BS TMP trouble

(Log on 21st)

As reported in the Elog, the BS TMP was failed on 18th. The pump was restarted, but it has problems.

1. Started the RP in the back side of TMP.
2. Opened the angle valve of RP.
3. Reset the TMP and started it.
4. The pressure in BS chamber was 1.8x10^-6 Torr.
5. After the rotation speed of TMP reached 600 Hz, opened the gate valve for BS chamber.
6. The pressure was increased to 2.0x10^-6 Torr. This means that the outgassing rate of TMP itself is too large.
7. Closed the gate valve and the angle valve.
8. Stoped the TMP and the RP.
9. Checked the RP. The inline filter between the TMP and the RP was filled by oile. The vapor of oile might contaminate the TMP.
10. Replaced the RP system (RP itself, inline filter, bellows, angle valve) to the other from the EW arm.
11. To declease the outgassing rate of TMP, a baking is necessary. Binded the TMP with the ribbon heater and attached the thermometer.
12. Restarted the TMP and the RP closing the gate valve.
13. The temparature must be lower than 120 ^oC. The heater was turned on/off manually around 110^oC for 30 min.
14. According to the manual of TMP, 4 hours are necessary for the baking. Will continue the baking on 23rd.

IR reflection measured on 20191015 after closing chamber

Injection: 234 uW

Reflection: 200 uW (off resonance)

Reflection: 135-139 uW (on resonance)

Off resonance reflectivity is 85.5%. 

SHG, OPO parameters after typhoon

SHG temperature: 3.115 kOhm

green power (mW)	OPO temperature (kOhm)	p pol PLL (MHz)	BAB maximum (V)
0		305	0.088
20	7.17		0.416
25	7.18		0.576
30	7.18		0.792
35	7.18		1.06
40	7.19	160	1.5
45	7.19		2.06
50	7.2		2.82
55	7.209		3.52
60	7.209	160	4.36

Correction of PD's position for FC_tra IR / IR transmission issue

Yao-Chin and Yuhang

We found filter cavity transmission IR PD is placed not on the waist of the detection beam. To avoid the clipping noise and also increase the detection ability, we corrected it. Look at the attached picture 1 and 2. Now the PD is closer to BS and the beam is well within PD.

Also, we measured the filter cavity transmitted power.  Compare this transmitted 0.5uW with incident 400uW. The measured transmission is about 0.6%. While the calculated value from the LMA specification is 1%. Since the measured IR after Dichroic and laser line filter goes thorough a lot of optics, this should be consistent.

Sudden failure of BS turbo pump

P-polarization Maps of OSTM (SK)

[Simon, Pengbo]

We have started to analyze the coated OSTM.

First, we took out the OSTM and inspected the mirror visually. We found pencil marks on the barrel and among them an arrow that indicates the thicker side of the wedged substrate and shows toward the HR side of the mirror.

The orientation which we choose for the measurement is to put the thicker side upside first and then rotate it by 90 degrees(with the thinner side facing the PC). Thus we only need to consider the horizontal deflection of the beam. After the adjustment of the sample, we found no output beam from the other side of the sample which mainly due to the coating absorption. So we removed one of the two ND filters to increase the laser power.

Second, we started the adjustment of the IU.

Due to the deflection effect, we first changed the lens position to make sure the beam goes through its center. Secondly, we rotated the second PBS slightly to have a perpendicular incident of the beam. Then, we adjusted the position of the two photodiodes so that the s-polarized beam and the p-polarized beam can hit the center area, respectively. Because we were doing the P-polarization measurement, the s polarized beam is nearly invisible, so we used the oscilloscope to finish the adjustment.

Then we ran the polarization map scan in the center with a 15mm radius. Attached to this report is the P-polarization map.

Not responding coil circuit seems open

[Yuhang, Yaochin, Eleonora]

We continued investiagation about not responding coil of the input mirror. (See entry #1742)

We measured the resistance across each coil loop at the flange. For all the working coils it was about 2.4 Ohm for the not working one we found the loop is open.

This suggests the magnet coud be fine and the problem could come from the coil circuit. 

Measurement of current loss budget

Aritomi, Eleonora, Yao-Chin, and Yuhang

We summarized the loss budget in the first attached figure.

From the power measurement as following, we could know this part contributes about 10% of loss

measured after first PBS after OPO	265uW
measured after last mirror before PR chamber	198uW

From the power measurement as following, we could know this part contributes about 1% of loss

measurement of IR reflection after PR chamber	221uW
measured before homodyne	218uW

We also measured the IR reflection(IR_ref) beam shape as in the attached figure 2. It has a bit ellipse in the horizontal direction. Because of this elongation, we had a second order peak when we match this IR_ref to alignment mode cleaner(as shown in the attached figure 3,4). By comparing this higher-order peak(12mV) and TEM00 peak(276mV), we could deduce the homodyne readout efficiency should be (1-12/276)^2 = 91.5%. But this IR_ref has beam jittering, so the real readout efficiency should be less than 91.5%.

We also measured the filter cavity round trip loss again. By using the same method we used last year, we found now this value is around 100ppm. We will do more investigation to make sure of this. The measurement is attached in the figure 5. 

In the last figure, the achievable squeezing without considering phase noise is attached.

Saturation of CC2 correction signal

[Aritomi, Yuhang, Yaochin, Eleonora]

We have saturation of CC2 correction signal (8Vpp) and it seems coming from servo since we have this saturation without connecting piezo. HV amplifier we are using can take up to 10V and the gain is 30 and piezo for phase shifter can take up to 1000V.  So we are using 240V out of 1000V piezo range.

Frequency dependent squeezing at 50kHz

[Aritomi, Yuhang, Yaochin, Eleonora]

Today we measured frequency dependent squeezing at 50kHz.

Setting is same as yesterday (AOM frequency is 109.13698MHz). Since CC2 lock is unstable, we made average number 5 while it was 50 yesterday.

Frequency dependent anti squeezing at 50kHz

[Aritomi, Yuhang, Yaochin, Eleonora]

Today we measured frequency dependent anti squeezing at 50kHz.

We set AOM frequency 109.13684 MHz which is larger than carrier frequency by 100kHz and corresponds to 50kHz detuning. CC2 demodulation phase is 105deg for squeezing, 155deg for anti squeezing and 135deg for intermediate.

Power measurement when PR chamber is open(measured on 20191011)

Aritomi, Matteo, Yao-Chin, and Yuhang

All the measurement is inside PR chamber. We sent BAB to do this measurement.

injection power	394uW
before dichroic	379uW
after dichroic	375uW
transmission from the first PBS	3.4uW
transmission from the second PBS	3uW
before going back to bench	334uW

Injection loss: 4.8%

Reflection loss: 10.9%

IR reflected from filter cavity cut issue solved

Aritomi, Matteo, Yao-Chin, and Yuhang

We opened the PR chamber on last Friday. But moving the last IR back reflection steering mirror, we solved the cut issue of this beam.

Attached figure 1: We just opened PR chamber, we checked with IR viewer. We could see a beam on the right side of chamber window.

Attached figure 2: After moving the last IR back reflection steering mirror, we made IR back reflection hitting on the window. And this IR back reflection is between IR(injection) and GR.

H1 INPUT mirror coil doesn't respond but coil driver is fine

Yesterday I investigated the issue of not responding coil of INPUT mirror.

I made some tests by swapping the coil driver channels and I concluded that the coil driver is working fine and the problem is inside the vacuum chamber.

The not responding coil is H1 (top coil). This unbalance in the pitch driving will excite length. This can be maybe counteracted with horizontal coils. 

We should consider if to open the chamber for further investigation/coil repairing. 

Set of iris on IR reflection path for tomorrow's chamber open

Yaochin and Yuhang

Since we will open the chamber tomorrow and move the last IR reflection steering mirror in-vacuum.

To define the beam reflection direction, we think we should check two points. The first point has put an iris. The second point is homodyne PD.

We will use these two points as a reference and then move the mirror.

Set of PD at the window of PR chamber

Aritomi. Yaochin and Yuhang

We set two PDs outside the PR chamber(attached figure 1), where we have the reference of two IR beams and two GR beams. The set two PDs are used to monitor two IR beams.

Mainly they have two functions:

1. Monitor the power (leakage from PBS before and after in-vacuum FI) injected into the filter cavity. Since we have the problem of IR transmission fluctuation, we want to see if this fluctuation also presents in the input beam. The problem can be found in this entry 1710. We just temporarily monitor this leakage power and the result is shown in the attached figure 2.

2. Use as a reference for the adjustment of in-vacuum FI.

The measurement of NIKHEF PD amplifier gain

Eleonora and Yuhang

Since we have the measurement of qubig PD and NIKHEF PD reported in entry 1728 we could use the measurement result to compute the NIKHEF amplifier gain. The signal transfer is like following:

qubig PD:       PDH signal pk-pk (164mV) = laser power (0.21mW) * photosensitivity (0.27 A/W) * amplifier gain (16e3 V/A) * RF amplification (14dB) * mixer gain * lowpass filter gain

NIKHEF PD:  PDH signal pk-pk (14.8mV) = laser power (1.2mW) * photosensitivity (0.2 A/W) * amplifier gain (we want to know V/A) *  RF amplification (14dB) * mixer gain * lowpass filter gain

Since we are using the same RF amplifier, mixer and lowpass filter, and we know the gain and photo-sensitivity of the quibig PD (16 kV/A) we could calculate the amplifier gain of NIKHEF PD:  ~ 340 V/A(50dB).

We shoud check if this is consistent with the calculation from the electronic schematics. If this is the case we conclude that the Nikhef quadrant has no problem but its gain is too small to be used without amplification.

Record IR and GR laser light height

[Yao-Chin, Yuhang]

We record GR & IR height before the light is sent into chamber window.
Pic. 1 shows the GR height of 76 mm and its beam size of 4 mm. Pic. 2&3 observed by infrared viewer show IR height of 76 mm and 75.5 mm. The observed position of Pic 2 is relatively close to chamber window than Pic 3.

Tuning of binary number of CC PLL and CC1 demodulation frequency (CCFC)

[Matteo, Aritomi]

Yesterday we found low frequency oscillation of CC2 error signal and tuned frequency of CC PLL and CC1,2 demodulation frequency to remove this, but we found that binary number of these frequencies were different. Frequency difference between CC PLL and CC2 demodulation was actually 0.04Hz and this may cause low frequency oscillation of CC2 error signal. Then we tuned binary number of CC PLL and CC1 demodulation frequency. Current setting is as follows. I attached the pictures to be sure. Note that CC PLL frequency is divided by 3 at PLL board.

channel	function	frequency (MHz)	binary number
CH0	CC PLL	20.99099988	1010101111110101010100010100
CH2	CC1 demod	13.99399992	111001010100011100010111000
CH3	CC2 demod	6.99699996	11100101010001110001011100

 

 

 

 

We'll check CC2 error signal and AOM frequency of CCSB next week.

Check of IR alignment

Current mode matching is 92% and should be fine.

Mode	IR transmission
TEM00	1500
HG01 (pitch)	180
IG20	130
offset	94