At Daniel's request I installed a neutral density filter (ND05A) on the photodiode located after the reference cavity in the ALS path. The power transmitted by the reference cavity was measured between the first turning mirror located after the step down periscope and a lens (see RC_Trans.jpg) to be 28.3 mW. At the time the FSS MEDM screen indicated a reference cavity transmission of 2.8 V. The 1 mW = 100 mV factor is consistent with a measurement taken a long time ago.
After Peter was done, I updated the calibration of H1:ALS-C_FIBR_EXTERNAL_DC_POWERMON.
When I was watching H1:PSL-FSS_TPD_DC_OUTPUT it was about 2.68V, so using Peter's measurement the power upstream of the BS for the PD should have been ~2.68/2.8*28.3=27.1mW.
I changed H1:ALS-C_FIBR_EXTERNAL_DC_SPLITTERR from 50% to 3% so the "power before pickoff" (H1:ALS-C_FIBR_EXTERNAL_DC_POWERMON) reads ~27mW.
(The BS should be 5% according to Peter's picture, ND05 means about a factor of 3 reduction, so it should be 5/3% ~ 1.7% instead of 3, something should be off by a factor of 2 or so but I wont investigate further because this is of no real consequence.)
ALS Y laser was turned on (EY is still laser safe, ISCTEY is locked). PLL was locked.
When I looked at the screen I thought the responsivity was off. This should be a Si diode.
I temporarily set responsivity to 0.36A/W assuming Thorlabs FDS1010 (but I don't know if it is), and "reflectivity of pick off beamsplitter" to 3%*0.64/.36=5.3% to keep the calibration correct. That's roughly a factor of 3 larger than expected from BS label and ND05.