J. Kissel

As I'm building up the MEDM screens for SPI, I'm finding some bugs in the Simulink model. This one: 
According to D2400111-v6, and the ADC/DAC Channel Usage Table in G2401479-v3, ADC3 CH20 = PWRIN_REF and ADC3 CH21 = PWRIN_MEAS. 

The initial release of the SPI model at LHO (LHO:89777, ${USERAPPS}/spi/h1/models/h1spih23.mdl userapps svn rev r34928, ${USERAPPS}/spi/common/models/SPI_LIBRARY.mdl rev r30005) had these flipped with ADC3 CH20 = PWRIN_MEAS and ADC3 CH21 = PWRIN_REF.
 
I've fixed the bug, and the top level and library are now at rev r35060 and r35059 respectively, and Dave's installing / restarting for me. No DAQ restart required, as it's just simulink wires getting uncrossed.

Closes FAMIS 39955

Last checked in alog 89599.

All trends look consistent and/or similar with last month's with the exception of the EX pump, which Jim commented on last time. This pump was seen going up in pressure. This pressure has since remained stable.

Closes FAMIS 39863. Last checked in alog 89728.

EY_FAN2_470_1 (green) and both EX fans are going through some quiet lapses that were only seen once in the last alog. Nothing is near the "noisy" threshold though.

Plots attached.

Thu Apr 16 10:10:09 2026 INFO: Fill completed in 10min 6secs

TITLE: 04/16 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
OUTGOING OPERATOR: None
CURRENT ENVIRONMENT:
    SEI_ENV state: MAINTENANCE
    Wind: 20mph Gusts, 13mph 3min avg
    Primary useism: 1.01 μm/s
    Secondary useism: 0.27 μm/s 
QUICK SUMMARY: More work in BSC2 planned for today to prepare for dome and cartridge removal. A M6.0 earthquake from New Zealand is passing through.

This afternoon Mitch and I went to the chamber and after putting a little extra pressure on the locking screws, we started taking tension off of the springs. All springs started around 2k lbs each, we left them with about 400lbs each. Before transferring load to the stops, I used the screws to push HEPI about 100 um above the reference position. RX,RY and RZ were all within 25urad, X&Y were about 80um away from the reference. Right now Z has settled about 100 um below the reference position, all of the other positions are nearly unchanged. First thing in the morning we will drop all of the springs to ~150lbs each and re-assess the position. If things are still good we will attach dial indicators and get ready to remove the actuators.

First attached image are the after (left side readouts) and before (right) IPS readings. Second image are the cartesian IPS locations in nm or nrad, left column is after spring relief, right is before.

TITLE: 04/15 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: None
SHIFT SUMMARY:

IFO is in IDLE and MAINTENANCE

Did not end up taking the dome off but went in for more measurements/FARO of the beamsplitter. Most of the setup from dome removal is done and dome removal is slated for tomorrow.

LOG:

This afternoon, Ryan C and I hopped into BSC2 (via BSC1) and took a few more measurements of the optic position relative to the cage for Jason and the IAS mapping.  Jason will periodically post some data which consolidates these more nicely visually.  In the end, we found the BS optic rotated clockwise in it's cage a small amount (likely how we installed it), namely 1mm out of plane from one side to the other.

I retook the ETM OPLEV charge measurements this afternoon after the changes we made last week (HV switched off for ETMY digitally, and the sign of the bias offset flipped for ETMX from alog89809). I'm not sure why the error is larger on EX than EY, the wind was medium high but that would effect EY more than EX, the ends were seeing similar ground motion in the 0.03 - 0.1 band. I had no issues running the script, I switched it from scipy to numpy last week.

The sign change appears to be successful in reducing the charge on ETMX, every DOF/quadrant saw its' charge reduced from last week. We will be trying to measure this weekly now to confirm and possibly fine tune the settings.

ETMYs' charge looks to be  approximately the same as the last measurement, although 2/3 of the DOF/qs saw a small trend away from zero of <5 [V] at most compared to last week. The trend of the charge on EY from Nov (previous measurement) to last week was toward zero on all DOF/qs but there was a much larger time difference between those two sets of measurements (~five months vs ~a week).

Following up on this report ALOG89822 and this work permit WP13164 we replaced two Kepco Power Supplies at EX.  These supplies source power to ISC and TCS racks in the EX CER room.

The +/- supplies were still running but the fan on the -18V supply had siezed up, and the supply was overheating.  The +18V supply seemed ok, but when I turned them both off it would not restart.

The +18V supply is now S1201904 has improved fan installed.
The -18V supply is now (no serial number) has improved fan installed.

Both supplies were set back to their original voltage configurations and power sequenced back on, this was completed yesterday afternoon.  The pulled supplies will be upgraded with new fans and placed back into spares inventory.

FAMIS 31133

On Friday, h1psl0 was rebooted as part of updates to its timing card firmware and RCG (see alog89859), which unlocked the PMC briefly and left the ISS off (on purpose, the autolocker is turned off in SDF so that it can be turned on later once the PMC thermalizes). It looks like the PMC (as usual) relocked at a slightly different temperature; it doesn't appear to have much effect this time on the transmitted power, but the reflected power is very slightly lower. I've turned the ISS autolocker back on.

Nyath, Jonathan:

Nyath has started the upgrade of the subversion repository server. This service will be down for a few hours.

Wed Apr 15 10:09:01 2026 INFO: Fill completed in 8min 58secs

Attahed and in the table below are oplev positions for ITMY, ITMX and BS since last DRMI lock on 03/19/2026 at 11:02 PDT

Note 1: The BS Oplev Damping loops were still on during some FARO work and were shaking the BS. We turned these off on morning of April 14.

Note 2: The right-most T-cursor marks the corner-station vent

(Travis, Randy, Gerardo)

On Monday, the -Y door went on the HAM1 chamber with some minor issues not noted on previous installations, for some reason we had two minor misalignments: one, the door did not hung parallel to the HAM's flange,small tilt towards the outside on the upper part of the door, maybe the pick point slid forward.  Two, the door required a clocking move, we pinned two holes spanning 5:12 O'clock, then rotated the door clockwise a bit until all of the holes aligned, then a small movement on the -Z to complete the alignment about 1/8".  After all of that the door looks pretty concentric to the chamber flange.

Something to note, a dirty glove was found inside the chamber, yes inside the chamber under the support tube on the -Y side of the chamber, as I write, I nod, because it does not take much time to get this dirty thing out of the chamber.

M. Todd, S. Dwyer, J. Driggers

Summary

We turned on inverse ring heater filters to speed up the heating for those (using nominal values for the settings). Because of the weekend mayhem with the earthquakes we did not get a SUPER long HWS transient measuring the full response, but we could get a pretty good estimate of the ring heater effect on the substrate thermal lens without any other heating in the measurement. This is good to compare to modeled values that we have.

I also turned on SR3 heater on Sunday to get estimates of the coupling of SR3 heating to the defocus of SR3. To do this, Jenne helped me untrip a lot of the SU watchdogs for the relevant optics to the HWS. About 3 hours after the SR3 was turned on the watchdogs must have tripped again and misaligned the optics. But fortunately we got the cooldown data for this as well and it's all pretty consistent. These measurement suggest a 4.7 uD/W coupling for SR3 heating, which is very similar to modeled coupling from Gouy phase measurements at different SR3 heater powers.

Overall, while these measurements provide more pieces to the puzzle, they make previous analyses a bit more confusing, requiring some more thought (as usual).

Jennie W, Camilla C

A while ago we heated up then cooled down the SR3 heater (alog #84749).

As part of measurements using this data I calculated the curvature change, following the approach at LLO by Aidan given iin alog #27262. Matlab code is below.

%calculate SR3 spherical lens

Pin = 2;%W

double_pass = 2;

SR3_t = (3*3600) + (11*60); % Time for cooldown in s.

delta_ITMY = -2.67e-5;% decrease in defocus of ITMY according to Hartmann sensor.

D_ITMY = delta_ITMY./double_pass;% defocus change in Dioptres

D_ITMY_error = 5e-6;% error on defocus in Dioptres.

R_SR3 = 36.013;% cold radius of curvature in m

delta_R = (2./((2/R_SR3)+D_ITMY))-R_SR3; % change in curvature during cooldown in m/

delta_delta_R = D_ITMY_error.*(2./((2./R_SR3)+D_ITMY)); % error on curvature change.

This means the rate of defocus change is 6.6750uD per Watt.

The final curvature change is + 0.0087 m +/- 0.0002 m as the mirror becomes less curved due to cooldown.