Sheila, Camilla
We unlocked OPOS platform and checked the location of the OFI_A PD (B:PD1) reflected beam. Couldn't take a clear photo but the beam is dumped about 3/4" from the back of the -y beam dump plate.
We removed the retro-reflecting mirror and quarter waveplate.
SQZ Throughput Measurements (10:15am - 10:35am)
Looks good but need to check CLF PD calibrations. Injecting 50mW seed light, PDs measured 23mW CLF refl and 1.5mW rej, this is not what we measured with Thorlabs power meter.
We seems to be getting a little less power on SQZT7 than through the IFO path. Maybe due to beam diverter reflectivity, plus view-port and periscope.
|
Position
|
Power (mW)
|
|
Out of CLF fiber
|
37
|
| CLF rejected |
8 |
| IR Out of OPO |
0.85 |
| Out of Gate Value (towards HAM5) |
0.83 |
| On SQZT7 |
0.81 |
| IR Out of OPO (re-check) |
0.84 |
FCGS Throughput Measurements (1st from 11:36am, 2nd from 11:03am)
More losses than expected here.
Putting 6.5mW into SQZT0 FCGS fiber. Getting 1.8mW into HAM7 (0.84 + 0.95) ~ 27% throughput as prev.
Expect SQZT7 SK to be 0.42mW but get 0.24mW (0.84mW x 50/50 s-pol SK:BS2). Maybe polarization is incorrect for BS. Is VP transmissive for 532nm and periscope mirrors coated for 532nm?
Using value after H:M4, expect SQZT7 FCGS to 0.33mW but get 0.2mW. This is a straight shot from after H:M4.
Expect more power before ZM1, we are only getting 87% though the VIP (0.83mW/0.95mw). Previously (
alog59213) adjusted A:M3 AIO to increase green
transmission reflection, we should re-check this or the polarization. We couldn't see a beam in the FC rejected polarization path but didn't check the PD.
|
Position
|
Power (mW) 1st
|
Power (mW) 2nd |
|
Transmitted though H:BS1 (to SK path)
|
0.84 |
0.82 |
| Reflected off H:BS1 (to FC) |
0.95 |
0.91 |
| After H:L2 |
0.94 |
|
| Before ZM1 |
0.83 |
0.8 |
| After H:M4 (refl off FC1) |
0.33 |
0.32 |
| SQZT7 SK (FCGS only, OPO green off) |
0.24 |
|
| SQZT7 FCGS* (*misaligned) |
0.20* |
|
*Since the H-path co-alignment adjustments (alog 65032), SQZT7 FC path is now misaligned and clipping the lower periscope mirror.
Room lights were < 50uW and ignored. D2000021 cartoon layout
Sheila and Daniel found that the loss on the SQZT7 measurements are due to the upper periscope mirror (labeled 2.0 - 0086): FCGS 0.32mW before and 0.21mW after UPM, SK (with only FC green) FCGS 0.40mW before and 0.24mW after UPM = 35- 40% loss. There was no measurable loss on the Lower periscope mirror (labeled 2.0 - 0080). We will swap the upper mirror.
Sheila found the CLF_FIBR_REJECTED PD (F:PD1) calibration was set to responsively of 0.3 when we measured 0.08 in alog 60344. Maybe we never changed it in previously, but it's now changed and accepted in medm. This still doesn't account for the 1.5mW on PD to 8mW on Thorlabs PM above, that would still be a factor of 2 off.
I found the 12% each way FCGS loss is mainly due to A:M3, but all beams look dumped. I will post detailed photos tomorrow.
Still to look at tomorrow: Recheck CLF PD, Work with Rahul on OPOS suspension as he found a unusal peak.
We are loosing ~12% of power in each direction on the FCGS path, total loss is ~ 0.2mW. I've followed all beams from this and think they are dumped well enough.
See attached photo. The majority of the light is lost from transmission of A:M3. This surprising as we reduced this in alog alog 59213 but moving A:M3 to reduce AIO from 55 to 45deg. It's possible the changes of H:M1 and H:M2 (59213, 60014) to co-align IR and green increased the AIO again. Polarization shouldn't be causing this as the FCGS light at A:M3 is all reflected off all H:POL1.
The other beams are dumped on the two aperture stops (SFI1 and A:L2) and the H:POL1 transmission is sprayed though the hole in the mirror mount but the power is very low and a beam dump can't easily be added.
I've attached annotated photos showing the green ghost beams we have in HAM7 shown on this photo. This is ~12% of power in each direction on the FCGS path so 22% of this path. Total loss is ~ 0.2mW when injecting 6.5mW into SQZT0 FCGS fiber. We are not planning to do anything about these, most light is caught by black DLC surfaces. Ghost beams caused by:
Photo of both A:M3 transmitted beams and the A:DC2 reflected beam.