The RGA is removed for repaire. It's volume at atmophere and sealed.. P4 reading of 38 Torr is not correct.
The IFO room temp is up
The IFO room temp is up a bit and it is coming down. The out side temp is not really high.
Found the MC autolocker kept failing, It turned out that c1iool0 and c1psl went bad and did not accept the epics commands.
Went to the rack and power cycled them. Burt resotred with the snapshot files at 5:07 today.
The PMC lock was restored, IMC was locked, WFS turned on, and WFS output offloaded to the bias sliders.
The PMC seemed highly misaligned, but I didn't bother myself to touch it this time.
I wanted to see what is the reason to have such large coupling between pitch and yaw motions.
The first test was to check orthogonality of the bias sliders. It was done by monitoring the suspension motion using the green beam.
The Y arm cavity was aligned to the green. The damping of ITMY was all turned off except for SD.
Then ITMY was misaligned by the bias sliders. The ITMY face CCD view shows that the beam is reasonably orthogonally responding to the pitch and yaw sliders.
I also confirmed that the OPLEV signals also showed reasonablly orthogonal responce to the pitch and yaw misalignment.
=> My intuition was that the coils (including the gain balance) are OK for a first approximation.
Then, I started to excite the resonant modes. I agree that it is difficult to excite a pure picth motion with the resonance.
So I wanted to see how the mixing is frequency dependent.
The transfer functions between ITMY_ASCPIT/YAW_EXC to ITMY_OPLEV_PERROR/YERROR were measured.
The attached PDFs basically shows that the transfer functions are basically orthogonal (i.e. pitch exc goes to pitch, yaw exc goes to yaw) except at the resonant frequency.
I think the problem is that the two modes are almost degenerate but not completely. This elog shows that the resonant freq of the ITMY modes are particularly close compared to the other suspensions.
If they are completely degenerate, the motion just obeys our excitation. However, they are slightly split. Therefore, we suffer from the coupled modes of P and Y at the resonant freq.
However, the mirror motion obeys the exitation at the off resonance as these two modes are similar enough.
This means that the problem exists only at the resonant frequencies. If the damping servos have 1/f slope around the resonant freqs (that's the usual case), the antiresonance due to the mode coupling does not cause servo instability thank to the sufficient phase margin.
In conclusion, unfortunately we can't diagnalize the sensors and actuators using the natural modes because our assumption of the mode purity is not valid.
We can leave the pitch/yaw modes undiagnalized or just believe the oplevs as a relatively reliable reference of pitch and yaw and set the output matrix accordingly.
The figures will be rotated later.
Local earth quake 3.7 mag trips PRM
What about the MC?
We did the following today morning:
It is unfortunate we have to do this dance each time c1susaux has to be restarted, but I guess it is preferable to repeated unsticking of the optic, which presumably applies considerable shear force on the magnets...
After Wednesday's locking effort, Eric had set the IFO to the PRMI configuration, so that we could collect some training data for the PRC angular feedforward filters and see if the filter has changed since it was last updated. We should have plenty of usable data, so I have restored the arms now.
Been non-functional for 3 weeks. Anyone else notice this? images missing since ~Sep 21.
I've re-submitted the Condor job; pages should be back within the hour.
I did a quick survey of the drive electronics for the PZT OM mirrors today. The hope is that we can correct for the clipping observed in the AS beam by using OM4 (in the BS/PRM chamber) and OM5 (in the OMC chamber).
Here is a summary of my findings.
I hope these have the correct in-vacuum connections. We also have to hope that the clipping is downstream of OM4 for us to be able to do anything about it using the PZT mirrors.
Tonight, and during last week's locking, we noticed something intermittently kicking the PRM. I've determined that PRM's LR OSEM is problematic again. The signal is coming in and out, which kicks the OSEM damping loops. I've had the watchdog tripped for a little bit, and here's the last ten minutes of the free swinging OSEM signal:
Here's the hour trend of the PRM OSEMS over the last 7 days a plot of just LR since the fix on the 9th of September.
It looks like it started misbehaving again on the evening of the 5th, which was right when we were trying to lock... Did we somehow jostle the suspension hard enough to knock the foil cap back into a bad spot?
Still no luck relocking, but got a little further. I disabled the output of the problematic PRM OSEM, it seems to work ok. Looking at the sensing of the PRMI with the arms held off, REFL165 has better MICH SNR due to its larger seperation in demod angle. So, I tried the slightly odd arrangement of 33I for PRCL and 165Q for MICH. This can indefinitely hold through the buzzing resonance. However, I haven't been able to find the sweet spot for turning on the CARM_B (CM_SLOW) integrator, which is neccesary for turning up the AO and overall CARM gain. This is a familiar problem, usually solved by looking at the value far from resonance on either side, and taking the midpoint as the filter module offset, but this didn't work tonight. Tried different gains and signs to no avail.
Summary pages will be unavailable today due to LDAS server maintenance. This is unrelated to the issue that Rana reported.
It started here
100 Sapphire prizms arrived.
Perhaps the problem is electrical? The attached plot shows a downward trend for the LR sensor output over the past 20 days that is not visible in any of the other 4 sensor signals. The Al foil was shorting the electrical contacts for nearly 2 months, so perhaps some part of the driver circuit needs to be replaced? If so a Satellite Box swap should tell us more, I will switch the PRM and SRM satellite boxes. It could also be a dying LED on the OSEM itself I suppose. If we are accessing the chamber, we should come up with a more robust insulating cap solution for the OSEMs rather than this hacky Al foil + kapton arrangement.
The PRM and SRM Satellite boxes have been switched for the time being. I had to adjust some of the damping loop gains for both PRM and SRM and also the PRM input matrix to achieve stable damping as the PRM Satellite box has a Side sensor which reads out 0-10V as opposed to the 0-2V that is usually the case. Furthermore, the output of the LR sensor going into the input matrix has been turned off.
Looks like what were PRM problems are now seen in the SRM channels, while PRM itself seems well behaved. This supports the hypothesis that the satellite box is problematic, rather than any in-vacuum shenanigans.
Eric noted in this elog that when this problem was first noticed, switching Satellite boxes didn't seem to fix the problem. I think that the original problem was that the Al foil shorted the contacts on the back of the OSEM. Presumably, running the current driver with (close to) 0 load over 2 months damaged that part of the Satellite box circuitry, which lead to the subsequent observations of glitchy behaviour after the pumpdown. Which begs the question - what is the quick fix? Do we try swapping out the LM6321 in the LR LED current driver stage?
GV Edit Nov 2 2016: According to Rana, the load of the high speed current buffer LM6321 is 20 ohms (13 from the coil, and 7 from the wires between the Sat. Box and the coil). So, while the Al foil was shorting the coil, the buffer would still have seen at least 7 ohms of load resistance, not quite a short circuit. Moreover, the schematic suggests that that the kind of overvoltage protection scheme suggested in page 6 on the LM6321 datasheet has been employed. So it is becoming harder to believe that the problem lies with the output buffer. In any case, we have procured 20 of these discontinued ICs for debugging should we need them, and Steve is looking to buy some more. Ben Abbot will come by later in the afternoon to try and help us debug.
The two 40 mm apeture baffles at the ends were replaced by 50 mm one. ITM baffles with 50 mm apeture are baked ready for installation.
Green welding glass 7" x 9" shade #14 with 40 mm hole and mounting fixtures are ready to reduce scatter light on SOS
PEEK 450CA shims and U-shaped clips will keep these plates damped.
Johannes acquired a crate to contain the Acromag setup and wiring, and installed a rail along the bottom panel so that the ADC units will be oriented vertically with the ehternet ports facing up. We briefly talkes about what the layout should be, and are thinking of using 2 rails, one for ADCs and one for DACs. We want to design a generic front panel to accept 25 pin D-Sub inputs and maybe also BNCs, which we can use for all the Acromag crates.
I got the epics session for the acromag to run on c1iscex and was able to access the channel values using caget on donatella. However, I get the following warning:
cas warning: Using dynamically assigned TCP port 48154,
cas warning: but now two or more servers share the same UDP port.
cas warning: Depending on your IP kernel this server may not be
cas warning: reachable with UDP unicast (a host's IP in EPICS_CA_ADDR_LIST)
It seems like there might be a way to assign a port for each unit, if this is a problem.
Also, c1iscex doens't have tmux; what's the best way to run the modbusApp and then detach? Right now I just left an epics session running in an open terminal.
We engaged the HV driver to the output port PZTs, hoping to mitigate the AS port clipping. Basically, the range of the PZT is not enough to make the beam look clean. Also, our observation suggested there are possible multiple clipping in the chamber. We need another vent to make the things clearly right. Eric came in the lab and preparing the IFO for it.
1. Before the test, the test masses have been aligned with the dither servo.
2. We looked at the beam shape on the AS camera with a single bounce beam. We confirmed that the beam is hard-clipped at the upper and left sides of the beam on the video display. This clipping is not happening outside of the chamber.
3. We brought an HV power supply to the short OMC rack. There is a power supply cable with two spades. The red and black wires are +150V and GND respectively.
4. The voltage of +/-10V was applied on each of the four PZT drive inputs. We found that the motion of the beam on the camera is tiny and in any case, we could not improve the beam shape.
5. We wondered that if we are observing ANY improvement of the clipping. For this purpose, we aligned AS110 sensor every time we gave the misalignment with the PZTs. Basically, we are at the alignment to have the best power we can get. We thought this was weird.
6. Then we moved the AS port spot with the ITMX. We could clearly make the spot more round. However, this reduced the power at the AS port reduced by ~15%. When the beam was further clipped, the power went down again. Basically, the initial alignment gave us the max power we could get. As the max power was given with the clipped beam, we get confused and feel safer to check the situation with the chambers open.
During this investigation, we moved the AS port opitcs and the AS camera. So they are not too precise reference of the alignment. The PZT HV setup has been removed.
I have completed the following non-Steve portions of the pre-vent checklist [wiki-40m.ligo.caltech.edu]
All shutters are closed. Ready for Steve to check nuts and begin venting!
- checked all jam nuts
- checked all viewports are covered
- turned oplev servos off
- took pictures of medm screens: sus summs, aligned oplev centering, IFO& MC alignment biases and vac configuration
- checked particle counts
- checked crane operational safety
- closed V1, VM1, annuloses
- opened VV1 and vented with Airgas brand, Industrial Grade Nitrogen [ 99.99% ] to 25 Torr
- switched over to Airgas brand compressed air, Alphagas " AI UZ300 " with Total Hydro Carbon 0.1 PPM
XLR(F)-XLR(M) cable for the blue microphone is missing. Steve ordered one.
We found one in the fibox setup. As we don't use it during the vent, we use this cable for the microphone.
Once we get the new one, it will go to the fibox setup.
IFO is at atmosphere. The MC can be locked in air now.
The doors will be coming off tomorrow 8am sharp.
Do we want to install the ITM baffles?
What about the found OSEM filters?
I did the following today to prepare for taking the doors off tomorrow.
I am leaving all shutters closed overnight.
So I think we are ready to take the doors off at 8am tomorrow morning, unless anyone thinks there are any further checks to be done first.
Should we look to do anything else now? One thing that comes to mind is should we install ITM baffles? Or would this be more invasive than necessary for this vent?
Steve reported to me that he was unable to ssh into the control room machines from the laptops at the Xend and near the vacuum rack. The problem was with pianosa being frozen up. I did a manual reboot of pianosa and was able to ssh into it from both laptops just now.
We re-checked IMC locking, arm alignments (we were able to lock and dither align both arms today, and also made the michelson spot look reasonable on the camera) and made sure that the AS and REFL spots were in the camera ballpark. We then proceeded to remove the heavy doors off ITMY and BS/PRM chambers. We also quickly made sure that it is possible to remove the side door of the OMC chamber with the current crane configuration, but have left it on for now.
The hunt for clipping now begins.
I say just fix the clipping. Don't worry about the PRM OSEM filters. We can do that next time when we put in the ITM baffles. No need for them on this round.
In the afternoon, we took the heavy door off the OMC chamber as well, such that we could trace the AS beam all the way out to the AP table.
In summary, we determined the following today:
Attachment #5 is extracted from the 40m CAD drawing which was last updated in 2012. It shows the beam path for the output beam from the BS all the way to the table (you may need to zoom in to see some labels. The drawing may not be accurate for the OMC chamber but it does show all the relevant optics approximately in their current positions.
EQ will put up photos from the ITMY and BS/PRM chambers.
Plan for Monday: Reconfirm all the findings from today immediately after running the dither alignment so that we can be sure that the ITMs are well-aligned. Then start at OM1 and steer the beam out of the chambers, centering the beam as best as possible given other constraints on all the optics sequentially. All shutters are closed for the weekend, though I left the SOS iris in the chamber...
Here is the link to the Picasa album with a bunch of photos from the OMC chamber prior to us making any changes inside it - there are also some photos in there of the AS beam path inside the OMC chamber...
Oct. 15, 2016
Another attempt (following elog 8755) to extract the oven transfer function from time series data using Matlab’s system identification functionalities.
The same time series data from elog 8755 was used in Matlab’s system identification toolbox to try to find a transfer function model of the system.
From elog 8755: H(s) is known from current PID gains: H(s) = 250 + 60/s +25s, and from the approximation G(s)=K/(1+Ts), we can expect the transfer function of the system to have 3 poles and 2 zeros.
I tried fitting a continuous-time and a discrete time transfer function with 3 poles and 2 zeros, as well as using the "quick start" option. Trying to fit a discrete time transfer function model with 3 poles and 2 zeros gave the least inaccurate results, but it’s still really far off (13.4% fit to the data).
1. Obtain more time domain data with some modulation of the input signal (also gives a way to characterize nonlinearities like passive cooling). This can be done with some minor modifications to the existing code on the raspberry pi. This should hopefully lead to a better system ID.
2. Try iterative tuning approach (sample gains above and below current gains?) so that a tune can be obtained without having to characterize the exact behavior of the heater.
Oct. 16, 2016
-Found the raspberry pi but it didn’t have an SD card
-Modified code to run directly on a computer connected to the TC 200. Communication seems to be happening, but a UnicodeDecodeError is thrown saying that the received data can’t be decoded.
-Some troubleshooting: tried utf-8 and utf-16 but neither worked. The raw data coming in is just strings of K’s, [‘s, and ?’s
-Will investigate possible reasons (update to Mac OS or a difference in Python version?), but it might be easier to just find an SD card for the raspberry pi which is known to work. In the meantime, modify code to obtain more time series data with variable input signals.
Ashley Fowler "high shool" student received basic 40m safety training and Lydia is her guarding angle.
[ericq, lydia, gautam]
IMC realignment, Arm dither alignment
I connected to the serial port using screen (through Terminal) and using Arduino's serial monitor and basically received the same strings that were received through python, so it's not a python issue. Checked the other TC 200 module and was also receiving nonsense, but it was all question marks instead of mostly K's and ['s.
This rules out a few possible reasons for the weird data. Next steps are to set up and configure the Raspberry Pi (which has been interfaced before) and see if the problem continues.
Lydia and I investigated the extra green beam situation. Here are our findings.
I can't think of an easy fix for this - the layout on the OMC chamber is pretty crowded, and potential places to install a beam dump are close to the AS and IMC REFL beam paths (see Attachment #1). Perhaps Steve can suggest the best, least invasive way to do this. I will also try and nail down more accurately the origin of these spots tomorrow.
Light doors are back on for the night. I re-ran the dithers, and centered the oplevs for all the test-masses + BS. I am leaving the PSL shutter closed for the night
Everybody is happy, except ITMY_UL or satalite box.
Gautam shows perfect form in the OMC chamber.
Tilted viewports installed in horizontal position. Atm2
[ericq, lydia, steve, gautam]
AS beam on OM1
Link to IMG_2337.JPG
AS beam on OM2
AS beam on OM3
AS beam on OM4
I didn't manage to get a picture of the beam on OM5 because it is difficult to hold a card in front of it and simultaneously take a photo, but I did verify the centering...
It remains to update the CAD diagram to reflect the new AS beam path - there are also a number of optics/other in-vacuum pieces I noticed in the BS/PRM and OMC chambers which are not in the drawings, but I should have enough photos handy to fix this.
Here is the link to the Picasa album with a bunch of photos from the OMC, BS/PRM and ITMY chambers prior to putting the heavy doors back on...
SRM satellite box has been removed for diagnostics by Rana. I centered the SRM Oplev prior to removing this, and I also turned off the watchdog and set the OSEM bias voltages to 0 before pulling the box out (the PIT and YAW bias values in the save files were accurate). Other Oplevs were centered after dither-aligning the arms (see Attachment #8, ignore SRM). Green was aligned to the arms in order to maximize green transmission (GTRX ~0.45, GTRY ~0.5, but transmission isn't centered on cameras).
I don't think I have missed out on any further checks, so unless anyone thinks otherwise, I think we are ready for Steve to start the pumpdown tomorrow morning.
These old specs are not so bad. But we now want to get replacements for the TRX and TRY and PSL viewports that are R <0.1% at 532 and 1064 nm.
I don't know of any issues with keeping BK-7 as the substrate.
Unfortunately, it seems that the large power supply which is used for the heater is dead. Or maybe I don't remember how to use it?
The AC power cord was plugged in to a power strip which seems to work for IO chassis. We also tried swapping power strip ports.
We checked the front panel fuses. The power one was 3 Ohms and the 'bias' one was 55 Ohms. We also checked that the EPICS slider did, in fact, make voltage changes at the bias control input.
Non of the front panel lights come on, but I also don't remember if that is normal.
Have those lights been dead a long time? We also reconnected the heater cable at the reference cavity side.
Pumping again after 7 days at atmosphere.
BS, ITMY and OMC chambers were open only.
Checked: jam nuts, viewport covers and beam shutters.
Oplev servo turned off and medm screens shots taken.
New item in vacuum: green shade 14 glass beam block at IR-input [ from the PSL ] viewport to block green reflection-scatter.
Reminder: viewport is not AR coated for green!
IFO pressure 1.7E-4 Torr on new not logged cold cathode gauge. P1 <7E-4 Torr
Valve configuration: vac.normal with anunulossess closed off.
TP3 was turned off with a failing drypump. It will be replaced tomorrow.
The pressure on the newly installed gauge on the X arm was 6E-5 torr when I came in today evening, so I decided to start the recovery process.
Dry pump of TP3 replaced after 9.5 months of operation.[ 45 mTorr d3 ]
The annulosses are pumped.
Valve configuration: vac normal, IFO pressure 4.5E-5 Torr [1.6E-5 Torr d3 ] on new ITcc gauge, RGA is not installed yet.
Note how fast the pressure is dropping when the vent is short.
All time stamps are blank on the MEDM screens.
I worked on recovering ALS today. Alignments had drifted sufficiently that I had to to the alignment on the PSL table onto the green beat PDs for both arms. As things stand, both green (and IR) beats have been acquired, and the noise performance looks satisfactory (see Attachment #1), except that the X beat noise above 100Hz looks slightly high. I measured the OLTF of the X end green PDH loop (after having maximized the arm transmission, dither alignment etc, measurement done at error point with an excitation amplitude of 25mV), and adjusted the gain such that the UGF is ~10kHz (see Attachment #2).
Given that most of the post vent recovery tasks were done, and that the ALS noise performance looked good enough to try locking, we decided to try PRFPMI locking again last night. Here are the details:
PRM alignment, PRMI locking
Post the most recent vent, where we bypass the OMC altogether, we have a lot more light now at the AS port. It has not yet been quantified how much more, but from the changes that had to be made to the loop gain for a stable loop, we estimate we have 2-3 times more power at the AS port now.
Great to hear that we have the PRG of ~16 now!
Is this 150ppm an avg loss per mirror, or per arm?
The autolocker was acting up today, Gautam traced it to EPICS channels ( namely C1:IOO-MC_LOCK_ENABLE and C1:IOO-MC_AUTOLOCK_BEAT ) served by c1iool0 not being responsive and keyed the crate. This restored it nominal operation.
We may have a mouse in the lab. Do not leave any food scrap in trash ! Traps will be set.
I realized that I did not have a Finesse model to reflect the current situation of flipped folding mirrors (I've been looking at 'ideal' RC cavity lengths with folding mirrors oriented with HR side inside the cavity so we didn't have to worry about the substrate/AR surface losses), and it took me a while to put together a model for the current configuration. Of course this calculation does not need a Finesse model but I thought it would be useful nevertheless.
In summary - the model with which the attached plot was generated assumes the following:
This calculation agrees well with the analytic results Yutaro computed here - the slight difference is possibly due to assuming different losses in the RC folding mirrors.
The conclusion from this study seems to be that the arm loss is now in the 100-150ppm range (so each mirror has 50-75ppm loss). But these numbers are only so reliable, we need an independent loss measurement to verify. In fact, during last night's locking efforts, the arm transmission sometimes touched 400 (=> PRG ~22), which according to these plots suggest total arm losses of ~50ppm, which would mean each mirror has only 25ppm loss, which seems a bit hard to believe.
It is also difficult to have a high arm transmission without having high PRG.
What about to plot the arm trans and the REFL DC power in a timeseries?
Or even in a correlation plot (X: Arm Trans or PRG vs Y: REFL Reflectivity)
This tells you an approximate location of the critical coupling, and allows you to calibrate the PRG, hopefully.
As Gautam mentioned, we had some success locking the PRFPMI last night. (SRM satellite box is still in surgery...)
Unsurprisingly, changing the loss/PRG/CARM finesse means we had to fiddle with the common mode servo parameters a little bit to get things to work. However, before too long, we achieved a first lock on the order of a few minutes. Not long afterwards, we had a nice half hour lock stretch where we could tune up the AO crossover and loop UGFs. The working locking script was committed to SVN. Really, no fundamentally new tactics were used, which is encouraging. (One thing I wondered about was whether a narrower CARM linewidth would still let our direct ALS->REFL11 handoff with no offset reduction work. Turns out it does)
However, the step where we increase the analog CARM gain isn't as bulletproof as it once had been. The light levels "sputter" in and out sometimes if the gain increases are too agressive, and can cause a lockloss. Maybe this is an effect of the narrower linewidth and injecting more ALS noise at high frequencies with the higher CARM bandwidth.
The spatial profiles of the light on the cameras is totally bananas. Here's AS and REFL.
As Koji suggested, here is a 2D histogram of TRY vs REFLDC. It appears that the visibility would max out at 75% or so at arm powers around 400. Indeed, we briefly saw powers that high, but as can be seen on the plot, we were usually a little under 300. Exploring the transmon QPD offset space didn't seem to have much effect here.
One thing that I hadn't looked at in previous locks is coherence with our ground seismometers. It would be cool to have more seismic feedforward, and looking at the frequency domain multiple coherence, it looks like we can win a lot between 1 and 20 Hz. I expected more of a win at 1Hz, though.
Following Koji's suggestion, I decided to investigate the relation between my Finesse model and the measured data.
For easy reference, here is the loss plot again:
Sticking with the model, I used the freedom Finesse offers me to stick in photodiodes wherever I desire, to monitor the circulating power in the PRC directly, and also REFLDC. Note that REFLDC goes to 0 because I am using Finesse's amplitude detector at the carrier frequency for the 00 mode only.
Both the above plots essentially show the same information, except the X axis is different. So my model tells me that I should expect the point of critical coupling to be when the average arm loss is ~100ppm, corresponding to a PRG of ~17 as suggested by my model.
Eric has already put up a scatter plot, but I reproduce another from a fresh lock tonight. The data shown here corresponds to the IFO initially being in the 'buzzing' state where the arms are still under ALS control and we are turning up the REFL gain - then engaging the QPD ASC really takes us to high powers. The three regimes are visible in the data. I show here data sampled at 16 Hz, but the qualitative shape of the scatter does not change even with the full data. As an aside, today I saw the transmission hit ~425!
I have plotted the scatter between TRX and REFL DC, but if I were to plot the scatter between POP DC and REFL DC, the shape looks similar - specifically, there is an 'upturn' in the REFL DC values in an area similar to that seen in the above scatter plot. POP DC is a proxy for the PRG, and I confirmed that for the above dataset, there is a monotonic, linear relationship between TRX and POPDC, so I think it is legitimate to compare the plot on the RHS in the row directly above, to the plot from the Finesse model one row further up. In the data, REFL DC seems to hit a minimum around TRX=320. Assuming a PRM transmission of 5.5%, TRX of 320 corresponds to a PRG of 17.5, which is in the ballpark of the region the model tells us to expect it to be. Based on this, I conclude the following:
In other news, I wanted to try and do the sensing matrix measurements which we neglected to do yesterday. I turned on the notches in CARM, DARM, PRCL and MICH, and then tuned the LO amplitudes until I saw a peak in the error signal for that particular DOF with peak height a factor of >10 above the noise floor. The LO amplitudes I used are
There should be about 15 minutes of good data. More impressively, the lock tonight lasted 1 hour (see Attachment #6, unfortunately FB crashed in between). Last night we lost lock while trying to transition control to 1f signals and tonight, I believe a P.C. drive excursion of the kind we are used to seeing was responsible for the lockloss, so the PRFPMI seems pretty stable.
With regards to the step in the lock acquisition sequence where the REFL gain is turned up, I found in my (4) attempts tonight that I had most success when I adjusted the CARM A slider while turning up the REFL gain to offload the load on the CARM B servo. Of course, this may mean nothing...