AnsweredAssumed Answered

AD5933 measurement value drifts over time

Question asked by richeekdey on Jun 29, 2018
Latest reply on Jul 13, 2018 by mcee

So, I have been doing wireless dispersion measurement using the AD5933 eval board hooked up to an external amplifier which then runs a set of parallel plates placed 9cm apart. 

Over the past week, I've noticed that if I take readings for a long time, the impedance values seem to drift quite a bit. For a sample run of 6 hours, the drift is quite noticable. Any idea what is causing this drift? My initial guess was that the biological sample that I put in between the plates dried up overnight and thus that caused a change of impedance. However, when I put nothing in between the plates but just air, the drift still seems to persist. I understand the clock of the eval board is not super-precise but even then the drift is a bit too much. And this becomes more pronounced for a run time of 13-14 hours. Each sweep takes about 2.5 seconds (approx), so in 6 hours, I got 4000 sweeps. And upon plotting the permittivity from that, I got the following graph. I have also attached the graph I got from the cucumber sweep (that was my biological sample that I left overnight). I binned it into 1000 samples and then plotted the dispersion effect. However, as you can see, it drifts over time...