpha spikes with exponential falls, correlated with sudden drops in the edge Te, as observed on GPC. These events tended to occur when the plasma equilibrium was shifted up, or rising. They were not correlated with sawtooth heat pulses, and in fact the edge dT/dt was generally negative leading up to each event. They look like brief lapses in edge confinement. Are they ELMs or brief H-L transitions? We didn't get a very satisfactory control discharge without the modulation because we were starting to get hit hard on the startup with runaways. K limiter was taking a beating, and radiation was getting to be out of control. We still had problems with machine conditions in the second part of the run, attempting to access I-mode. Mo radiation was just too severe, we think, to achieve a substantial Pnet, and no sustainable temperature pedestal was achieved in USN cases which were loaded from very successful I-mode shots (1091203015, 1100204014). We did manage to test the ZCUR modulation scheme in these configurations and found that 100Hz modulations were easy to apply, with programmed ramp rates of 2ms, and peak-to-peak amplitude of 4mm. The realized amplitude of the oscillation was 6mm in the 1.2MA case, and 4mm in the 1.35MA case. Perhaps there is more headroom to increase the oscillation. At 15:00 I turned over the run, as planned, to Matt Reinke for equally exciting Hirex Sr calibration shots. |
| Entered: Mar 2 2010 06:03:30:920PM |
| Author: Matt Reinke |
| Shots 025-032 were MP502 - HIREXSR Calibration (Thanks Jerry!)
The repeatability took more time to setup as the gas puff from 1100226 seemed to cause an upper x-point MARFE, leading to disruptions. As per The Wolfe's suggestion, moved to a PWM of B-SIDE-LOWER at 1 psi, with at first 40% duty cycle (to small) and 70% (just about right). 029-032 were repeatable seeded discharges with 025 being the baseline unseeded. We were able to do a top 1/2, bot 1/2 cover of H-like crystal and got similar but cleaner results than 1100226. We weren't able to do any vignetting tests on the He-like system since we overcorrected for the declination problem and spent the rest of few shots trying to get it centered and aligned. The system should have good up/down alignment but the z-line is still very vignetted. We'll have to try and address this on Thursday by enclosure movements. |
| Physics Operator Summaries |
| Entered: Mar 2 2010 06:02:24:843PM |
| Author: Steve Wolfe |
Physop summary for Tuesday, Mar 2, 2010
MP#593 Vertical equilibrium shifts for triggering ELMs
SL: Hughes (shot#1-24)
MP#502 HIREX_SR calibration
SL: Reinke (shot#25-32)
The vertical modulation experiment made it through essentially all
of the planned cases, but not with notable success. The radiated
power was about equal to the input power for all but a couple of
the high current USN shots, and ELMs were hard to come by. Despite
the high prad and apparently high Mo levels, the LSN cases based
on 1100209028 looked pretty EDA-ish, although this changed when we
reduced the density, and ended up with very short, ugly, ELM-free
H-modes. There were a few promising looking events early in some RF
pulses that might be certifiable as ELMs, as noted in the SL_SUMMARY.
However, compared to the results from 1100209, the correlation with the
vertical modulation was not nearly as definitive.
The Upper null cases, based on two I-mode shots from Dec and Feb, never
really developed into I-modes, despite the fact that the RF ran to
over 5MW; except for shot#23, Prad kept pace with the input. The
vertical modulation worked successfully on the upper null high current
cases, but there was no sign of ELMs, which is not surprising considering
the low Pnet's and lack of a temperature pedestal. These shots seem to
be a counter-example to the proposition that I-mode does not require
a boronized machine. Note that the upper null cases were in segments 3 and
4.
The startups in the first part of the run were based on the ones
I ran last week (1100224021) which had been fizzle-free, bounce-free,
and runaway-free for most of the day. However, this was not the case today.
We had bad runaway problems starting early in the run, and got a fizzle on
shot#7. I tried reducing the early OH1_VOLT programming (point at .05sec,
pivoting on the point at .035sec), but this was unsuccessful. Increasing
RCUR offset to 3000 and making EF4 more positive (-1280) were also
unsuccessful. Note that the OH1_VOLT and IP traces in the active segments
(3 and 4) after shot#16 were also adjusted to avoid a lurch at segment switch
with the OH1_VOLT set to 150V (instead of 250V).
Jerry turned the run over to Matt at ~15:00 as planned. Since I'd had
no success getting rid of the runaways all day, I imported both segment#1 and
#2 from Matt's reference shot 1100226015. The startup was about as different
as possible, having very negative EF4(-1315) , RCUR offset=2000, BR_0=-.0005 instead of +.0005, and of course none of the voltage adjustments.
The PG4 was at -.5 sec instead of -.3sec, and was 77msec instead of the
50 to 60msec I had been using. Of course, the first shot with these settings
ran with no runaways and no bounces. However, both came back later in the
afternoon, though not nearly as bad as some of the earlier shots. I left this
startup unchanged for the rest of the day (#25-32), with no fizzles.
This part of the run had some difficulties getting started because Matt
was asking for rather low plenum pressures of Argon (0.5psi) which proved
difficult to set. Shots #26 and 27 disrupted, apparently due to excessive
argon. On shot#28 I suggested we put closer to 1psi argon in the plenum (we
ended up with .8psi) and use modulation on the valve drive, adjusting
the duty factor to get the correct amount. This seemed easier to
setup than using the dpcs_piezo_chopper output_pro, which was created
earlier to deal with this sort of thing. I used a 20msec period,
and on #28 set a 40% duty factor with 1msec ramps, with ModMax=100.
This seemed to work but didn't give enough argon, and we increased to
70% on #29-32. All of these shots ran full length. There were cell accesses
before every shot for this part of the run, and some progress was made
on aligning the spectrometers. They believe that the "hyrdogen-like"
system which is to be used for the runs on Thursday and Friday to look at
Calcium, is now correctly setup, but more work needs to be done
to the helium-like system. See Matt's SL_SUMMARY for details.
Scorecard:
----------
Tests/No power shots 0
Duds 0
Fizzles (or < 50msec) 4
Plasmas 28 (2 current rise disruptions)
================= ==
Shots 32
|
| Session Leader Comments | |||
| Mar 2 2010 09:16:28:653AM | 1100302001 | Jerry Hughes | First shot is a reload of 1100209028, which had an EDA H-mode with itty bitty
ELMs, synced to a 200Hz ZCUR modulation. 5.4T, 1MA. nl04=1e20. cryopumped. We will run it and see what we get. If Amanda requests a Btor tweak to optimize ECE positions, we will implement that on subsequent shots. Result: I thought I saw the WIDE2 view rotate during the shot. Probably my imagination. Reasonable shot with what looks like an EDA H-mode. Starting at 0.9s, we see some hash on D-alpha, though not as large as on the target shot. RF is reacting to the jogs. Lost E-port. Yijun will try to improve tune on next shot. Also, fast EFIT needs to be re-run at a shorter period in order to make sense of the modulation. |
| Mar 2 2010 09:31:34:217AM | 1100302002 | Jerry Hughes | Shot 2:
This was a retry of shot 1 with more power. RF behavior was a bit worse, with faults on E and J during the initial H-mode phase. There are some H-mode phases with up to 3MW of ICRF, some signs of ELMs but no real standout signals. Several H-modes, density dropping with each successive one. Next: RF guys will try to improve the tune. |
| Mar 2 2010 10:39:39:873AM | 1100302003 | Jerry Hughes | Shot 3:
Got close to the RF demand on that one. Plasma is radiating like crazy. There are large bursts in the D-alpha early, before the modulation, not seen on the RF traces. After the modulation, you see smaller more discrete D-alpha spikes, and fuzz on the RF. Radiation being what it is, the edge is much colder than on 1100209028, so maybe it's no surprise we see smaller "ELM" activity. t_mod,start - t_L-H = 0.18s Next: Repeat, but with modulation starting at 0.85s rather than 0.9s. |
| Mar 2 2010 10:38:51:687AM | 1100302004 | Jerry Hughes | Shot 4:
Very similar to shot 3. Modulation begins at 0.85s. t=0.85 to 0.90s is the period of peak temp, neuts in this shot. There are also frequent D-alpha spikes in there. Radiation gets high by 1s and kills the confinement. t_mod,start - t_L-H = 0.141s |
| Mar 2 2010 10:36:49:230AM | 1100302005 | Jerry Hughes | Shot 5:
Started the modulation at 0.8s. Looks like 200Hz ELM activity on the D-alpha, at least for the first ~100ms of the mod. Same confinement degradation by 1s. D-port didn't go, so total RF was at the 2.5MW level. t_mod,start - t_L-H = 0.047s |
| Mar 2 2010 10:51:10:887AM | 1100302006 | Jerry Hughes | Shot 6:
Starting modulation at 0.75s. Radiation rise not really affected. A very brief period of ELM-like activity on the fast D-alpha. Longer period of spikes on the slower D-alpha signal, all the way out to 1.4s and beyond. After the modulation ends, the bursts seem larger and broader. t_mod,start - t_L-H = 0.06s Earl sees the rotation on the TV screen too. We speculate that it is the magnetic field acting on the CRT. Next: move modulation to start at 0.7s |
| Mar 2 2010 10:40:16:813AM | 1100302007 | Jerry Hughes | Shot 7:
No plasma. |
| Mar 2 2010 11:24:32:880AM | 1100302008 | Jerry Hughes | Shot 8:
Got L-H at 0.7s, almost contemporaneous with the onset of ZCUR modulation. Looking carefully, I think what is happening is broad bursts in the HA_2_BRIGHT signal (looking at upper part of plasma) are associated with the phases with ZCUR shifted up, and that the faster spikes in MCP_VIS_SIG1 (collimated near midplane) are associated with the downward strokes. So what we're seeing right now may just be recycling light moving about as we jog the topology. |
| Mar 2 2010 11:27:12:350AM | 1100302009 | Jerry Hughes | Shot 9:
Lost density feedback, got a locked mode. Will try again. |
| Mar 2 2010 11:34:15:297AM | 1100302010 | Jerry Hughes |
Shot 10: No plasma |
| Mar 2 2010 11:59:57:200AM | 1100302011 | Jerry Hughes | Shot 11:
Lowered target nl04. Got a lot of ELM-free H-modes -- not worth counting them all -- gradually decreasing in density. Within the first couple of H-modes there is activity on the D-alpha that seems correlated with the ZCUR shifts, although, it is not obviously ELMy. |
| Mar 2 2010 12:43:51:597PM | 1100302012 | Jerry Hughes | Shot 12:
Lower nl04. . . . And it made it through. Similar to shot 11. But there is an interesting phase between 1.0 and 1.1s where I see spikes in D-alpha, with exponential decays, accompanying sudden drops in the edge GPC. The events are happening when Te,GPC8~800eV, and predominately when ZCUR is rising or near its local maximum. In the period just before the "ELM", Te,ped is ~500eV, ne,ped is ~8e19. (e.g. t=1.06025s) Afterwards there is significant pedestal relaxation. One wonders if they are not just a flavor of H-L transition. Next: Try this shot without modulation. |
| Mar 2 2010 12:26:52:640PM | 1100302013 | Jerry Hughes | Shot 13:
Attempt to repeat the low density case in 12 without ZCUR modulation. Radiation is very high on this shot. Not enough H-mode duration. |
| Mar 2 2010 12:27:25:560PM | 1100302014 | Jerry Hughes | Shot 14:
No plasma |
| Mar 2 2010 12:54:56:717PM | 1100302015 | Jerry Hughes | Shot 15.
Similar to 14. Next: Move on to I-mode cases. |
| Mar 2 2010 01:15:17:810PM | 1100302016 | Jerry Hughes | Shot 16:
Reloaded shot 1091203015 (5.7T, 1.2MA), which had an I-mode. Got the desired shot, complete with 5.2MW of ICRF. No I-mode, unfortunately. Prad is a bit higher than it was in the target. The target went into I-mode following a sawtooth crash with deltaT~1.8keV. Maybe the size of the sawteeth matters. Next: try some seeding with Ne. Can't hurt (?) |
| Mar 2 2010 01:17:13:500PM | 1100302017 | Jerry Hughes | Shot 17:
No plasma. Matt observes that the 2pi foil is saturated on the last few shots. Cell access to correct. |
| Mar 2 2010 01:29:22:260PM | 1100302018 | Jerry Hughes | Shot 18:
Disastrous startup. Disrupts at 0.3s. |
| Mar 2 2010 01:53:32:900PM | 1100302019 | Jerry Hughes | Shot 19:
This shot went through. RF power is there, and so is the high Prad. No obvious I-mode, but Arturo notes a temporary decrease in broadband turbulence in the 75GHz reflectometer on this shot and on 16, right after RF comes up, correlated with increased edge Te. Maybe if the radiation was less severe, we'd be there. Next shot: Let's try shaking, starting at 0.70s. At least see whether this configuration can tolerate the jogs. |
| Mar 2 2010 01:58:18:340PM | 1100302020 | Jerry Hughes | Shot 20.
Runaways and early disruption. |
| Mar 2 2010 02:18:25:270PM | 1100302021 | Jerry Hughes | Shot 21.
I-mode attempt with modulation beginning at 0.7s. The ZMAGX trace looks very good. J-port is lost shortly after the modulation starts. Next shot: retry with full power. |
| Mar 2 2010 02:40:19:420PM | 1100302022 | Jerry Hughes | Shot 22:
Full power in the I-mode target, with modulation. Successful modulation at ~100Hz. |
| Mar 2 2010 02:56:49:863PM | 1100302023 | Jerry Hughes | Shot 23
Recalled a nice I-mode trophy. Prad is down, but but still no I-mode formation. 1.35MA, 5.8T. It seems that I-mode is definitely not a cure for high Z disease. Next: try to modulate this condition. |
| Mar 2 2010 03:04:19:950PM | 1100302024 | Jerry Hughes |
Shot 24: Repeat the 1.35MA USN shot with modulation. Lost J-port early, as we did on shot 21. Modulation was successful, with 4mm of oscillation. We've exhausted the run plan, and it's 3pm, so over to Matthew Logan. |
| Mar 2 2010 03:08:01:563PM | 1100302025 | Matt Reinke | Switchover to MP502 HIREXSR calibration.
Load 1100226015 but change to 0.4-0.5 sec Ar puff to get baseline. Before the shot, gave He-like crystal a large change in declination. No masks installed. IMPINJ of CaF2 at 0.7 sec. |
| Mar 2 2010 03:42:50:180PM | 1100302026 | Matt Reinke | Repeat 025 but w/ 0.4-1.4 sec Ar puff
No masks This disrupts early, loses density control. Looks like too much Ar. We're at 0.7 psi compared 0.5 last Friday. |
| Mar 2 2010 03:58:33:010PM | 1100302027 | Matt Reinke | Repeat w/ Ar back down at 0.6 psi
No masks Disrupted early but made it a bit later. Same gas pressure is not working the same as it did Friday. Looks like I get some sort of MARFE at the upper xpoint (WIDE2) and loose density control |
| Mar 2 2010 04:00:48:030PM | 1100302028 | Matt Reinke | Repeat but with 0.8 psi w/ 40% duty cycle (didn't know we could do that).
No masks. |
| Mar 2 2010 04:24:27:993PM | 1100302029 | Matt Reinke | Repeat w/ 70% duty cycle on puff
No masks Much better Ar seeded shot. |
| Mar 2 2010 04:24:53:027PM | 1100302030 | Matt Reinke | Repeat
Bot 1/2 H-like Ar mask |
| Mar 2 2010 04:40:59:060PM | 1100302031 | Matt Reinke | Repeat
1/2 mask on top of H-like |
| Mar 2 2010 04:55:23:123PM | 1100302032 | Matt Reinke | Repeat
remove mask on H-like |
| Physics Operator Comments | |||
| Mar 2 2010 07:46:26:597AM | 1100302001 | Steve Wolfe | DPCS Setup for MP#593 Load from 1100209028 Import seg1 from 1100224021, with modified RCUR offset, IC_EF4U from last week which seemed to have better runaway suppression Change pg4 puff from [-.525,-.5] to [-.350,-.3] for cryopump Seg2: Note irregular modulation on zcur is a consequence of 1msec snapgrid with odd period, leave it alone, since we are trying to reproduce the results of 1100209028 (200Hz modulation, MaxMod=.004). Zero out the (unused) ssep waveforms, just so the PhysAout scope will readout directly Note there is a little jog in rcur at 1.43 to 1.5sec; don't know what it's for, must have been a piggyback experiment. Leave it for now, ask Jerry when he gets in. Import seg3 from seg2 of 1091203015 (1.2MA, 5.6T, USN) to be used for step#5. Import seg4 from seg2 of 1100204014 (1.3MA, 5.8T, USN) to be used for step#6. Note that the title says it's 1.2MA and it's actually 1.3MA Remember to increase TF current limit when/if we get to this step. Load at 2-Mar-2010 07:32:27.00 (be safe while I check the segs from those two shots) In a separate konsole, unset pcs_scratch and bring up pcs /safe Verify seg2 was active on 1091203015 and on 1100204014. quit pcs /safe session Seg3: Change PG1 puff in seg3 from [.55,.75] to [.3,.375] Do a build_segment Seg2: Note that there is a feedforward puff on pg3 and the NL04 gains are [8,0,0]. This is not ideal for cryopump, but it worked before and we want to reproduce the shot, so leave it. Change seg titles, load it up. Load at 2-Mar-2010 07:46:00.00 |
| Mar 2 2010 09:09:36:143AM | 1100302001 | Steve Wolfe | Shot#1: plasma to 1.35sec, no significant ELMs EDA-ish H-mode until 1.18sec, E faults early, others trip at 1.18 and then recover, but plasma doesn't. Next: repeat |
| Mar 2 2010 09:28:57:917AM | 1100302002 | Steve Wolfe | Shot#2: Drop tfcurrent from 150000 to 148500 No change to startup, probably should have dropped the gas plasma into rampdown, several H-modes not steady. Some elms but not as obvious as on 1100209028. RF less happy than on shot#1. Excursions on zcur near the H-L-H transitions (six H-mode), several almost lose it. Next: RF trying to get better match |
| Mar 2 2010 09:43:22:683AM | 1100302003 | Steve Wolfe | Shot#3: RF trying to get better match No dpcs changes. plasma full length, runaways early (so much for the startup) ELMs during three H-modes, first one mostly eda-like, but lots of prad, Prf=3.5MW, Prad(2pi foil)=3.3MW Next: Move modulation earlier instead of delaying RF |
| Mar 2 2010 09:56:32:310AM | 1100302004 | Steve Wolfe | Shot#4: Move modulation earlier instead of delaying RF Modstart=.85, period=.005, duration=.551, ramps=.002, max=.004, min=0., duty=.5 Do this to both wire 2 and 3, since paste screws up grid No startup changes, see if runaways get worse Load at 2-Mar-2010 09:47:06.00 plasma full length, early runaway burst dumps at .118sec Prad similar, so is RF, so is halpha trace, which seems to have some strong activity before the modulation starts. Next: move mod earlier |
| Mar 2 2010 10:13:20:733AM | 1100302005 | Steve Wolfe | Shot#5: move mod start to .80, duration to .601 Leave other parameters the same. Load at 2-Mar-2010 09:58:08.00 plasma into rampdown, disrupts at 1.67sec Prf down a little? Neuts down by a factor of two from 4, 2pi is below 3MW. D never made it through the initial round of faulting. Things that look sort of like ELMs start with modulation, but peter out, and there are still some events after modulation ends between 1.4 and 1.45sec Some early runaways again. Next: move modulation start to .75sec |
| Mar 2 2010 10:29:26:360AM | 1100302006 | Steve Wolfe | Shot#6: move mod start to .75, duration to .651 Load at 2-Mar-2010 10:14:31.00 Start fast efit at 0.70sec No startup changes. plasma to 1.65sec, similar to #5 Got all the RF's, still only 3MW. and still radiating 3MW. Still seeing ELM-like things through EOF (after mod is off!) runaways better. Next: Modstart to .70sec |
| Mar 2 2010 10:40:37:093AM | 1100302007 | Steve Wolfe | Shot#7: move mod start to .70, duration to .701 Load at 2-Mar-2010 10:30:38.00 fast efit start to .65sec Move oh2 reverse current limit to -30,30; it was close to the rail on #6. Fizzle. looks like it moved in, slow burnout. Next: drop pg4 a smidge |
| Mar 2 2010 10:54:37:080AM | 1100302008 | Steve Wolfe | Shot#8: pg4 from 50 to 47msec Changes on shot#7: move mod start to .70, duration to .701 Load at 2-Mar-2010 10:30:38.00 fast efit start to .65sec Move oh2 reverse current limit to -30,30; it was close to the rail on #6. change fast_efit burst_steps from 1750 to 2000 plasma full length, bad runaways early, multiple H-modes, possibly more elmy things. Runaways still seem to get started around .05sec, no sign of bounces. Next: drop target density to 0.8e20 |
| Mar 2 2010 11:27:26:383AM | 1100302009 | Steve Wolfe | Shot#9: drop target density to 0.8e20, drop pg3 feedforward Decrease seg1 oh1_volt at [.05,.1] sec from 250 to 200V; no change to seg2 value, which is at 250, so there may be a lurch at seg switch. TORVAC RSview problem, calling Joe &/or Yuri Need a cell access to fix TORVAC Time in RECOOL 26:45 disrupts at .95sec. no H-mode at 3MW RF. Looks like a locked mode around 0.7sec TCI feedback was screwed up, bouncing rail to rail and losing fringes. Early density was 3e19 at .4sec, that's probably where it got into trouble. Next: repeat, with TCI hopefully |
| Mar 2 2010 11:38:48:550AM | 1100302010 | Steve Wolfe | Shot#10: repeat with target density to 0.8e20, drop pg3 feedforward On#9, the jump in oh1 voltage at seg switch wasn't serious, not clear if reduction in runaways was related to lower voltage or not. Jim says TCI is "better than it was". Runaway early disruption, big injection at .041sec, 100ka in trouble earlier, CIII light showing spikes before 20msec. xrays seem to start around 0.30sec Next: Blame it on a flake and try again |
| Mar 2 2010 11:51:17:363AM | 1100302011 | Steve Wolfe | Shot#11: repeat with target density to 0.8e20, drop pg3 feedforward plasma into rampdown 1.62sec, lots of short ugly h-modes. There are some neuts, 3 to 4e13. Just over 3MW RF, 2pi foil is similar Next: nl04 down to 6e19 |
| Mar 2 2010 12:09:27:883PM | 1100302012 | Steve Wolfe | Shot#12: nl04 down to 6e19, pg3 decrease again NL04 PID gain from [8,0,0] to [4,8,1.5] Plasma to 1.6sec, huge runaways. So much for dropping the oh1 by 50V. Ugly short H-modes, I think. Nothing looks like ELMs to me, but Jerry has the gogs on. Prf 3.5MW, Prad about the same. Next: tweak rampdown rcur and clearin to get rid of disruption Turn off modulation and see what changes |
| Mar 2 2010 12:20:45:757PM | 1100302013 | Steve Wolfe | Shot#13: tweak rampdown rcur and clearin to get rid of disruption RCUR flat instead of ramping in, clearin ramping from +.003 to -.003 Turn off modulation and see what changes Mod amp to 0, period to .1s No startup change, will runaways get better or worse? Runaways worse, plasma disrupts at 1.43sec, before my rampdown changes. If there's an H-mode it ends by 0.9sec. Need to get rid of the runaways. There's actually synch rad around 0.5sec. Runaways are building up from .04sec, Next: program RCUR in by 2cm, ef4 10amps more positive |
| Mar 2 2010 12:32:18:770PM | 1100302014 | Steve Wolfe | Shot#14: program RCUR in by 2cm,from [.62,.64] to [.61,.62] at [.02,.05]sec ef4 from -1290 to -1280 No seg2 changes fizzled. That didn't work. Ratiomatic was off, but it looks like this one went up and in after a bounce at .022sec Next: Put rcur waveform back, increase offset from 2500 to 3000 |
| Mar 2 2010 12:45:47:677PM | 1100302015 | Steve Wolfe | Shot#15: Put rcur waveform back, increase rcur offset from 2500 to 3000 leave ef4 at -1280, pg4 from 47 to 50msec Another seriously bad runaway shot, disrupts at 1.5sec Again, no bounce or other obvious position problem. I can try raising the pg4 more, or dropping the oh1_volt more, or both. Next: go on to the I-mode cases, and try both of the above |
| Mar 2 2010 01:04:24:810PM | 1100302016 | Steve Wolfe | Shot#16: go on to the I-mode case, turn off seg2, turn on seg3 This one is at 1.2MA, 8e19, TF=-158400 PG4 from 50 to 55msec Drop OH1_volt at [.05,.1] from 200V to 150V still no change in seg3 oh1_volt No modulation yet Plasma full length, still lots of runaways, but no synch rad Startup looks like it almost choked. No ELMs According to BTOR we got -5.74T, -159.8ka on TF rog Current is down at seg switch (470kA), and there is a big step in OH1 voltage. Next: decrease Ip program at .1sec add 200msec of neon |
| Mar 2 2010 01:12:07:070PM | 1100302017 | Steve Wolfe | Shot#17: decrease Ip program from 600 to 500 at .1sec Decrease seg3 oh1_volt from 250 to 150 at start of seg4 (try to smooth out the seg switch, not that the runaways are gone) add 200msec of 1psi neon starting from .55sec, to make it like the 1091203015 target. fizzle. Next: pg4 back down to 50msec |
| Mar 2 2010 01:36:48:863PM | 1100302018 | Steve Wolfe | Shot#18: pg4 back down to 50msec BR_0 from .0005 to .0003 (#17 moved up before it crashed) Changes from #17: decrease Ip program from 600 to 500 at .1sec Decrease seg3 oh1_volt from 250 to 150 at start of seg4 (try to smooth out the seg switch, not that the runaways are gone) add 200msec of 0.8psi neon starting from .55sec, to make it like the 1091203015 target. Runaway disruption early .29sec, oh1 crowbars after the disruption. Runaways start very early, but no bounces or position problems. There's also a big marfe from .18sec or so on. Zave is over 4. Next: PG4 from 50 to 55msec |
| Mar 2 2010 01:43:54:967PM | 1100302019 | Steve Wolfe | Shot#19: pg4 back up to 55msec BR_0 from .0005 to .0003 (#17 moved up before it crashed) Changes from #17: decrease Ip program from 600 to 500 at .1sec Decrease seg3 oh1_volt from 250 to 150 at start of seg4 (try to smooth out the seg switch, not that the runaways are gone) add 200msec of 0.8psi neon starting from .55sec, to make it like the 1091203015 target. plasma full length, runaways reduced but still there. decent neuts. Next: try turning on modulation |
| Mar 2 2010 01:58:03:030PM | 1100302020 | Steve Wolfe | Shot#20: turning on modulation from .7 to 1.4sec period=.01, .002 ramps, ModMax=-.004, ModMin=0 Another runaway-dominated disruption around 0.2sec very similar to 18, except the self-powered circuit didn't fire. Next: raise pg4 again |
| Mar 2 2010 02:17:08:520PM | 1100302021 | Steve Wolfe | Shot#21: pg4 from 55 to 60msec, move start of pg3 puff to -.020sec Zcur modulation from .7 to 1.4sec period=.01, .002 ramps, ModMax=-.004, ModMin=0 plasma full length, some runaways, not so bad as previous. Got modulation, did not get ELMs, RF lost at least one xmitter Only 2.5MW after .77sec (J dropped out) Getting about 5 to 6mm zmagx modulation. Next: repeat, hope for less runaways, more rf |
| Mar 2 2010 02:51:18:043PM | 1100302023 | Steve Wolfe | Shot#23: move seg1 pg3 puff 5msec earlier (-.025) Move on to case 6 (1.3MA, 1e20, 5.8T): Seg3 off, seg4 on (from 1100204014) Drop Ip at 0.1sec from 600 to 500ka at 0.1sec, drop OH1_volt to 150V Disable B-side lower (no neon), Raise TF current to 170kA No modulation Still humongous runaways. D didn't go, got 3.8MW, no elms Not a good I-mode, radiation still too high, but down to 2.5MW on prad_main Next: repeat with modulation, hope the RF happens |
| Mar 2 2010 03:03:45:090PM | 1100302024 | Steve Wolfe | Shot#24: repeat with modulation, hope the RF happens Modmax=-.004m, period=.01s, ramps=.002s move pg3 puff to -.030sec, hope the xrays stop happening plasma to 1.12sec, runaways. Prad growing after 1sec, maybe melting something? J drops out after .77sec, D and E survive Modulation ran as programmed, doesn't seem to go unstable Gets about 4mm, which is what we asked for. Might take more, although the chopper current is bouncing off zero. Next: Go to MP#502 experiment |
| Mar 2 2010 03:24:36:330PM | 1100302025 | Steve Wolfe | Shot#25: MP#502 Import seg2 from 1100226015 Seg4 off, seg2 on B-side lower filled to 0.5psi Argon Change pg1 puff from [.4,1.4] to [.4,.5]sec Import seg1 from 1100226015 as well, since this one isn't working well. This one has pg4=75msec at -.5sec, rcur offset=2000, BR_0=-.0005, EF4=-1315 About as different as can get, but it ran on Friday so try it. plasma full length, NO RUNAWAYS! None! And no bounce either. BR_0 looks too negative, but I'm not touching it. nl04=7e19, got argon puff from .4 to .5sec, .7psi Next: cell access and repeat |
| Mar 2 2010 03:44:02:887PM | 1100302026 | Steve Wolfe | Shot#26: cell access PG1 puff from .4 to 1.0sec (mistake, SL asked for 1.4sec) Runaways are back, small bounce at 12msec. disrupts at 1.05sec. prad is big, looks like too much argon, but maybe something else. Pump out bside-lower, refill with 0.5psi argon, and change puff to go to 1.4sec |
| Mar 2 2010 03:54:47:600PM | 1100302027 | Steve Wolfe | Shot#27: cell access PG1 puff from .4 to 1.4sec, bside lower re-filled to 0.58psi argon No startup change this time, since it disrupted maybe 77msec is enough plasma disrupts at 1.3sec, no runaways. rising prad Next: Decrease bside-lower pressure to 0.4psi |
| Mar 2 2010 03:57:46:950PM | 1100302027 | Steve Wolfe | Shot#27: cell access PG1 puff from .4 to 1.4sec, bside lower re-filled to 0.58psi argon No startup change this time, since it disrupted maybe 77msec is enough plasma disrupts at 1.3sec, no runaways. rising prad Next: Fill bside-lower to 0.8psi and use duty factor and modulation to get 40% of that |
| Mar 2 2010 04:10:57:183PM | 1100302028 | Steve Wolfe | Shot#28: Fill bside-lower to 0.8psi and use duty factor and modulation to get 40% of that Modulate from .4 at .02sec period, duration=1.001 ramps=.001, ModMax=100, ModMin=0, duty=0.4 on top of zero base on pg1 plasma full length. No runaways, Got some argon, Matt says not enough. Next: increase duty to 0.7 |
| Mar 2 2010 04:23:49:963PM | 1100302029 | Steve Wolfe | Shot#29: increase duty to 0.7, keeping period=0.02 plasma full length, no runaways. Prad rises to ~700kW Matt seems happy with the argon level, so leave it alone until he changes his mind. Next: repeat after another cell access |
| Mar 2 2010 04:38:56:823PM | 1100302030 | Steve Wolfe | Shot#30: repeat after another cell access resisting temptation to drop the fill plasma full length, no runaways. 2pi bolo looks similar. Next: repeat after cell access |
| Mar 2 2010 04:53:19:920PM | 1100302031 | Steve Wolfe | Shot#31: repeat after another cell access plasma full length, with runaways (5e14) Couple of bounces in startup. Ratiomatic was off again, seems to fail every shot now, probably on cryo regens, and the auto-restart is non-functional. Next: repeat after cell access |
| Mar 2 2010 05:06:53:200PM | 1100302032 | Steve Wolfe | Shot#32: repeat after another cell access plasma full length, some runaways (2e14) END OF RUN Boronization to start after a (hopefully brief) cell access |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 08:57:43:257AM | Plasma | Ok | |
| 2 | 09:15:06:137AM | Plasma | Ok | |
| 3 | 09:31:56:030AM | Plasma | Ok | |
| 4 | 09:48:40:330AM | Plasma | Ok | |
| 5 | 10:05:57:930AM | Plasma | Ok | |
| 6 | 10:19:16:180AM | Plasma | Ok | |
| 7 | 10:35:14:010AM | Plasma | Ok | |
| 8 | 10:48:05:087AM | Plasma | Ok | |
| 9 | 11:17:58:560AM | Plasma | Ok | |
| 10 | 11:31:32:700AM | Plasma | Ok | |
| 11 | 11:44:31:370AM | Plasma | Ok | |
| 12 | 11:58:13:370AM | Plasma | Ok | |
| 13 | 12:11:56:683PM | Plasma | Ok | |
| 14 | 12:25:11:450PM | Plasma | Ok | |
| 15 | 12:38:25:793PM | Plasma | Ok | |
| 16 | 12:51:43:960PM | Plasma | Ok | |
| 17 | 01:09:15:807PM | Plasma | Ok | |
| 18 | 01:25:00:647PM | Plasma | Bad | OH1 self powered |
| 19 | 01:39:24:867PM | Plasma | Ok | |
| 20 | 01:52:37:210PM | Plasma | Ok | |
| 21 | 02:05:53:070PM | Plasma | Ok | |
| 22 | 02:19:22:740PM | Plasma | Ok | |
| 23 | 02:37:00:370PM | Plasma | Ok | |
| 24 | 02:54:52:593PM | Plasma | Ok | |
| 25 | 03:16:54:573PM | Plasma | Ok | |
| 26 | 03:32:59:263PM | Plasma | Ok | |
| 27 | 03:50:17:893PM | Plasma | Ok | |
| 28 | 04:03:50:987PM | Plasma | Ok | |
| 29 | 04:17:02:563PM | Plasma | Ok | |
| 30 | 04:32:56:940PM | Plasma | Ok | |
| 31 | 04:46:57:973PM | Plasma | Ok | |
| 32 | 05:01:48:067PM | Plasma | Ok | |
| System Availability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Mar 2 2010 08:57:05:427AM | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||