Abstract:
I changed the frequency of local oscillator for IFO RF sidebands.
RF AM seemed to be reduced by a factor of 10 or more by changing LO frequency from 5.6246235 MHz to 5.62437105 MHz.
Detail:
First, I subtract the dark noise offsets of each PD on POP and REFL for the preparation.
After that, I misaligned both ITM and ITMY for obtaining pure reflection from PRM on REFL PDs.
Then, to reduce the amount of AM sideband as like klog4831, klog7281, klog9073, and , I modulated the laser frequency.
For the frequency moduration, I used DGCARM path and injected the signal to error point of IMC.
Detailed configurations are as follows:
1. Open switches of SUMMING CMS IN1 and IN2 not to feedback the signal from GR PDH CMS.
2. Change slow output offset of SUMMING CMS to zero.
3. Close CARM CMS IN2 to feedback DGCARM signal to IMC.
4. Close IMC CMS IN2 to add the DGCARM signal to error point of IMC.
5. Turn off all FMs in DGCARM filter bank and injected the excitation signal (1023 Hz, 1000 cnts) from DGCARM filter bank.
Then, I tweeked LO frequency to find the best freqeucy for each PD signal for 1f sideband.
Following table shows the best LO frequencies for each PD.
| Best frequency (MHz) | |
| REFL 45I | 5.6243675 |
| REFL 45Q | 5.6243728 |
| REFL 56I | 5.6243708 |
| REFL 56Q | 5.6243731 |
| Average | 5.62437105 |
Since the difference of best frequencies for each PD are larger than the measurement errors (~0.2 Hz), I used the average value this time.
Figure 1 shows spectra of each PD before and after changing LO frequency.
Black and red line show the spectra of each RF PD signal at REFL before and after changing te LO frequency from 5.6246235 MHz to 5.62437105 MHz.
As you can see, AM sideband of 1f decreased by a factor of 10 or more, and AM sideband of 3f became smaller than the PD noise.
After changing the LO, I haven't checked the phase of each PD yet.
So, I will do it for the next step.
Note:
Measured data was stored at /users/Commissioning/data/IFO/2022/1202/.
The LO frequency before changing was 5.6246235 MHz but I cannot find the log to change this frequency except for the miyoki-san's recovery log (klog16155 and klog23010).
So, I don't know from when this value is used.
