ADG1606 Multiplexer used for impedance measurements at low frequencies

Question

Dear

I was looking for a multiplexer to speed up my impedance measurements. I have metal traces with a coating as samples to test. I want to test the insulation integrity over time. On my sample, I have one low-impedance channel and multiple high impedance channels, they are measured to be > 100 MOhm at 100 Hz and linearly increase to GOhm at 1 Hz. When measuring with the multiplexer, it is in this region that the measurements are influenced by the multiplexer for only the high imepdance channels. In terms of grounding and removing noise, I use a farday cage with the floating ground of the potentiostat connected to it. I also use the same earth ground for the ground of the potentiostat, arduino and multiplexer.

I looked at the Idc values of my potentiostat when measuring and they look like this: 
 
 Mux high impedance channel 
 
 -8,859613E-005 op 1 MHz 
 -1,672427E-008 op 1 kHz 
 -1,680497E-008 op 1 H z 
 
 High impedance channel without multiplexer

-7,194425E-007 op 1MHz 
 -2,132992E-011 op 1 kHz 
 5,289324E-014 op 1Hz 
 
 The cannel-to-channel capacitance or crosstalk capacitance: 
 
 Cd off: 248 pF 
 Cd, Cs On: 271 pF 
 
 The Channel On Leakage, ID, IS (On): +/- 0.3 nA 
 
 Could this cause the influence or is there something else i need to take into account? Are there better suitable types of multiplexers do these kind of measurements at low frequencies, like ADG1208 for example? Kind regards

Lintexed · Accepted Answer

Hi Christian 
 Thanks for getting back to me.

So I am testing an insulating coating on a metal trace as device under test (DUT). I was mentioning 1Hz and 100 Hz as this is where the impedance spectrum changes significantly when adding the multiplexer in my system. I want to do impedance measurements for 1 Hz to 1 MHz ( and even 0.1 Hz if deemed possible). 
 
 I tested the multiplexer first with a sample randles model of 10kOhm and 10 uF in parallel. For this the spectrum is fine, so I doubt the Ron influences the measurement. As at high frequencies the reries resistance can be found (this is the resistance in the connections, the PBS wherein the DUT is submerged, the mutliplexer resistance, measured to be 28 Ohm). At low frequencies, the resistance of the parallel circuit is found (10 kOhm).

The Gamry potentiostat used a 100 mV AC rms voltage signal as input. I can change that to higher or lower values, but I have not seen a difference altering this in the case where i measured the samples without a multiplexer. 
 
 Ideally I would have a mutiplexer to measure 8 channels, but the more the better. My goal is not to switch fast or measure the impedances fast, but to measure over a large amount of DUTs sequentially. I want to start the impedance measurement of multiple DUTs sequentially so I do not lose time characterizing them, but the total time of measurement is not my major concern. I think for the voltages a supply of 3.3 or 5V is sufficient. The currents are all low, the Idc for the log files of the potentiostat are mentioned below and range from pA to a couple of uA I think they will never be higher than a mA anyway. 
 
 Also doing an open circuit measurement (so not connecting the RE/CE and WE to anything but leaving them floating in the farday cage) with and without the multiplexer gives a difference as can be seen below: 
 
 Blue is with the mux and red without. 
 Thanks for the help. 
 Kind regards)

Electronic Design Services - PartnerZone

Products Mentioned

ADG1606 Multiplexer used for impedance measurements at low frequencies