| Miniproposals | ||||||||||||||
|
| Operators | |
| Session leader(s): | Earl Marmar,Stewart Zweben,Jim Terry |
| Physics operator(s): | Steve Wolfe |
| Engineering operator(s): | Bill Parkin,Ed Fitzgerald |
| Engineering Operator Run Comment |
| MP#215 J-port conditioning SL:Boivin PO:Wolfe EO:Fitzgerald,Parkin |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:26:570PM |
| Author: To Be Determined |
| Session Leader Run Summary 1010622 MP# 281 Edge Turbulence Imaging SL: Terry, Zweben PO: Wolfe EO: Fitzgerald, Parkin Edge turbulence data was obtained on 6/22/01 with the Gas Puff Imaging system, the fast photodiode system, and the scanning Langmuir probes. The probes will also provide edge profile data which can be used for theoretical modeling of the turbulence. The GPI data was taken with D puffs at an exposure time of 2 microsec/frame, with turbulence patterns similar to those seen previously. GPI results from shots 2,3,4, and 6 will provide a good set of data for calculating correlation lengths for comparison with theory, and the photodiode and Langmuir probe data will provide frequency spectra and turbulence levels. As a piggyback experiment on the MP, Lithium pellets were injected on 4 discharges (shots 2, 4, 5 and 6). The exposure on the ultra-fast PSI camera was adjusted, and good cigar data was obtained, especially on shot 5. The camera was being triggered by CAMAC, and because of variations in pellet arrival time, the integration was setup to average over the entire ablation event. For future experiments we will trigger the camera using the ablation light, as detected on a separate photodiode. In this way, it should be possible to obtain 12 separate Li^+ ablation trails with time resolution of about 20 microseconds, corresponding to radial spatial resolution of 2 cm. |
| Physics Operator Summaries |
| Entered: Jul 7 2004 04:37:13:490PM |
| Author: To Be Determined |
| Physics Operator's Summary for Friday, 1010622 MP# 281 Edge Turbulence Imaging SL: Marmar,Terry,Zweben PO: Wolfe EO: Fitzgerald, Parkin This was supposed to be a continuation of MP#215-Jport conditioning, but the #4 phase shifter had another arc, and at about 1:00PM the RF group decided to punt. The substitute MP, advocated by Marmar, was Edge Turbulence Imaging. Two experiments were tried using the new and old "fast" cameras. The new camera was used to image lithium pellets in an attempt to revisit the old cigar measurement of field angle, and the old camera, in conjuction with the scanning probe, looked at turbulence and "blobs". The first shot of the day was a dud, which I blamed on the vacuum conditioning done on J-port in the morning, while the RF'ers were looking for the arc. Contrary to the stated plan, we had omitted the half hour of ECDC prior to starting the run, in the hope that enough time had passed. It hadn't, so we did 30 minutes ECDC between shots #1 and #2. Plasmas started happening about 2:15PM. I started with shot 1010621014, a 1MA diverted plasma from yesterday, and moved RXU in (shot#2) and ZXU up (shot#4) by 1cm each, and changed the current rise to get to 1MA earlier. The idea was to get back to a more usual C-Mod shape, with higher upper triangularity, but I had to stop tweaking after shot 5, because the session leaders requested constant shots. The last change, increasing ZXU, actually decreased triangularity, while increasing SSEP. Shots #2 and 4 disrupted just after the pellet, about .92sec. For shot#5, Earl delayed the lithium pellet to 1.2sec, which seemed to be a more robust target. Shots #5 and #6 had good pellets and p-modes, and the plasmas survived through rampdown. Some good turbulence data was obtained, but the pellet trigger did not work, so only a couple of cigar frames were obtained at low rep rate. Shot #3 disrupted with an upward-going VDE at about .77sec, for no obvious reason. The n/nc was relatively low, and there was no sign of an initial oscillation and no supply had gotten near a rail. It almost looked like the chopper just quit. The same thing happened on shot #7, and after closer examination of the traces, it was determined that there is indeed a chopper power supply problem. It looks like either a device (or set of devices) failed to commutate off when it got its off pulse, or perhaps the off pulse was too small. In either case, the voltage stayed high and the current rose quickly to the trip point, which is somewhere between 2500 and 3000A. At that point the control voltage slammed full off and the IOC fired. The control voltage goes to -10V in two digitizer samples, i.e. < 120usec. The hybrid demand heads in the same direction, but much slower and to no effect since by the time it reacts the supply has packed it in. The failure on shot 7 began at 0.944sec, the one on shot #3 at .7705sec, for those interested in looking at the traces. Since it was already 5:00PM, we decided to stop the run without risking a last shot. Scorecard: ---------- Tests 0 Duds 1 Fizzles 0 Plasmas 6 4 disruptions: 2 pellet-related, 2 Chopper failures ======= = Shots 7 |
| Session Leader Comments |
| Physics Operator Comments | |||
| Jun 22 2001 07:34:39:837AM | Steve Wolfe | Coldstart hybrid at 7:30; all matrices respond, no errors | |
| Jun 22 2001 01:28:45:320PM | Steve Wolfe | RF is bagging it for the day. We will run a short run for Stewart Zweben (MP unknown) Start from 1010621014 in seg 2. Move RXU from .01 to 0.0 | |
| Jun 22 2001 01:27:13:287PM | 1010622001 | Steve Wolfe | Shot #1: Start from 1010621014 in seg 2. Move RXU from .01 to 0.0 Dud. They were doing J-port conditioning this morning, and we never did the ECDC. Now we will. |
| Jun 22 2001 02:22:08:713PM | 1010622002 | Steve Wolfe | Shot #2:ECDC for 30 minutes While I'm waiting, change the current programming so it gets to 1MA at 0.5sec instead of 0.6. 28 and 14 levels now look higher than before we started ECDC, maybe a leak or maybe just junk kicked up. base pressure down to 1.4e-6 in check plasma, disrupted at about 0.9sec, probably Marmar's pellet. pretty good startup, pressure a little high. ecdc for 10 minutes between shots. next shot: lower the density |
| Jun 22 2001 03:15:57:250PM | 1010622003 | Steve Wolfe | Shot #3: lower the density nl_zmeter from .6 to .36 also bring the current up a little faster. plasma, disrupts at .77sec - NOT the pellet. nl4=8e19, about what we wanted. Nothing obviously wrong. just a brick? Granetz points out that this was a VDE, with a very sudden onset. Not clear why, n/nc is only 1.03 next shot:raise zxu by 1cm, drop the pre-puff. |
| Jun 22 2001 03:14:33:340PM | 1010622004 | Steve Wolfe | Shot #4: raise zxu by 1cm, drop the pre-puff from 22->19msec ecdc for 10min. JT and Blip a cell access. plasma disrupts at 0.9sec (pellet) nice looking startup this time. ssep to 15mm, doutu down to .26 |
| Jun 22 2001 03:16:51:800PM | 1010622004 | Steve Wolfe | Shot #4: raise zxu by 1cm, drop the pre-puff from 22->19msec ecdc for 10min. JT and Blip a cell access. plasma disrupts at 0.9sec (pellet) nice looking startup this time. ssep to 15mm, doutu down to .26 next shot: Earl will delay pellet to 1.2sec, no other changes. |
| Jun 22 2001 03:54:51:293PM | 1010622005 | Steve Wolfe | Shot #5: Earl will delay pellet to 1.2sec, no other changes. plasma, full length, got a pellet and survived. startup ok. saw the pellet with the fast camera, untriggered. didn't get turbulence images for some reason? H/D .06. next shot: no PCS changes |
| Jun 22 2001 04:23:58:997PM | 1010622006 | Steve Wolfe | Shot #6: no PCS changes plasma with pellet, full length. startup ok. next shot: Earl changed the triggering arrangement for the fast camera. JT changed aperture on the other camera. No PCS changes. |
| Jun 22 2001 04:59:15:003PM | 1010622007 | Steve Wolfe | Shot #7: Earl changed the triggering arrangement for the fast camera. JT changed aperture on the other camera. No PCS changes. plasma disrupted at 0.95, not the pellet. This is like shot #3, a big excursion on the chopper current There is a supply problem on the chopper. The dynamic voltage takes off, with out being commanded to do so, and the current hits its internal limit, causing the control voltage to slam off in <.2msec. This also happened on shot#3. The result is a VDE caused by the supply. next shot: drop the EFC voltage from 800 to 700V, just in case it might help. |
| Jun 22 2001 05:10:59:833PM | 1010622007 | Steve Wolfe | Looking closer at the EFC traces, got a little concerned. Since it is already after 5PM, stop the run. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 01:23:52:030PM | Plasma | Ok | |
| 2 | 02:11:45:480PM | Plasma | Ok | |
| 3 | 02:37:25:050PM | Plasma | Ok | |
| 4 | 03:05:04:467PM | Plasma | Ok | |
| 5 | 03:34:56:653PM | Plasma | Ok | |
| 6 | 03:59:58:287PM | Plasma | Ok | |
| 7 | 04:30:25:410PM | Plasma | Bad | EFC fault |