40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  PSL  Not logged in ELOG logo
Entry  Wed Feb 1 15:46:51 2012, tara, DailyProgress, Electronics Equipment, acav characterization inloop.pngnb_2012_02_01.pnginputrange.pngACAVOLGTF.png
    Reply  Thu Feb 2 23:51:32 2012, frank, tara, DailyProgress, Electronics Equipment, acav characterization 2012_02_02_block.png2012_02_02_pll.pngmarconi_TF.png
       Reply  Sat Feb 4 03:18:08 2012, tara, DailyProgress, Electronics Equipment, acav characterization 2012_02_03_pll.pngPLL_TF.png
          Reply  Mon Feb 6 22:35:49 2012, frank, tara, DailyProgress, Electronics Equipment, acav characterization TF.png
Message ID: 808     Entry time: Wed Feb 1 15:46:51 2012     Reply to this: 809
Author: tara 
Type: DailyProgress 
Category: Electronics Equipment 
Subject: acav characterization 

I'm working on characterizing ACAV loop to check if the loop is good enough for the coating noise measurement. The results show that there we need better improvement on this part.

 

I realigned the cavity, and the visibility is only 85%. I have not figured out yet why the visibility decreases from 90%. Then I measured dark noise @ error point, inloop noise with different gain setting, and error signal's slope ( all with 50 ohm system).

 

 Error slope (with 2V DC on RFPD) =122.7 kHz. The visibility is @ 85%

1) Dark noise and inloop noise:  The in loop noise changes with gain setup on UPDH box. I changed the gain from 3 to 7 and made sure that there was no oscillation (oscillating @ gain7.5).

inloop.png

 With the error signal slope, I can calibrate in loop noise to absolute frequency noise from ACAV loop and plot them on noise budget. On the figure below, we can see that at high frequency noise from ACAV loop is dominating.

nb_2012_02_01.png
 

Notice the bump around 3kHz. It comes from the phase noise of the marconi. If I change the tuning range from 10khz to 1kHz the in loop noise also change (blue to green in the below figure). It is a trade off between lower frequency noise, less gain.   The peaks around 80 Hz go up because we have less gain to suppress the noise.

inputrange.png

 

2) OLG TF at different gain setup

ACAVOLGTF.png

 

Back to the 10kHz input range. Since the in loop noise is getting worse with more gain, I look into the OLG TF of the loop. The UGF is around 3-10kHz, depending on the gain. TF looks ok at gain 3-5, the magnitude increases along with the gain. At gain 7, the TF starts to deviate at high frequency (it might be oscillating at high frequency).

The in loop noise should be suppressed with 1/(1+G) factor, but the results above do not tell the same story. The gain increases but the in loop noise gets worse. So the next step is to look deeper into this problem. I'll use SR650 instead of UPDH and check if the same problem occur or not. This should verify if the UPDH box is bad or not.

Note: I'll calculate what we need for ACAV loop, and check if LIGO homemade VCO will be good enough for ACAV loop or not.

ELOG V3.1.3-