Displaying report 1-1 of 1.
Reports until 17:05, Tuesday 07 February 2023
H1 ISC (CSWG, ISC)
jeffrey.kissel@LIGO.ORG - posted 17:05, Tuesday 07 February 2023 - last comment - 17:14, Tuesday 07 February 2023(67272)
OMC DCPD Transimpedance Amplifier Dark Noise
J. Kissel

While I was out cleaning up the OMC DCPD Signal Chain (LHO:67267), and there was no light on the DCPDs for a full-signal-chain dark noise measurement (LHO:67262), I figured I'd take advantage of the super-convenient BNC spigots on the new D2200215-style OMC DCPD whitening chassis that read out the OMC DCPD Transimpedance Amplifier. This both tests the usefulness / functionality of the spigots as well as helping us add another individual term to the dark noise budget, in hopes to confirm than indeed the transimpedance amplifier noise is the dominant noise in the transamp + whitening + ADC signal chain.

Attached are the "raw" results***.

In short, within the gravitational wave band, between 100 Hz and ~3 kHz, each DCPD has a noise floor of around 300 nV/rtHz. 

Interestingly,
    - This is is higher than the ~200 nV/rtHz that was measured of the LLO unit, SN01 or S2100830, (see LLO:61913, cast into these TIA output voltage noise units on page 3 of G2201909), and
    - and also higher than the ~150 nV/rtHz predicted by a Rich Abbott spice model (also on page 3 of G2201909)
    - and also higher than the ~100 nV/rtHz measured when they were first built by the Cardiff group (see Figures 6 and 7 of T2100403)
    - BUT, less noisy than the ~800 nV/rtHz Louis and I measured the noise with a breakout board and clip-doodles in front of the D2100631-style whitening chassis (LHO:63967) -- which leads me to believe that was a pretty junk measurement that should be disregarded.
where, the pointers above all show the noise in the same "differential voltage of the TIA at the output" units, so the commentary that we seem to have the noisiest is an apples to apples comparison.

That being said, Camilla and Sheila's measurement of the full chain (again, LHO:67262, specifically the OMC_DCPD_darkandNLN.png plot) implies that this is still well *above* the ADC noise (though we don't yet have an isolated measurement of it directly), and well *below* the noise when there is light on the DCPDs.

***I'll work over the next few days to cast these TIA, Whitening, and ADC noise sources into the *same* noise units at a single point in the chain, rather than at specific, incomparable parts of the signal chain that are confused by frequency-responses and gains of each part.
Non-image files attached to this report
Comments related to this report
jeffrey.kissel@LIGO.ORG - 17:14, Tuesday 07 February 2023 (67274)CAL
Here're some photos of the now super-trivial set up with the D2200215 style chassis: BNC cables go straight from the chassis into the SR785, in a one-to-one map. No breakout boards, no clip-doodles, no worry about a common electrical ground plane between chassis and measurement device; just really nice and easy. 

Great work, Dean!

IMG_9308.jpg shows the zoom out of the set-up,
IMG_9307.jpg shows the chassis connections, and
IMG_9306.jpg shows the SR785 connections.

1000x better than the hot mess we'd needed to set up before.
Images attached to this comment
Displaying report 1-1 of 1.