Multiple Analog input channels affecting each other (DAQ_mx)

[64Bader]

Hi, I use NI BNC-2110 / PCIe-6363 + Labview 2012 for my DAQ.

In out task, we acquire two analog samples(these two samples are timed by on board clock, 1000000 rate) from each analog input (AI_0 and AI_4) whenever DAQ receives a digital trigger from PFI0 (rising edge and retriggerable). We will receive 200 triggers and our program will finish until we receive 400 samples on each channel.

The interesting thing is that different analog inputs seem to affect each other. As shown in the figure below, the white curve is analog input AI_0 and the red curve is analog input AI_4.  AI_0 is connected to a triangle input and the curve

below seems right. AI_4 is actually shorted (some result if open) but the acquired data is affected by AI_0.

However, this problem does not occur when only one channel is used in vi (Only the "create channel" and the "read" vi's are changed to be multiple sample and we repeat the experiment with everything else unchanged). So I believe it is not an actual voltage problem before the BNC-2110 board receives the analog input.

Any idea about how to solve this? Thanks in advance! Happy new Year.

Yang

[Dennis_Knutson]

What do you short ai4 to? When open, the result you have is normal. The effect is called ghosting and there have been numerous posts on it.

[64Bader]

I used a wire to connect the BNC core and BNC shell on AI_4 when I shorted it(it is set on floating source mode).

The effect happens even if I have some other kind of input in AI_4(the input is like a stair case but it is always modulated by AI_1). AI_4 always appears modulated by AI_0.

Actually it will not bother me at all if this only happens when AI_0 is open.

I guess the sample clock (1000000) is too high to cause this(Yep, probably it is Ghosting...). When this rate is 1000, it does not happen as I have just tried.