| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Ian Hutchinson |
| Physics operator(s): | Steve Horne |
| Engineering operator(s): | David Gwinn,Vinny Bertolino |
| Engineering Operator Run Comment |
| Kappa scan to determine RF antenna req'ts |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:08:600PM |
| Author: To Be Determined |
| Run Summary for 930916 Leader Hutchinson Asst Horne Engineers Gwinn, Bertolino. Plan: Objective: map out the controllable boundaries of operation at high elongation. Establish feasibility of matching shape to dipole antennas while providing adequate inboard clearance. Starting with a shot from 930817. Multiple changes. 1. Get reliable plasmas. Try to overcome the startup difficulties of the last two days and get 0.5 MA plasmas for close to 1s. 2. Use the plasmas we have had recently, with kappa increasing as a function of time. However, with the x-point kept in the correct position. Either by hand or with the zxl controller. 3. Vary the OH2/EF2 contribution to this equilibrium. 4. Do this at higher current up to 1 MA to see if this affects the stability. 5. Do it with more vertically centered plasmas. Notes on control setup. We are using OH2 average and difference controllers. The Ip controller is OH1,2,EF1,3 in proportion to the turns ratio with initial gain P=0.7. RCUR S=0.67. ZXL initially P=0. Scorecard: 2 duds 3 PS tests 9 Fizzles 2 Early VDEs 16 Full Plasmas 32 Shots Summary: We achieved a fair fraction of the plan, no serious EF2 exploration, and only 1 shot at higher current but the x-point was fairly well controlled by hand (no zxl feedback) and some higher kappas were explored. Shots 8 to 10 were influenced and 11 to 17 lost because of engineering difficulties with OH1 phasing wrong on one phase. This problem returned later in the run, but by then we had found a more robust current ramp configuration which was not quite so sensitive to the bad ripple in the current ramp so we kept going. The solution to the current ramp problem (which has been with us all this week) was primarily to lower the rate of ramp of Bz (negative). This was done by making the OH1 voltage more negative. Eventually we reached the limit of index where VDEs were experienced in the current rise. Then OH2 was used instead. It might be necessary to tweak this some more. Shape improvement to give true single null plasmas was done and the plasma was then moved back up after shot 25 to test more centered plasmas. Everything we saw is consistent with the result that kappa = 1.6 is stable but that significantly above is not. This is a problem for the dipole antenna. More analysis is needed but the first result is not encouraging. Shots after 25 had a test blip introduced at .5s where the vertical feedback gains were switched to zero for 10-20ms and then turned back on. This seems to give both the passive growth rate and the active response (after the turn back on). Sensible result were seen but will need complete analysis. Detailed log: Shot Result 1 p=1.2e-5 Null looked so-so. Dud 2 p=1.5e-5 EF4 raised to 1075 from 1050A. Dud 3 EF4 to 1100. BR to 7.0 from 7.5 mT. 400kA .8s Plasma 4 BR to 6.5 mT, Gas puff to 0s from .02, EF4 S to -1000 during shot Good shot. radiation less. Plasma 5 Gas to -.01, EF1 S to 0A from 1000 at end. Ip P=2 Plasma. 6 EF1 S to 1500from 500 @.4s, OH2 V to 25V from 20V. Gas to 30V Fizzle from ~20. Fizzle not caused by any changes. 7 Repeat. Runaways. Poor bkdwn and rise. Plasma 8 p=1.8e-5, EF1l S raised to 2700/2000 during flattop. Plasma still poor breakdown. Shape is quite good late. 9 Raise chopper bias 200V. EF4 to 1075. EF1Ls=3000 EF1U=1800 (-300), EF2U=1050 @ .15, Fizzle 10 Gas pulse to -.02. OH1 V to -60 (was -70) @ 5ms. OH2s V to -30 @ ~20ms. EFC gain P=5. Poor current rise. Higher elongation. Growing oscillations. Plasma 11 OH2s V=0 @.8s. OH1 V to -150 @24ms, and to -40 from 0 at 70ms. Fizzle OH1 PS wrong phasing discovered. Might be cause of irreproducibility. Problem happened between shots 7 and 8. 12 Repeat but p=1.6e-5. Fizzle Breakdown fields are obviously bad. This seems to be the PS. 13 Switch off one OH1 bridge, Fizzle 14 Switch off the other. TF crowbar. OH ripple terrible. Fizzle 15 Power supply test. Put OH1 V back to -70 at 24ms. Test 16 PS test after turning power to gate board off. Fixed! Test 17 OH only. Engineering failed to switch stuff back on Test 18 Runaway with disruption. Otherwise OK. Plasma 19 Gas to -0.03s, EF1U to 1000A, EF2U to 1500 for shaping later. Fizzle 20 Gas to -.025s, OH1 to -100V @5ms, OH2s to -20V @24ms. Fizzle 21 Gas to -15ms, Oh1 to -130 V @ 5ms. Ef1U to -500 EF2U to 1500 @.8s. p=1.3e-5. Current hesitation gone! Slowed the Bz rise. Plasma 22 EFC extra wire to P=10, off at .7s. OH2s to 0/-20 V at crowbar, from -20/-40 to increase early elongation. Plasma Runaways. Current pause. Horrid little 250-300 kA plasma. 23 Gas to -.02, OH1 to -160V at 5ms. Good. Plasma Switching off EFC extra gain clearly led to instability. 24 OH1 to -190V and 5ms. EFC kept on. VDE at same time. Plasma 25 EF1u reduced more later. All zcur feedback off 0.500 to 0.520s. Increased gas. All channels at 30 during shot. Plasma The growth rate should be readily measurable during the blip. 26 Put zcur to -1cm from -3 cm. EF1u reduced later some more. Plasma Runaways. Bad current rise. 27 Gas to -30ms. OH1 -220 V at 5ms. Plasma VDE at 37ms. Too much gradient. 70kA 28 OH1 to -200V, OH2s to -10 (from 0) @5ms. Oh2 kept at -10. Plasma Slow current rise. Turnoff blip killed EFC. 29 Tweaked the EFC turn off blip to make it less drastic. Gas to -25ms. Oh2 to -15V at 5ms, 5V at 24ms. No current pause. Plasma Survived the blip. 30 Raised Ip demand to 800kA. Blip to 15ms duration. Ip P set to 2 as before but at 130ms vs 200ms, i.e. feedback on earlier. Plasma OH1 has been single phasing during the last few shots. This is a VDE up at 42ms. EFC is known to cause such an excursion. 31 EFC turn-on delayed to 60ms. That fixed the excursion Plasma 680kA lasted 0.8s. 32 Increase the Ip P to 4. zcur to 0. OH1's problem got us. Fizzle |
| Physics Operator Summaries |
| Session Leader Comments |
| Physics Operator Comments |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 09:16:55:370AM | Plasma | Ok | no plasma |
| 2 | 09:30:43:670AM | Plasma | Ok | no plasma |
| 3 | 09:43:32:980AM | Plasma | Ok | 435 ka plasma |
| 4 | 09:59:44:510AM | Plasma | Ok | 440 ka |
| 5 | 10:13:54:720AM | Plasma | Ok | 490ka |
| 6 | 10:28:11:000AM | Plasma | Ok | fizzle |
| 7 | 10:44:07:520AM | Plasma | Ok | no faults |
| 8 | 10:59:41:720AM | Plasma | Ok | 460 ka |
| 9 | 11:12:36:500AM | Plasma | Ok | fizzle |
| 10 | 11:26:52:840AM | Plasma | Ok | 440 ka |
| 11 | 11:43:08:850AM | Plasma | Bad | fizzle, oh1 60 hz, looks like one gate turning on early |
| 12 | 11:56:02:900AM | Plasma | Bad | 60 hz oh1 |
| 13 | 12:11:28:180PM | Plasma | Ok | fizzle, oh1 east convertor only ok |
| 14 | 12:27:47:500PM | Plasma | Ok | plasma shot OH1 60 Hz on pulse tf c-bar. |
| 15 | 12:46:06:310PM | Test | Bad | OH1 60 Hz |
| 16 | 01:30:40:850PM | Test | Ok | OH1 good |
| 17 | 01:40:54:270PM | Plasma | Bad | oh1 only all other supplies off in camac |
| 18 | 01:51:01:200PM | Plasma | Ok | no faults |
| 19 | 02:04:14:380PM | Plasma | Ok | no faults |
| 20 | 02:23:57:730PM | Plasma | Ok | no faults |
| 21 | 02:36:34:030PM | Plasma | Ok | no faults |
| 22 | 02:50:22:510PM | Plasma | Ok | oh1 >1scr not conducting late in shot |
| 23 | 03:02:55:270PM | Plasma | Ok | no faults |
| 24 | 03:15:56:520PM | Plasma | Ok | no faults |
| 25 | 03:29:55:400PM | Plasma | Ok | no faults |
| 26 | 03:44:14:450PM | Plasma | Ok | no faults |
| 27 | 03:58:44:740PM | Plasma | Bad | fizzle -- oh1 60Hz again |
| 28 | 04:11:52:390PM | Plasma | Ok | oh1 still has 60Hz |
| 29 | 04:27:18:740PM | Plasma | Ok | oh1 60Hz |
| 30 | 04:40:23:630PM | Plasma | Ok | oh1 >1scr not conducting, 60Hz |
| 31 | 04:55:21:780PM | Plasma | Ok | no faults -- 60Hz on oh1 still |
| 32 | 05:08:13:370PM | Plasma | Ok | oh1 60Hz |