| Miniproposals | ||||||||||
|
| Operators | |
| Session leader(s): | Steve Wolfe |
| Physics operator(s): | Ron Parker |
| Engineering operator(s): | Andy Pfeiffer |
| Engineering Operator Run Comment |
| MP298a C-Mod Startup SL: Wolfe PO: parker Magnetics guru: granetz |
| Session Leader Plans |
| Entered: Mar 15 2007 06:57:58:983PM |
| Author: Steve Wolfe |
Session Leader plan for Friday, March 16,2007 MP298a C-Mod Startup SL: wolfe PO: parker Magnetics guru: granetz On Thursday we got through all the power system tests but did not get to try for plasma for the last shot because the PF breaker would not close. This is expected to be replaced before we start on Friday. Numerous flux loops on the same connector (B-bottom#1) were observed to be intermittent or bad during the tests on Thursday. Of particular concern is f23, which actually gave the wrong polarity on the TF tests, and f19, which shorted or opened during several pulses. Also having shown intermittent problems on Tuesday and Thursday were f01, f21, and probably f25. All of these are on the same feedthrough. On several shots f01 and f21 were actually showing zero signal (bit noise). Either prior to the run or after the first shot, Bob will try removing and replacing the air-side connector on this feedthrough. If this fixes the problem, we will proceed with a normal first day of plasma ops, with the main goal of getting plasmas and beginning cleanup. If one of these flux loops remains bad, we will need to modify a lot of control software to accommodate it. Ian has pointed out that we can reduce the weight of any signal to zero (or negligible) in the fluxihh routines by making a new IAN tree using a routine derived from /home/hutch/idld/wtdfil.pro , and we can modify the routine ianup (in /usr/local/cmod/codes/pcs/user_algs/contihh/) to call a different tree number (presently using IAN_001). This is preferable to trying to generate a synthetic value for a missing loop in magrawmesh, especially for loops that do not have a true synthesis out of other signals, including f23 and f19. Note that the Pcs directories are under CVS now, so in principle all software damage is reversible. In addition to the fluxihh predictors, we would need to change recon_vac and filament predictors. The filament predictor (ZCUR) just requires turning off the relavant input nodes. The RECON_VAC will need a new MHD_STATIC tree, created using fix_recon_vac.pro. I'll need to consult my notes from May 19, the last time we had to do this. I estimate that replacing and testing software would take several hours. If more than one loop remains bad, then we should probably get ready to break vacuum and repair the inner (vacuum-side) connector. Assuming we overcome the flux loop problems, we will be carrying out a program similar to that outlined in the PO_PLAN for 1070312: Load from 1060613015 (first shot after last field reversal to normal direction, LSN, 5.4T). Probably get a dpcs config warning - verify dpcs_fizzle source shot for normal current direction (1060308026). Reduce Ip from 1MA to 800kA, change NL04 from 9e19 to 6e19. TCI seemed to be working during the tests; we assume it will work for plasma. If TCI not working, then turn off NL04 feedback and set Segment 2 pre-programmed PG3 to zero volts, after an initial puff. Tune startup as required, try for plasma shots to clean up machine. Consider lower Ip target (600kA) depending on results. Tuning is not going to be helped by the fact that we have not yet figured out how to fix MFLUX if we are actually forced to lose a flux loop. If we actually get plasmas, one thing to monitor carefully is the f11 loop, which appeared to have a real in-vessel failure in May-June last year, but then recovered and has not appeared to have a problem during tests so far. If we can recover this loop, then we should. Even if we can't, we should consider the option of making a new ian_tree for this case as well. |
| Physics Operators Plans |
| Entered: Mar 15 2007 06:47:59:653PM |
| Author: Ron Parker |
| MP298a C-Mod Startup
Run begins at 9:00 and ends at 17:00 Expect some delay in AM due to EF breaker replacement Also, condition of magnetic pickups will be assessed -- may require longer term maintenance. The main purpose of this run is to develop startup plasmas to initiate the new run campaign. Gas Setup: ---------- fill B-top 6psi D2 Hybrid Enabled(PG4) fill B-side lower 1 psi Argon Hybrid Disabled (PG1) fill B-side Upper as is, Hybrid Disabled (PG2) fill B-main 40psi D2 Hybrid Enabled(PG3) NINJA pumped out, Disabled The following engineering setup should be used: ******************************************************************** Power Systems as on 1060613015 Acoil Configuration +Dtop -Dbot -Jtop +Jbot, Enabled Check on TF invert time relative to back-porch programming (see Irby) Gas setup as above, except ENABLE B-top and B-main. TORVAC: Pumping Gate valves open Enable the following gate valves and shutters, assuming no vacuum problems: ECE, Z-bolo, VUV ICRF setup: None ******************************************************************** Physop plan for MP#298a: ------------------------ Load from 1060613015 (first shot after last field reversal to normal direction, LSN, 5.4T). Probably get a dpcs config warning - verify dpcs_fizzle source shot for normal current direction (1060308026). Reduce Ip from 1MA to 800kA, change NL04 from 9e19 to 6e19 (assuming TCI is working); if TCI not working, then turn off NL04 feedback and set Segment 2 pre-programmed PG3 to zero volts, after an initial puff. Tune startup as required, try for plasma shots to clean up machine. Consider lower Ip target (600kA) depending on results. |
| Session Leader Summaries |
| Entered: Mar 16 2007 01:12:21:840PM |
| Author: Steve Wolfe |
Session Leader summary for Friday, Mar 16, 2007 MP#298a "C-Mod Startup" SL: wolfe PO: Parker EO: Pfeiffer The goal was to get plasmas. We got a breakdown flash on shot#1, then ran into a new problem as well as some previously existing ones, and shut down after two more test shots. I suppose the run was a partial success. Prior to the run (this morning and last night), the PF breaker responsible for ending yesterday's run was replaced, and Bob Granetz removed the air-side connector on the B-bottom feedthrough that has the intermittant flux loops on it. He verified that f19 and f23 have shorts to the vessel and the other 8 loops do not. He then replaced the connector. The access control programming that caused a breach on opening the big door to the cell under controlled access was not fixed, limiting cell access to the lower doorway. Shot#1 was a full-up plasma attempt, based on 1060613015, modified as noted in the SL_PLAN and PO_PLAN. This resulted in a fizzle with good early breakdown, but no current pickup. The apparent flux plots were horrible, and at least two flux loops, F19 full loop and F15 partial, were clearly bad, not coming back to zero after the shot and having the wrong magnitude. We also encountered a problem with the RSVIEW/PLC programming for the OH2 supplies. The delay on the fizzle detector turns out not to have been implemented for the OH2's, although it is there and functional for the TF and OH1. On the OH2 screen the box to enter the delay is there, but it doesn't connect to any PLC logic. Yuri and Joe will need some time to re-implement this feature. Because we rely on this delay to force the oh's to reverse swing in order to keep the oh coax resistance down we will not attempt any more plasma shots until this programming is done and tested. We ran two additional test shots in order to check the magnetics. Shot#2 was a 10kA oh1 PLC test, and on this shot f25 and f21 were broken, and f15 and f19 looked ok (note that f15 is on a different connector). Shot#3 was a 100kA TF PLC test, and on this shot F21 was completely dead (bit noise) but f19 looked ok, and the other loops also looked ok. In particular, F23 which had reverse polarity on yesterday's TF tests looked normal. We are debating whether we should break vacuum in order to attempt a repair of the magnetics connector on the vacuum-side of B-bottom #1 feedthrough. Aside from the loss of time, the problem is that we aren't sure how to effect a reliable repair, and also don't know how to verify any repair without closing up and trying to run again. Note that after shot#3 on which f21 was either open or closed throughout the shot, Bob measured it as being continuous back to the patch panel, and not shorted to the machine. There is also a risk of damaging other magnetics signals in the course of accessing this connector. The alternative is to continue next week in the hope that additional shots will reseat the connectors in a more robust state. While this sounds like it would be a violation of Second Law, it has happened before. So far f11 has not had a problem this year (or for the last weeks of operation last year). |
| Physics Operator Summaries |
| Entered: Mar 16 2007 12:06:57:200PM |
| Author: Ron Parker |
| Run aborted after first shot, which did in fact breakdown.
Two problems: Fizzle detector not driving OH 2 into reverse swing. This is a software problem that needs some time to repair. Run aborted due to the time this will take. Second problem: Several flux loops flakey. This makes t~ 0 flux plots unreliable. Not clear if this can be fixed without opening. |
| Session Leader Comments | |||
| Mar 16 2007 08:13:52:797AM | Steve Wolfe | Note to physop: Irby says not to bother with backporch on the TF unless we see something on the scanner data, which looked good during TF tests yesterday. Therefore, the timings from the 1060613015 engineering setup (no backporch, f.d=[.08,1.8],tf invert=1.6) are acceptable. | |
| Mar 16 2007 09:08:38:750AM | Steve Wolfe | Bob reports that he did in fact remove and replace the air-side connector last night after the run. That is probably the best we can do without a vacuum break, unless we can find a problem back at the rack (unlikely). We'll see. JT is arranging for some camera views. We should have wide1 but not wide2 today. | |
| Mar 16 2007 09:45:13:953AM | 1070316001 | Steve Wolfe | Shot#1: The bigboard display is not showing the toroidal field. I don't remember if it did yesterday or not. |
| Mar 16 2007 10:02:47:527AM | 1070316001 | Steve Wolfe | Shot#1: Plasma attempt. Got a flash! Null looks terrible. f19 is clearly bad. f15 partial also didn't come back to zero. Did not get double swing on the oh2's; demand is there, and the fizz delay is supposedly at 500 on them as well as the oh1, which did swing. |
| Mar 16 2007 10:25:08:827AM | 1070316001 | Steve Wolfe | For Shot#2: We will run an oh2u/l test under hybrid control PLC limits will be +/- 12kA, we draw +/- 10kA on the avg oh2 current. Gains on zcur and rcur to zero, P gain extended to 2sec on OH2 and Br0 There will be a brief cell access for Ron to check the port heaters on the bottom, and Bob is going to check on the rogowski signal. We had a small leak because the vessel got over-cooled. Yuri is investigating the PLC code on the fizzle delay for the OH2's. |
| Mar 16 2007 11:21:20:670AM | 1070316001 | Steve Wolfe | Shot#2: Change of plans. Yuri has found that the fizzle delay code is missing from the OH2 PLC/RSView program. It will not be fixed today. We will have a 10kA PLC test to check the current state of the magnetics. Then we will probably do a 100kA TF test, and then depending on results, we will shut it down for the day. |
| Mar 16 2007 11:26:19:967AM | 1070316001 | Steve Wolfe | I have saved the test shot that Ron and I prepared for the oh2's as 1070316666 in /home/wolfe/pcs_scratch We may use it next week when we test the f.d. |
| Mar 16 2007 11:42:43:950AM | 1070316002 | Steve Wolfe | Shot#2: Change of plans. Yuri has found that the fizzle delay code is missing from the OH2 PLC/RSView program. It will not be fixed today. We will have a 10kA PLC test to check the current state of the magnetics. Then we will probably do a 100kA TF test, and then depending on results, we will shut it down for the day. Note that the shot programming for the oh2 double swing test will be in this pulse file, as well as my pcs_scratch dir. This time, f25 is broken! f19 looks ok. So does f15. F21 has a jump; |
| Mar 16 2007 11:57:53:793AM | 1070316003 | Steve Wolfe | Shot#3: 100kA TF test. This time f21 is zero, but f23 is normal polarity and f19 is ok again. |
| Physics Operator Comments | |||
| Mar 16 2007 09:02:33:970AM | 1070316001 | Ron Parker | Machine ready at 9:00. JT requires access, short delay. |
| Mar 16 2007 09:21:42:873AM | 1070316001 | Ron Parker | Flash! Probably downhill from here. |
| Mar 16 2007 09:41:28:437AM | 1070316001 | Ron Parker | OH 2 did not double swing as it should have. Delay while this is troubleshot.
Also, calculated field null is not consistent with the fact that it broke down. |
| Mar 16 2007 11:20:43:293AM | 1070316001 | Ron Parker | The software problem causing the failed double swing has been found, but will take a while to fix. Our plan is to do some magnetics tests, after which we'll scrub the run. Happy shoveling. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 09:17:43:920AM | Plasma | Ok | |
| 2 | 11:37:07:230AM | Test | Ok | oh1 10Ka magntest |
| 3 | 11:52:29:217AM | Test | Ok | TF 100KA magnetics |
| System Availability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Mar 16 2007 09:17:08:873AM | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Mar 16 2007 11:36:32:420AM | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||