| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Ian Hutchinson |
| Physics operator(s): | Steve Wolfe |
| Engineering operator(s): | Ed Fitzgerald,Bill Parkin,Bill Cochran |
| Engineering Operator Run Comment |
| MP#199 Upper Null Triangularity Scan in Reverse Field. Hutchinson/Wolfe |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:25:367PM |
| Author: To Be Determined |
| Session Leader's Summary 1001106 MP 199 Upper X-point H-modes with triangularity scans. SL Hutch PO Wolfe Summary _______ Not many good plasmas were obtained today and practically no decent heating because there were fizzle problems and H/D ratio was unpleasantly high. Nevertheless some dynamic triangularity scans were obtained in H-modes that showed QC modes and other EDA characteristics even though they were repetitively cycling in and out of H-mode because of excessive radiation (cf heat). The trends of x-ray width versus triangularity observed in the lower x-point scans were substantially reproduced. X-ray width decreases as triangularity is decreased or as current is decreased. The triangularity dependence remains in cases where the toroidal field is ramped up to maintain q constant. (Confirming the independence of the effect on B_t). In short, all the evidence is that the upper diverted case is the same as the lower diverted case, so the effects we see are not a divertor hardware proximity effect, but truly shaping. Useful shots: 19 Sustained EDA case at high triangularity. 20 Triangularity dynamic scan delta upper .8 to .4. Shows width narrowing. 21 Triangularity scan with simultaneous Ip scan to keep q constant. Much less x-ray width variation 22 Triangularity scan, Ip constant, Bt ramped up, q constant. Shows narrowing. 23 Higher power (1.8MW vs 1MW) version of 22. Similar result. 24 Fixed high triangularity, fixed Bt, Ip scan. X-ray width undetermined. |
| Physics Operator Summaries |
| Entered: Jul 7 2004 04:37:12:613PM |
| Author: To Be Determined |
| Physics Operator Summary for Monday, 1001106 MP#199 Upper null triangularity scan with reversed field SL: Hutchinson PO: Wolfe EO: Cochran, Fitzgerald, Parkin Plan: ----- Start from an upper null like 1001006008 (flipped for reverse field, and current increased to 1MA). Start with a lowish density nl4=7e19. Get enough RF in to produce an H-mode, and scan triangularity (upper and lower) to look for EDA boundary. If an EDA is obtained, try dynamic triangularity scans. Repeat at higher density, up to 1.5e20. Note: This run was previously attempted on 991104, but RF wasn't operating well, so results were unsatisfactory. Machine Conditioning: --------------------- Boronization and ECDC in D2 over the weekend, with ECDC ending 2 hours before the run. Run should start at 9:00 AM Power systems as on 1001103002 except EFC start time should be -.030sec EFC Voltage to 900V Gas setup fill B-top with 6 psi of D2 Hybrid enabled fill B-side lower with 1 psi of Ar Hybrid DISABLED fill B-side upper with 35 psi of D2 Hybrid DISABLED, for ECDC fill C-side with 30 psi of D2 Hybrid enabled leave J-bot as is Hybrid DISABLED Assuming no vacuum problems the following gate valves should be enabled: ECE, VUV, Beast PCS Setup: Seg 1 from 1001103013; change chopper voltage for early ZCUR control flattop IC_EF4 at 1340. Seg 3 upper null, increase Ip to 1MA, change tf current to 150kA; nl4 to 7e19; turn on. Seg 4 from 991104xxx and re-compute predictors; Probably only use for waveforms. Other: The fizzle detector threshold was increased from .191 to .223 prior to the run, to try to avoid having the level go low again after a disruption, as observed on 1001103022. We really need to tweak up the iso-amp offset voltages going to the hybrid! Results: ------- The startup problems that plagued last week's runs have not gone away. We began with a series of fizzles, and tweaking to try to get a plasma. Ian believes that the main problem is extreme sensitivity to Br; when the nominal value of fluxihh_zcur is positive, it seemed not to go. It looked to me like the problem was radial position, and I also changed the RCUR offset, ending up at -3500 (starting from -5500). Later in the day, I tried adjusting the pre-puff to keep the position in a "good" range; widths were from 18-24msec. After disruptions, it took two or three fizzles, and a tweak to get a plasma back; the current rise was initially slow, then increased progressively, until it ended up being too fast. Br and zcur also seemed to change, and I ended up at a Br offset of .0025, although I think this may really be too high. All shots were run with Segments 1 and 3 active. Clearin from the initial segment 3 was much too negative at -.015; we ended up at +.005 to get a sensible gap, which implies that an offset of zero is good for the upper nulls. However, the normal field shots in October (e.g. 1001006008) had a reasonable gap at this setting; the difference is perhaps the large negative offset Ip signal into the wavegen. We got the upper triangularity above 0.8, corresponding to RXU of -.06. RXL of the same value gives delta_l of .4. High triangularity tends to drive the EF2U to zero unless the EF4 demand is programmed positive, i.e. fighting the EF3. Dynamic triangularity scans from hi to low were carried out, with ramps programmed on EF4 as well. We ran a few shots with ramping triangularity and either current or Bt ramps to maintain constant q. The PF waveforms for these shots were "interesting"; caligraphy anyone? H/D at the beginning of the first shot was only 0.02, but went high once the RF came on, and during the day was up to 0.2. Max RF power was 2.5MW, but the only shots which were close to constant were down in the 1.2-1.7MW range. We had injections (mostly D-port?) and faulting on all systems all day. Heating efficiency, with the possible exception of shots 11-13, was poor. "Successful" shots were: #6 - 0.7e20, 1MA inner wall limited, no H-mode #7 - 0.7e20, 1MA 0.5cm inner gap, disrupted during RF #11 0.7e20, 1MA, H-modes #12 0.7e20, .9MA, larger inner gap, #13 0.9e20, .9MA, earlier diversion, EDA until injection #14 tried for high triangularity, disrupted early #18 high triangularity, 4 EDA-ish H-modes #19 higher triangularity, longer EDA with lower RF power #20 triang. scan (hi-low) #21 triang scan with Ip ramp, const. q #22-23 traing. scan with Bt ramp, const. q #24 IP ramp at constant hi triang. Scorecard: --------- Tests 0 Duds 0 Fizzles 11 Plasmas 13 ------- -- Shots 24 |
| Session Leader Comments | |||
| Nov 6 2000 09:15:44:610AM | Ian Hutchinson | MP 199 Upper X-point Abnormal Field Direction, Triangularity scans for EDA characteristics etc. Get EDA H-mode. Do triangularity scans. | |
| Nov 6 2000 10:48:48:093AM | Ian Hutchinson | Shot 6 first plasma. Not diverted 1MWfrom E-port. H/D 2-6%. | |
| Nov 6 2000 10:59:50:467AM | 1001106007 | Ian Hutchinson | Disruption apparently caused by RF sparklers showing on D-antenna. |
| Nov 6 2000 12:21:13:937PM | 1001106011 | Ian Hutchinson | Plasma. First H-mode 2MW, not EDA. The safety factor is rather low 3.6. 1MA. |
| Nov 6 2000 12:47:00:403PM | 1001106012 | Ian Hutchinson | Plasma 0.9MA. q=4. H-mode seems trying to be EDA. Lower x-point is rather close ask to move it away. nl04=.7 target. |
| Nov 6 2000 01:18:18:793PM | 1001106013 | Ian Hutchinson | Plasma. EDA until hit by injection. |
| Nov 6 2000 01:19:29:340PM | 1001106014 | Ian Hutchinson | Attempt at a higher triangularity case disrupts early because of power supply problems. |
| Nov 6 2000 02:51:35:077PM | 1001106018 | Ian Hutchinson | Plasma. Higher triangularity up to .65 at top. Ragged short radiative H-modes. H/D ratio up to 14%. |
| Nov 6 2000 03:23:02:250PM | 1001106019 | Ian Hutchinson | Plasma. Two long H-modes. Killed by an injection at the end. 1.2MW. Upper triangularity 0.75. This is a sustained EDA case at high triangularity. |
| Nov 6 2000 03:45:40:123PM | 1001106020 | Ian Hutchinson | Plasma. A triangularity scan. Many H-modes with impurity accumulation. Delta upper from .8 to .4. The xray width shows a narrowing as the triangularity decreases. 5mm dropping to around 2.7mm. Density width on edge Thomson also shows a factor of about 2 narrowing. Quasi coherent mode present on PCI till about 1.3s then seems to disappear in favor of some broader band stuff. In so far as this shot looks like the previous triangularity scans in its pedestal width variation, it is prima facie a verification that the result is not a divertor interaction. That was a major part of the proposal. |
| Nov 6 2000 04:13:56:640PM | 1001106021 | Ian Hutchinson | Plasma. Shot to ramp the current down so as to keep safety factor constant. succeeded at that very well. Multiple H-modes as before. Very low stored energy. Heating efficiency is ghastly. H/D ratio is very high, 15%. X-ray width decrease is markedly less than the previous shot. Showing that the current-dependence is cancelling out the triangularity effect, or else that the effect is really safety factor. Density width is decreasing with time similar to previous shot. It needs to be investigated if this is related to a flux mapping/expansion effect. PCI sees a quasi-coherent mode only weakly up to about 1s and not significantly thereafter. |
| Nov 6 2000 04:32:17:497PM | 1001106022 | Ian Hutchinson | Plasma with constant current but magnetic field ramped up to keep q constant as the triangularity sweeps. Three H-modes. Little heating. Xray width is similar to shot 20, i.e. it decreases as the triangularity drops. Density width behaves the same. PCI does see QC mode even to low triangularity but it becomes more incoherent. |
| Nov 6 2000 04:48:25:903PM | 1001106023 | Ian Hutchinson | Plasma. Higher power 1.8MW otherwise similar, with Bramp to keep q constant. The extra power is hardly noticeable. H/D is still creeping up. |
| Nov 6 2000 05:18:28:403PM | 1001106024 | Ian Hutchinson | Plasma. Plasma current ramp-down at constant high triangularity. X-ray widths can't be determined because of low signal. |
| Physics Operator Comments | |||
| Nov 3 2000 05:44:19:467PM | Steve Wolfe | Setting up a shot for Monday morning Seg 1 from 1001103013; change chopper voltage for early ZCUR control flattop IC_EF4 at 1340. Seg 3 upper null, increase Ip to 1MA, change tf current to 150kA; turn on. Seg 4 from 991104xxx and re-compute predictors; Probably only use for waveforms. Save as pcs$wolfe:pcs_1001106000 | |
| Nov 6 2000 07:12:13:827AM | Steve Wolfe | load from previously prepared pcs$wolfe:1001106000 coldstart at 7:09 - successful, all matrices respond. | |
| Nov 6 2000 07:31:36:543AM | Steve Wolfe | Adjust fizzle detector threshold from .191 to .223 (< 1/4 turn clockwise) On shot 22 on Friday the fd toggled after a disruption, causing a problem for the RF. The offset on the rogowski iso-amp seems to have drifted even further than when it was necessary to make a similar adjustment in March (see notes for 1000322). | |
| Nov 6 2000 08:16:15:577AM | Steve Wolfe | For shot#1, adjust nl4 program from 7.89e19 (?) to 7.0e19 | |
| Nov 6 2000 10:25:32:170AM | 1001106001 | Steve Wolfe | shot#1: Try for upper null 1MA plasma. fizzle. early breakdown, but a slow current rise. Ian thinks it has a br problem. Irby needs to adjust the TCI laser, wouldn't have had feedback if it had gone. next shot: tweak OH2 voltages at .01 and .015sec by -10V and -5V (corresponds to OH1 change from last week, try to improved index evolution. |
| Nov 6 2000 10:25:43:747AM | 1001106002 | Steve Wolfe | shot#2: : tweak OH2 voltages at .01 and .015sec by -10V and -5V (corresponds to OH1 change from last week, try to improved index evolution. fizzle. No obvious change in fields, Br is ~+.0005, It moves out farther this time, but I/Bz isn't too big. Next shot: EF4 from 1340 to 1325. |
| Nov 6 2000 10:26:02:450AM | 1001106003 | Steve Wolfe | shot#3: EF4 from 1340 to 1325. fizzle: next shot Br more positive |
| Nov 6 2000 10:26:29:340AM | 1001106005 | Steve Wolfe | shot#5: RCUR offset from -5500 to -4000. And another one. This one looks like a vertical problem. next shot: Br back to .0015 |
| Nov 6 2000 10:21:01:310AM | 1001106006 | Steve Wolfe | shot#6: Br back to .0015 plasma. finally! Still has a little lurch at 13msec EF2's are doing something weird before .5sec Got 1MW rf, puts in gas but no H-mode. It's on the inner wall. CLEARIN has an offset problem I guess. next shot: Change clearin |
| Nov 6 2000 10:42:35:217AM | 1001106007 | Steve Wolfe | shot#7: Change clearin from -.015 to -.005 disruption at 0.84sec, brick. left gap is only .5cm, outer gap ~1.3 next shot: increase clearin, rcur |
| Nov 6 2000 11:07:03:810AM | 1001106008 | Steve Wolfe | shot#8: Change clearin from -.005 to 0., rcur from .6575 to .661 fizzle. slow current rise, stays inside. next shot: repeat |
| Nov 6 2000 11:16:27:950AM | 1001106009 | Steve Wolfe | shot#9: Change clearin from -.005 to 0., rcur from .6575 to .661 another fizzle. almost identical to 8. next shot: rcur offset to -3500, pg4 from 20msec to 18msec |
| Nov 6 2000 11:43:22:653AM | 1001106010 | Steve Wolfe | shot#10: rcur offset to -3500, pg4 from 20msec to 18msec Change clearin from -.005 to 0., rcur from .6575 to .661 another fizzle. very late breakdown. Ian points out when fluxihh_ZCUR is positive it doesn't go. |
| Nov 6 2000 12:12:15:390PM | 1001106011 | Steve Wolfe | shot#11: pg4 back to 20msec, Br to .002 from .0015 Change clearin from -.005 to 0., rcur from .6575 to .661 plasma, full length. Bounced once at 15msec, but survived. ZCUR was more positive, but went negative. some runaways. H-mode, looks pretty ELM-free. EF2's to zero in rampdown. q is only 3.7 |
| Nov 6 2000 12:24:28:297PM | 1001106012 | Steve Wolfe | shot#12: INcrease clearin from 0 to .005, rcur to.6635,ip to 900kA plasma, some early runaways again. another bouncer, gets to 50Ka and hits outside. next shot: raise fill, nl4 to 9e19. |
| Nov 6 2000 12:55:34:247PM | 1001106013 | Steve Wolfe | shot#13: raise fill to 22msec, raise nl4 to 9e19. clearin ramp earlier to divert earlier, ZXl from -.02 to -.03 turn on Argon. plasma, full length, eda until injection at .87sec, current rise is fast, but doesn't bounce, no runaways. D faults out, so does J, E only a few faults. next shot: higher triangularity |
| Nov 6 2000 01:23:30:640PM | 1001106014 | Steve Wolfe | shot#14: rxl,rxu to -.04 early disruption. EF2U goes to zero early. Startup nearly didn't make it, hesitates at 50kA then hits at 20msec and barely recovers. next shot: drop ic_ef4 in seg 3 from -2000 to -1000, slow down ramp in of RXx |
| Nov 6 2000 01:43:46:873PM | 1001106015 | Steve Wolfe | shot#15: drop ic_ef4 from -2000 to -1000, slow down ramp in of RXx fizzle. Not surprising after the disruption. Breakdown a little late, zcur positive again. slowish current rise, snuffs inside. |
| Nov 6 2000 01:53:51:937PM | 1001106016 | Steve Wolfe | shot#16: drop ic_ef4 from -2000 to -1000, slow down ramp in of RXx and another fizzle. still looks like it's on the inner wall to me. I could wait one more, or tweak something. Br is coming back up, but zcur still reads positive, whatever that means. |
| Nov 6 2000 02:04:33:530PM | 1001106017 | Steve Wolfe | shot#17: drop seg 3 ic_ef4 from -1000 to -500, slow down ramp in of RXx until .5sec It doesn't come back on its own. Need to do something |
| Nov 6 2000 02:06:39:857PM | 1001106017 | Steve Wolfe | shot#17: drop seg 3 ic_ef4 from -1000 to -500, slow down ramp in of RXx until .5sec It doesn't come back on its own. Need to do something This one got to over 40kA, shouldn't lose that! Still inside, still positive zcur. |
| Nov 6 2000 02:32:10:670PM | 1001106018 | Steve Wolfe | shot#18: Br to .0025 drop seg 3 ic_ef4 from -1000 to -500, slow down ramp in of RXx until .5sec plasma, full length, four h-modes. This one ran too far out, bounced, but survived. zcur still starts positive. next shot: increase triang. again |
| Nov 6 2000 03:28:32:657PM | 1001106020 | Steve Wolfe | shot#20: Sweep triangularity after .8 sec (-.06 -> +.04) drop pg4 from 24 to 22 msec (post-disruption) After shot 19 State aborted and SPANNA crashed; therefore we have had a long delay. plasma, full length; 6 h-modes. du from .8 to .4, dl from .4 ->.2 |
| Nov 6 2000 03:51:21:590PM | 1001106021 | Steve Wolfe | shot#21: reduce upper sweep to +.03 to keep off gusset, increase lower sweep to .06, also change Ip 900->800, to keep q nearly constant. No change to startup this time. plasma, full length, multiple H-modes. next shot: constant Ip, ramp field up. |
| Nov 6 2000 04:18:11:920PM | 1001106022 | Steve Wolfe | shot#22:Ip=900, ramp tf from 150->165; increase puff to 24msec plasma, full length. four h-modes, q looks pretty constant. next shot: same, ask for more RF power. |
| Nov 6 2000 04:41:37:280PM | 1001106023 | Steve Wolfe | shot#23:Ip=900, ramp tf from 150->165, a little faster; plasma, full length, four h-modes; 1.7MW next shot: high triang., with current ramp |
| Nov 6 2000 05:07:01:640PM | 1001106024 | Steve Wolfe | shot#24: seg3 from #19 (hi triang, no sweep), rap Ip from .9->.8 plasma, full length, breakdown slightly late. three H-modes, all EDA. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 08:59:50:920AM | Plasma | Ok | |
| 2 | 09:16:20:000AM | Plasma | Ok | |
| 3 | 09:29:41:420AM | Plasma | Ok | |
| 4 | 09:43:43:483AM | Plasma | Ok | |
| 5 | 09:56:54:420AM | Plasma | Ok | |
| 6 | 10:12:14:717AM | Plasma | Ok | |
| 7 | 10:35:52:670AM | Plasma | Ok | |
| 8 | 10:58:29:997AM | Plasma | Ok | |
| 9 | 11:12:00:043AM | Plasma | Ok | |
| 10 | 11:34:33:217AM | Plasma | Ok | |
| 11 | 11:55:41:107AM | Plasma | Ok | |
| 12 | 12:19:34:340PM | Plasma | Ok | |
| 13 | 12:44:37:513PM | Plasma | Ok | |
| 14 | 01:10:45:000PM | Plasma | Ok | |
| 15 | 01:34:42:310PM | Plasma | Ok | |
| 16 | 01:48:45:983PM | Plasma | Ok | |
| 17 | 02:02:57:373PM | Plasma | Ok | |
| 18 | 02:22:16:997PM | Plasma | Ok | |
| 19 | 02:48:22:403PM | Plasma | Ok | |
| 20 | 03:20:10:840PM | Plasma | Ok | |
| 21 | 03:44:43:623PM | Plasma | Ok | |
| 22 | 04:10:45:047PM | Plasma | Ok | |
| 23 | 04:36:17:153PM | Plasma | Ok | |
| 24 | 05:01:31:640PM | Plasma | Ok | |