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
Message ID: 1984     Entry time: Wed Nov 15 14:54:30 2017
Author: Craig 
Type: DailyProgress 
Category: BEAT 
Subject: FSS Fast Gain Effect on Beatnote Spectrum 

I was wondering what is causing our huge mountain at 10 kHz.  awade said it might be the PZT EOM crossover with gain peaking.  I did a quick test in which I turned down the Fast gain sliders to -2V, then incremented them up by 1V while taking spectra at every point. 

We start to get diminishing returns from increasing the Fast gain at around 4V.  Before then, we can see some FSS gain peaks moving around and breathing at around 2-3 kHz.

We have a broad peak at 350 Hz.  This one is weird in that it likes to breath, or come in an out of existence, on a timescale of a few seconds. 

Similar descriptions for the 640 Hz and 1040 Hz lines, but to a lesser extent.  There is some sort of creepy-crawly noise floor we're hitting between 300-2000 Hz once we're no longer FSS limited.  I blame the australian.

Also the mountain at 10 kHz is not affected by any of this.

Next step: FSS error signal spectra, FSS OLGs, and FSS calibration for both cavities.  This will help with determining our current limiting noise floor.

 

Some interesting points:

The North cavity has weird breathing in it's error signal on the order of a few milliseconds.

The South cavity does not like a Fast gain of 4V: the error signal tends to blow up randomly, then be fine for a minute, then do it again.  Could be the crossover between the EOM and PZT acting up.

The North cavity and South cavity have significantly different responses to extremely high Fast gain.  The South cavity, when the Fast gain is too high, causes the EOM to start fighting it, increasing the EOM response until it rails.  The North cavity has no limit to how high you can turn the Fast gain; the EOM response never increases in response to a huge PZT gain.  Cause unknown, maybe the North has good phase at the crossover while the South is bad?

Some settings:

The Fast and Common gain sliders are calibrated to 32 dB/V.

The Common gain sliders were set to -3.5V.

The number of averages was 25.

The PLL was locked at FM Devn of 1 kHz/Vrms, Preamp Gain of 2000, and Carrier Freq around 109 MHz.

Edit: Remade plot with more resonable labels and ylabel.

 

Attachment 1: TransBeatnote_BothPathsFastGain_0V_15-11-2017_134517_Spectra.pdf  189 kB  | Hide | Hide all
TransBeatnote_BothPathsFastGain_0V_15-11-2017_134517_Spectra.pdf
Attachment 2: TransBeatnote_BothPathsFastGain_0V_15-11-2017_134517.tgz  89 kB
ELOG V3.1.3-