We received the custom cables to test the new suspension electronics. They are under my desk. So we are ready.

This batch was a small one - the company says that they can make molded cables if we have a minimum order, something to consider I gues.s.

Update 1900 11 Feb: I verified that the pin outs of the cables are as we intended (for one set of each type of cable). Because this was a small order, the connectors have metal shells, and so for cable #2 (sat box to flange), the two shells are shorted to each other. I can't verify if the shield is isolated from the shell on J5 without cutting open the cable. One thing that occurred to me is that we should give pins 5,8,11 on J4 and 16,20,24 on J5 (respectively) unique identifiers. They should only be shorted to GND on the circuit board itself. To be fixed for the next iteration. I uploaded some photos here.

I was unable to measure the capacitance of the cable using the LCR meter, and didn't opt to try any other method.

Summary:

I did what I consider to be a comprehensive set of tests on the production version of the high voltage coil driver circuit. I think the performance is now satisfactory and the circuit is ready for the production build. Barring objections from anyone, I will ask Chub to place an order for components to stuff the 4 necessary units + 1 spare on Friday, 12 Feb (so that people have a full day to comment). A big thanks to Chub and the folks at JLCPCB for dealing with my incessant order requests and patiently supporting this build and letting me turn this around in 10 days - hopefully this is the end of this particular saga.

Schematic is here. All references to component designations are for v4 of the schematic.

Important design changes:

1. All I/O to this board will be via D9 connectors. This will allow bypassing the high voltage stage in future suspensions while retaining the same cable config in the suspension drive, if that is desired. Some re-arrangement of the grouping of coils was also done for consistency with the planned upgrade.
2. Differential receiving for the input signal from the Acromag. The nominal quad opamp is LT1125 but if we find noise issues (which I didn't), the OP497 has compatible PCB footprint.
3. Added input protection dual diode D6 to protect the PA95 as recommended in the datasheet. This should protect the IC if (for example) the HV line isn't plugged in but the Acromag input is non-zero.
4. Increased the feedback resistance from 30kohms to 12kohms. R16 through R21 are now 20k, while the old design had 5k. The purpose is to reduce the current demand in the feedback path, hopefully this opens up the number of DCPS we can use. To keep the overall gain of 31, the resistor R15 was upped from 1kohms to 4kohms.
5. Feedback capacitance reduced from 15 uF to 3 uF. This was largely for space saving / ease of layout on the PCB. The resulting corner frequency is increased slightly from 0.35 Hz to 0.45 Hz but this doesn't have any imapct on the performance of the circuit at frequencies of interest (1/2/pi/R/C had R=30k, C=15uF, now R=120k, C=3uF).
6. Added an R-C-R network at the output of the PA95, before the fast actuation signal is summed and sent to the OSEM coil.
   • This is probably the most important change, noise-performance wise.
   • The purpose of the network is to passively filter out the excess noise we saw at ~100 Hz (the corner from the 4kohm resistor + 3uF cap is at ~13 Hz, so factor of 10 filtering at 100 Hz, which was deemed sufficient, see earlier elogs in the thread). 
   • The Johnson noise contribution of the 20 kohm resistor is slightly higher than the original design which had a 25 kohm series resistor (but no R-C-R passive filter at the output of the PA95). But once again, this was deemed to have negligible effect on the performance at frequencies of interest to us.
   • The total current driving capability of the circuit is almost unchanged since the 20kohm + 4kohm nearly equals the old 25kohm resistance.
7. Made the Vmon paths for monitoring the HV output of the PA95 differential sending, seems like a good practise to follow for all new designs.
8. Added on-board bypass capacitors (2 x 10uF WIMA film caps) for cleaning up the HV supply noise.

Tests:

A series of tests were done. Note that only 1 channel was stuffed (I am out of PA95s), and the HP power supplies borrowed from Rich were used for the HV rails. For the +/-18V, a regular bench-top unit was used.

1. Low voltage stage tests
   • TF of the differential receiving stage was measured and the overall unity gain and corner at 24kHz were verified, see Attachment #1.
   • With the input of the circuit grounded, I measured the noise of the circuit at various points (see legends on Attachment #2). I didn't bother to do a detailed verification against a SPICE model as the levels seemed roughly what is expected.
2. Overall noise performance with HV stage enabled
   • For a range of drive currents, generated by applying the appropriate voltage using an Acromag XT1541 DAC module to the J1 connector, I measured the voltage on the circuit side of the 20 kohm resistor (by clipping onto the resistor leg. Note that the path to ground for the current was provided by connecting a 20 ohm resistor between pins 1 and 6 on J3a, and then grounding pin 6.
   • Results are shown in Attachment #3. 
   • For the drive currents at the edge of the range of operation, there is a small excess relative to lower drive currents. My best hypothesis for why this is happening is that the HV rail is too close to the requested output voltage (the HP units are rated for 320V, which is borderline if we want 300V at the output of the PA95). In any case, the R-C-R passive filter means that above ~60 Hz, there is excellent agreement between model and measurement.
3. Time domain tests
   • Used a function generator. to drive a 50 mHz, 3Vpp sine wave to the "Bias Input" (=J1), and monitored (i) pickoff of drive signal, (ii) High voltage output at the circuit side of the 20kohm resistor, and (iii) the Vmon output (=pins 1/6 on J4), all on a 100 MHz Tektronix scope.
   • Results shown in Attachment #4. Once again, I see no red flags.
   • While I had the unit hooked up to the scope, I also checked the time domain signal with the scope set to 100 ns/div time base. I saw no evidence of any oscillatory features, either when no input signal was applied, or when a DC signal was provided (in which case the scope was set to AC coupling). 👍 
4. Checked that the protection diodes at various locations in the circuit work.
5. Checked the pin-mapping on all 6 D9 connectors is consistent with the top level diagram in the schematic.

PCB remarks:

As I was stuffing the board, I noticed a few improvements that can be made. Just noting these here for documentation purposes - these changes are mostly aesthetic and I personally see no need to order another set of PCBs.

1. In some places, the silkscreen designators don't have the correct "orientation" relative to the component they are designating. I didn't find any serious ambiguity in terms of being misled to stuff the wrong component onto the wrong pads, but in the spirit of doing a professional job...
2. The current limiting resistors on the +/-430V LEDs (R37/R38) have footprints for leaded components rather than SMT (which is what the +/-15V LEDs have).
3. R45 and R46, the current limiting resistors for the rear panel power indicator LEDs, have 0805 footprint rather than 1206.
4. When I drew up the PCB, R23, the 4kohm resistor in the R-C-R network, was set up as a 1W resistor. Let's say there can be 15 mA flowing through this resistor - the power dissipated is 15e-3 ^2 * 4e3 is 0.9W, which is uncomfortably close to the limit. For all the tests above, I used a 3W resistor, and didn't find any serious noise issues. The drilled holes are a little tight for this higher power rated resistor, but I don't think this is a showstopper.

Communications with Apex:

I've been talking to support at Apex, and pointed out that I couldn't match the SPICE model performance even for a simple non-inverting amplifier with the PA95. The feedback I got from them was that 

1. They don't optimize the SPICE models for input noise and so it was a nice coincidence that model and measurement are somewhat close (but not exactly).
2. They recommend the PA194, which is actually advertised as "low-noise". The PA95 is apparently not a "low-noise" part, with its 2uVrms input noise. 

Whiel the PA194 is compatible with our voltage and current requirements for this application, it is ~3x the cost, and seems like the R-C-R output filter allows us to realize the goal of 1pA/rtHz, so I'm inclined to stick with the PA95.

Production assembly:

I'd prefer to get as much of the board stuffed by Screaming Circuits as possible. It took me ~3 hours to stuff 1 channel + the power supply parts, standoffs etc. So I estimate it'll take me ~6 hours to stuff the entire board. So not the end of the world if we have to do it in-house.

MC1 suspension is glitching again, so this is a good chance to install the new sat box and test it in the field.

As it currently stands the Center of Mass of the Adapter Ring/Optic assembly is 0.0175" out of the plane formed by the suspension wire. See Attachments. The side plate, along with the EQ stops are hidden to show the CoM and the plane.

Note: The changes discussed in the meeting with Calum have not been added and are a work in progress. These changes include:

- Adding a 45 deg chamfer to the both parallel faces of the adapter ring. This along with a modified bracket for the EQ stops will allow for easier adjustment of the screws. 

- Potentially changing material of adapter ring to stainless stell to more accurately emulate the mass of a 3" optic.

- Different adjustment mechanism of the "dumbell" at bottom of adapter ring to something similar to the VOPO suspension (will need to consult Calum further)

Gautam pointed out that there are extra Sm-Co magnets stored in the clean optics cabinet.

I took the magnet box out and put it on the rolling table next to south flow bench. The box contains 3 envelopes with magnets.

They are labelled as following:

FLUX 94 - 50 parts

FLUX 93 - 10 parts

FLUX 95 - 40 parts

(What is FLUX??)

The box also contains some procurement documents.

The clean and bake dcc says :

1. Ultrasonic clean in methanol for 10 minutes

2. Bake in vacuum at 177 C° for 96 hours

Should we go ahead with the C&B?

[jordan, gautam]

• Ran 60ft long cables from 1X4 to MC1/MC3 chamber flange, via overhead cable tray, and top of PSL enclosure for the last ~20ft. Note that it may be that the overhead cable trays cannot support the weight of the cables for 15 SOSs (total 30 shielded cables with 37 wires as twisted pairs) when we eventually add the optics for the BHD upgrade.
• Installed aLIGO satellite amplifer in 1X4.
• Tapped +/-20 V (which is the available voltage closest to the required +/-18V). For this, the Sorensens were powered down, and the actual taps were made from the fusable blocks powering the Trillium interface box. We made sure to leave an extra slot so that this kind of additional headache is not required for the next person doing such work.
• Once installed, I plugged in the dummy suspension box and verified that the unit performs as expected. 
• Some photos of the installation are here.

After this work, the IMC locked fine, the AS camera has the Michelson fringing, the fast CDS indicators are all green, and the seismometer BLRMS all look good - therefore, I claim no lasting damage was done as a direct result of today's work at 1X4. I will connect up the actual suspension at my leisure later today. Note that the MC1 glitches seem to have gone away, without me doing anything about it. Nevertheless, I think it's about time that we start testing the new hardware. 

Unrelated to this work: while I was testing some characteristics of the MC1 suspension (before we did any work in the VEA, you can see the timestamp in the ndscope), I noticed that the MC1 UL coil channel cannot actually be used to actuate on the optic. The coil driver Vmon channel demonstrates the appropriate response, which means that the problem is either with the Satellite box (it is just a feedthrough, so PCB trace damaged?) or with the OSEM itself (more likely IMO, will know more once I connect the new Satellite Amplifier up). I only show comparison for UL vs UR, but I checked that the other coils seem to be able to actuate the optic. This means we have been running for an indeterminate amount of time with only 3 face actuators on MC1, probably related to me having to do this work. 

Also unrelated to this work - while poking around at 1X5 rear, I noticed that the power connections to the existing Satellite Boxes are (understatedly) flaky, see connections to T1-T4 in Attachment #2..

The curie temp of SmCo seems about x2 (in K) of the one for NdFeB. i.e. 600K vs 1000K. So I believe 177degC = 450K is not an issue. Just make sure the curie temp, referring the specific property for the magnets from this company. (You already know the company from the procurement doc). It'd be great if you upload the doc on the 40m wiki.

Done.

Also, the magnets are nickel-plated. I guess that doesn't matter for the baking (Curie temp of 355 °C)?

The summary pages had failed because of a conda env change. We are dependent on detchar's conda environment setup to run the scripts on the cluster. However, for some reason, when they upgraded to python3.9, they removed the python3.7 env, which was the cause of the original failure of the summary pages a couple of weeks ago. Here is a list of thoughts on how the pipeline can be made better.

1. The status checking is pretty hacky at the moment.
   • I recommend not using shell via python to check if any condor jobs are "held".
   • Better is to use the dedicated python bindings. I have used this to plot the job durations, and it has worked well.
   • One caveat is that sometimes, there is a long delay between making a request via a python command, and condor actually returning the status. So you may have to experiment with execution times and build some try/except logic to catch "failures" that are just the condor command timing out and not an actual failure of the summare jobs.
2. The status check should also add a mailer which emails the 40m list when the job is held. 
   • With htcondor and python, I think it's easy to also get the "hold reason" for the job and add that to the mailer.
3. The job execution time command is not executing correctly anymore - for whatever reason, the condor_history command can't seem to apply the constraint of finding only jobs run by "40m", although running it without the constraint reveals that these certainly exist. Probably has to do with some recent upgrade of condor version or something. This should be fixed.
4. We should clear the archive files regularly. 
   • The 40m home directory on the cluster was getting full. 
   • The summary page jobs generate a .h5 archive of all the data used to generate the plots. Over ~1 year, this amounts to ~1TB.
   • I added the cleanArchive job to the crontab, but it should be checked.
   • Do we even need these archives beyond 1 day? I think they make the plotting faster by saving data already downloaded locally, but maybe we should have the cron delete all archive 
5. Can we make our own copy of the conda env and not be dependent on detchar conda env? The downside is that if something dramatic changes in gwsumm, we are responsible for debugging ourselves.

Remember that all the files are to be edited on nodus and not on the cluster.

Note from today's meeting:

1. Can we adjust the thickness of the cylindrical hole for the mirror to move the COM in the plane of the wires. (We should be able to do that)

and

2. Please check how much we can displace the COM by the bottom dumbbell.

There is some non-trivial sign flipping in the sensors/coils in this new setup because it is a hybrid one with the old interfacing electronics (D000210, D010001) and the new Satellite Amplifier (D080276). So I haven't yet gotten the damping working. I am leaving the PSL shutter closed and will keep working on this today/tomorrow. I have made various changes to the c1mcs realtime model and the c1susaux database record where MC1 is concerned. I have backups of the old ones so we can always go back to that if we so desire.

In the meantime, the PSL shutter is closed and there is no light to the IFO.

Update 1700: I've implemented some basic damping and now the IMC is now locked. The WFS loop runs away when I enable it, probably some kind of weird interaction with the (as of now untuned) MC1 local damping loops. I will write up a more detailed report later.

Update 2300: Did the following:

1. Re-calibrated the cts2um filter in all SENSOR filter banks to account for the increased transimpedance and LED drive current. I judged the overall scaling to be x0.25 but this can be calibrated against the bounce peak height for example (it lines up pretty well).
2. Re-measured the input matrix - it was very different from what was loaded. I am measuring this again overnight for some consistency.
3. Re-tuned the damping gains. Now the optic damps well, and the loops seem file to me, both via broadband noise injection TF and by step response metrics.
4. Yet, the WFS servo cannot be enabled. The WFS signal is summed in before the output matrix so I don't know why this would have a different behavior compared to the local damping, or indeed, why this has to be changed. Will need some (WFS) sensing/actuation matrix measurements to know more.

Dropping this for tonight, I'll continue tomorrow. Meanwhile, the OSEM input matrix measurement is being repeated overnight. PSL shutter is closed.

I am setting up a testing rig for the OSEMs we recently obtained. I found the schematic for the OSEM assembly from which the pin assignment can be read.

I connected the OSEM's pin plate to a female DB15 on a breakout board. I find the pin assignment (attachment 1, sorry for the image quality) to be:

There are several things that need to be done for each OSEM.

1. Measuring inductance of the coils. I checked that the measurement wires don't add any measurable inductance.

2. Check that the PDs and LEDs are alive (e.g. check forward voltage drop with fluke)

3. Energize the LED and PD.

4. Check PD DC level. For this, I might need the satellite box amplifier.

5. Check LED spot position on the PD.

6. Re-engrave OSEM S/N if needed.

I still need to figure a sensible scheme for points 3-5.

The WFS servo was recommissioned. The matrix can be tuned a bit more, but for now, I've recovered the old performance and the alignment doesn't seem to be running away, so I defer further tuning for later. The old Satellite box was handed over to Yehonathan for his characterization of the "spare" OSEMs.

This finishes the recovery of the MC1 suspension, I am now satisfied that the local damping loops are performing satisfactorily, that the WFS servo is also stable, and that POX/POY locking is recovered. On MC1, we even have 4 actuatable face OSEMs and the PIT(YAW) bias adjust slider even moves the optic in PIT(YAW), what a luxury. 

I've SDFed all the changes, and have backup of the old realtime model and C1SUSAUX_MC1 database files if we want to go back for whatever reason. The changes required to make this suspension work are different from what will eventually be required for the BHD suspensions (because of the hybrid iLIGO/aLIGO electronics situation), so I will not burden the readers with the tedious details.

Before installation, I performed a bunch of tests on the aLIGO sat amp. All the measurements were made with the dummy suspension box substituting for an actual suspension. Here are the results.

Attachment #1: Transimpedance amplifier noises.

• Measurement setup: J7 of the Satellite Amp goes to J9 on D1900068 front end (even though the connector is actually labelled "J3" on the box we have - maybe a versioning problem?). The outputs then go to a G=100 SR560 in AC coupled mode (the main purpose here was to block the large DC from the SR785, but I tacked on G=100 while I was at it).  
• Top panel shows the raw measured voltages.
• The bottom panel does a bunch of transformations:
  • Undoes the z:p = 3:30 Hz whitening on board the sat amp.
  • Undoes the G=100 gain of the SR560, and the AC coupling poles/zeros of SR560 and SR785.
  • Converts from voltage to current by dividing by the transimpedance gain, 242 kohms. 
• Some model curves are shown for comparing to the measured spectra. It may be possible that we don't need to modify the nominal z:p = 0.4:10 Hz - I don't think the nominal seismic level will saturate the output even with the 0.4:10 Hz whitening, and it gives us even more clearance to the ADC noise (although we don't need it, we are gain limited at those frequencies, this is mostly a suggestion to reduce the workload).
• The neon green curve is measured with the actual MC1 suspension plugged in, local damping enabled. It doesn't line up with the nosie floor of the bench tests, probably because the cts/um conversion factor could be off by some factor? Around 1 kHz, you can also see some broad peaks that are reminiscent of those seen in the MC_F spectrum after the c1psl Acromag upgrade. I hypothesize this is due to some poor grounding. Hopefully, once we get rid of the single-ended sending/receiving components in the suspension electronics chain, these will no longer be an issue.

Attachment #2: LED drive current source noises. I mainly wanted to check a claim by Rich in a meeting some time ago that the LED intensity fluctuations are dominated by inherent LED RIN, and not by RIN on the drive current. 

• Measurement setup: a pair of pomona mini-grabbers was used to clip onto TP3. I found the voltage noise to be sufficiently high that no preamplification was required, and the DC level was <1V, so I just used the SR785 in AC coupled mode. 
• The dummy suspension box was being driven while the measurement was being made (so the current source is loaded).
• One channel (CH6) shows anomalously high nosie. I confirmed this was present even after the box was plugged in for ~1 day, so can't be due to any thermal / equilibriating transients.
• I didn't check for consistency at the monitor testpoint, but that is exposed even with the MC1 suspension plugged in, so we can readily check. Anyways, from the corresponding photodiode curve in Attachment #1, it would seem that this excess RIN in the drive current has no measurable effect on the intensity fluctuations of the LED (the DC value of the paired PD is consistent with the others, ~6V DC). I must say I am surprised by this conclusion. I also checked for coherence between TP3 and the PD output using the SR785, and found none. 🤔 
• Nevertheless, for the remaining channels, it is clear that the drive current is not shot noise limited for <1kHz. This isn't great. One possible reason is that the collector voltage to Q1 is unregulated (my modeling suggests only ~10dB rejection of collector voltage fluctuations at the output). I believe the current source designed by Luis for A+ makes some of these improvements and so maybe Rich was referring to that design, and not the aLIGO Satellite Amplifier flavor we are using. Anyways, this is just academic I think, the performance is the unit is fine for our purposes.

I will update with the MC1 suspension characterization (loop TFs, step responses etc) later.

Yeah, it's really inconsistent. You had 35mA LED drive and the current noise of the noisy channel was 5e-7 A/rtHz at 1Hz. The RIN is 1.4e-5 /rtHz. The approx. received photocurrent is 30uA as we discussed today and this should make the noise around 4e-10 A/rtHz at 1Hz. However, the readout noise level is better than this level. (well below 1e-10 A/rtHz)

BTW, the IMC seemed continuously locked for 5 hours. Good sign.

A summary of things that need to be fabricated/purchased/done:

I forgot that I had already done some investigation into recovering the PRFPMI lock after my work on the RF source. I don't really have any ideas on how to explain (or more importantly, resolve) the poor seperation of MICH and PRCL sensed in our 3f (but also 1f) photodiodes, see full thread here. Anyone have any ideas? I don't think my analysis (=code) of the sensing matrix can be blamed - in DTT, just looking the spectra of the _ERR_DQ channels for the various photodiodes while a ssingle frequency line is driving the PRM/BS suspension, there is no digital demod phase that decouples the MICH/PRCL peak in any of the REFL port photodiode spectra.

Adjusting the thickness of the cylindrical hole for the mirror on the 2" optic sleeve, from .6875" to .61" thick, moves the CoM to 0.0003" out of plane from the suspension wire. This is with the dumbell at its neutral point.

How close to zero do we need this to be? More fine tuning of that thickness can get it to zero, but this would require much tighter machining tolerance on that hole depth.

Moving the dumbell towards the back of the SOS assembly (noted as negative direction, with origin at the plane formed by the wires), moves the CoM to -0.002" from the plane.

Moving the dumbell towards the front of the SoS assmebly (positive direction wrt the plane formed by the suspension wire), moves the CoM to +0.0022" from the plane.

So the total adjustment range with the dumbell is -0.002"to 0.0022", with the plane formed by the wires as the origin.

See Attachments

We want to move the CoM with the adjustment range so that the residual deviation is adjusted by the bottom dumbbell. 0.0003" is well within the range and good enough.

I briefly talked with Jordan about this. This suspension will have OSEMs right? With 400ohm series resistance for the coil drivers, we will have ~+/-20mrad actuation range. Of course we'd like to use as much of this for interferometry and not static pitch alignment correction (possibly even increase the series resistance to relax the dewhitening requirements). But what is the target adjustability range in mrad with the dumbell/screw config? My target in the linked elog is 500urad (not any systematic optimum, but will allow us to use most of the DAC range for interferometry). Are these numbers in inches commensurate with this 500urad?

On a related note - are there grooves for the wires to sit in on the side of the sleeve? We looked at the solidworks drawing, and noticed that the groove doesn't extend all the way to the top of the clamp. Also, the material of both the clamping piece and the piece onto which the wire is pressed onto is SS. Don't we want them to be Aluminium (or something softer than the wire) so that the wire makes a groove when the clamp is tightened?

Jordan's screenshot actually shows that the vertical distance (Y) is 0.0000". We want to have the vertical distance of CoM from the wire clamping point to be 0.9mm in the nominal SOS design (this might need to be adjusted to have a similar pitch resonant freq for the different inertia of moment). Let's say it is ~mm ish.

The full range of the bottom dumbbell adjustment gives us the CoM adjustment range of +/-0.002” = +/-50um. This corresponds to an alignment range of +/-50mrad. And we want to set it within +/-500urad.
So we need to adjust the dumbbell position with the precision of 1/100 of the full range (precision of 0.5um).

The groove does not extend to the top of the clamp. The groove shallower than the wire diameter cause the hysteresis of the alignment. Also, the material of the pieces should be stainless steel. Al clamp is softer than the wire and will cause the groove to be dug on the material, causing increased bending friction and hysteresis again.

Saying, all of our suspended masses with Al stand-offs are suffering this issue to some extent. That was the reason to buy the ruby standoffs.

I would:

1. look at the free swingin michelson. Should be able tu null that siggnal in all ports to define the Q phase.
2. If things are weird, put an RF signal nto the demod board mhich is offset from the LO by ~100 Hz and verify the demod/whitening chain is kosher.
3. Lock PRMI and drive lines > 200 Hz. If PRC/MICH are not orthogonal, then there may be a mis tuning of RF SB wavelength and cavity lengths.
4. IF PRMI is sort of healthy, we could be having a weird SB resonance in the arms.

In prep to try some of these debugging steps, I did the following.

1. ndscope updated from 0.7.9 to 0.11.3 on rossa. I've been testing/assisting the development for a few months now and am happy with it, and like the new features (e.g. PDF export). v0.7.9 is still available on the system so we can revert whenever we want.
2. Arms locked on POX/POY, dither aligned to maximize TRX/TRY, normalization reset.
3. PRMI locked, dither aligned to maximize POPDC.
4. All vertex oplevs re-centered on their QPDs.

While working, I noticed that the annoying tip-tilt drift seems to be worse than it has been in the last few months. The IPPOS QPD is a good diagnostic to monitor stability of TT1/TT2. While trying to trend the data, I noticed that from ~31 Jan (Saturday night/Sunday morning local time), the IP-POS QPD segment data streams seem "frozen", see Attachment #1. This definitely predates the CDS crash on Feb 2. I confirmed that the beam was in fact incident on the IPPOS QPD, and at 1Y2/1Y3 that I was getting voltages going into the c1iscaux Acromag crate. All manner of soft reboots (eth1 network interface, modbusIOC service) didn't fix the problem, so I power cycled the acromag interface crate. This did the trick. I will take this opportunity to raise again the issue that we do not have a useful, reliable diagnsotic for the state of our Acromag systems. The problem seems to not have been with all the ADC cards inside the crate, as other slow ADC channels were reporting sensible numbers.

Anyways, now that the QPD is working again, you can see the drift in Attachment #2. I ran the dither alignment ~4 hours ago, and in the intervening time, the spot, which was previously centered on the AS camera CRT display, has almost drifted completely off (my rough calibration is that the spot has moved 5mm on the AS CCD camera). I was thinking we could try installing the two HAM-A coil drivers to control the TTs, this would allow us to rule out flaky electronics as the culprit, but I realize some custom cabling would be required, so maybe not worth the effort. The phenomenology of the drift make me suspect the electronics - hard for me to imagine that a mechanical creep would stop creeping after 3-4 hours? How would we explain the start of such a mechanical drift? On the other hand, the fact that the drift is almost solely in pitch lends support to the cause being mechanical. This would really hamper the locking efforts, the drift is on short enough timescales that I'd need to repeatedly go back and run the dither alignment between lock attempts - not the end of the world but costs ~5mins per lock attempt.

On to the actual tests: before testing the hardware, I locked the PRMI (no ETMs). In this configuration, I'm surprised to see that there is nearly perfect coherence between the MICH and PRCL error signals between 100Hz-1kHz 🤔 . When the AS55 demodulated signals are whitened prior to digitization (and then de-whitened digitally), the coherence structure changes. The electronics noise (measured with the PSL shutter closed) itself is uncorrelated (as it should be), and below the level of the two aforementioned spectra, so it is some actual signal I'm measuring there with the PRMI locked, and the coherence is on the light fields on the photodiode. So it would seem that I am just injecting a ton of AS55 sensing noise into the PRCL loop via the MICH->PRM LSC output matrix element. Weird. The light level on the AS55 photodiode has increased by ~2x after the September 2020 vent when we removed all the unused output optics and copper OMC. Nevertheless, the level isn't anywhere close to being high enough to saturate the ADC (confirmed by time domain signals in ndscope).

To get some insight into whether the whole RF system is messed up, I first locked the arm cavities with POX and POY as the error signals. Attachment #3 shows the spectra and coherence betweeen these two DoFs (and the dark noise levels for comparison). This is the kind of coherence profile I would expect - at frequencies where the loop gain isn't so high as to squish the cavity length noise (relative to laser frequency fluctuations), the coherence is high. Below 10 Hz, the coherence is lower than between 10-100 Hz because the OLG is high, and presumably, we are close to the sensing noise level. And above ~100 Hz, POX and POY photodiodes aren't sensing any actual relative frequency fluctuations between the arm length and laser frequency, so it's all just electronics noise, which should be incoherent.

The analogous plot for the PRMI lock is shown in Attachment #4. I guess this is telling me that the MICH sensing noise is getting injected into the PRCL error point between 100Hz-1kHz, where the REFL11 photodiode (=PRCL sensor) isn't dark noise limited, and so there is high coherence? I tuned the MICH-->PRM LSC output matrix element to minimize the height of a single frequency line driving the BS+PRM combo at ~313Hz in the PRCL error point. 

All the spectra are in-loop, the loop gain has not been undone to refer this to free-running noise. The OLGs themselves looked fine to me from the usual DTT swept sine measurements, with ~100 Hz UGF.

I'm curious to see if the demod phase for MICH in REFL & AS chamges between thi simple Mcihelson and PRMI. IF there's a change, it could point to a PRCL/f2 mismatch.

But I would still bet on demod chain funniness.

Continuing with the OSEM testing. I measure the resistance of the wires from the RLC meter to the coil to be ~ 0.9ohm. I will subtract this number from the subsequent coil resistance measurements.

I took the old MC1 satellite box for powering the PD and LED in the OSEM assemblies. I connected an idc breakout board to J4 and powered the box with a DC supply according to the box's schematics.

After getting a bit confused about some voltage reading from one of the PD readouts Gautam came and basically redid the whole rig. Instead of using breakout boards, he powered the amplifier circuit directly from the DC supply. Then, to connect the OSEM pinboard directly to the J1 connector he made a DB25 ribbon cable where the two connectors are opposite to one another to mimic the situation with the vacuum feedthru. He also connected a DB25 to BNCs breakout cable, specific to the satellite box, to the J3 port to read the individual PDs through a BNC connector. We managed to confirm the normal operation of one OSEM (Normal PD voltage and LED light spot hitting the PD using a camera with no IR filter).

It was getting a bit late. Going to start checking the OSEMs tomorrow.

The defaults cds-crtools didn't come with some of the older ezcautils (like ezcaread, ezcawrite etc). This is now packaged for debian, so I installled them with sudo apt update && sudo apt install dtt-ezca-tools on rossa. Now, we don't have to needlessly substitute the commands in our old shell scripts with the more modern z read, z write etc.

I am wondering if there is a relative implicit minus sign between the z servo and ezcaservo commands...

I measured the conversion efficiencies for all the RFPD demod boards except the POP port ones. An RF source was used to drive the PD input on the demod board, one at a time, and the I/F outputs were monitored on a 300 MHz oscilloscope. The efficiency is measured as the percentage ratio V_IF / V_RF. 

I will upload the full report later, but basically, the numbers I measured today are within 10% of what I measured in 2017 when I previously did such a characterization. The orthogonality also seems fine. 

I believe I restored all the connections at 1Y2 correctly, and I can lock POX/POY and PRMI on 1f signals after my work. I will do the noise characterization tomorrow - but I think this test already rules out any funkiness with the demod setup (e.g. non orthogonality of the digitized "I" and "Q" signals). The whitening part of the analog chain remains untested.

Update 2/23 1215: I've broken up the results into the demod boards that do not (Attachment #1) and do (Attachment #2) have a D040179 preamp installed. Actually, the REFL11 AO path also has the preamp installed, but I forgot to capture the time domain data for those channels. The conversion efficiency inferred from the scope was ~5.23 V/V, which is in good agreement with what I measured a few years ago.

• The scope traces were downloaded.
• The resulting X/Y traces are fitted with ellipses to judge the gain imbalance and orthogonality.
• The parameter phi is the rotation of the "bounding box" for the fitted ellipses - if the I and Q channels are exactly orthogonal, this should be either 0 or 90 degrees. There is significant deviation from these numbers for some of the demodulators, do we want to do something about this? Anyways, the REFL11 and AS55 boards, which are used for PRMI locking, report reasonable values. But REFL165 shows an ellipse with significant rotation. This is probably how the CDS phase rotator should be tuned, by fitting an ellipse to the digitized I/Q data and then making the bounding box rotation angle 0 by adjusting the "Measured Diff" parameter.
• The gain imbalance seems okay across the board, better than 1dB.
• The POX and POY traces are a bit weird, looks like there is some non-trivial amount of distortion from the expected pure sinusoid.
• I measured the LO input levels going into each demod board - they all lie in the range 2-3dBm (measured with RF power meter), which is what is to be expected per the design doc. The exception the the 165 MHz LO line, which was 0.4 dBm. So this board probably needs some work. 
• As I mentioned earlier, the conversion efficiencies are consistent with what I measured in 2017. I didn't break out the Eurocards using an extender and directly probe the LO levels at various points, but the fact that the conversion efficiencies have not degraded and the values are consistent with the insertion loss of various components in the chain make me believe the problem lies elsewhere. 

For completeness, I will measure the input terminated I/F output noise levels later today. Note also that my characterization of the optical modulation profile did not reveal anything obviously wrong (to me at least). 

I can't obtain a consistent view between the existing drawings/photographs and your pin assignment. Please review the pin assignment again to check if yours is correct.

Looking from the back side and the wires are going down, the left bottom pin is "Coil Start" and the upper right adjacent pin is "Coil End". (See attachment)
So in your picture 1 should be the coil start and 4 should be the coil end, but they are not according to your table.

Yes, my phone camera mirrored the image. Sorry for the confusion.

I see you already uploaded the correct pin assignment.

Continuing with the new rig, I measure the resistance of the cable leading to the coil to be 0.08+(0.52-0.08)+(0.48-0.08)=0.9ohm.

Total: 63 OSEMS. Centered working OSEMS: 42. Will upload a more detailed summary to the wiki soon.

Note: The Olympus camera is eating the AA camera very quickly (need to replace every 1.5 days). I'm guessing this is because of the corrosion in the battery housing.

I measured the noise of the I/F outputs of all the LSC demodulators. I made the measurement in two conditions, one with the RF input to the demodulators terminated with 50 ohms to ground, and the other with the RFPD plugged in, but the PSL shutter closed (so the PD dark noise was the input to the demodulator). The LO input was driven at the nominal level for all measurements (2-3 dBm going in to the LO input, measured with the RF power meter, but I don't know what the level reaching the mixer is, because there is a complicated chain of ERA amplifiers and attenuators that determine what the level is). 

As in the previous elog, I have grouped the results into boards that do not (Attachment #1) and do (Attachment #2) have the low noise preamp installed. The top row is for the "Input terminated" measurements, while the bottom is with the RFPD plugged in, but dark. I think not a single board shows the "expected" noise performance for both I and Q channels. In the case where the preamp isn't installed and assuming the mixer is being driven with >17dBm LO, we expect the mixer to demodulate the Johnson noise of 50 ohms, which would be ~1nV/rtHz, and so with the SR785, we shouldn't measure anything in exceess of the instrument noise floor. With the low noise preamp installed, the expected output noise level is ~10nV/rtHz, which should just about be measurable (I didn't use any additional Low Noise front end preamp for these measurements). The AS55_I channel shows noise consistent with what was measured in 2017 after it was repaired, but the Q channel shows ~twice the noise. It seemed odd to me that the Q channels show consistently higher noise levels in general, but I confirmed that the SR785 channel 2 did not show elevated instrument noise at least when terminated with 50 ohms, so seems like a real thing.

While this is clearly not an ideal state of operation, I don't see how this can explain the odd PRMI sensing.

I did the characterization discussed at the meeting today.

1. RF signal at 100 Hz offset from the LO frequency was injected into the PD input on the demod boards.
2. The digitized CDS channels were monitored. I chose to look at the C1:LSC-{PD}_I_OUT and C1:LSC-{PD}_Q_OUT channels. This undoes the effect of the analog whitening, but is before the digital phase rotation.
3. Attachments #1 and Attachments #2 are for the case where the analog whitening is not engaged, white Attachments #3 and Attachments #4 are for when the whitening is engaged, and they look the same (as they should), which rules out any crazy mismatch between the analog filter and the digital dewhitening filter.
4. I have absorbed the flat whitening gain applied to the various PDs in the cts/V calibration indicated on these plots. So the size of the ellipse is proportional to the conversion gain.

I think this test doesn't suggest anything funky in the analog demod/whitening/AA/digitization chain. We can repeat this process after the demod boards are repaired and use the angle of rotation of the ellipse to set the "D" parameter in the CDS phase rotator part, I didn't do it today.

While working at the LSC rack, I lost the input pointing into the IFO (the TT wiring situation is apparently very fragile, and this observation supports the hypothesis that the drifting TTs are symptomatic of some electronics issue). After careful beam walking, it was recovered and the dither alignment system was used to maximize TRX/TRY once again. No lasting damage done. If I can figure out what the pin-mapping is for the TT coils in vacuum, I'm inclined to try installing the two HAM-A coil drivers to control the TTs. Does anyone know where I can find said pin-out? The wiki page links seem broken and there isn't a schematic available there...

Ok it should be possible to back it out from the BOSEM pin out, and the mapping of the in-vacuum quadrupus cable, though careful accounting of mirroring will have to be done... The HAM-A coil driver actually already has a 15 pin output like the iLIGO coil drivers that are currently in use, but the pin mapping is different so we can't just replace the unit. On the bright side, this will clear up 6U of rack space in 1Y2. In fact, we can also consider hooking up the shadow sensor part of the BOSEMs if we plan to install 2 HAM-A coil drivers + 1 Dual satellite amplifier combo (I'm not sure if this number of spares is available in what we ordered from Todd).

I've started writing up a rough testing sequence for getting the three new front-ends operational (c1bhd, c1sus2, c1ioo). Since I anticipate this plan undergoing many updates, I've set it up as a Google doc which everyone can edit (log in with LIGO.ORG credentials).

Link to planning document

Please have a look and add any more tests, details, or concerns. I will continue adding to it as I read up on CDS documentation.

I've set up one 2U server unit received from KT in the control room area (the fans in it are pretty loud but other than that, no major issues with it being tested here). The IPMI interface is enabled and the machine is also hooked up to the martian network for diskless boot (usual login creds). I also installed a Dolphin interface card and the one-stop-systems host side card, and both seem to be recognized (the OSSI card has the "PWR" LED on, the Dolphin card shows up in the list of PCIe devices, but has no LEDs on at the moment). I actually can't find the OSSI card in the list of PCI devices, but maybe I'm not looking for the right device ID, or it needs the cable to be connected to the I/O chassis side to be recognized. Anyways, let the testing begin.

The machine previously christened c1bhd has been turned off and completely disconnected from the martian network (though I didn't bother removing it from the rack for now).

BTW - I think most of our 19" racks are deformed from years of loading - I spent 5 mins trying to install the rails (at 1Y1 and 1X7) to mount the supermicro on, and couldn't manage it. I could be missing some technique/trick, but i don't think so.

After all the work at the LSC rack over the last couple of days, I re-locked the PRMI (ETMs misaligned), and measured the sensing matrix once again. The PRMI was locked using 1f error signals, with AS55_Q as the MICH sensor and REFL11_I as the PRCL sensor. As shown in Attachment #1, the situation has not changed, there is still no separation between the DoFs in the REFL signals. I will measure the MC lock point offset using the error point dither technique today to see if there is something there.

The Marconi frequency was tuned by looking at 

1. The ~3.68 MHz (= 3*f1 - fIMC) peak at the IMC servo error point, TP1A, and
2. The ~25.8 MHz (= 5*f1 - fIMC) peak at the MC REFL PD monitor port. The IMC error point is not a good place to look for this signal because of the post-demodulation low pass filter (indeed, I didn't see any peak above the analyzer noise floor).

The nominal frequency was 11.066209 MHz, and I found that both peaks were simultaneously minimized by adjusting it to 11.066195 MHz, see Attachment #1. This corresponds to a length change of ~20 microns, which I think is totally reasonable. I guess the peaks can't be nulled completely because of imbalance in the positive and negative sidebands. 

Then, I checked for possible offsets at the IMC error point, by injecting a singal to the AO input of the IMC servo board (using the Siglent func gen), at ~300 Hz. I then looked at the peak height at the modulation frequency, and the second harmonic. The former should be minimized when the cavity is exactly on resonance, while the latter is proportional to the modulation depth at the audio frequency. I found that I had to tweak the MC offset voltage slider from the nominal value of 0V to 0.12 V to null the former peak, see Attachment #2. After accounting for the internal voltage division factor of 40, and using my calibration of the IMC error point as 13 kHz/V, this corresponds to a 40 Hz (~50 microns) offset from the true resonant point. Considering the cavity linewidth of ~4 kHz, I think this is a small detuning, and probably changes from lock to lock, or with time of day, temperature etc.

Conclusion: I think neither of these tests suggest that the IMC is to blame for the weirdness in the PRMI sensing, so the mystery continues.

Koji asked me to test the production version of the coil driver with the KEPCO HV supplies. See Attachment #1 for the results. For comparison, I've added a single trace from the measurements made with the HP supplies. I continue to see excess noise with the KEPCO supplies. Note that in the production version of the board that was tested, there are a pair of 10uF bypass capacitors on the board for the HV supply lines. It is possible that one or both KEPCO supplies are damaged - one was from the ASY setup and one I found in the little rack next to 1X2. The test conditions were identical to that with the HP supplies (as best as I could make it so).

This is very disappointing. Even with KEPCO linear supply with the improved HV driver circuit, the noise level is significantly higher than the 20kOhm R thermal noise.

What is special with the HP supplies? Can you replace KEPCOs with the HP supply, one by one to specify which one is making the noise bad?

I will try the test of switching out KEPCOs one at a time for the HP. Given that the passive RC filter doesn't filter out the excess, I am wondering if the KEPCO is somehow polluting the circuit ground? The measurement was made between the circuit side of R24 (see schematic) and a ground testpoint, so the passive R23/C15 pole should filter the noise above ~15 Hz.

On Friday evening we checked out a few more things, somewhat overlapping with previous tests. All tests done with PRMI on carrier lock (REFL11_I -> PRC, AS55_Q-> MICH):

• check that PRC drive appropriately minimizes in REFL55_Q. I:Q ratio is ~100:1; good enough.
• put sine waves around 311 and 333 Hz into PRCL and MICH at the LSC output matrix using awggui and LSC osc. not able to adjust LSC/OSC output matrix to minimize the MICH drive in REFL_I.
• measured the TF from BS & PRM LSC drive to the REFL55_I/Q outputs. very nearly the same audio frequency phase, so the problem is NOT in the electronics || mechanical transfer functions of the suspensions.

Further questions:

1. is this something pathological in the PRMI carrier lock? we should check by locking on sidebands to REFL55 and REFL165 and repeat tests.
2. Can it be a severe mode mismatch from IMC output to PRMI mode? the cavity should be stable with the flipped folding mirrors, but maybe something strange happening. How do we measure the mode-matching to the PRC quantitatively?
3. huge RAM is ruled out by Gautam's test of looking at REFL demod signals: dark offset vs. offset with a single bounce off of PRM (with ITMs mis-aligned)
4. if there is a large (optical) offset in the AS55_Q lock point, how big would it have to be to mess up the REFL phase so much?
5. what is going on with the REFL55 whitening/AA electronics?

unrelated note: Donatella the Workstation was ~3 minutes ahead of the FE machines (you can look at the C0:TIM-PACIFIC_STRING on many of the MEDM screens for a rough simulacrum). When the workstation time is so far off, DTT doesn't work right (has errors like test timed out, or other blah blah). I installed NTP on donatella and started the service per SL7 rules. Since we want to migrate all the workstations to Debian (following the party line), lets not futz with this too much.

gautam, 1 Mar 1600: In case I'm being dumb, I attach the screen grab comparing dark offset to the single bounce off PRM, to estimate the RAM contribution. The other signals are there just to show that the ITMs are sufficiently misaligned. The PRCL PDH fringe is usually ~12000 cts in REFL11, ~5000cts in REFL55, and so the RAM offset is <0.1% of the horn-to-horn PDH fringe.

P.S. I know generally PNGs in the elog are frowned upon. But with so many points, the vector PDF export by NDS (i) is several megabytes in size and (ii) excruciatingly slow. I'm proposing a decimation filter for the export function of ndscope - but until then, I claim plotting with "rasterized=True" and saving to PDF and exporting to PNG are equivalent, since both yield a rasterized graphic.

I looked into this a bit more and crossed off some of the points Rana listed. In order to use REFL 55 as a sensor, I had to fix the frequent saturations seen in the MICH signals, at the nominal (flat) whitening gain of +18 dB. The light level on the REFL55 photodiode (13 mW), its transimpedance (400 ohm), and this +18dB (~ x8) gain, cannot explain signal saturation (0.7A/W * 400 V/A * 8 ~ 2.2kV/W, and the PRCL PDH fringe should be ~1 MW/m, so the PDH fringe across the 4nm linewidth of the PRC should only be a couple of volts). Could be some weird effect of the quad LT1125. Anyway, the fix that has worked in the past, and also this time, is detailed here. Note that the anomalously high noise of the REFL55_Q channel in particular remains a problem. After taking care of that, I did the following:

1. PRMI (ETMs misaligned) locking with sidebands resonant in the PRC was restored - REFL55_I was used for PRCL sensing and REFL55_Q was used for MICH sensing. The locks are acquired nearly instantaneously if the alignment is good, and they are pretty robust, see Attachment #1 (the lock losses were IMC related and not really any PRC/MICH problem).
2. Measured the loop OLTFs using the usual IN1/IN2 technique. The PRCL loop looks just fine, but the MICH loop UGF is very low apparently. I can't just raise the loop gain because of the feature at ~600 Hz. Not sure what the origin of this is, it isn't present in the analogous TF measurement when the PRMI is locked with carrier resonant (REFL11_I for PRCL sensing, AS55_Q for MICH sensing). I will post the loop breakdown later. 
3. Re-confirmed that the MICH-->PRCL coupling couldn't be nulled completely in this config either.
   • The effect is a geometric one - then 1 unit change in MICH causes a 1/sqrt(2) change in PRCL. 
   • The actual matrix element that best nulls a MICH drive in the PRCL error point is -0.34 (this has not changed from the PRMI resonant on carrier locking). Why should it be that we can't null this element, if the mechanical transfer functions (see next point) are okay?
4. Looked at the mechanical actuator TFs are again (since we forgot to save plots on Friday), by driving the BS and PRM with sine waves (311.1 Hz), one at a time, and looking at the response in REFL55_I and REFL55_Q. Some evidence of some funkiness here already. I can't find any configuration of digital demod phase that gives me a PRCL/MICH sensing ratio of ~100 in REFL55_I, and simultaneously, a MICH/PRCL sensing ratio of ~100 in REFL55_Q. The results are in Attachments #5. 
5. Drove single frequency lines in MICH and PRCL at 311.1 and 313.35 Hz respectively, for 5 minutes, and made the radar plots in Attachments #2 and #3. Long story short - even in the "nominal" configuration where the sidebands are resonant in the PRC and the carrier is rejected, there is poor separation in sensing. 
   • Attachments #2 is with the digital REFL55 demod phase set to 35 degrees - I thought this gave the best PRCL sensing in REFL55_I (eyeballed it roughly by looking at ndscope free-swinging PDH fringes).
   • But the test detailed in bullet #4, and Attachments #2 itself, suggested that PRCL was actually being sensed almost entirely in the Q phase signal.
   • So I changed the digital demod phase to -30 degrees (did a more quantitative estimate with free-swinging PDH fringes on ndscope, horn-to-horn voltages etc).
   • The same procedure of sine-wave-driving now yields Attachments #3. Indeed, now PRCL is sensed almost perfectly in REFL55_I, but the MICH signal is also nearly in REFL55_I. How can the lock be so robust if this is really true? 
6. Attachments #4 shows some relevant time domain signals in the PRMI lock with the sidebands resonant. 
   • REFL11_I hovers around 0 when REFL55_I is used to sense and lock PRCL - good. The m/ct calibration for REFL11_I and REFL55_I are different so this plot doesn't directly tell us how good the PRCL loop is based on the out-of-loop REFL11_I sensor.
   • ASDC is nearly 0, good.
   • POP22_I is ~200cts (and POP22_Q is nearly 0) - I didn't see any peak at the drive frequency when driving PRCL with a sine wave, so no linear coupling of PRCL to the f1 sideband buildup, which would suggest there is no PRCL offset.
   • Couldn't do the analogous test for AS110 as I removed that photodiode for the AS WFS - it is pretty simple to re-install it, but the ASDC level already doesn't suggest anything crazy here.

Rana also suggested checking if the digital demod phase that senses MICH in REFL55_Q changes from free-swinging Michelson (PRM misaligned), to PRMI aligned - we can quantify any macroscopic length mismatch in the PRC length using this measurement. I couldn't see any MICH signal in REFL55_Q with the PRM misaligned and the Michelson fringing. Could be that +18dB is insufficient whitening gain, but I ran out of time this afternoon, so I'll check later. But not sure if the double attenuation by the PRM makes this impossible.

The PRM violin filter seems very suboptimal - the gain peaking shows up in the MICH OLTF, presumably due to the MICH-->PRM LSC output matrix. I plot the one used for the BS in comparison in Attachment #1, seems much more reasonable. Why does the PRM need so many notches? Is this meant to cover some violin modes of PR2/PR3 as well? Do we really need that? Are the PR2/PR3 violin modes really so close in frequency to that for the 3" SOS? I suppose it could be since the suspension wire is thinner and the mass is lighter, and the two effects nearly cancel, but we don't actuate on PR2/PR3? According to the earlier elog in this thread, this particular filter wasn't deemed offensive and was left on.

Indeed, as shown in Attachment #2, I can realize a much healthier UGF for the MICH loop with just a single frequency notch (black reference trace) rather than using the existing "PRvio1,2" filter (FM2), (live red trace). The PR violins are eating so much phase at ~600 Hz.

agreed, seems excessive. I always prefer bandstop over notch in case the eigenfrequency wanders, but the bandstop could be made to be just a few Hz wide.

There were multiple problems with the REFL55 demod board. I fixed them and re-installed the board. The TFs and noise measured on the bench now look more like what is expected from a noise model. The noise in-situ also looked good. After this work, my settings for the PRMI sideband lock don't work anymore so I probably have to tweak things a bit, will look into it tomorrow.

I finished testing the OSEMs. I put all the OSEMs back in the box. The OSEMS were divided into several bags. I put the OSEM box next to the south flow bench on the floor.

I have uploaded the OSEM catalog to the wiki. I will upload the LED spot images later.

In summary:

Total 64 OSEMS, 31 long, 33 short.

Perfectly centered LED spots, ready for C&B OSEMS: 30, 12 long, 18 short.

Perfectly centered LED spots, need some work (missing pigtails, weird screws) OSEMS: 7, 5 long, 2 short.

Slightly off-centered (subjective) LED spots, ready for C&B OSEMS: 20, 7 long, 13 short.

Slightly off-centered (subjective) LED spots, need some work (missing pigtails, weird screws) OSEMS: 4 long

Defective OSEMS or LED spot way off-center: 3.

After this work, I measured that the orthogonality was poor. I confirmed on the bench that the PQW-2-90 was busted, pin 2 (0 degree output) showed a sensible signal half of the input, but pin 6 had far too small an output and the phase difference was more like 45 degrees and not 90 degrees. I can't find any spares of this part in the lab - however, we do have the equivalent part used in the aLIGO demodulator. Koji has kindly agreed to do the replacement (it requires a bit of jumper wiring action because the pin mapping between the two parts isn't exactly identical - in fact, the circuit schematic uses a transformer to do the splitting, but at some unknown point in time, the change to the minicircuits part was made. Anyway, until this is restored, I defer the PRMI sideband locking.

I see no evidence of anything radically different from my PSL table optical characterization in the IMC transmitted beam, see Attachment #1. The lines are just a quick indicator of what's what and no sophisticated peak fitting has been done yet (so the apparent offset between the transmission peaks and some of the vertical lines are just artefacts of my rough calibration I believe). The modulation depths recovered from this scan are in good agreement with what I report in the linked elog, ~0.19 for f1 and ~0.24 for f2. On the bright side, the ALS just worked and didn't require any electronics fudgery from me. So the mystery continues.

A new hybrid splitter (DQS-10-100) was installed. As the amplification of the final stage is sufficient for the input level of 3dBm, I have bypassed the input amplification (Attachment 1). One of the mixer was desoldered to check the power level. With a 1dB ATTN, the output of the last ERA-5 was +17.8dBm (Attachment 2). (The mixer was resoldered.)

With LO3dBm. RF0dBm, and delta_f = 30Hz, the output Vpp of 340mV and the phase difference is 88.93deg. (Attachment 3/4, the traces were averaged)

0 dBm ~ 0.63 Vpp. I guess there is ~4dB total loss (3dB from splitter and 1dB from total excess loss above theoretical from various components) between the SMA input and each RF input of the JMS-1-H mixer, which has an advertised conversion loss of ~6dB. So the RF input to each mixer, for 0dBm to the front panel SMA is ~-4dBm (=0.4 Vpp), and the I/F output is 0.34Vpp. So the conversion loss is only ~-1.5 dB? Seems really low? I assume the 0.34 Vpp is at the input to the preamp? If it's after the preamp, then the numbers still don't add up, because with the nominal 6dB conversion loss, the output. should be ~2Vpp? I will check it later.