Chris S. Joe D. Greg G. With the help of Joe and Chris the SM1 mirrors (SN 3,4) were cleaned and are air-baking now, they should be done Monday evening. The SM2 mirrors were also completed recently and are wrapped and bagged and ready for install.
(Stefan Ballmer - not Jeff) I set up a test for diagnosing front end to front end communications via 4 separate routes: 1) IPC 2) EPICS read (pull data) 3) EPICS write (push data) 4) EPICS write and read, via a third epics server (push to 3rd server, pull from 3rd server) The setup has a counter on H1odcmaster, this is sent via temporary IPC to h1ascimc, and then returned back to h1odcmaster via the 4 routes described above. For IPC and epics read operations I get an error flag, which I am reporting. No such thing is available for an epics write operation. h1odcmaster is currently running at 32k, h1ascimc is running at 2k. I calculated the following quantities - DELAY : round trip delay in sec - DELAY mean : averaged DELAY - DELAY max: : largest DELAY observed - JITTER MAX : largest time interval between updates - JITTER MIN : shortest time interval between updates - JITTER MAX-MIN : = JITTER MAX - JITTER MIN, the peak-to-peak jitter - JITTER RMS : root-mean-square of the time interval between updates Conclusion for now: - No errors observed in the ~1h this was running - No jitter on IPCs - epics read has a faster update rate (~8Hz) than epics write (~2Hz) - as a result the delay is shorter for epics read - the jitter MAX-MIN is about 40msec for EPICS access - the RMS jitter is about 4msec I turned on recording for all the fast return channels over the weekend. The modifications to h1ascimc and h1odcmaster will be backed out on Monday.
Here is another snapshot after running for 2 days. No qualitative change.
The ref cav was not locked all day, I didn't get a chance to look at it until just now, but it seems like the problem is actually with the PMC.
The indicators on the medm screen seem to think that the PMC is locked, but it does not look locked in the camera, and the reflected power is 12.5 W.
TF's running on ETMX ISI. More reps, so this test will take longer. Running on opsws6.
Sometime yesterday one of the leads for the DC power cable between a lab power supply and the Comtrol at end Y got disconnected. Richard went out and fixed it. I restarted the IOC for the attached dust monitor at 13:37. I later burtrestored it.
Roofers on the roof! Removal of east and west doors from Ham 3 – Apollo Checking dust monitors in LVEA – Richard Dust monitor (unit #10) installed by HAM 3 – Richard Baffle work in LVEA (West bay) – Mitchel/Thomas Vo Working at End Y – Alexa WBSC9 ETMx Alignment - Jason Cable work from HAM3 to HAM6 – Filiberto Delivery for Richard – PLATT Electric Supply Vented sleeve surrounding GV6's gate annulus piping leak - Kyle
Took a look at the ETMx this morning to asses the alignment after the HEPI actuator install was completed. As a reminder the large X axis positional error has been signed off by SYS, no correction will be performed.
IAS is now moving full time to the ITMx in WBSC3. We will have to return to the ETMx to fine align the optic pitch and align the ETMx ACB.
This morning, I locked the BSC3 ISI and Greg went into the chamber and installed the corner cube. I'm not sure what state the ACB is normally in when the cube is attached, but it is in the lowered position right now, which made install of the cube precarious. Greg was able to do it, but it didn't look easy. Otherwise, we are ready for IAS to start working.
HAM's2 and 3 ISI's were locked this morning. Originally went out just for HAM2, but I saw that Apollo had just finished pulling the last door at HAM3, so I took advantage. I think immediate activities are planned only at HAM2, so if HAM3 needs to be released for some reason, let me know.
J. Kissel Because the DAQ was needed this morning for measurements, we didn't resurrect the dead h1sus56 computer until the afternoon. That meant I didn't get nearly as far as I'd hoped before leaving. My apologies. The current status is reported below, and will most likely remain as such until I get back from vacation on Tuesday. Status: - Mods to the HSTS_MASTER and HLTS_MASTER library parts (and all the library parts they use) are complete -- the HSTS required a little extra thinking, unfortunately - The MC_MASTER remains unchanged, didn't get to it - The top-level HLTS models' (H1SUSPR3, H1SUSSR3) mods are complete, and they compile successfully - The newly compiled front end code for H1SUSSR3 has been installed, and is now running; PR3's has not yet been installed - Of the HSTS', only H1SUSMC1's top level model has been completed and successfully compiled. The remaining HSTS models must be reconnected as h1susmc1's has in order to compile. They are guaranteed to fail compiliation for the time being. - None of the necessary MEDM screen modifications have been made for the HLTS or HSTS. Because I do not commit to the repository until I have a complete package for Stuart, the outstanding svn status I reported yesterday remains.
Transfer functions running on ETMX ISI, starting now. Should end by 8 AM. Running on opsws6.
The LVEA test stands (h1susquadtst, h1susbstst, h1seitst) have been powered off, removed from the DAQ, and removed from the dolphin configuration. All equipment has been left in place for future use.
Kyle Pressure response immediate on PT114, PT124 etc...
We got all the Actuators connected. The last one, V3, took a while as we swapped out the suspect parker valve noted by Jim earlier. The first one we swapped in L170 also appeared to do the same pull down Jim noted. But this may have been a ringer. We had not bled the actuator yet as we wanted to compare the performance and then we learned this valve had been removed from WHAM6. However, we thought we were in trouble as the swap at WHAM6 was possibly unneeded because the problem there was that the Supply & Return hose were miss plumbed. Still, we pulled L170 and swapped in L206 and bled the Actuator for 1.5+hours. After we put the Actuator back in run mode, the Actuator drove back to a near center position so we deemed the valve problem addressed and attached this final Actuator. Dial Indicators suggested we had tilted the table a few tenths, C2/4, so we followed with some tweeks of the big HEPI Springs to bring that back to start position. Have alerted IAS/Jason that we are ready for a post Actuator look--he may be looking now.
On caveat here: since the ISI was floating we couldn't attach the corner cube to get an accurate longitudinal measurement. The total station can use a "Sheet Refelctor" mode to measure the distance to the optic from the HR face directly, but this introduces more error in the measurement. For the purpose of calculating the lateral and vertical errors a few mm of error in the longitudinal measurement means almost nothing at this distance (>33 m; a 5 mm shift in the longitudinal direction only imparts a change of <1 µm in the lateral/vertical error calculation), but I would not use this longitudinal measurement as a basis for the actual optic position (and adjust HEPI when it might not be necessary). When the ISI is locked down we will attach the corner cube and get an accurate measurement of the longitudinal position error of the ITMx (and make any decisions about HEPI adjustments at that point).
These are still running the test code. I changed the sample time to 20 seconds and the hold time to 280 seconds for the dust monitors in the optics lab, vacuum prep lab and bake oven room. These were set through the 'expert' medm screen for each.
I made PSL model changes this week to add channels to the science frame. Pre-empting next week's upgrade to matlab2012b, I modified all the mdl models used to build the code using 2012b (in some cases just adding an SVN $Id$ tag in order to make a change).
Today I tried to open the psl models using matlab2010a. I had no problems with h1psliss and h1pslfss, but h1pslpmc crashed 2010a. There are no problems opening this with 2012b. Presumably PMC uses some matlab parts with are not backwardly compatible with 2010a.
When I made the transition, the original 2010a versioned file was save by matlab with a 2010a prefix. We should always do this.
You have to use the 'Export Model to > Previous Version' if you want to save a model in the earlier format. Incompatible blocks that postdate the export file format version are converted to empty masked subsystem blocks in the exported file:
http://www.mathworks.com/help/simulink/ug/saving-a-model.html#f2-86382
The default file format is set in the preferences, or via command line:
http://www.mathworks.com/help/simulink/gui/simulink-preferences-window-main-pane.html#bs_hdzw
Cheryl asked if we could get TFs for IM2, IM3 and IM4 to complete the IM1 set from 8287 before the HAM door is removed, so I set them going in parallel in three Matlab sessions on opsws7. They should be done by about 11:30 am.
All data taken OK. Individual plots attached.
There's still a bug in the script that does comparison plots so stay tuned for that.
Note that the model now includes eddy-current damping, but the values are from by-eye fitting to IM1 data, not actual estimates from magnet and copper block properties.
Chamber is released for door work.
Here's the comparison plot, including one old trace for IM1 taken by Szymon with DTT on 1/23, and the four new traces for IM1-4 taken with Matlab recently.
The consistency from measurement to measurement is very good, even between DTT and Matlab.
There appears to be a different gain error for each DOF, which is probably an issue with the model:
L: model is about 0.6 of measurement
P: model is about 1.4 of measurement
Y: model is about 2.4 of measurement
Arnaud spotted that the script error mentioned above turned out to be that plothaux_matlabtfs.m and plotHAUX_dtttfs.m were writing intermediate data files of slightly different formats. We tweaked plotHAUX_dtttfs.m and regenerated ^/trunk/HAUX/H1/IM1/SAGM1/Results/2013-01-23_1600_H1SUSIM1_M1.mat.
