J. Oberling, R. Short

Today was spent recovering the PSL stabilization systems after we got the amplifiers up and running yesterday.

PMC

We started with the PMC.  We did a quick alignment check at lower power (turned down with the High Power Attenuator (HPA) after Amp2) and all looked good here.  The power incident on the PMC was slowly increased to max, which we measured at 130.1W with our roving power meter.  The PMC ramp was activited and we say TEM00 flashes, so we hit the lock button and it locked without issue.  It only had ~100W in transmission so the alignment was tweaked with the picos and we also tweaked the pump diode currents for the amplifiers.  In the end we had 108.7W in transmission and 21.5W in reflection, and were running Amp1 at 9.0A and 8.8A and Amp2 at 9.1A (both power supplies).  The alignment onto the PMC locking PD was tweaked and we took a visibility measurement (correctly accounting for the 0.015V of dark voltage on this PD):

• Unlocked: -0.556 V
• Locked: -0.067 V
• Visibility: 87.9%

The PMC throughput was calculated at (Pout/Pin) 83.6%.  We also calibrated the PMC Trans and Refl PDs.  From the shape of the reflected spot we know there's more mode matching work to do with the PMC, but we have no shortage of power available to the IFO as is so we moved on to recovering the ISS.

ISS

We started by measuring the amount of power in the diffracted beam with the ISS OFF, so it should be ~3% of our maximum power.  We expected ~4.1W but measured 4.8W, so not quite right on the diffracted power % calibration.  The ISS AOM alignment was tweaked to max the power in the diffracted beam; this was 6.4W.  So we set about collecting the data needed to recalibrate the diffracted power % (change the Offset slider from 0 to 25 in steps of 1, recording the AOM voltage and power in the diffracted beam along the way).  The power into the ISS AOM was measured at ~135.5W.  To calibrate the diffracted power % we use the ISS AOM power in and the measured diffracted power to calculate the percentage of the PSL beam diffracted by the AOM, plot the AOM voltage vs this percentage, and fit a 2nd order polynomial to the resulting plot.  The plot is attached and the new polynomial is: 391.4x² - 199.05x + 25.492.  We then set the Offset slider so that we had a "bank" of ~3% of our total Amp2 output for the ISS to use; the new Offset is 3.3, which gives us ~4.1W in the diffracted beam by default.  We looked at our overnight trends of Amp2 output and saw a total drift of ~1.5% of the laser power, so we set the ISS bank to be double that; if we find we need more power in the band the Offset can be changed.  We then adjusted the voltage on the ISS PDs to be near to 10V; PDA was reading 9.95V and PDB was reading 9.97V (PDB is currently our in-loop sensor and PDA is our out-of-loop sensor).  To finish, we tweaked the ISS QPD alignment.  Done with the ISS, we moved on to the FSS.

FSS

We started by trying to lock the FSS RefCav, and it locked on its own with zero issues.  The TPD was reading 0.88V so we did nothing to the FSS.  Should we need to tweak things up at a later date we will, but for now all appears healthy.

At this point we cleaned up the work space in the PSL enclosure and left, putting the enclosure into Science Mode once we had left.  On the outside we measured TFs for the PMC, ISS, and FSS, as well as took a look at the FSS crossover.

Transfer Functions

We started with the PMC.  We measured the UGF at ~1.6kHz with a phase margin of 59.9 degrees, see 2nd attachment.

Next was the ISS.  We measured the UGF at ~66.8kHz with a phase margin of 23.4 degrees, see 3rd attachment.  This was much higher than we had measured in right before the NPRO swap, so we dropped the inner loop ISS gain by 4dB (from 11dB to 7dB).  This put the UGF at ~41.9kHz with a phase margin of 41 degrees, much closer to where we were at before the NPRO swap (see 4th attachment).

Finally, the FSS.  We found the UGF around 420kHz, so we raised the Common Gain to 16dB (a change of 1dB) to put the UGF at ~519kHz with a phase margin of 65 degrees (see 5th attachment).  We then looked at the NPRO PZT/PSL EOM crossover (take a spectrum of the IN1 port on the TTFSS box).  We ended up lowering the Fast Gain from 7dB to 5dB to even out the crossover hump at ~20kHz, see final attachment.  We did notice a new peak at ~32.7kHz that could be the PZT resonance of the new NPRO.  We left this as-is for now, but if this becomes a problem we can tune the notch for the NPRO PZT to hopefully squash this peak; if that doesn't work we would have to pull the TTFSS and install a new notch.  This finished LVEA work so we went to the Control Room.

Control Room

Back in the Control Room Ryan did a remote alignment of the PMC and FSS RefCav.  Not much improvement on the RefCav, still have a TPD of ~0.88V with the IMC unlocked, but there was a good improvement in PMC Trans; the PMC is transmitting ~109.2W now.  We also updated the diffracted power % calibration and set the ISS AOM to diffract ~3% (tagging OpsInfo, this should be around 3% now).  We noticed the ISS AOM appears to be working harder with this NPRO, as the diffracted power % is moving more than with the previous NPRO.  We'll keep an eye on this over the coming days.

DIAG_MAIN was also complaining about the NPRO noise eater being out of range.  Looking at the channel it appeared the noise eater was off, even though we had it set in the PSL Beckhoff software to ON.  Toggling the software switch did not change anything, so I went out to the LVEA and physically switched the noise eater ON (it should be OFF at the power supply for the software switch to work).  This fixed the problem, so it appears there's something still not quite right with the remote diagnostics board in this NPRO, specifically with the remote noise eater switch.  So for future info, if we have to toggle the NPRO noise eater it has to be done at the NPRO power supply in the LVEA, not in the PSL Beckhoff software.

We went through and accepted all of the SDF diffs from the NPRO swap, and Ryan did a rotation stage calibration (he'll post these results as a comment).  At this point the NPRO swap was complete and the PSL is recovered and delivering light to the IFO again.  At the time of writing the IFO is locked at Nominal Low Noise and Observing once again.  This closes WP 12155.