Abnormal spikes in the analog output in BNC 2110.

[Haisenberg]

Before I talked about the problem, Let me briefly introduce the background. Basically, I applied a voltage across a membrane with a pore and measured the current passing through the pore (Resistive pulse sensing). Usually the current should be a flat current with a noise band around 200 to 300 pA if voltage is controlled by other DC power supplies (I tried two different ones and named them as 1 and 2)

Since I have to control it in real-time. So I used a DAQ board(PCIe 6321) and BNC board (BNC 2110) for DC analog output.

The problem is: when I used BNC board to generate DC analog output, the current measured is much more noisy (>1000pA, see attached pic 1 for comparison).

Then I compared the three DC output voltages (9 volt). I found out that there are lots of spikes in the output (See attached pic 2). However, when I plot all of them and did a spectrum analysis. It shocked me that the DC output from BNC board actually has a much better precision, and the smallest fluctuation and high-frequency component (see attached pic 3).

Does anyone know why? Greatly appreciated! (I am new in DAQ)

[Bob_Schor]

Sounds like a question for a good Electrical Engineer, equipped with oscilloscope and knowledgable about impedance matching, proper grounding, and shielding.

Bob Schor

[Haisenberg]

Thanks for the reply. But a good electric engineer is always hard to find.

[Bob_Schor]

Well, if you have access to a library, you can probably find some textbooks.  Check the hobbyist Web sites.  Do a search for "Grounding and Shielding".  Look up "Impedance Matching".

Bob Schor