aLIGO LHO Logbook

H1 ISC

elenna.capote@LIGO.ORG - posted 13:55, Friday 08 April 2022 - last comment - 12:46, Monday 11 April 2022(62526)

What power can we put on the OMC DCPDs?

[Sheila, Craig, Koji]

We performed OMC scans at 50 W (alog 62514) and we would like to know how much higher input power OMC scans we can do without burning the DCPDs. We are not able to lower the modulation depth of the RF45 band (alog 62507) past a certain point without losing lock.

At 50 W input we put 30 mA of current on the DCPDs. At 60 W we would put 40 mA.

We contacted Koji to ask what the maximum power the DCPDs can tolerate. This is the PD we are using in the OMC. Although the DCC says that we are using Excelitas 30655, we are actually using lasercomponents IGHQEX3000. Koji has tested both PDs up to 332 mW of incident power (IGHQEX3000 high power test).

At 100 W we would likely be putting 60 mA which may be pushing it, but it not entirely impossible.

However, we may saturate the amps. We do not have information about the response of the DCPD amps with continuous saturation.

Craig has documented all of this at: https://git.ligo.org/aligo_commissioning/labutils/-/tree/master/omc_scan

Comments related to this report

jenne.driggers@LIGO.ORG - 11:46, Monday 11 April 2022 (62558)

Link

I think it's clear, but just to remind myself, these are all per-PD currents (so, at 50W for example, it's 30 mA per photodiode, for a total of 60mA).

jeffrey.kissel@LIGO.ORG - 12:46, Monday 11 April 2022 (62563)

Link

If it's helpful in this study, I've recently put together a lot of material in efforts to understand every detail of the new O4 GW DPCD transimpedance amplifier (TIA) and whitening chassis, see G2200551.

Perhaps most relevant to you and the amount of power you can put on the DCPDs: each PDs chain has a newly, very frequency-dependent transimpedance amplifier system. In short, the maximum transimpedance is ~140 kOhms around the complex pair of poles ~20 Hz, leveling off to 100 kOhms around 50 Hz all the way out to ~10 kHz. 
    (And I should note, what I quote here as "the" transimpedance is the transfer function from the single-ended photodiode current to the differential voltage coming out of the TIA, out of chamber entering into the whitening chassis. The actual singular resistive component that serves as "the transimpedance amplifier" on the gravitational wave readout path is R5 = 10 kOhm, but there's a factor of 5 * 2 = 10 in the subsequent "internal whitening" of that voltage out of the op-amp component U3, and again, the influence of the big honking inductor L4 causes some amplification of that 100 kHz at 20 Hz.)

The first attached plot conveys this: simpleanalyticmodel_OMC_DCPD_TIA_Final.png

As for the whitening chassis, whose gain is also frequency dependent, you get a maximum gain of either 25 or 50 of that differential output voltage to the ADC.
The second attached plot conveys this: simpleanalyticmodel_OMC_DCPD_Whitening_Final.png

The simple analytic model that produced the plots of both of these systems can be found here:
    https://git.ligo.org/Calibration/ifo/-/blob/main/scripts/electronics/omctransimpedanceamplifier/models/simpleanalyticmodel_OMC_TIA_.m

    https://git.ligo.org/Calibration/ifo/-/blob/main/scripts/electronics/omcwhitening/models/simpleanalyticmodel_OMC_Whitening_.m

Images attached to this comment