| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Ian Hutchinson |
| Physics operator(s): | Steve Horne |
| Engineering operator(s): | Vinny Bertolino |
| Engineering Operator Run Comment |
| tf hybrid control |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:14:227PM |
| Author: To Be Determined |
From: MHD2::HUTCH Date: 30-NOV-1995 21:43:05 Subj: Session 951130 Session Leader's Report, 951130 _______________________________ Discharge Development, Bt under hybrid control, Rapid current ramps. MP 126 PO: Horne (Swan Song!) SL: Hutchinson EO: Bertolino Summary _______ TF hybrid control worked as commanded from shot 2. This is now apparently completely routine. We ran the rest of the run in this mode without incident. It should probably be used from now on. Shot 2 was an accidental low-q disruption. Might be interesting. Shot 4 was an accidental low density shot. Ditto. By shot 6 we had the chopper coming on at 50ms without problems. It is recommended that this shot be used as the starting point for obtaining standard current-ramp plasmas in the future. The extra stability ought to give greater programming flexibility. Shot 6 was the nearest thing to a fiducial shot that we had. Shape and density seem about right. From shot 8 onward the OH switches were closed at 0.2s. This means that they are essentially out of the picture. It seems that their effect was not all that strong anyway by the time (0.1) that they used to be closed. EF1 switches were still closed at 0.1s so as to allow their currents to remain negative. Shots, 8/9, had a big Rcur excursion from 0.11 to 0.12s, and a big radiation rise coincident. Part of this was fixed by removing (shot 10) some of the programming on the OH voltages that was associated with switches. The big rise in radiation was then essentially smoothed away. We spent a lot of effort later trying to fix the rest of the lurch, with incomplete success. Remaining possible culprits include Ip feedback. From 9 on we had faster current ramps during 0.1-0.4s. Associated with this seemed to be a rise in the Prad_bolo and the Moly. The direct cause of this is not obvious. The plasmas tended also to have greater Rcur by a few mm. However, the more we pushed the current up faster the more radiation and Mo we saw. From 10 onward we had RF at the end of the shot. Good-looking H-modes were obtained on many shots, with low or absent ELMs. However, as usual, the radiation rose rapidly during ELM-free periods. Shots 12 onward steadily pushed the elongation and diversion earlier. It became increasingly clear that the Mo and radiation start to drop as the plasma diverts. Shot 13 had diversion at .22s, shot 19 at 0.2s. shot 24 at .18s. Shot 18 is a particularly clear example of how the 2pi and Mo monitor comes down after diversion, up again when the OH1 crossover bangs the plasma against the inner wall, and the down as it diverts again. This is a strong indication of the value of divertor for minimizing the Mo impurity. Shots 20 onward had reduced early gas puff. This seemed to allow the temperature to rise, as intended. However, the Mo influx was still further increased. No evidence of persistent hollow current profiles was obtained. On the contrary, ECE sawteeth were observed on essentially all shots from 0.15s onward. This is right at the top of the first fast current ramp as we are still in a slower current ramp stage. Therefore we certainly are not getting high values of q_0. We have not found a viable operating space in density between density sufficient to prevent excessive Mo influx and yet not so great as to cause rapid current penetration. It appears that one possibility of opening such a window might be to obtain diversion early before the main current rise, at low current. This will likely entail tuning of segment 1 (or at least prior to 0.1s). This could be done by putting clearin on wire 4 of segment 1. A concern will be to obtain sufficient loop voltage to give a hope of producing hollow current. Since typical resistive drop is perhaps up to 2 V at low current, we shall need at least 5V, or else get the RF on at low current. At these voltages it will be very hard to get sufficient control from the power supplies. Startup was very reliable after shot 8. But no disruptions had to be endured. Note: ECE Michelson absolute calibration was low today. Details ________ 1 Fiz Attempt at fiducial. 2 Pla TF under Hybrid control. Accidentally loaded a ramp down shot. Got 4.69T with ramp down to 2T at 1 sec. Plasma disrupts at 0.5s when B_t=3.2T and I_p = 0.8MA. Looks as if TF is doing what it is told. This disruption ought to be very interesting for analysis for safety factor variation and scaling. Programmed B_t at 151kA initially step down to 114kA at 0.2s, then flat 3 Fiz Got the TF we programmed. 151 (5.4T) and 116kA (4.2T) flat sections. Gets down to second level at about 0.5s. All three of these shots have a B_r error of about +2mT which possibly is the cuase of the fizzles. Chopper programmed to come on earlier. B_r reduced 1mT. Raise PLC TF limit to 165 in case that is influencing us. 4 Pla Disrupts at .66s because the gas programming had been messed up got a very low density about 2.e19. Chopper came on at 50ms correctly. But had a big step at segment switch. Fixed gas programming. Not sure what the chopper problem is. 5 Fiz CII .025. Bz and B_r a little too high also. -15A EF4. -.5mT Br. 6 Pla All the way. OH switch closings to .2s. TF help constant on hybrid. Lowered initial gas puff 0-30ms. 7 Fiz Initial Flux R_edge way high for no obvious reason. B_z error not so clear but the vac-field surfaces are messed up. Suspect this might be a magnetics data problem. 8 Pla All the way. Reprogrammed Ip for faster rise to 800kA at about .25s. 9 Pla Fine. Got more rapid rise. OH1 cross-over at .23s, when Ip is about .67MA. Outward excursion on Rcur at .13s to about .666m coincides with at big Mo influx. This seems to be caused by the left-over voltage programming form the segment switch. Cleaned up the OH voltage program a lot. Also started the z-cur down ramp at 70ms. 10 Pla The rcur excursion is reduced and so is the dramatic Mo influx. Program Ip to 500kA at .1s. Increased OH1 voltage at ~.1s Lowered zcur in seg 1 and 2 11 Pla Getting slightly more elongation at .1s. Increase clearin ramp; move zcur down some more. 12 Pla Diversion at .3s. Nearly diverted from .22s Increased early clearin gain. Increased rcur gain by factor 2 at 100ms. Adjust I_p gain. Set TF to -145kA. 13 Pla Diversion at .22s. Still got our Rcur problem at .11s Lost the spectroscopy camac. Delayed the zxl gain switch out of the way. Delayed the segment switch to .105s to test if the lurch is caused by it. 14 Pla Rcur lurch does not appear to follow the segment switch, although not as bad on this shot. Got a zcur lurch. Put segment switch back to .1, delayed EF1 crowbar to .11s. 15 Pla Got our Rcur lurch still. Adjusted zxu, clearin, zcur for earlier diversion. Removed Rcur derivative gain notch at .1s. 16 Pla Rcur lurch still there. RCUR gain to 7 in segment 2 matching seg 1. 17 Pla Still lurches. Increase the IP to 800kA at .2s. Increased OH1 to 250V during the ramp. 18 Pla Not such a smooth rise. We still get only Ip=670kA at crossover. Segment switch to .095s. Eased off on the Ip demand. Increased clearing to 0 at .2s. 19 Pla Rcur lurch is at the same time. Vertical lurch also. Diverted at 0.2s. Mo signal drops earlier accordingly. Returned segment switch to .1s. Reduced gas puff to 25V (from 35) @ 10-70ms. 20 Fiz Caused by segment switch set to -.1s!!! Lots of camac errors. Try again. 21 Pla Camac problems. No H-mode. Yag density is lower. No interferometer. Gas is still coming in. We are using the linear controller. reduce gas to zero for 30ms-70ms. 22 Pla slightly less diverted early. draw gas V to zero till .2s. program zxl lower during .2s region. 23 Pla More Mo observed. Program Clearin to 0 from .16s. OH2 Vs to -25V at .12s. 24 Pla Diverted at .18s. Sawteeth still quite early. Scorecard _________ Shot,Duration(s),Ipmax(MA) Shot,Duration(s),Ipmax 1 0.000 0.024 2 0.504 0.914 3 0.000 0.021 4 0.677 0.943 5 0.000 0.021 6 1.527 0.827 7 0.000 0.021 8 1.505 0.833 9 1.535 0.836 10 1.529 0.833 11 1.535 0.833 12 1.543 0.833 13 1.538 0.835 14 1.543 0.835 15 1.544 0.835 16 1.538 0.833 17 1.540 0.835 18 1.546 0.835 19 1.548 0.838No Ip 20 21 1.510 0.830 22 1.545 0.833 23 1.548 0.838 24 1.547 0.835 Finished summary for 951130 Total of 24 shots. 24 shots. 19 plasmas. 4 fizzles. 1 duds. Fizzle = gt 17 lt 50kA. |
| Physics Operator Summaries |
| Session Leader Comments |
| Physics Operator Comments |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 08:37:47:940AM | Plasma | Ok | |
| 2 | 09:06:57:670AM | Plasma | Ok | |
| 3 | 09:26:07:030AM | Plasma | Ok | |
| 4 | 09:52:37:790AM | Plasma | Ok | |
| 5 | 10:15:08:200AM | Plasma | Ok | |
| 6 | 10:28:18:450AM | Plasma | Ok | |
| 7 | 10:40:20:070AM | Plasma | Ok | |
| 8 | 11:00:19:710AM | Plasma | Ok | |
| 9 | 11:14:16:910AM | Plasma | Ok | |
| 10 | 11:50:02:010AM | Plasma | Ok | |
| 11 | 12:14:28:690PM | Plasma | Ok | |
| 12 | 12:32:31:400PM | Plasma | Ok | |
| 13 | 12:59:03:040PM | Plasma | Ok | |
| 14 | 01:21:09:580PM | Plasma | Ok | |
| 15 | 01:37:03:670PM | Plasma | Ok | |
| 16 | 02:00:47:130PM | Plasma | Ok | |
| 17 | 02:24:18:800PM | Plasma | Ok | |
| 18 | 02:39:09:740PM | Plasma | Ok | |
| 19 | 03:07:18:180PM | Plasma | Ok | |
| 20 | 03:19:19:120PM | Plasma | Ok | |
| 21 | 03:29:12:020PM | Plasma | Ok | |
| 22 | 03:53:57:050PM | Plasma | Ok | |
| 23 | 04:08:19:640PM | Plasma | Ok | |
| 24 | 04:27:21:360PM | Plasma | Ok | |