KAGRA Logbook

IOO (IMC)

tomotada.akutsu - 1:29 Friday 14 May 2021 (16744)

Power budget at MCE

K. Tanaka, Akutsu

Together with the other works to be (hopefully) reported by Kenta, we measured light beam power in the MCE clean booth. The power meter was the Thorlabs' pd-type one (see here). The

Power at the MCE viewport window: 120 uW.
In front of the DC-QPD1 on the MCE_TRANS table: 19.5 uW
In front of the DC-QPD2: 29.3 uW
In fron of the DCPD: 30.0 uW

When the power at the MCE viewport window was measured, K1-IMC_CAV_TRANS_OUT was 0.421 (counts?? or kind of calibrated??).

By the way, according to MEDM, the input power to IMC would be around ~300mW as usual. Hmmm... theoretically, I guess the IMC_MCE_TRANSMIT is 0.03 of the input for some reason, so, 300mW*0.03= 9 mW. But the measured one is 120 uW. Did I miss something?

According to the wiki, the MCe transmission@1064nm is <200ppm, while the database says it is 4.7ppm. With (200ppm,4.7ppm) of MCe transmittance, the net transmittance P(MCE_TRANS)/Pin corresponds to maybe (0.03, 0.0007), respectively. With Pin=300mW, 0.0007 net transmittance means still 200uW, so something inconsistent with the measured value 120uW. Loss??

We observed a halo-like stray light surrounding the main transmit beam (See the attached figures) Needs further investigation. The halo seemed happen at the edge of the viewport window, but this would not be the direct source... hmm... Anyway, the power budgetting is still on the way. Need to check the energy conservation at least on the MCE table.

Images attached to this report

Comments to this report:

yuta.michimura - 22:17 Wednesday 19 May 2021 (16785)

I found that I did a similar calculation back in 2016. It says that P(MCE_TRANS)/Pin ~ 0.00027 at that time. Note that the measurement was done using one of the IMC TRANS DC PDs and not the power meter at the viewport, so losses between the viewport and the PD is included in my measurement. Considering this, I think your measurement is roughly consistent with my measurement in 2016, if the throughput from the viewport to the PD is ~70%.
https://klog.icrr.u-tokyo.ac.jp/osl/?r=618

tomotada.akutsu - 3:33 Thursday 20 May 2021 (16787)

Hmm... Although I need to prepare the better method, but according to my recent measurement, the power through put from the PSL room to STM2 is very good. For example in 16728, we measured the power in-between STM2 and IMMT1, and was 0.348-0.345 W, which is almost the same value measured in the PSL room (see also 16729). So there should not be severe power loss due too mode-mismatch... these measurements had been done on the same day but different time, and the PMC output power has been varying, so we may not be able to simply compare the measured values. This varying is one of the "non-urgent issues to be resolved in the end" in my mind, to be shared among you hopefully.

To resolve the inconvenience, as I have already mentioned in 16729, I would like to put an additional power meter behind the lower steering mirror of the last periscope in the PSL room. To do precise power budgetting, this PD would be useful to provide a reference number without shuting down the laser beam.

tomotada.akutsu - 3:55 Thursday 20 May 2021 (16788)

By the way, the Thorlabs PD-type power meter that I sometimes refer recently is PM121D, which cosists of a sensor head S121C and a driver PM100D. As shown in the link, the sensor head S121C is made of Si, sensitive to the light of 400-1100nm, up to 500mW.