| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Robert Granetz |
| Physics operator(s): | Joseph Snipes |
| Engineering operator(s): | Steve Fairfax |
| Engineering Operator Run Comment |
| tests for real time data |
| Session Leader Plans |
| Physics Operators Plans |
| Session Leader Summaries |
| Entered: Jul 7 2004 03:49:06:663PM |
| Author: To Be Determined |
| Run summary for 11 May 1993: ---------------------------------------------------------------------------- The first six shots were engineering tests to verify that problems which cropped up on Saturday had been fixed (OH2 commutation, TF crowbar). These tests were successful. Shots 7 & 8 were our first attempts at a null today, with a toroidal field of 4.4 T. There was an extended delay due to difficulty recooling the TF magnet. A faulty GN2 pressure transducer was tracked down and replaced. From about shot 9 on, all the power supplies work extremely well. Ditto for the computer systems, alternator, etc. This gave us the opportunity to "tweak" the programming of PF coil currents, gradually resulting in an improving null structure. The hybrid computer controlled the OH1, OH2's, EF1's, and EF4 using feedback on the coil currents (although PLC current limits were being reached on some supplies), and EF3 was controlled with simple voltage preprogramming. To avoid radial magnetic fields, the OH2 and EF1 coil pairs had to be well-balanced (i.e. equal to within 50-100 amps.) This balance seemed to drift apart over several shots, even though the hybrid was feeding back on the current. (In addition, it seems that the calibration of the bus rogowskis is only good to about +/-100 amps (that's ~0.5%). It also became clear to me that in order to get the actual PF currents within a few percent of the demand currents (even for near steady state conditions), the voltage preprogramming has to be very consistent. (Looking back on my notes, I see I made the same observation a week ago when testing out individual supplies.) As the nulls improved, we briefly tried varying the H2 gas pressure by a factor of 2 in both directions (4e-5 --> 2e-5 --> 8e-5 --> 4e-5 again) without achieving breakdown. Small adjustments in the vertical field continued to be made using EF4 and then EF3. By shot 19, the null structures were as good as we had last year, with a somewhat higher toroidal field of 4.4 Tesla. Loop voltage was about 13-14 volts. Shot 23 Shot 23 was perhaps even better, but still no flash. On the continued advice of several people, including our visiting `experts' from D-IIID, the gas pressure was raised back up to 9e-5 torr for shot 25, resulting in the first plasma of 1993. The last 3 shots (26, 27, 28) all had plasma, and the TF was gradually raised to 5.0 Tesla, while the SCR re-close was moved from 25 to 40 msec after commutation. Fill pressure was also slowly raised (0.9 --> 1.0 --> 1.1e-4) The plasma current was only 5-10 kA. Radiated power was 15-20 kW. The horizontal tomography array and other visible detectors indicate that the light is moving around inside the chamber. On the last shot (#28) the EF1's suddenly became unbalanced by 500 amps for some unknown reason, resulting in a relatively large radial field and poor null structure. Interestingly, we still had breakdown, although just barely, with only 10 V of loop voltage. It seems that after the first few breakdowns have been achieved, it becomes easier and easier to get plasmas. Special thanks to Steve Fairfax and the entire engineering crew for getting the power supplies to perform so well. The all-too-necessary "tweaking" process is impossible without reliable operation. ************ Suggested improvements/changes for next run ******************* Improve vertical field programming after the flash using EF3. I think this might be made simpler if we turn on hybrid feedback on Bz. Improve voltage preprogramming of some of the PF supplies so that they don't hit the PLC limits, and so that the hybrid feedback control of coil currents becomes even more accurate. Move EF3 ground back to the supply cabinet. |
| Physics Operator Summaries |
| Session Leader Comments | |||
| May 11 1993 08:55:48:200AM | Joseph Snipes | Starting with test shots to check the fixes made yesterday. Then, we expect
to start trying to get plasmas. | |
| May 11 1993 08:56:50:430AM | Joseph Snipes | Real time scope display kludged into rogowski inputs to hybrid.
| |
| May 11 1993 09:59:55:690AM | Robert Granetz | Run 930511. Session leader: R. Granetz
Still trying to get breakdown. First 6 shots were engineering tests of OH1 and OH2U&L commutation switches. Everything worked. First plasma shot: 007 This has been programmed to be a repeat of 930508003. | |
| May 11 1993 10:03:36:150AM | Robert Granetz | Run 930511. Session leader: R. Granetz
Still trying to get breakdown. First 6 shots were engineering tests of OH1 and OH2U&L commutation switches. Everything worked. First plasma shot: 007 This has been programmed to be a repeat of 930508003. | |
| May 11 1993 09:57:18:490AM | 930511007 | Joseph Snipes | Commutation resistors at start of run are:
OH1 156 mohm OH2U 45.97 mohm OH2L 46.07 mohm EF1U 200.8 mohm EF1L unlabeled (analogous tap point as EF1U Shots 1-6 engineering test shots Shot 7 Repeat of 930508003 on Saturday except B_T = 4.4 T OH1 voltage pre-programmed current limit = 15.5 kA OH2U+L current feedback tau = 0.1 sec EF1U+L " " " " EF3 + 4 voltage preprogrammed |
| May 11 1993 10:08:44:110AM | 930511007 | Robert Granetz | Semi-okay. OH1 and EF1U&L power supplies faulted after re-close.
OH2 worked okay, except the current drooped slightly about at -0.1 sec. Btor=4.4 T. |
| May 11 1993 10:14:04:540AM | 930511007 | Robert Granetz | Note: The PF breaker pulled out at t=-0.1 s. This presumably caused the droop
in OH2's and EF1's, and the slow coastdown of the OH1 and EF1's. |
| May 11 1993 10:59:35:340AM | 930511007 | Joseph Snipes | Apparent leak in the LN2 cooling since the TF coils are not cooling as
expected. Short break to check out the problem. |
| May 11 1993 11:42:11:840AM | 930511008 | Joseph Snipes | Repeat of pulse 7 except voltage programming on OH1 reduced.
All breakers came out before commutation. Power supplies failed after reclosing the switches. |
| May 11 1993 12:07:34:160PM | 930511008 | Joseph Snipes | Repeat of pulse 7 except voltage programming on OH1 reduced.
All breakers came out before commutation. Power supplies failed after reclosing the switches. Slight current imbalance in OH2's. |
| May 11 1993 12:26:40:350PM | 930511008 | Joseph Snipes | Equalizing OH2's, making their flattops 0.5 s long, reprogrammed voltage on
OH1 to have an inversion. Hit integral limit on OH1 |
| May 11 1993 12:27:09:110PM | 930511009 | Joseph Snipes | Equalizing OH2's, making their flattops 0.5 s long, reprogrammed voltage on
OH1 to have an inversion. Hit integral limit on OH1 |
| May 11 1993 12:56:44:600PM | 930511010 | Joseph Snipes | Shot 10
Slightly lower OH2 voltage programming and made them equal. OH1 current feedback turned on. |
| May 11 1993 01:02:16:030PM | 930511010 | Joseph Snipes | Shot 10
OH2U+L switch reopened at -1.5 s due to OH1 turning on hard. |
| May 11 1993 01:21:45:320PM | 930511011 | Joseph Snipes | Shot 11
Delayed OH1 turn on by 15-20 ms. Reasonably good shot, but too much VF. |
| May 11 1993 01:40:16:660PM | 930511012 | Joseph Snipes | Shot 12
Turn on current feedback control of EF4. Initial rate of rise of OH1 reduced. Dropped voltage demand a little on OH2's. Shot improved a bit, but the null still needs some work. |
| May 11 1993 01:54:31:440PM | 930511013 | Joseph Snipes | Shot 13
Raised EF4U by 100 A. Invert OH1 a little harder. |
| May 11 1993 01:59:48:460PM | 930511013 | Joseph Snipes | Shot 13
Reasonably good shot, somewhat improved, but still needs tweaking to get the null right. |
| May 11 1993 02:10:00:110PM | 930511014 | Joseph Snipes | Shot 14
Add another 100 A to EF4U on programming demand. Lower EF1's by 500 A each. |
| May 11 1993 02:18:19:170PM | 930511014 | Joseph Snipes | Shot 14
Not bad. HUTCH thinks we need to reduce the radial field. |
| May 11 1993 02:26:47:250PM | 930511015 | Joseph Snipes | Shot 15
Increase gains on EF1's and EF4's to try to match programming. Extended front porch on OH2's. |
| May 11 1993 02:36:07:650PM | 930511015 | Joseph Snipes | Shot 15
Looks good. Need to tweak down the horizontal field. |
| May 11 1993 02:39:52:250PM | 930511016 | Joseph Snipes | Shot 16
Lowering EF1L by 100 A. Raising gain a little more on EF4 to try to match programming. |
| May 11 1993 02:45:46:290PM | 930511016 | Joseph Snipes | Shot 16
Looks reasonably good, but still too much horizontal field. |
| May 11 1993 02:53:29:560PM | 930511017 | Joseph Snipes | Shot 17
Lowering EF1L by 100 A to reduce horizontal field. Large errors after 2.5 s are causing cross-talk on the hybrid. Current programming changed after 2.5 s to more closely match what we are getting. |
| May 11 1993 02:58:08:800PM | 930511017 | Joseph Snipes | Shot 17
Reasonably good null. Still, no breakdown. |
| May 11 1993 03:07:19:920PM | 930511018 | Joseph Snipes | Shot 18
Vertical field to ramp negative more slowly by changing EF3 voltage pre-programming. |
| May 11 1993 03:14:09:610PM | 930511018 | Joseph Snipes | Shot 18
Not bad but still no breakdown. |
| May 11 1993 03:21:25:910PM | 930511019 | Joseph Snipes | Shot 19
Reducing fill pressure to 2 x 10**-5 torr. Tweaking EF1U up by 50 A. |
| May 11 1993 03:26:27:740PM | 930511019 | Joseph Snipes | Shot 19
Still no breakdown. Tony Taylor suggests increasing the fill pressure. |
| May 11 1993 03:32:59:440PM | 930511019 | Joseph Snipes | Shot 19
Best null yet! |
| May 11 1993 03:36:55:980PM | 930511020 | Joseph Snipes | Shot 20
Increasing fill pressure to 8 x 10**-5 torr. |
| May 11 1993 03:46:31:990PM | 930511020 | Joseph Snipes | Shot 20
Still no breakdown. |
| May 11 1993 03:56:44:870PM | 930511021 | Joseph Snipes | Shot 21
Lowering OH2L by 200 A to rebalance upper and lower. Fill pressure returned to 4 x 10**-5 torr. EF3 voltage pre-programming lowered by 25 V. Gains on EF1's changed from 2 to 2.5. |
| May 11 1993 04:01:03:470PM | 930511021 | Joseph Snipes | Shot 21
Still no breakdown. |
| May 11 1993 04:06:46:770PM | 930511022 | Joseph Snipes | Shot 22
OH1U increased 100 A, EF4 increased by 100 A. |
| May 11 1993 04:25:02:300PM | 930511022 | Joseph Snipes | Shot 22
Still no breakdown. Fair null. Some PF power supplies faulted after commutation. Discovered that we were programming EF4L but feeding back on EF4U, so feedback was unaffected by our changes in pre-programming. |
| May 11 1993 04:27:51:200PM | 930511023 | Joseph Snipes | Shot 23
Feedback is now connected to EF4L. OH2U increased 100 A, OH2L decreased 100 A. |
| May 11 1993 04:34:50:660PM | 930511023 | Joseph Snipes | Shot 23
Best null yet!! Still, no breakdown. |
| May 11 1993 04:41:11:080PM | 930511024 | Joseph Snipes | Shot 24
Increased EF4 current by 50 A, increasing vertical field. Decreased the rate at which EF3 drives vertical field negative again. Raised current limit on OH2L by 200 A. |
| May 11 1993 04:50:32:990PM | 930511024 | Joseph Snipes | Shot 24
Pretty null!! Still, no breakdown. Seems like we need more Vloop. |
| May 11 1993 04:55:49:440PM | 930511024 | Joseph Snipes | Shot 25
Increase fill pressure to 9 x 10**-5 torr. First Plasma!!!! |
| May 11 1993 05:09:46:300PM | 930511026 | Joseph Snipes | Shot 26
Switches set to close at 40 ms rather than 25 ms. TF increased about 7%. |
| May 11 1993 05:17:17:820PM | 930511026 | Joseph Snipes | Shot 26
Second plasma!!!! Almost 8 kA of plasma current for about 15 ms. |
| May 11 1993 05:20:28:760PM | 930511027 | Joseph Snipes | Shot 27
Fill pressure increased to 1 x 10**-4 torr. Voltage pre-programming on EF3 has been adjusted so that B_z does not go as negative as it has been going. |
| May 11 1993 05:26:00:850PM | 930511027 | Joseph Snipes | Shot 27
Third plasma!!! We've set things up pretty well for the boss tomorrow. Increasing fill pressure makes plasma break down earlier now at about 7 msec and lasts about 23 msec. |
| May 11 1993 05:32:26:450PM | 930511028 | Joseph Snipes | Shot 28
EF3 programming changed to agree better with shot 26 so that B_Z goes negative a little faster. B_T increasing by another 7.5% to about 5 T. |
| May 11 1993 05:35:57:720PM | 930511028 | Joseph Snipes | Shot 28
Wimpy plasma compared to the first three. Still, not a bad day's work. |
| May 11 1993 05:42:27:590PM | 930511028 | Joseph Snipes | Shot 28
EF4 power supply fault resulted in the wimpy plasma. |
| May 11 1993 05:43:52:740PM | 930511028 | Joseph Snipes | Shot 28
Correction!!! EF1 power supply fault resulted in the wimpy plasma. |
| Physics Operator Comments |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 07:35:40:480AM | Test | Ok | test for real time data |
| 2 | 07:41:29:290AM | Test | Ok | test for real time data |
| 3 | 07:56:50:100AM | Test | Ok | test for real time data |
| 4 | 08:02:44:820AM | Test | Ok | test for real time data |
| 5 | 08:47:44:450AM | Test | Bad | Test of Cop Bank Voltage feedback OK; Magnet resistance and ground falut |
| 6 | 09:10:18:020AM | Test | Ok | OH1 closed circuit and oh2 commutation test OK |
| 7 | 09:53:49:730AM | Plasma | Bad | PF circuit breaker tripped; no power supply faults |
| 8 | 11:24:42:650AM | Plasma | Ok | TF breaker tripped late in inversion |
| 9 | 12:11:23:640PM | Plasma | Ok | All suplies OK; low AC tripped units after bilp |
| 10 | 12:37:06:220PM | Plasma | Ok | Alternator field breakersa tripping; reset limits from 1200 to 1800 amps |
| 11 | 12:59:56:750PM | Plasma | Ok | All power supplies OK |
| 12 | 01:27:44:810PM | Plasma | Ok | All power supplies OK; physics operator tuning null |
| 13 | 01:49:56:000PM | Plasma | Ok | Good shot with more TF inversion; still no flash |
| 14 | 02:07:45:400PM | Plasma | Ok | TF inverting harder; still no flash |
| 15 | 02:22:51:920PM | Plasma | Ok | TF inversion margin now 35 degrees; everything OK, still no plasma |
| 16 | 02:37:38:300PM | Plasma | Ok | All power supplies OK |
| 17 | 02:52:20:950PM | Plasma | Ok | "Good Null" sez physics, but still no plasma. All power supplies OK |
| 18 | 03:06:19:560PM | Plasma | Ok | OH1 1 scr out; all others OK; no plasma |
| 19 | 03:20:37:100PM | Plasma | Ok | gas fill changed to 2e-5; still no plasma |
| 20 | 03:35:59:550PM | Plasma | Ok | Null is so good Grad-Shafranov solver shows 3 kA plasma current but stil |
| 21 | 03:52:52:030PM | Plasma | Ok | EF1L commutation fault |
| 22 | 04:06:02:460PM | Plasma | Ok | ef1l crobar after blip |
| 23 | 04:25:20:380PM | Plasma | Ok | EF1L Commutation fail after blip; excellent null; no flash |
| 24 | 04:39:41:110PM | Plasma | Ok | No faults; no plasma |
| 25 | 04:52:45:440PM | Plasma | Ok | flash! |
| 26 | 05:08:51:680PM | Plasma | Ok | Flash; about 10 kA |
| 27 | 05:19:13:070PM | Plasma | Ok | flash; 136 kA in TF; no faults |
| 28 | 05:31:16:520PM | Plasma | Ok | flash despite 500 amp imbalance in EF1 curreents at commutation |