Differences between revisions 1 and 16 (spanning 15 versions)
⇤ ← Revision 1 as of 2018-08-04 14:24:58 →
Size: 775
Editor: ChristopherwipfATligoDOTorg
Comment:
← Revision 16 as of 2018-08-07 01:34:55 → ⇥
Size: 2233
Editor: DuotaoATligoDOTorg
Comment:
Deletions are marked like this. Additions are marked like this.
Line 10: Line 10:
  * The worst-case dewhitening state for the noise monitor is ACQ off, LP on.  * The worst-case dewhitening state for the noise monitor is ACQ off, LP on.
Line 13: Line 13:

=== Coil driver transfer function ===

 * 1G: [[attachment:CDtrans-1G.pdf]]
 * 1k: [[attachment:CDTrans-1k.pdf]]
 * The coil driver attenuates about 1/5 in the 1k domain.

=== Some design thoughts 8/5 ===
 * The current design uses a couple passive filters, interleaved by amplifiers. Passband 0.482Hz - 482Hz, gain ~200. Combining with the coil driver transfer function, the signal is amplified about 40 times from the input of the coil driver. Compared with the given input signal, it seems the saturation is mainly caused by low frequency signals below 10Hz.
 * Following Rana's idea on Friday: a pair of low/high pass filter + instrument amplifier (where we get the gain), what else do we need?
  * Sharper cutoff (Higher Q)? Perhaps higher order passive filters?
  * Noise? Input impedance?
 * If we have a good filter, according to the input spectrum, maybe applying a gain of 500? This allows the 300nV DAC noise to overwhelm the 4uV ADC noise. This also keeps the signal away from saturation, according to the given coil driver input spectrum. Do we have to worry about the strong mode at 500 Hz in both H1 and L1?
 * Trying to learn about active filters. Would we consider using them?

=== Decisions 8/6 ===
 * Start from the existing design but
  * spread the gain across the stages to manage risk of saturation
  * Move the low frequency cutoff to 20Hz
 * Move on to active filters for better stop band to pass band transition

SUS Noise Monitor

Design goals

  • No saturations as long as the coil driver input signals are at the 99th percentile or below.
  • Provide enough gain (at 20 Hz and above) to boost the DAC noise above ADC noise.

Design inputs

  • DAC noise model: page 6 of G1401399.

  • PUM coil driver transfer function. LISO models are available here.

  • The worst-case dewhitening state for the noise monitor is ACQ off, LP on.

  • ADC noise level, about 4 uV/rtHz. (Reference: T070213)

  • Coil driver input spectra, G1801540.

Coil driver transfer function

Some design thoughts 8/5

  • The current design uses a couple passive filters, interleaved by amplifiers. Passband 0.482Hz - 482Hz, gain ~200. Combining with the coil driver transfer function, the signal is amplified about 40 times from the input of the coil driver. Compared with the given input signal, it seems the saturation is mainly caused by low frequency signals below 10Hz.
  • Following Rana's idea on Friday: a pair of low/high pass filter + instrument amplifier (where we get the gain), what else do we need?
    • Sharper cutoff (Higher Q)? Perhaps higher order passive filters?
    • Noise? Input impedance?
  • If we have a good filter, according to the input spectrum, maybe applying a gain of 500? This allows the 300nV DAC noise to overwhelm the 4uV ADC noise. This also keeps the signal away from saturation, according to the given coil driver input spectrum. Do we have to worry about the strong mode at 500 Hz in both H1 and L1?
  • Trying to learn about active filters. Would we consider using them?

Decisions 8/6

  • Start from the existing design but
    • spread the gain across the stages to manage risk of saturation
    • Move the low frequency cutoff to 20Hz
  • Move on to active filters for better stop band to pass band transition

Electronics/NoiseMonitor (last edited 2020-02-06 00:41:08 by DuotaoATligoDOTorg)