I changed the group permissions of all files under the /opt/rtcds/userapps/release area to permit individual accounts modification access to these controls owned files. Please report any file permission problems to me, thanks.
This is the first step in implementing this change. Not all users accounts belong to the controls group.
I restarted the DAQ at 13:12PDT today to support SUS and ASC model changes made this morning.
h1dc0 monit did not restart the daqd process after I shut it down. I logged into h1dc0 as root and restarted monit (which seemed to be running) which in turn started the daqd process.
After the DAQ restart h1susprm and h1suspr3 showed a 0x2000 DAQ error. I pressed the DAQ RELOAD button on their GDSTP MEDM screens and cleared the error.
Mark B. and Gerardo
Yesterday (4/22/14) Gerardo and I measured the vertical mode Q of the OFIS for three different positions of the ECD block to try to get the Q in the specified range of 25-30.
The OFIS was set up on the optical bench in the H2 laser enclosure. The ECD block sits on a tray below the payload which is supported by four groups of vertically pointing screws at accessible positions around the edge of the structure. Gerardo had earlier attempted to set the ECD block at the nominal height using the spacer tool provided but found that this was too high and caused interference. He therefore lowered the block until it was just barely clear plus approximately an extra two turns of the 1/4-20 screws. This was our starting point for further adjustments.
To measure the vertical Q, we used the laser pointer and QPD from the monolithic violin mode setup and used a convenient screw on the top of the payload to partially block the beam. We displayed the "pitch" output of the QPD box on a digital oscilloscope with a 1 sec/div timebase and photographed the screen to capture the data. I read off the peak positions with GraphClick, and worked out the logarithmic decrement and Q with Mathematica.
For the initial position, the Q was 10.7. We lowered the ECD by one turn on all the screws and got Q = 18.9. We lowered the ECD another half turn and got Q = 23.3. Finally, today (4/23/14) we lowered the ECD another 3/4 turn and got 27.8, which is in spec.
Attached is a JPG of the setup, a PDF of all the screenshots, and for the fourth and final run, the Mathematica notebook, PDF thereof and the raw data.
We went back on Friday 4/25 and used the same method to measure the longitudinal (parallel to the OFI beam axis) and transverse mode Qs. For the longitudinal measurement we were able to keep the laser in almost the same position, just clipping a different edge, but for the transverse we had to send the beam on an odd diagonal path clipping one corner and then passing through the hole in the beam dump at the end (see photo). The results were
L: 21.3
T: 15.3
V: 27.8 (from 4/23, above)
The spec is <30 per T1000308-v1, p36, so these Qs look good and we propose to leave it like this.
I measured the gap between the copper plate and the magnets, 4 mm and all 4 corners.
Per request of Jeff Kissel, I extracted the frequencies from the data of 4/23 and 4/25 for the final configuration of the dampers:
In addition, I pulled the Dial Indicators from the East side, Turned off the ISI Sensor/Driver electronics, unplugged all the the East air-side cabling, and removed those Feed-Thru protection shrouds. This will give access to the final bolts of the Septum flange. Be mindful of the CPS Feedthrus on the North side--long and vulnerable.
For the operators or future HEPI Lockers, attached are trends from before starting thru after stopping the HEPI down. We shoot for < +-200 IPS Counts (655cts/mil) and it looks like we did pretty well here. Also are trends of the Cartesian values. These are nanos(meters/rads) for the horizontal/rotation dofs. We shifted 7um in X, less for Y & Z and 5urads RX and less for RY & RZ. RZ is likely the most important for the upcoming alignments, it shifted 2urads.
Bartlett, Thomas
Notes for Saturation counter bleeding design:
to note: after a day of heavy installation work around this chamber, the HPI L4C WD is still a long way from saturation. I (Brian L) believe that this is not a good day to consider when thinking about the auto-bleed-down rate. This sort of shenanigans should be cleared by an operator - under the belief that if this sort of thing were happening to the HPI, then someone should know about it.
Aidan, Dave H, Matt H, Greg G.
We aligned the X-arm CO2 laser through the first 40% of the optics on the table, including:
We measured the DE of the AOM as a function of applied voltage (via the front end model). The results are posted in the table below. The AOM is quoted as having a maximum DE of ~85% at the exact Bragg angle.
The Y-arm table laser is ready to test on Wednesday.
[this is something new. We would like the Operator to post minutes for the morning meeting]
Morning Installation Meeting Minutes
Attendees: Vern, Keita, Jeff K, Jeff B, Hugh, Corey, Justin, Jim W, Jodi, Aiden, Mark (apollo), Cyrus, Aaron, Richard, Pablo, John.....(sorry if I'm missing others, this is from memory over an hour after the fact)
Topics Brought Up:
Soft closed GV5 at 9:00 am, and GV7 at 9:07 am. They are to remain soft closed for TCS table work near BSC1 and BSC3.
Valves are open now, GV5 and GV7 were open by 1:37 pm.
The new drivers installed on the DAQ yesterday as part of WP4583 did not improve performance as expected, and in fact made it worse, so I have rolled back the changes to h1nds1, h1fw1, and h1broadcast0. I have left h1dc0 as it is since the change there does not seem to have had any ill effects. It's not clear to me why the same change showed an improvement on the DAQ test stand, other than the traffic level is significantly higher on the production system and must influence the driver behavior or the system configurations have diverged enough to affect the results of testing of this nature. Here are the downtimes for reboots:
h1nds1: 15:40:35 - 15:45:54 UTC
h1fw1: 15:48:23 - 15:52:30 UTC Frames starting/ending in this period will be missing - use h1nds0/h1fw0 for data in this gap.
h1broadcast0: 16:00:40 - 16:02:40 UTC
J. Kissel (Approved by M. Landry, K. Kawabe, V. Sandberg) I'm taking the opportunity while gate-valves are closed for TCS table moving to install ASC dither alignment IPC channels into all core-optic SUS, (see ECR ). This will replicate what has been done for PR2 (see LHO aLOG 10346). Note, for now (so the SUS models will compile without complaint), I plan to add the IPC sender channels at the top-level of the ASC model, with zeros sent into them, and I'll leave it up to the ISC team to hook them up to their main library block as they wish. The models affected with be (to add new IPC sender parts) h1asc (to add new IPC receiver parts) h1susprm h1suspr3 h1susbs h1susimty h1susitmx h1susetmx h1susetmy h1sussrm h1sussr2 h1sussr3 [because it uses the HSTS_MASTER, MC_MASTER library part, so new unused inputs must be terminated] h1susmc1 h1susmc2 h1susmc3 (to add dither paths to each stage of each SUS type) SIXOSEM_F_STAGE_MASTER FOUROSEM_STAGE_MASTER FOUROSEM_STAGE_MASTER_OPLEV SIXOSEM_T_STAGE_MASTER QUAD_MASTER BSFM_MASTER HLTS_MASTER HSTS_MASTER MC_MASTER Work beginning 1082304916 2014-04-23 16:15 UTC 2014-04-23 09:15a PDT I'll make sure to capture a new h1asc_safe.snap before getting started.
model restarts logged for Tue 22/Apr/2014
2014_04_22 09:14 h1dc0
2014_04_22 09:15 h1broadcast0
2014_04_22 09:15 h1fw0
2014_04_22 09:15 h1fw1
2014_04_22 09:15 h1nds1
2014_04_22 09:16 h1nds0
2014_04_22 09:29 h1nds1
2014_04_22 09:40 h1fw1
2014_04_22 09:53 h1fw1
2014_04_22 10:10 h1broadcast0
2014_04_22 10:16 h1broadcast0
no unexpected restarts.
There is a desire to have Spool cameras looking at the ITMs--One to show only Green light & one to show only IR light. Currently we only have one camera looking at the ITMs, we want to have (2) GigE Digital cameras with proper filters to look at the ITMs. This is to aid in alignment with the Green & IR lasers.
I received filters (from MidOpt Systems) this afternoon and measured transmission of Green & IR light through each of the filters. I had the laser point at an Ophir Power Meter, noted power, then placed filter in front of power meter and noted power. Here are the results:
IR laser set to ~0.474W
Transmission through the filters = power with filter (w/o filter)
Green laser set to ~100.5 mW
Transmission through the filters = power with filter (w/o filter)
The links for the filters above give info on them and Transmission v. Wavelength charts. The Filters are M62 (62mm thread). They can be screwed on the output of our Butterfly electric zoom lens. These lens would then be installed on our Digital GigE cameras (with a beam expander in between?).
Next thing to be done would be to see if any of these filters will work.
A wiki showing information for cameras is on the H1 Camera Page.
Fixed (not that they were "broken") the unsoldered connectors.
PDH CM board was modified so that it should be identical to X arm.
For some reason we found that the demod output for the PDH was railed at 10V (or was that -10V?). LO and PDH RF were both OK, but the I output as well as IMON were both railed. The demod chassis has two I/Q demod and the unused one was still alive, so we just switched. The broken one is still left there.
Though the only Hartman path optic on the table was a pickoff for that path, I installed another pickoff downstream of the first pickoff and directed the beam to REFLB diode so we have something to monitor.
We started locking the arm again. But the CM board output to VCO would rail really quickly. We might have to enable LSC path to offload the VCO.
When locked, the arm transmission is not maximized, we're probably talking about 10% or so smaller transmission than the true peak. Looking on the oscilloscope, it's clear that the PDH error signal is not symmetric around the 00 carrier transmission when locked. We swapped the RF source with IFR generator and changed the modulation frequency +-30kHz or so (transverse mode spacing is 28.59kHz for Y arm) with a few kHz step at a time. We were able to observe clear steps in the transmission when we changed the modulation frequency, but this was not a big effect, we were never able to fix the asymmetry of the PDH error around the lock point. In the end we changed the oscillator back to the original.
Since it seems like we have a relatively large 20 mode, one thing to try is to improve the mode matching and see what happens.
This morning Pablo and I worked on getting (separate) cameras on both of the arm transmitted beams on ISCT1. We approximately followed the layout posted in llo alog 12198
To do this we had to add a lens in the Y arm path before the pick off for a camera, we used a Laser Optic +50mm roc. We used R=10% beam splitters for the pick offs for the cameras. We used analog cameras temporarily. I borrowed the POP cable (labeled POD) for the X arm trans camera, and used the REFL_PD camera that we have been using for the X+Y arms for the Y arm. The Y arm spot is on a baffle, the camera has a lens and is focused on the spot on the baffle, this makes a nicer image where we can clearly see the different cavity modes. We will do this for the X arm camera as well once we find a good lens.
Once we saw that both arms had nice IR flashes, I worked on aligning the diff beatnote. On the diff BBPD we have about 126 uW from the Y arm when it is locked and about 40uW from the X arm. We get -30dBm on the beatnote, (at 160MHz) I din't really try to optimize this since the cavity keeps dropping out of lock making it difficult to adjust.
[Sheila Arnaud]
EY oplev signal got somehow really noisy during the afternoon. It doesn't seem to be seimic noise. It is not clear why, but we should check this when possible.
Attached is a comparison spectra of EY oplev pitch at two different times this afternoon, with ISI isolated. Pink curve is the most recent one.
We started looking at this OpLev because we saw 1urad pp in pitch, which may have been one reason the arm cavity wasn't locking stably. We tried a few configurations with different blends.
With Tbetter everywhere except RZ, we see almost 1urad pit pp (over 16 seconds) and 0.4urad yaw.
With Tcrappy on Stage 2 and Tbetter on stage except for RZ we see 0.25pp PIT, which is now dominated by motion around the suspension resonances instead of the low frqeuency motion we saw with Tbetter.
Turning on and off sensor correction didn't seem to change the situation much, it is off right now.
(Keita, Alexa)
A more in depth alog will be posted later about what we did today at EY. For now I have attached the data for the OLTF we measured (mag --> 122.txt, phase --> 123.txt). The UGF was about 2.3kHz with a phase margin of a little less than 60 deg. The phase shifter was adjusted to 23ns. The PDH REFL servo board settings were as follows:
We also measured the peak to peak of the PDH error signal and found it to be ~816mVpp.
We remeasured the green power along ISCTEY (its possible we had the wrong laser setting on the powermeter when we measured this last):
Modified the output stages of the ETM UIM coil drivers, to give a factor of 4 more range per ECR E1400164. All four channels of each unit were modified as followed: R10 & R15 were changed to 2.2K Ohm R5 & R23 were changed to 2.0K Ohm Unit from EX - S0900303 Unit from EY - S0900304
The theoretical transconductance (@ DC) for each channel in the "High Dynamic Range Mode" is 0.625mA/V = -64dB Amps/Volt (from p2 of T140023 assuming a maximum 10V input).
This is a tabletop photodetector interface (D1002932-V5, S1103811) in ISCTEY that was misbehaving yesterday. I opened it and found this.
None of the power connectors for Thorlabs PDs and BBPDs were soldered to the board. And one of the connectors was loose (right-most one in the video). See how the legs move as I wiggle.
It's a miracle that it worked at some point. I soldered all of the connectors.
(If avi doesn't play, try mov file, though the time of the mov file seems to be out of whack.)
I opend and checked two randomly selected "Table-Top" boxes (D1002932-v4) from the twelve boxes in Mid-Y storage, serial numbers S1103796 and S1103810, for quality of build and found them to be in good shape. The front circuit card was soldered onto the panel connectors. All screws and fasteners were tight. The general quality of work looked good. Let's hope the box Keita found (S1103811) was an oversight.
I am about to start upgrading drivers on the DAQ as specified in WP4583. There will be NO DATA recorded during the period when the data concentrator is rebooted - as it is due for a full FSCK at boot, this will be a period of 10-15 minutes. For the remaining system upgrades, there will be intermittent access to the recorded data as those systems (h1nds1, h1fw1, h1broadcast0) are rebooted. Changes to h1nds0 and h1fw0 will happen at a later time. Further updates will be posted to this entry with specific downtime.
DAQ Downtime Report
h1dc0: 16:07:40 - 16:18:00 UTC There is NO data recorded by the DAQ during this period.
h1nds1: 16:26:00 - 16:28:20 UTC
h1fw1: 16:33:55 - 16:55:50 UTC There is NO data available via h1nds1/h1fw1 for this period. Use h1nds0/h1fw0 for frames ending/starting in this timeframe.
h1broadcast0: 17:06:55 - 17:16:00 UTC
Most installs were uneventful. However, on h1fw1, the MTU was not set to 9000 in /etc/conf.d/net as it is on h1fw0, which prevented daqd from running after restart. I changed /etc/conf.d/net to match and rebooted to fix; I have no idea how it ever worked before. On h1broadcast0, I disabled the items in local.start that are only useful for a data concentrator; h1broadcast0 being a clone of a data concentrator had these unnecessary additions.
Technical Details
(l inadvertently left these out of the original entry)
The change is to upgrade the Myricom ethernet adapter drivers for the DAQ broadcast network to version 1.5.3.p3, compiling them with the MYRI10GE_ALLOC_ORDER=2 option and using the big_rxring firmware at driver load. This is to attempt to reduce the number of dropped frames that are seen occasionally, most often on the framewriters, that also trigger 'retransmission request' errors in the daqd log. And additionally, on the data concentrator to make use of the MYRI10GE_THROTTLE option to see if tuning the packet emission rate has any effect for the receiving systems. The primary method of measuring any change is to use the SNMP monitoring of the DAQ broadcast switch to monitor the dropped/paused frames per host port. The same changes on the test stand indicate some improvement.
