| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Jim Terry |
| Physics operator(s): | Joseph Snipes |
| Engineering operator(s): | Andy Pfeiffer,Gary Dekow |
| Engineering Operator Run Comment |
| MP# 435 EO Pierson PO Snipes SL Terry |
| Session Leader Plans |
| Entered: Aug 23 2005 06:06:00:573PM |
| Author: Jim Terry |
| We will implement MP# 435 "Long Pulse AT Target Plasma Operation".
The main goals of this MP are: 1) to produce a long pulse (multiple L/R times) discharge with parameters suitable for AT targets, specifically an EDA H-mode edge at low-to-moderate densities and at moderate to high auxilliary heating powers; 2) to examine the effects of the long pulse operation on the divertor, specifically divertor structure temperature excursions and metal (Mo and W) impurity production; 3) to examine the gas inventory in long pulse operation Run Plan: A) Begin with shot 1050708029, but with the following changes Ip at flattop = 0.6MA (instead of 0.4MA); Bt=5.0T (instead of 5.4T); RF power waveform reduced by a factor of 2 (1.3 MW instead of a 2.5 MW peak power) and lengthened to 2.0 sec (Note we want to keep the initial power rampup); End-Of-Flattop (EOF) set to 2.0 sec (instead of 1.5s) Use the B21_PROJ feedback on the A-coil; Set fizzle detector for ON at 0.08 s and OFF at 2.3 sec, with the OFF time moved progressively later as the pulse length is extended. B) Lengthen the EOF shot-by-shot in increments of 0.5s until we reach a flux-swing limit/coil temperature limit. (If I extrapolate the volt-sec consumption of shot 1050708023 during its 0.54MA phase to the OH volt-sec consumption limit found in the longest pulse to date, i.e. 1010803018, I "predict" that we may reach ~4.2 sec.) Thus we will probably use ~5 shots in this pulse-length extension phase. A possible option at this point is to reduce the flattop current from 0.6MA to 0.4MA. Monitor coil temperatures and cooling closely. Monitor divertor temperatures and metal impurity levels. Monitor gas inventory. C) At maximum EOF, increase RF power to 2.5MW, though keep the initial power ramp. At this point we hope to demonstrate a long pulse with an EDA edge. At this point there is also a decision point on whether to employ between-shot-boronization to improve/maintain the H-mode. D) At maximum EOF and RF power, compare divertor heating and metal impurity level in the case where the strike points are held fixed as long as possible and the case where the strike points are swept. |
| Physics Operators Plans |
| Entered: Aug 23 2005 02:49:05:260PM |
| Author: Joseph Snipes |
| Physop Plan 1050824 MP#435 Long Pulse AT target SL: J. Terry PO: J. Snipes overnight ECDC in helium Engineering setup as on 1050708029 but with extended pulse length out to 5 s plasma Set fizzle delays on TF and OH to zero. Set fizzle detector on at 0.08 s and off at 2.3 s. Set TF invert time to 4.4 s and all other invert times to 2.5 s. A-coil configuration: normal [+Dtop -Dbot -Jtop +Jbot] and enabled. Set CORRECT_COIL on time to 0.2 s and the off time to 2.2 s. Gas Setup: fill B-Top with 6 psi D2 Hybrid enabled (PG4) fill B-side lower with 1 psi Ar Hybrid disabled (PG1) leave B-side upper alone fill B-main with 40 psi D2 Hybrid enabled (PG3) Enable the following: ECE and VUV gate valves, Z-bolo shutters ICRF Setup: transmitters in run-on at 8:30 Start from 1050708029 but maintaining Ip = 0.6 MA flattop initially out to 2 s, then slowly increasing shot to shot to 4.5 s flattop. TF will be at 5.0 T. Attempt to get long steady EDA H-modes at low density. Between shot boronization may also be used. |
| Session Leader Summaries |
| Entered: Aug 26 2005 12:45:52:490AM |
| Author: Jim Terry |
| We fulfilled many of the goals of the MP (#435), "Long Pulse AT Target Plasma Operation"
1) We produced plasmas of record length (to date) - 3.5 sec EOF, 3.94 sec End-Of-Plasma (EOP) 2) We produced H-modes of record length ~1.7 sec (shots 016 and 019), although we did not sustain the EDA H-mode edge until the EOF, as desired. 3) Plasma operation was limited by the power supply limits, notably on EF1, not by coil temperatures 4) Moly levels did not rise during the long pulse L-mode plasmas, and did not increase with time within the H-mode phases (though this may not be evidence that the source is not increasing, since confinement is probably the dominant effect). No tungsten was observed in the core. 5) Divertor power fluxes are highly toroidally asymmetric, with leading edges getting much hotter than the average. 6) Most diagnostics successfully extended their observation periods to the 4 s (EOP). (TS was limited to 3.5 sec.) A shotlist of the important shots is below: shot EOF(s) EOP(s) H-mode period(s) RF energy input(MJ) 006 3.0 3.47 1.53-2.99 2.97 007 3.5 3.95 3.2-3.5 2.23 008 3.5 3.5 1.47-2.87 4.86 009 3.5 3.5 2.66-2.02 1.74 013 3.5 3.89 - 2.05 016 3.5 3.54 1.4-3.11 4.91 019 (after 1 between-shot boronization) 3.14 3.14 0.68-2.42 3.83 Shot 013 was the shot for which we have the best chance for gas inventory evaluation, since for that shot the gate valves were closed and the plasma disrupted at 130kA in rampdown. The strike points could not be held fixed after about 1.5 sec (it drifted up the outer divertor plate). IR camera data showed this movement. The points touched by the strike point cooled after the strike point passed, indicating that in principle strike point sweeping will mitigate divertor temperature rises. It was also interesting to note that relatively low power shots transitioned to H-mode quite late, e.g. shot 009 at 2.66 s when the RF power had been on for almost 2 sec and was only 0.68 MW at the transition. (See also shot 007.) What are the implications of this for thresholds? |
| Physics Operator Summaries |
| Entered: Aug 24 2005 05:15:51:293PM |
| Author: Joseph Snipes |
| The run started out well though I had some difficulty getting a complete rampdown without disruption. The setup required for the long pulse shots is very long indeed and prone to mistakes because so many things need to be changed. Still, by shot 4, we had a good rampdown. Shot 7 was the longest shot with a flattop of more than 3 s and a total pulse length of just under 4 s, which is a C-Mod record. Adding substantial RF power actually reduced the achievable pulse length because the density increased greatly during H-mode, which cooled the plasma and increased the volt-s consumption. Still, we were limited more by the shape control with EF1 regularly on its 5 kA rail after 2.8 s. Then, we did a number of shots closing the gate valves for 300 s from INIT until after the pulse. Shot 11 failed due to computer problems and we had to wait until we could find Tom or Josh to resolve them. Shot 12 had a problem with a PF breaker that resulted in no breakdown. Shot 13 ramped nearly all the way down to 3.9 s, but then had a problem with OH2U that led to the disruption in rampdown. Shot 14 had again a problem with OH2U and L that led to no breakdown. Shot 15 then disrupted at 0.69 s apparently due to a more subtle problem with OH2. Shot 16 finally went close to full length but had difficulties with EF1 near the end of the pulse. Then, we boronized for 20 minutes and had difficulty getting breakdown, but the last shot went to 3.09 s before disrupting.
Shot,Duration(s),Ipmax(MA),Result Shot,Duration(s),Ipmax,Result | 01 2.076 0.598 Plasma | 02 2.031 0.608 Plasma | | 03 2.520 0.589 Plasma | 04 2.504 0.589 Plasma | | 05 2.997 0.588 Plasma | 06 3.463 0.588 Plasma | | 07 3.933 0.587 Plasma | 08 3.505 0.680 Plasma | | 09 3.158 0.618 Plasma | 10 3.016 0.665 Plasma | | 11 0.000 0.002 Dud | 12 0.000 0.015 Fizzle | | 13 3.893 0.580 Plasma | 14 0.000 0.023 Fizzle | | 15 0.691 0.626 Plasma | 16 3.544 0.664 Plasma | | 17 0.000 0.024 Fizzle | 18 0.000 0.023 Fizzle | | 19 3.107 0.706 Plasma Date 1050824: 19 shots. 14 plasmas. 1 fizzles. 1 duds. 1 System Error. 2 power supply faults. (Plasma = Ip gt 50.0 kA Dud = H_alpha lt 0.40 V) Total plasma duration: 40.44 s (exceeding -60 kA) |
| Session Leader Comments | |||
| Aug 24 2005 10:19:18:890AM | 1050824001 | Jim Terry | EOF at 2.0, then disrupts
RF good - ramping to 1.3 at 1 sec, on thru 2.0 s gas puff at 1.4 sec - seen by diodes Strike point is too high - seen on probe 5 before 1.4 s, then moves up to probe 7 IR data good (NUC=1) - definite hot spot TV cameras ON thru >4.8 sec - hot spots in WIDE1 and JANT No DANT |
| Aug 24 2005 10:11:40:890AM | 1050824002 | Jim Terry | EOF at 2.0, then disrupts
RF good - ramping to 1.3 at 1 sec, on thru 2.0 s gas puff at 1.4 sec - seen by diodes runaways early Strike point is lower - seen on probes and IR IR data good (NUC=4) - definite hot spot TV cameras ON thru >4.8 sec - hot spots/dust on JANT No DANT |
| Aug 24 2005 10:45:34:937AM | 1050824003 | Jim Terry | repeat - trying to get rampdown
EOF at 2.2s, disrupts at 2.5s at 0.3 MA RF good - ramping to 1.3 at 1 sec, on thru 2.0 s 3 small injections - core Mo low (26 units for 116 A line - it's L-mode), no obvious W. Strike point is good IR data good (NUC=4) - definite hot spot on leading edge of module TV cameras ON until 4.175 sec - hot spots/dust on JANT and IR No DANT |
| Aug 24 2005 11:51:40:797AM | 1050824004 | Jim Terry | repeat - good got thru rampdown
max Volt-sec is 1.83 EOF at 2.0s, end-of-pulse is 2.52 s RF good - ramping to 1.3 at 1 sec, on thru 1.8 s (as programmed) Strike point is good -moves up from probe 1 to probe 4 at 2.0 s Core Mo low (23 units for 116 A line - it's L-mode), no obvious W. IR data good (NUC=4) - definite hot spot on leading edge of module TV cameras ON until 4.175 sec - hot spots/dust on JANT No DANT |
| Aug 24 2005 11:51:13:407AM | 1050824005 | Jim Terry | attempt to extend EOF to 2.5s - good, got thru rampdown
max Volt-sec is 2.36 short H-mode periods (3 ELMfree) starting at 1.988s EOF at 2.5s, end-of-pulse is 3.01 s RF good - ramping to 1.3 at 1 sec, on thru 2.5 s (as programmed) Strike point is good - moves up from probe 1 to probe 5 at 2.5 s Core Mo low (24 units for 116 A line - in L-mode), no obvious W. IR data good (NUC=4) - definite hot spot on leading edge of module TV cameras ON until 4.175 sec - hot spots/dust on JANT No DANT |
| Aug 24 2005 11:57:27:733AM | 1050824006 | Jim Terry | attempt to extend EOF to 3.0s - good, got thru rampdown
EF1_U and EF1_L are railed max Volt-sec is 3.17 ties a record for EOF short H-mode periods (starting ELMfree at 1.526s then Type III ELMs or dithers) EOF at 3.0s, end-of-pulse is 3.47 s RF good - ramping to 1.3 at 1 sec, on thru 3.0 s (as programmed) Strike point is good - moves up from probe 1 (until 1.5 s) to probe 7 at 3.0 s Core Mo low (24 units for 116 A line - in L-mode, 45 in dithering period), no obvious W. IR data good (NUC=4) - definite hot spot on leading edge of module TV cameras ON until 4.175 sec - hot spots/dust on JANT No DANT |
| Aug 24 2005 01:00:26:267PM | 1050824008 | Jim Terry | attempt to have H-mode with an EOF of 3.5s at max RF power - good, disrupted at 3.5s
EF1_U is railed max Volt-sec is 3.76 Vs H-mode periods starting 1.45 s ending 2.84s EDA or Type III ELMs) EOF at 3.5s, end-of-pulse is 3.5 s RF not as programmed - one antenna trips for good at 2.277s, at least 1.0 MW thru 3.5 s Strike point is OK - but moves up from probe 1 to probe 7 at 3.5 s Core Mo low (180 units for 116 A line in H-mode period), no obvious W. IR data not written to tree (NUC=4) -should have gone thru 4.5 s TV cameras ON until 4.175 sec - hot spots/dust on JANT There is a timing delay on WIDE1 and JANT |
| Aug 24 2005 01:30:25:560PM | 1050824009 | Jim Terry | attempt to repeat of shot 007 in order to do gas inventory test, i.e. gate valves closed
an EOF of 3.17s - disrupted at 3.17s H-mode periods starting 1.45 s ending 2.84s EDA or Type III ELMs) EOF at 3.5s, end-of-pulse is 3.5 s RF good - ramping to 0.7 MW at 1 sec, on thru 3.0 s (as programmed) Strike point is better, moving less - but moves up from probe 1 to probe 5 at 3.1 s Core Mo low (25-20 on 116 A line in L-mode phase), no obvious W. IR data not written to tree (NUC=4) TV cameras ON until 4.175 sec - hot spots/dust on JANT There is a timing delay on WIDE1 and JANT |
| Aug 24 2005 04:22:33:230PM | 1050824010 | Jim Terry | another attempt to repeat of shot 007 in order to do gas inventory test, i.e. gate valves closed
an EOF of 3.01s - disrupted at 3.01 s no H-mode no RF (as programmed) Strike point is better, moving less - but moves up from probe 1 to probe 5 at 3.0 s No VUV data IR data good (NUC=4) - 240 frames seems to work, 250 fails lots of good dust/specks on IR TV cameras ON until 4.175 sec - hot spots/dust on JANT There is a timing delay on WIDE1 and JANT |
| Aug 24 2005 04:24:33:103PM | 1050824012 | Jim Terry | Dud
199 frames of IR data |
| Aug 24 2005 03:12:05:637PM | 1050824013 | Jim Terry | another attempt to repeat of shot 007 in order to do gas inventory test
ramp RF power down, since it seems to be helping in the flux swings, as long as it doesn't go into H-mode. an EOF of 3.5s - disrupted well into rampdown at 3.89 s no H-mode RF to 1 MW at 1. s then rampdown to 0.5 MW by 3.5 s(as programmed) Max Volt-sec consumption is 3.46 Vs Early MARFE hurts the OH Vs Strike point is better, moving less - but moves up from probe 1 to probe 5 at 3.0 s VUV data shows OK Moly (30 decreasing to 20 units for the 116 A line in L-mode) IR data good (NUC=4) - 240 frames Martin points out the the moving strike point shows within pulse cooling after strike point has left a given point - demonstrating in principle that strike point sweeping will alleviate divertor temperature rise. TV cameras ON until 4.175 sec - hot spots/dust on JANT There is a timing delay on WIDE1 and JANT |
| Aug 24 2005 04:25:18:310PM | 1050824014 | Jim Terry | Dud
no IR data |
| Aug 24 2005 04:26:14:823PM | 1050824015 | Jim Terry | attempt at long pulse with H-mode edge - disrupted at 0.68 s
Early MARFE is gone IR triggered early (before shot) (NUC=4) - no IR data There is a problem with OH2L |
| Aug 24 2005 04:34:05:543PM | 1050824016 | Jim Terry | attempt to have H-mode with an EOF of 3.5s at max RF power - good, disrupted at 3.54s
max Volt-sec is 3.50 Vs H-mode periods starting 1.5 s ending 2.84s EDA or Type III ELMs) EOF at 3.5s, end-of-pulse is 3.54 s RF not as programmed - E trips repeatedly Strike point is OK - but moves up from probe 1 to probe 7 at 3.0 s, higher after that Core Mo low (160 units for 116 A line in H-mode period - falls after 1.8 sec, but still in H-mode according to the density), no obvious W. IR data not taken (NUC=4) - PC disk was full TV cameras ON until 4.175 sec - hot spots/dust on JANT (144 frames) There is a timing delay on WIDE1 (144 frames) and JANT |
| Aug 24 2005 06:12:29:603PM | 1050824019 | Jim Terry | 20 min of boronization after shot 016
50 cm min/70 cm dwell/103 cm max attempt to have H-mode with an EOF of 3.5s at 2.0 MW RF power - good, disrupted at 3.14s H-mode periods starting 0.68 s ending 2.84s EOF at 3.14s, end-of-pulse is 3.14 s RF not as programmed - E trips repeatedly Strike point is OK - but moves up from probe 1 IR data good (NUC=4) - 270 frames TV cameras ON until 4.175 sec - hot spots/dust on JANT (144 frames) There is a timing delay on WIDE1 (144 frames) and JANT |
| Physics Operator Comments | |||
| Aug 24 2005 09:19:39:390AM | 1050824001 | Joseph Snipes | Shot 1: Started with shot 1050708029 and set Ip = 0.55 MA and TF current to 139 kA for 5.0 T in flattop with a back porch at 50 kA from 2.35 to 4.39 s. Pulse
length extended to 2.0 s EOF (end of flattop). Setup segment 2 to be a normal pulse but then to switch segments to segment 3 at 1.4 s and then setup segment 3 to have only the tail end of the pulse so that it can be easily extended to longer pulse lengths. nl04 set to 5 x 10^19 m^-2. Setup feedback on the A-coils with B21_PROJ in wire 11. Argon gas puff enabled. Problem with the Thomson scattering shutter holding up the first shot. Problem also with a mechanical timeout timer on the EFC that is set to 4.9 s, which will eventually limit the pulse length. |
| Aug 24 2005 09:52:17:280AM | 1050824001 | Joseph Snipes | Shot 1: Good shot. Disrupted just after the end of flattop at 2.07 s. |
| Aug 24 2005 09:56:11:810AM | 1050824002 | Joseph Snipes | Shot 2: Corrected a jump in RXL and RCUR at segment switch at 1.4 s. Lowered
ZXL by 1 cm to -0.01 m in segments 2 and 3. Moved RXL in by 0.5 cm to -0.005 m in segments 2 and 3. Modified the rampdown to keep it diverted throughout the rampdown. Good shot, but still disrupted at 2.02 s. |
| Aug 24 2005 10:22:17:623AM | 1050824003 | Joseph Snipes | Shot 3: Corrected nl04 in segment 2 so that it stays at 5 x 10^19 m^-2 out past
the segment switch. Changed the R and Z of the X points to attempt to keep the plasma diverted into rampdown. Set RCUR, ZCUR, and CLEARIN integral and derivative gains out to 5 s. Also, corrected RCUR and ZCUR predictors for CLEAR from 3 s to 5 s. Good shot. Disrupted at 2.52 s. |
| Aug 24 2005 10:47:19:373AM | 1050824004 | Joseph Snipes | Shot 4: Increased prepuff by 3 ms to 24 ms duration. Added some pre-programmed
gas between 0.05 and 0.1 s to avoid the runaways. Extended all derivative, integral, and proportional gains out to 5 s even though they were set to zero. Good shot ramped all the way down. |
| Aug 24 2005 11:14:36:547AM | 1050824005 | Joseph Snipes | Shot 5: Increased pulse length to 2.5 s EOF. Decreased prepuff to 19 ms duration.
Good shot ramped down to 3 s but more runaways. |
| Aug 24 2005 11:34:07:670AM | 1050824006 | Joseph Snipes | Shot 6: Increased prepuff by 3 ms to 22 ms duration. Reduced EF4 current by
10 A to -1325 A. Extending EOF to 3.0 s. Good shot. Disrupted in rampdown at 3.45 s. |
| Aug 24 2005 12:05:43:110PM | 1050824007 | Joseph Snipes | Shot 7: Extended pulse length out to 3.5 s. Added preprogrammed gas from 0.02 s
to 0.2 s in segments 1 and 2 to avoid the runaways. Also increased EF4 current by 20 A to -1305 A. Good shot. Ramped all the way down to near 4 s. |
| Aug 24 2005 12:29:12:373PM | 1050824008 | Joseph Snipes | Shot 8: Added 4 ms to the prepuff to 26 ms duration. Removed pre-programmed gas
puff in segment 2. Reduced pre-programmed gas in segment 1. Changed starting times in segment 3 to 1 s to avoid missing times between segments 2 and 3. Good shot. Disrupted at 3.5 s. |
| Aug 24 2005 01:03:47:077PM | 1050824009 | Joseph Snipes | Shot 9: Closing gate valves in INIT and keeping them closed for 300 s after the
shot. So, I've reduced the prepuff by 4 ms back to what it was on shot 7, 22 ms duration. Put in a ramp in RXL 1 cm inward between 1.4 and 3.5 s and a ramp in ZXL downward by 1 cm over the same time interval. Disrupted at 3.14 s. |
| Aug 24 2005 01:21:37:950PM | 1050824010 | Joseph Snipes | Shot 10: Repeat without RF. Disrupted at 3 s. |
| Aug 24 2005 01:40:04:750PM | 1050824011 | Joseph Snipes | Shot 11: Reloaded shot 7. Put a downward ramp in the RF. No power shot due to
computer problem. |
| Aug 24 2005 02:12:00:170PM | 1050824012 | Joseph Snipes | Shot 12: Try again. PF breaker pulled at -0.7 s. No plasma. |
| Aug 24 2005 02:49:15:840PM | 1050824013 | Joseph Snipes | Shot 13: One more time . . . Good shot. Almost made it all the way down.
Disrupted at 3.9 s at 100 kA. PF crowbar from OH2U just before the disruption. |
| Aug 24 2005 03:20:58:137PM | 1050824014 | Joseph Snipes | Shot 14: Removed once again the pre-programmed gas puff between 0.1 and 0.2 s in
segment 2 and some of the early pre-programmed gas in segment 1. Removed the back porch on the TF waveform and set the TF invert time to 4.0 s. Dud! All gate valves closed on shot 13 and were still closed for 14. OH2U and L did not behave correctly just before breakdown. |
| Aug 24 2005 03:28:42:323PM | 1050824015 | Joseph Snipes | Shot 15: Try again. Disrupted at 0.69 s. Just barely went. Still a subtle
problem with the OH2s. |
| Aug 24 2005 04:09:52:090PM | 1050824016 | Joseph Snipes | Shot 16: Try again. Disrupted at 3.54 s. The plasma went double null by 3.2 s.
The EF1 supplies are railed after about 2.8 s. |
| Aug 24 2005 04:10:25:683PM | 1050824016 | Joseph Snipes | Shot 17: Boronizing for 20 minutes before this shot. |
| Aug 24 2005 04:44:36:870PM | 1050824017 | Joseph Snipes | Shot 17: Reduced prepuff to 8 ms duration to account for a fresh boronization.
Dud! |
| Aug 24 2005 04:59:16:247PM | 1050824018 | Joseph Snipes | Shot 18: Increased prepuff to 20 ms duration and reduced EF4 current by 15 A to
-1320 A. Fizzle! |
| Aug 24 2005 05:10:41:310PM | 1050824019 | Joseph Snipes | Shot 19: Decreased prepuff to 17 ms duration. Disrupted at 3.09 s. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 09:29:36:063AM | Plasma | Ok | |
| 2 | 09:54:01:720AM | Plasma | Ok | |
| 3 | 10:19:58:610AM | Plasma | Ok | |
| 4 | 10:45:44:030AM | Plasma | Ok | |
| 5 | 11:10:08:813AM | Plasma | Ok | |
| 6 | 11:30:44:110AM | Plasma | Ok | |
| 7 | 12:00:37:593PM | Plasma | Ok | |
| 8 | 12:27:17:843PM | Plasma | Ok | |
| 9 | 01:00:21:437PM | Plasma | Ok | |
| 10 | 01:17:35:470PM | Plasma | Ok | |
| 11 | 01:34:29:903PM | Plasma | Bad | igor fail no shot |
| 12 | 02:08:18:250PM | Plasma | Bad | OH2L Failed at start |
| 13 | 02:35:57:340PM | Plasma | Bad | OH2u IOC Fault |
| 14 | 03:13:10:480PM | Plasma | Bad | Diagnostic Gate Valve Not Opened To A Fault |
| 15 | 03:25:49:010PM | Plasma | Bad | OH2L Problem with oscillation |
| 16 | 03:49:00:433PM | Plasma | Ok | |
| 17 | 04:40:45:560PM | Plasma | Ok | |
| 18 | 04:53:15:433PM | Plasma | Ok | |
| 19 | 05:05:42:027PM | Plasma | Ok | |