I am looking for some guidance with setting up a preamplifier for biopotentials (10µV to 100 µV signal range, BW 1Hz to 3 kHz). The electrodes have a quite high impedance (@1kHz) of 1 MOhm. The gain should be between 500 - 1000.
The signal will be digitized at 10 kHz with an NI-PCI6040e DAQ card, with 12-bit resolution with the tiniest input range around +/-100 mV.
For this purpose I've used so far an AD8221 for each channel with good results. Now I'd like to expand the number of channels (~8 channels) but minimize the space (something in a LSCFP format?). So I thought about multiplexing the electrode inputs. But honestly I have no clue where to start and how to avoid pitfalls.
Only the +IN will be multiplexed, the -IN is fixed to one electrode.
1) I don't know if there's some kind of switching noise due to the multiplexer and if I should amplify the signals before I route them to the multiplexer to increase the SNR signal/switching noise? And as this noise would only occur on one channel, should I rout +IN and -IN through the multiplexer?
2) as DEMUX I use ideally the same IC as the MUX, right? and control both IC's with the signal out of one signal source.
3) I would use at least 30 kHz as multiplexing frequency, right?
4) I guess it is quite important to reduce any capacitance at the input path
5) does it make sense to switch inbetween every channel-switch always to -IN to aid settling?
6) is it a straight-forward approach or should I implement the whole story differently?
Thank you in advance,