NIDAQ USB 6212 voltage measurement unstable with simple high pass filter circuit

[jford]

I am attempting to measure the voltage across a high pass filter.  In my setup, C = 5pF, R = 1Gohm, and Vin = 100mVDC + Noise.  I am expecting Vout to measure 0 + noise, but instead it appears that the signal is floating often reaching the rails of the DAQ.  If I instead apply my voltage at Vout and try to measure at Vin, the results are even worse, with the measured signal often reaching the rails and saturating the DAQ.

Is this behavior expected?

[santo_13]

These DAQs are general purpose and not good to measure noise as you intend. Typical DAQ requires a low impedance signal source, in your case, it is a very high impedance and hence it cannot digitize the signal correctly.

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000000P9mKSAS&l=en-US

https://www.ni.com/en-us/support/documentation/supplemental/16/specifications-explained--ni-multifun...

[Henrik_Volkers · Accepted solution]

What type of source(s) do you want to measure ?

Do you want to use more than one channel?  (this is a multiplexed device, so low impedance is needed in that case)

and the input impedance is 100pF > 10 GOhm (for single channel measurement) so you build mainly capacitive voltage divider 😉

The spec notes a input bias current of +-100 pA , now multipy that with your 1G Ohm .... 😄    no way to expect 0 Volt!

If you want to measure with high impedance (and low input capacity) you need to use an active FET-Probe (or DIY) 

https://elektrotanya.com/files/forum/2009/10/e04a036.pdf

https://www.eevblog.com/forum/testgear/hi-z-probe-for-50-ohm-spectrum-analyzer/

Edit: dual gate FETs are harder to find these days, but nice OPs are available