I am trying to design a voltage follower that is relative phase independent with frequency (<100KHz) with a high impedance input/ mV signal detection problem.
The signal input is a 5cm long saline filled 2-10MegOhm glass AgCl saline microelectrode, mounted right on the BNC input (I am not using the 250$ eval board cause we are poor), to a circuit in a shielded probe box. The electrode/probe box sits in a shielded Faraday cage. Electrode, probably 3pF load too.
The instrumentation I am currently using uses guarded AD8065s on a guard coax cable, about 20cm long, (nA bias current) but bringing the guard to the electrode is not enough to get a good signal with small mV sine wave at f>1KHz. Above 1KHz, there are huge phase errors.
1. Guarding, what should we be doing to the board, connector, (shield over part of electrode)? (Make the surfboard have guards like the eval board?)
2. Feedback resistor in voltage follower...what would you choose? 100K, 1Meg?
PS is +/- 3.3V DC, battery, 9V I guess.
I have some very high value tissue I need to analyze well in a bioimpedance spectroscopy application, and need to get something that is phase linear to at least 10KHz, 50KHz hopefully.