| Miniproposals | ||||||||||
| ||||||||||
|
| Operators | |
| Session leader(s): | Istvan Cziegler,Steve Wolfe |
| Physics operator(s): | Robert Granetz |
| Engineering operator(s): | Sam Pierson,Bill Parkin,Bill Byford |
| Engineering Operator Run Comment |
| MP # 362 & 487 S/L Cziegler (487) Wolfe (362), P/O Granetz, E/O Pierson |
| Session Leader Plans |
| Entered: May 16 2007 03:24:58:127PM |
| Author: Steve Wolfe |
Session Leader Run Plan for Second half of 1070517
MP#362 "Shaping and Control Development for high elongation, high I_N"
SL: wolfe
The main goal of this experiment is to develop an ITER-like shape for use in
Marmar's proposed high current, high performance experiments at 5.4T. Shaping
requirements may be somewhat relaxed from the original MP goal, which was a
match to the nominal ITER shape. The target is 1.6MA at 5.4T, which
corresponds to I_N=I/aB = 1.35, marginally less than the ITER value of 1.4.
Previous runs on MP#362 were 1031126 and 1040127. The best case to date is
1040127010, which succeeded in getting to 1.6MA at kappa=1.86, but disrupted
at 1.15sec with an apparent vertical instability. These shots are fighting three
disruptive effects: low q, vertical instability and locked modes. The approach
will be to address the locked mode issue by implementing the B21_PROJ feedback
algorithm and to sneak up on the other limits by starting with an equilibrium
that is less demanding than 1040127010 and working our way up.
Initial PCS setup (note - Setup is saved in /home/wolfe/pcs_scratch shot 1070517000 segment 2):
------------------
Aiming for 1.35MA, elongation~1.7 to start.
Start with programming from 1040127010 segment 2 (and recalculate magnetic
predictors, including Ip; acknowledge warnings about DPCS tree features).
Put B21_PROJ feedback on wire 9, in place of "bolofeed":
1. In EDIT_WIRE screen, Load wire from a suitable recent source
2. On V#01-16 screen, zero the ACOIL_CUR waveform
3. Integrator Gates: Change start of C gate to 0.2sec
Move fast ZCUR feedback from wire 10 back to wire 3:
1. In Edit_Wire screen on wire 3, load wire from wire 10
2. Change derivative gain from to 5.0, P gain to 2.5
3. Zero wire 10.
Reduce Ip program from 1.65 to 1.35
Reduce CLEARIN program from +.005 to -0.005 (back off elongation)
Increase IC_EF4U program from 2000 to 3500A
Change Ref 9-12 from "Ip" to "10"
Shot sequence:
--------------
1. Run the above setup and tune. Since the starting point is so ancient there
will likely be adjustments required to CLEARIN and the xpoint locations. I
wouldn't be surprised if the first try came out limted. Check q95, n/ncrit,
locked mode signatures, flux swing (particularly oh1). Things to adjust,
depending on results:
a) Ip (reduce for higher q95), Ip gain switch at 0.6 sec (delay it for volt demand)
b) CLEARIN, RCUR for gaps
c) ZCUR (make negative to reduce total elongation)
d) ZXU (make more positive to reduce total elongation)
e) nl04 (increase to help locked mode, decrease for volt-sec); suggested value=1.2e20?
Tuning the starting point should take 3-5 shots.
2. Begin increasing current, elongation, keeping q95>3. Take ten shots or
remainder of run, trying to get up to at least 1.5MA.
Steps:
A) Ip=1.5MA, kappa~1.75
B) Ip=1.6MA, kappa~1.85
Things to try (hopefully won't need all of them):
Equilibrium:
a) Raise IC_EF4U program to 4000 (or even 4500) as current increases
b) Slow down or delay CLEARIN, elongation changes relative to current rise.
c) Avoid voltage saturation due to too rapid current rise.
d) If we're running out of volt sec, try adding some ICRF, but we probably
don't want to push power if it's faulting, and we probably don't need a
lot of L-H-L transitions.
e) Consider shortening the Ip flattop to avoid power supply current limits
Stability:
f) increase slow ZCUR derivative gain (from 0.7 to 2.5, probably not all at
once), monitoring vertical stability traces - response to slow branch
g) increase P gain on IC_EF4, and/or decrease P gain on EF2_by_EF4 to try to
suppress EF2 oscillation contribution to slow branch of Zcur instability
h) Tweak fast ZCUR gains to suppress fast branch of vertical mode, if needed
Locked modes:
i) Watch out for early locked modes. Consider slowing down current ramp if
these appear to be present
j) Increase density to combat locked modes later in pulse.
|
| Entered: May 16 2007 05:53:45:200PM |
| Author: Jim Terry |
| This part of the run is described in MP 487 - "Effect of the Plasma
Configuration on Edge Turbulence". The purpose of this set of discharges is to characterize the broadband edge turbulence spectrum of L-mode and pre-H-mode plasmas in different (LSN, USN, DN) magnetic configurations. We have observed previously that this broadband turbulence feature chanes character depending on magnetic configuration and exists at and insde of the separatrix. One of the things we wish to measure is the radial profile of the feature in each of the LSN, USN, DN magnetic configurations. SETUP: Set up NINJA with 10 psi of D2 puff thru capillaries "A-B lim" and "K-bolo" (the one for the "X-pt" camera view) NINJA timing is a trigger at 0.35 s with a 0.25 s duration The plan has 3 parts: Part i - "Radial Scan of Turbulence Feature by Scanning the Outer Gap" 2 good shots scanning the outer gap from 3 cm to 0.3 cm starting at 0.5 sec and reaching 0.3 at 1.0 s. Do this in LSN, USN, and DN. NL04=1.1e20 Load shot 1070511025 for the LSN and put in the gap scan (and Bt=5.6 T) Load shot 1070511022 for the USN and put in the gap scan (and Bt=5.6 T) Load shot 1070511029 for the DN and put in the gap scan (and Bt=5.6 T and remove small SSEP scan Part ii - "Response of Turbulence Feature upon L-to-EDA H-mode Transition" 2 shots (LSN) with a steady EDA H-mode produced by turning on 2MW of ICRF at 0.75 s for 0.5 sec. NL04 (target)=1e20. Outer gap=1.2 cm. Observe QCM feature during H-Mode phase. Part iii - 'Response of Turbulence Feature to SSEP scan. Set outer gap to 1.2 cm and scan the configuration from LSN thru DN to USN in 0.4 s starting with SSEP at -1.0 cm and ending with 1.0 cm. Set up fast-diode view as below: data window was from 0.50 to 1.024 sec diode # trans fiber # view F/no 1 GPI #3 OUTER SOL 36 F/4.0 2 GPI #2 OUTER SOL 22 F/4.0 3 GPI #1 OUTER SOL 5 F/1.4 4 GPI #4 OUTER SOL 23 F/5.6 5 GPI #5 OUTER SOL 10 F/5.6 6 #39 OUTER SOL 9 F/5.6 7 #40 OUTER SOL 7 F/1.4 8 #41 OUTER SOL 11 F/5.6 9 #42 OUTER SOL 12 F/5.6 10 #43 OUTER SOL 13 F/5.6 11 #44 OUTER SOL 14 F/8.0 12 #45 OUTER SOL 6 F/1.4 13 #46 OUTER SOL 29 F/5.6 14 #47 OUTER SOL 31 F/2.8 15 #48 OUTER SOL 32 F/2.8 16 #49 OUTER SOL 33 F/4.0 17 #50 OUTER SOL 34 F/4.0 18 #51 OUTER SOL 35 F/4.0 19 #52 OUTER SOL 3 F/1.4 20 #20 OUTER SOL 37 F/4.0 21 #21 OUTER SOL 38 F/4.0 22 #22 OUTER SOL 39 F/4.0 23 #26 OUTER SOL 40 F/5.6 Set up "X-pt" viewing camera with 150kHz frame rate with 4 microsec integration time and a data window from 0.5 s to 0.7 s. Other key diagnostics (besides GPI): Edge Thomson Scattering ASP and FSP for SOL flow measurements edge channels of FRCECE fast magnetics |
| Physics Operators Plans |
| Entered: May 16 2007 03:22:25:127PM |
| Author: Robert Granetz |
Engineering setup for Thursday 17 May 2007 MP#487 Co-current fluctuations (1/2 day) MP#362 Shaping & control development for high triangularity, I_n (1/2 day) Session leaders: I. Cziegler (#487), S. Wolfe (#362) Physics operator: R. Granetz ECDC overnight in D2, pressure=2e-4 Torr. Rmin/Rdwell/Rmax = 44/45/103 cm. Bake at 60 C. Power systems as on 1070511022. A-coil configuration: normal [Dtop -Dbot -Jtop +Jbot] and enabled. Gas Setup: fill B-Top with 6 psi D2 Hybrid enabled (PG4) fill B-side lower with 1 psi Ar Hybrid DISABLED (PG1) leave B-side upper as is Hybrid DISABLED (PG2) fill B-main (C-side) with 40 psi D2 Hybrid enabled (PG3) fill NINJA with 10 psi D2 ENABLED Enable the following gate valves and shutters, assuming no vacuum problems: ECE, VUV, Z-bolo shutters Torvac setup: Gate valves open during pulse. Special diagnostics: Fast H-alpha diodes; other fluctuation diagnostics ------------------------------------------------------------------------------- ICRF Setup: 2+ MW needed on some shots LH Setup: not needed ------------------------------------------------------------------------------- Run plan for first half-day: Part i - ""Radial Scan of Turbulence Feature by Scanning the Outer Gap"" 2 good shots scanning the outer gap from 3 cm to 0.3 cm starting at 0.5 sec and reaching 0.3 at 1.0 s. Do this in LSN, USN, and DN. NL04=1.1e20 Load shot 1070511025 for the LSN and put in the gap scan (and Bt=5.6 T) Load shot 1070511022 for the USN and put in the gap scan (and Bt=5.6 T) Load shot 1070511029 for the DN and put in the gap scan (and Bt=5.6 T) Part ii - ""Response of Turbulence Feature upon L-to-EDA H-mode Transition"" 2 shots (LSN) with a steady EDA H-mode produced by turning on 2MW of ICRF at 0.75 s for 0.5 sec. NL04 (target)=1e20. Observe QCM feature during H-Mode phase. Part iii - 'Response of Turbulence Feature to SSEP scan. Set outer gap to 1.2 cm and scan the configuration from LSN thru DN to USN in 0.4 s starting at ------------------------------------------------------------------------------- Run plan for 2nd half-day: The main goal of this experiment is to develop an ITER-like shape for use in Marmar's proposed high current, high performance experiments at 5.4T. Shaping requirements may be somewhat relaxed from the original MP goal, which was a match to the nominal ITER shape. The target is 1.6MA at 5.4T, which corresponds to I_N=I/aB = 1.35, marginally less than the ITER value of 1.4. Previous runs on MP#362 were 1031126 and 1040127. The best case to date is 1040127010, which succeeded in getting to 1.6MA at kappa=1.86, but disrupted at 1.15sec with an apparent vertical instability. These shots are fighting three disruptive effects: low q, vertical instability and locked modes. The approach will be to address the locked mode issue by implementing the B21_PROJ feedback algorithm and to sneak up on the other limits by starting with an equilibrium that is less demanding than 1040127010 and working our way up. |
| Entered: May 16 2007 04:40:44:170PM |
| Author: Robert Granetz |
| For first half run day, Part i of SL_PLAN: Loaded seg1 from seg1 of 1070516001, seg2 from seg2 of 1070511025 (LSN), seg3 from seg3 of 1070511022 (USN), seg4 from seg4 of 1070511029 (DN, SSEP sweep), Changed Bt to 5.6 T in segs 2,3,4 (156 kA). Confirmed nl_04 = 1.1e20 m-2 in segs 2,3,4. Programmed an RCUR sweep from t=0.5 to 1.0 s to scan outer gap from 3 cm to 3 mm in segs 2,3,4. Removed SSEP scan from seg 4 (set SSEP to 0.0). ** SAVE ABOVE IN /HOME/GRANETZ/PCS_SCRATCH shot 1070517000 ** For Part ii of SL_PLAN: I will have to reload seg 2 from seg2 of 1070511025, and change Bt to 5.6 T (156 kA), and reduce nl_04 to 1.0e20 m-2. For Part iii of SL_PLAN: I will load seg 4 from seg 4 of 1070511035, and increase the SSEP scan range to -1 --> +1 cm, and adjust nl_04 as desired. For 2nd half of run day: Load seg 2 from seg 2 of /home/wolfe/pcs_scratch shot 1070517000 |
| Session Leader Summaries |
| Entered: May 17 2007 05:32:34:760PM |
| Author: Steve Wolfe |
Summary for high current ITER shape development MP#362 This part of the run started at 15:00, shot#18. We got a 1.35MA disruption on this shot, which deranged flux loop F12 (~40mWb baseline offset after shot). We had fizzles with bad field structure on the next four shots, after which Bob went down and found that F12 had high resistance and was NOT shorted to the vessel. Zapping the connector seemed to fix it and we got a normal startup and plasma on shot 23, for which we had reloaded a normal shape low current shot. Since there would only have been time for one more shot anyway, we decided not to risk another high current disruption and shut down the run early. The disruption on 18 came early, around 0.6sec, and seemed to be a slow mode 50Hz vertical instability, with the EF2 coils showing the same frequency but up-down symmetric (in phase). We did not get a chance to test the ZCUR gain changes, and other tweaks that were proposed to address this problem. For reference, a first cut at these changes are programmed in seg2 of shot#19. Marco suggests that in addition to these, increasing the gain on IC_EF4U to 8 (instead of dropping to 2) after 0.5sec would move the pole and help the EF2 oscillation, although this would probably change the resulting equilibrium some as well. |
| Entered: May 18 2007 11:44:46:353AM |
| Author: Istvan Cziegler |
| In this run we wanted to characterize the broadband edge-SOL turbulence of L-mode and pre-H-mode plasmas. In particular, we were looking for a more quantitative and systematic comparison of the differences between different plasma configurations that we found in the runs 1060531 and 1060601.
The experiment had three major points i. Moderate gap scan in each configuration (LSN, USN, DN) in L-mode. shots -001 LSN, nl_04~1.1e20m^-3 (high), no transition, (slight SSEP shift during the) gap scan from 25mm to 4mm A-lim puff at .45s X-point puff at .7s -002 LSN, nl_04 increasing: 1.05-1.2e20m^-3 (high) no transition gap scan from 30mm to 5 mm A-lim puff at .45s X-point puff at .7s SP @ 0.8, 1.0, 1.2s, ASP(+0) -003 USN, nl_04 ~1.25e20m^-3 (high) no transition gap scan from 30mm to 4 mm A-lim puff at .45s X-point puff at .7s -004 no shot, stuck in init -005 USN, nl_04 ~1.25e20m^-3 (high) no transition gap scan from 30mm to 4 mm A-lim puff at .45s X-point puff at .6s SP @ 0.8, 1.0, 1.2s, ASP(+3) -006 DN (very close) nl_04=1.2-1.3e20m^-3 (high) no transition gap scan from 30mm to 4 mm A-lim puff at .45s X-point puff at .55s signal on diodes drops significantly -007 DN nl_04=1.0e20 m (decreased to eliminate much of the noise)(didn't have the desired effect, in fact probably changed the plasma) no transition gap scan from 30mm to 4 mm A-lim puff at .45s X-point puff at .55s signal on diodes still quite low SP @ 0.8, 1.0, 1.2s, ASP(+7) -008 LSN with changed plasma parameters and shot loaded: 1070511025, no RF power no transition nl_04~1.0e20 m^-3 A-lim puff at .44s X-point puff at .72s We found very clear radial structure in all three of the above plasma configurations, and the comparison of the different X-point positions seems to be fruitful too. ii. EDA H-mode plasmas with a long pre-H-mode observation period (~0.3 s) and with an observation period that covers both the L-to_H transition and extends for at least 0.1 s into the steady-QC-freq-period. shots -009 LSN, nl_04~.9e20m^-3 rf=1-2 MW transition at .85s EDA H-mode QC mode observable SP @ 0.54, 0.74s, ASP(+7)-first scan too deep, FSP(-3)-ok -010 LSN, nl_04~.95e20m^-3 rf=2.0+ MW transition at .815s EDA H-mode QC mode observable outer gap is 12 mm (too big) hard to find the modes SP @ 0.54, 0.74s, ASP(+5)-good target, FSP(-3)-good -011 LSN, nl_04~.95e20m^-3 rf=2.0+ MW transition at .805s EDA H-mode QC mode observable outer gap is 10-11 mm decrease requested to see beyond separatrix SP @ 0.54, 0.74s, ASP(+5)-good target, FSP(-3)-good -012 LSN, nl_04~.9e20m^-3 goes to 1.5e20m^-3 in H-mode rf=2.0+ MW transition at .805s EDA H-mode QC mode observable outer gap is 15 mm (increased, decrease requested) SP @ 0.54, 0.74s, ASP(+0)-far SOL -013 LSN, nl_04~.9e20m^-3 goes to 1.7e20m^-3 in H-mode rf=2.0+ MW transition at .81s EDA H-mode QC mode observable outer gap is 6-7 mm (small excursions) SP @ 0.54, 0.74s, ASP(+0)-good target We got sufficient data on the transition and pre-H-mode turbulence character, though the signals are very noisy which can hinder later data processing. We observed a drag-like phenomenon around .9s in every single H-mode shot in which the ion-drift turbulence (almost or completely) disappeared while the QCM broadened and suffered a significant frequency shift (spin down suspected). -014 disrupts at .32s, good background shot for the diodes, preferred over the one from 1070516 (shot #1070517020) iii. Shots with an SSEP scan from -1 to +1 cm to study direct and immediate response to configuration variation -015 SSEP scan from LSN to DN to USN, outer gap is 22mm to study in-SOL turbulence part of the spectrum disrupts at 1.2s strong gap fluctuation starts at .89s before USN is properly achieved nl_04~.85e20m^-3 (low) -016 SSEP scan LSN DN USN outer gap is at 22mm proper SSEP scan nl_04~.85e20m^-3 (low) -017 SSEP scan LSN DN USN outer gap is at 7mm (steady in the period of interest) proper SSEP scan nl_04~.85e20m^-3 (low) end of MP 487 Pedestal--in-SOL turbulence seems to respond more strongly to SSEP scans than internal modes -- as expected. Comparison with probe and X-point Phantom views will be important. |
| Physics Operator Summaries |
| Entered: May 17 2007 05:13:06:437PM |
| Author: Robert Granetz |
Run summary for Thursday 17 May 2007
MP 487: Co-current fluctuations (1/2 day)
MP 362: Shaping & control development for high triangularity, I_n (1/2 day)
Session leaders: I. Cziegler (#487), S. Wolfe (#362)
Physics operator: R. Granetz
The first half of the run (shots 1-17) was for studying fluctuations in
LSN, USN, and DN with scans of the outer gap, all with normal field and
current direction. The run was very successful, and we managed to
accomplish everything that was desired, which was quite a lot. This was
despite several delays due to the shot cycle hanging in INIT. In
particular, the scan of the outer gap in LSN, USN, and DN went very
smoothly, and so did the SSEP scan with fixed outer gap. ICRF worked
reasonably well from all three antennas, with a total power of 2+ MW.
The second half of the run was devoted to development of an ITER-like shape
for future use in high performance experiments. This started on shot 18
with an attempt at 1.3 MA. It went vertically unstable and disrupted at
t=0.6 s, resulting in a baseline offset error on F12. This was then
followed by 4 fizzles, all with weird looking fields and breakdown
characteristics. F12 was showing signs of having a small error. So after
shot 22, I went down into the cell to check F12. It was not shorted to the
machine, but it had a resistance of several tens of ohms. So I zapped it
with 50 volts, and its resistance returned to the normal range of a few
ohms. I then reloaded the startup from the last normal shot (17), and we
got a normal, full-length plasma. End of run.
Scorecard: 18 plasmas
duds
4 fizzles
power supply tests
2 no-power MDS shot cycles
--------------
Total: 24 shot cycles**
** One of the no-power shot cycles was not counted by MDS. (There were
really two shot 4's, both of which were no-power shot cycle.)
|
| Session Leader Comments | |||
| May 16 2007 02:25:51:917PM | Steve Wolfe | Begin setting up the PCS for second part of 1070517 MP#362 ITER-like shape development Following prescription in draft run plan. Start from PCS_MODEL between shots 1070516015 and 016 Import segment 2 from 1040127010 Call alg for wires 1,2, 4,5,6,7,8,10,16; ignore factor overrun on RCUR; don't reload any offsets. Load wire2 from wire 10; adjust max D gain to 5.0, drop final P from 3 to 2.5 Zero wire 10 Import seg3 from 1070511036, just to get the B21_PROJ wire! IN seg2, load wire 9 from seg3 wire 9 "B21_PROJ" Change Ref 9-12 from Ip to 10 Zero ACOIL_CUR on V1-16 Move integrator C gate to start 0.2sec Touch a point in V17-32, just to establish the new output names ON all_p screen: Decrease Ip from 1.65 to 1.35 CLEARIN flattop from .005 to -.005 IC_EF4U from 2000 to 3500 Sanity check on clearin: On 1070511001 we were running clearin=+.0035 and getting inner gaps of 12mm, so -.005 might end up with 3 to 4mm gap. Rcur on 1070511001 was 0.6595, compared to .6605 in seg2 now, and we were getting 12.5mm outer gap, so that's going to be kind of close, but we'll give it a try. While I'm at it, go ahead and recall the nl04 and IC_EF4U as well. save as /home/wolfe/pcs_scratch/*_1070517000 | |
| May 17 2007 08:48:47:400AM | Steve Wolfe | Minor tweak to saved PCS setup: The TF was set a little low. In PCS /safe Load From /home/wolfe/pcs_scratch/ shot 1070517000 In seg2 V01-16: change TF current from -147000 to -150000 Save As /home/wolfe/pcs_scratch/ shot 1070517000 | |
| May 17 2007 10:25:38:157AM | 1070517001 | Istvan Cziegler | Reloaded shot 1070511025 for LSN gap scan from 25 mm to 4 mm (a little smaller than we originally wanted, especially on the 3 cm side, because the signal is very short lived in sight) between .5s and 1s. NINJA puffed from .45 seconds. We have saturation on several diodes, need to readjust them. The anticipated double lobe structure still appears, though somewhat blurred. I set the high-pass filter on the diodes at 50 kHz, but the broad band signal still appears weak after ~.6s when the signal still seems to be aligned with the vertical array. Hard to tell - yet - if this is due to the radial intensity variation that we are looking for. |
| May 17 2007 10:44:14:367AM | 1070517001 | Istvan Cziegler | In the first .2s of the GPI data, I can see a broad though quite dim feature reaching up to 300 kHz with fixe3d wave number around 2 cm^-1. No trace in the magnetics though. It may be some pickup or it can be due to some diodes saturating. |
| May 17 2007 11:01:00:390AM | 1070517002 | Istvan Cziegler | Excellent shot. Starting gap increased to 30 mm. Noise bar at k_theta~2 cm^-1 is still present unfortunately, but we do positively see a clear variation in the signal intensity on the electron direction. We can move on to USN. |
| May 17 2007 11:14:07:300AM | 1070517003 | Istvan Cziegler | Excellent shot. USN plasma, very strong radial structure in broadband turbulence.
Exact repeat requested. |
| May 17 2007 05:02:59:060PM | 1070517005 | Istvan Cziegler | Good repeat of shot -004 in USN configuration. The signal is somewhat weaker than in -004, but the radial structure was qualitatively reproduced. Requesting DN for next shot. |
| May 17 2007 12:28:20:127PM | 1070517006 | Istvan Cziegler | Turbulence amplitude is much weaker than in either LSN or USN shots earlier. I need to look into the next shot (exact repeat) to see if it is reproduced. Otherwise the structure is more like that of the LSN than a USN the feature in the electron direction being (Doppler) shifted to higher frequencies (weaker rotation); the transition from ion to electron direction also appears somewhat deeper. |
| May 17 2007 12:33:56:550PM | 1070517007 | Istvan Cziegler | Strong k=0 cm^-1 noise, probably still due to high density. A radial structure is still observable, will need more work later. Move on to power shots. Try to achieve a transition with LSN in the same geometry. |
| May 17 2007 12:37:33:533PM | 1070517007 | Istvan Cziegler | I am now seeing differences between the co-propagating features of the LSN and DN plasmas. The wave numbers at low frequencies are shifted to higher values for DN. It wil need more quantitative analysis. |
| May 17 2007 12:46:20:380PM | 1070517008 | Istvan Cziegler | No power, more LSN data may still be useful. Power for next shot? |
| May 17 2007 12:55:23:447PM | 1070517008 | Istvan Cziegler | k=0 noise much stronger than in the morning shots. The character of the features changes significantly with signal like that of the (second) DN shot. Is it the density that is causing it? |
| May 17 2007 02:47:33:037PM | 1070517015 | Istvan Cziegler | Ssep scan was good before the gap got wild at around .895 -- there is a slight difference between the LSN and the DN-USN plasmas, but the position of the nominal separatrix at the outer midplane should still be checked. Move on to ssep scan at a gap of .7cm |
| May 17 2007 02:52:51:070PM | 1070517016 | Istvan Cziegler | Repeat of shot 15 with proper ssep scan this time. Diode data are all good, no saturation, but the intensity and the character changes through the shot. We'll need to compare this with changes in the plenum pressure. |
| May 17 2007 03:22:40:753PM | 1070517018 | Steve Wolfe | Shot#18: Beginning the part of the run for MP362, ITER-shape Have Bob load pre-programmed segment 2 and change engineering setup to power systems as on 1040127010. Trying for 1.35MA, tall with small gaps all round. See what happens and tuned from there. Early disruption, around .6sec. Big vertical instability. Oscillation at 50Hz. Oscillations on clearin, ef2, xpoints, etc. Starting at 0.5sec Next: make clearin flat at -.005. zcur 2 D to 1.5 from 0.7 delay Ip gain switch from 0.6 to 0.8sec |
| May 17 2007 03:52:53:500PM | 1070517019 | Steve Wolfe | Shot#19: make clearin flat at -.005. zcur 2 D to 1.5 from 0.7 delay Ip gain switch from 0.6 to 0.8sec fizzle - forgot to drop the gas. Fields look bad. Magnetics problem? Initial Bz0 is at least 5 mT too positive Ip jump is twice normal. Lots of Halpha, no zmeter or CIII light. |
| May 17 2007 04:29:51:763PM | 1070517022 | Steve Wolfe | Shot#22: Br0 down to -.004. Still bad, OH2 currents still not changing much. |
| May 17 2007 04:33:20:127PM | 1070517022 | Steve Wolfe | Shot#22: After shot#22, Bob checked resistance on F12 and found not shorted to machine, but R="tens of Ohms". Have cap, will zap. |
| May 17 2007 04:41:59:770PM | 1070517022 | Steve Wolfe | Shot#22: Signal on F12 at t=0- is about 50mWb less than F01 and F16, while before the disruption on shot #18 these were all equal. Not sure if this is the only problem, but it seems to be bad. Fortunately, we could synthesize F12 if we have to. Meanwhile, the zap seems to have worked (resistance back to normal) so we'll try another shot. |
| May 17 2007 04:52:46:507PM | 1070517023 | Steve Wolfe | Shot#23: Startup reload from shot 17, seg2 from shot 13, trying for a normal discharge (or even a normal fizzle). Plasma. startup looks good. The zap seems to have worked. Quit while we're ahead (or at least not behind) |
| Physics Operator Comments | |||
| May 17 2007 10:02:31:850AM | Robert Granetz | There was a 50-minute delay in starting the run because of computer server problems with the ICRF CPCI. | |
| May 17 2007 10:43:13:523AM | 1070517001 | Robert Granetz | Shot 01 -- Plasma. Good shot. LSN with outer gap scan from 25 to 4 mm from t=0.5 to 1.1 s. Bt=5.65 T, Ip=0.8 MA, nl_04=1.1e20 m-2. Very close to the requested discharge. Inboard gap is held steady at 1.2 cm. Cell access needed by JT after this shot. NOTE: Steve Wolfe discovered that the ZXU estimator in seg 2 had not been updated to exclude F10. That is probably why this shot eventually becomes USN at late times. I called the algorithm before the next shot, and I confirmed that everything else in segs 1,2, 3 and 4 are already up-to-date. Next shot: Update to ZXU estimator will change the equilibrium evolution, hopefully only a small amount. Tweak RCUR programming to start scan with a larger outer gap. Enable argon puff (100 ms starting at t=0.3 s). Lithium pellet will be fired at t=1.8 s. |
| May 17 2007 11:00:15:230AM | 1070517002 | Robert Granetz | Shot 02 -- Plasma. Good shot. Outer gap scans from about 3 cm to about 0.5 cm. Bt, Ip, and nl_04 are as requested. Next shot: Switch to USN (seg 3). Made a couple of small tweaks to RCUR programming (outer gap scan). |
| May 17 2007 11:07:56:500AM | 1070517003 | Robert Granetz | Shot 03 -- Plasma. Good shot. USN with outer gap scan; about 3 cm to 0.4 cm. Bt=5.65 T, Ip=0.8 MA, nl_04 about 1.2e20 m-2. Next shot: repeat, no DPCS changes. |
| May 17 2007 11:46:20:017AM | 1070517004 | Robert Granetz | Shot 04 ? -- No-power cycle because we hung in INIT. MDS does not know about this shot, so it didn't really happen. The next shot will also be shot 4. |
| May 17 2007 11:53:30:160AM | 1070517004 | Robert Granetz | Shot 04 (again) -- Another hang in INIT. The next shot should be 5. |
| May 17 2007 12:00:23:397PM | 1070517005 | Robert Granetz | Shot 05 -- Plasma. Good shot. USN with gap scan; same as shot 3, as requested. Next shot: go to double-null (switch seg 4 on, 3 off) |
| May 17 2007 12:21:15:700PM | 1070517006 | Robert Granetz | Shot 06 -- Plasma. Good shot. Double-null (very close) with a good outer gap scan. Bt=5.64 T, Ip=0.8 MA, nl_04=1.2-1.3e20 m-2. Next shot: repeat, except decrease nl_04 to 0.8e20 m-2, since the NINJA puff is fueling the plasma also. |
| May 17 2007 12:38:59:097PM | 1070517007 | Robert Granetz | Shot 07 -- Plasma. Good shot. Double-null with outer gap scan. nl_04=1.0e20 m-2. Next shot: go to step ii of the run plan for the first half of the day. reload seg 2 from 1070511025 seg 2 (LSN, outer gap=12 mm). Increase Bt to 5.6 T, decrease nl_04 to 0.8e20 m-2. Turn on seg 2 and turn off seg 4. 2 MW ICRF requested, starting at t=0.75 s for duration of 0.5 s |
| May 17 2007 12:47:22:587PM | 1070517008 | Robert Granetz | Shot 08 -- Plasma. Good shot. LSN, no gap scan (approx 13 mm). Bt=5.64 T, Ip=0.8 MA, nl_04=1.0e20 m-2. But no ICRF. Next shot: repeat, no DPCS changes. Get the ICRF going. |
| May 17 2007 01:11:15:033PM | 1070517009 | Robert Granetz | Shot 09 -- Plasma. Good shot. LSN, EDA H-mode. Prf=1-2 MW. D & E good, J had problems. Next shot: repeat, no DPCS changes. Request more ICRF power from D & E in lieu of J. |
| May 17 2007 01:35:39:933PM | 1070517010 | Robert Granetz | Shot 10 -- Plasma. Good shot. LSN, EDA H-mode. Prf=2.0+ MW, fairly clean over desired time window. Next shot: repeat, no DPCS changes |
| May 17 2007 01:36:59:997PM | 1070517011 | Robert Granetz | Shot 11 -- Plasma. Good shot. LSN, outer gap is 10-11 mm. EDA H-mode. Prf=2+ MW. Next shot: decrease outer gap to 6 mm. Move LPI to t=1.6 s (since it's not disrupting the plasma). |
| May 17 2007 01:48:36:967PM | 1070517012 | Robert Granetz | Shot 12 -- Plasma. Good shot, but I mistakenly programmed RCUR to make the outer gap bigger, not smaller. Oops. But this shot will be useful anyway. Next shot: program RCUR to achieve an outer gap of 6 mm, not 15 mm. |
| May 17 2007 02:13:17:713PM | 1070517013 | Robert Granetz | Shot 13 -- Plasma. Good shot. LSN, outer gap=6-7 mm, but with some excursions. Prf=2.0+ MW. nl_04 didn't work properly, but the feedback did. We had an I2t fault, but it was apparently not real. Next shot: switch to part iii of this MP. Turn on seg 4 and turn off seg 2. Set RCUR so that outer gap is 2.2 cm. Set SSEP scan to go -1 to +1 cm in the time interval from t=0.6 to 1.0 s. Confirm that nl_04=0.8e20 m-2. |
| May 17 2007 02:25:04:530PM | 1070517014 | Robert Granetz | Shot 14 -- Plasma, but disrupted at t=0.32 s. My error...I programmed the SSEP sweep from -1.0 to +1.0, but the DPCS units are meters, not centimeters. Another oops. (I started from a trace that was all zeroes, so there were no obvious vertical scale references.) This time, the shot is not useful. Next shot: change SSEP programming to be -0.01 to +0.01 meters. Reduce pre-fill by 2 ms. |
| May 17 2007 02:39:51:863PM | 1070517015 | Robert Granetz | Shot 15 -- Plasma. Disrupted at t=1.22 s, but it did the SSEP scan as requested. However, the outer gap did not stay constant at 22 mm. Steve Wolfe points out that the problem is because the inner and outer gap control normally references the flux at the lower x-point, and when the equilibrium changes from lower to upper null, the estimator becomes invalid. Next shot: Steve showed me how to have the estimator switch reference points during the discharge. I've set this up for CLEARIN to change to ZXA=+0.39 m at t=0.8 s. |
| May 17 2007 02:46:31:070PM | 1070517016 | Robert Granetz | Shot 16 -- Plasma. Good shot. Disrupted in rampdown. Good SSEP scan. Outer gap was reasonably steady at 22 mm during the time of interest (t=0.6-1.0 s) Next shot: Same SSEP programming, but set outer gap at 7 mm (was 22 mm) by increasing RCUR by 7.5 mm. |
| May 17 2007 03:03:16:107PM | 1070517017 | Robert Granetz | Shot 17 -- Plasma. Good shot. Got the SSEP scan, and the outer gap was held constant at 7 mm during the time range of interest (t=0.6 to 1.0 s). End of MP 487. Start MP 362. Turn on seg 2, turn off seg 4. Load seg 2 from /home/wolfe/pcs_scratch shot 1070517000 seg 2. Load into model. |
| May 17 2007 03:24:02:443PM | 1070517018 | Robert Granetz | Shot 18 -- Plasma. Disrupted at t=0.6 s. Ip=1.30 MA. F12 had a baseline offset of 46 mWb. Looks like it went vertically unstable. Definitely not a locked mode. Next shot: Delay gain switch on Ip to 0.8 s (from 0.6 s), Reduce early CLEARIN to -0.005 m. Increase slow ZCUR derivative gain to 1.5 (from 0.7). |
| May 17 2007 03:55:36:097PM | 1070517019 | Robert Granetz | Shot 19 -- Fizzle. Fields changed, but there weren't any magnetics errors reported. F12 looks fine. Next shot: drop pre-fill 2 ms. Add -20 amps to IC_EF4U. Reload seg 2 from seg 2 of shot 13 today, which was the last one that used seg 2. |
| May 17 2007 04:06:58:053PM | 1070517020 | Robert Granetz | Shot 20 -- Fizzle. Closer to a normal fizzle. F12 signal may be slightly in error? Next shot: add the 20 amps back to IC_EF4 (to -1335 A). Add -0.001 T to Br0 (to -0.004 T) |
| May 17 2007 04:14:54:103PM | 1070517021 | Robert Granetz | Shot 21 -- Fizzle. Next shot: reduce Br0 to -0.005 T. |
| May 17 2007 04:44:37:940PM | 1070517022 | Robert Granetz | Shot 22 -- Fizzle. I'm going into the cell to check on F12. F12 is not shorted to the machine. Neither is F16. But the resistance of F12 is in the several 10's of ohms, while F16 is the normal few ohms (measured with an analog meter). So I zapped F12 with 50 volts and its resistance came right down to the normal range of a few ohms. It is still not shorted to the machine. Next shot: reload seg 1 from 1070517017 seg 1. Seg 2 is from 1070517013. |
| May 17 2007 04:53:41:633PM | 1070517023 | Robert Granetz | Shot 23 -- Plasma. Good shot. F12 now matches F16 at commutation. Looks like everything is back to normal. End of run. |
| Engineering Operator Comments | ||||
| Shot | Time | Type | Status | Comment |
| 1 | 09:50:34:300AM | Plasma | Ok | |
| 2 | 10:45:50:943AM | Plasma | Ok | |
| 3 | 11:01:34:233AM | Plasma | Ok | |
| 4 | 11:14:12:020AM | Test | Bad | |
| 5 | 11:53:23:767AM | Plasma | Ok | |
| 6 | 12:07:54:257PM | Plasma | Ok | |
| 7 | 12:21:14:780PM | Plasma | Ok | |
| 8 | 12:39:27:240PM | Plasma | Ok | |
| 9 | 12:57:58:777PM | Plasma | Ok | |
| 10 | 01:12:43:720PM | Plasma | Ok | |
| 11 | 01:25:45:647PM | Plasma | Ok | |
| 12 | 01:39:55:293PM | Plasma | Ok | |
| 13 | 01:52:36:673PM | Plasma | Ok | I2T fault after the shot |
| 14 | 02:12:53:493PM | Plasma | Ok | |
| 15 | 02:25:56:327PM | Plasma | Ok | |
| 16 | 02:38:34:613PM | Plasma | Ok | |
| 17 | 02:50:57:913PM | Plasma | Ok | |
| 18 | 03:04:46:327PM | Plasma | Ok | |
| 19 | 03:23:47:363PM | Plasma | Ok | |
| 20 | 03:56:56:830PM | Plasma | Ok | |
| 21 | 04:09:27:147PM | Plasma | Ok | |
| 22 | 04:22:04:010PM | Plasma | Ok | |
| 23 | 04:45:44:660PM | Plasma | Ok | |
| System Availability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| May 17 2007 09:48:24:377AM | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||