J. Kissel scribe for S. Koehlenbeck, J. Freed, R. Short, and guest star J. Oberling
ECR E2400083
IIET 30642
WP 12453

Toward the end of day two of install (LHO:83961), the team set the power at the input of the ALS-FC2 by rotating ALS-HWP2, to be 50.5 [mW] (with PMC TRANS at 103.5 [W]). The reported power for the on-table ThorLabs SM1PD1A, however, measured consistently lower at "31 [mW]," lower than the goal for this PD, a reported power of "45 [mw]" (LHO:83927), when the power of the beam between ALS-M9 and ALS-FC2 was physically measured at 48.7 [mW] (with PMC TRANS at 103.0). 

We think lower reported power on the PD is related to 
(1) The slight translation of the what-used-to-be ALS-L5 to ALS-M9 beam, because of the thickness of the SPI-BS1 that was inserted into that path, and 
(2) a change in the amount of s- and p- polarization in reflection of the ALS-PBS01 after rotating ALS-HWP2 to increase the power in the ALS/SQZ/SPI path, because the polarization state of the reflection of a PBS can be power dependent.

To support (2), we measured the polarization today between ALS-M9 and ALS-FC2 with a temporary PBS and power meter. On 2025-04-17 at 19:02 UTC with PMC power at 103.7 [W], we measure
    s-pol = 49.7 [mW]
    p-pol = 1.8 [mW]
    (total = 49.7 + 1.8 = 51.5 [mW])
Regrettably, we did not measure this polarization state prior to changing anything.

However, because we've *measured* equal total power at ALS-FC2 before vs. after, we're confident that at least the total power going into ALS-FC2 is the same. 

As such, we called in Jason to help couple the beam into ALS-FC2 using ALS-M9 and what alignment screws are on ALS-FC2 itself. Attached is the trend of that process, where we used the SPI distribution chassis PD in two forms: 
    H1:ALS-C_FIBR_INTERNAL_DC_POWER         never calibrated version, used to be 0.1 [cts]
    H1:ALS-C_FIBR_INTERNAL_DC_POWERMON      version we'd been using, calibrated into [mW]

In the end, we landed with the values of these channels at 

    H1:ALS-C_FIBR_INTERNAL_DC_POWER         0.1 [cts]
    H1:ALS-C_FIBR_INTERNAL_DC_POWERMON      31.4 [mW]
which matches our goals from LHO:83927.

The team also adjusted the position of the SM1PD1A in transmission of ALS-M9 to account for (1). However, even after moving the PD out of maximum and back, they were not able to find any position where the PD readout exceeded
    H1:ALS-C_FIBR_EXTERNAL_DC_POWERMON      31.0 [mW]
This why we expect that something more like (2) is going on. Though we looked through 
    - E1300483 (doesn't list ALS-M9), 
    - E1900246 (calls the optic ALS-M6, which is not for 1064 nm in E1300483), 
    - E0900325 (isn't new enough to have this mirror), 
    - Peter King's purchase orders circa 2019 (Req 122638553 for the fiber collimator parts, and Req 122583017 for the SM1PD1A parts)
we couldn't find anything that was clearly and obviously the ALS-M9 optic, so we can't validate or reconstruction the reflectivity in each polarization to model the factor of 31 [mW] / 45 [mW] = 0.7x (or 30% drop) reported value change.
In any event, this EXTERNAL PD channel will need to be recalibrated to better reflect the real power in this path.

Also, when we re-start up anything that uses the output of the ALS / SQZ fiber distribution, it may need to be adjusted to account for the polarization change described in (2)