analog generation discontinuity

[Denes]

Hello all, and thanks in advance for the help.  Here is my situation.  I am using Labview to output a continuous waveform to an NI-USB 6221 BNC DAQ and inputting it into an analog in channel on the same device.  Everything seems to work fine for relatively high amplitudes (above 100mV) but when I output below this level, specifically below 80mV peak, I see a series of spikes in the waveform.  When I change the frequency of the signal (sine wave) I change the number of spikes -->it's on the output side.  The generation signal appears to be continuous, beginning and ending at 0V.  I thought this was a problem with my code, but then tried it out on the sample codes (Acq&Graph Voltage-Int Clk.vi & Cont Gen Voltage Wfm-Int Clk.vi) and had the same problem.  Thanks for your help.

        

Sam

[Christian Mergl]

So when you say that you don't think it's a programming issue, why didn't you post that in the "Multifunction DAQ" forums where it belongs to?

Maybe you can give us another screenshot where you zoom into the graph with the measured data until we can see the sine and the locations

of the spices. Due to this you might be able to determine a frequency of the spices and then check the sourrounding of your measurement system,

because maybe another device is making noice.

Christian

[Denes]

Thanks Christian.  I've posted to the Multifunction DAQ forum.  But just to answer your questions, I disconnected the BNC from the generation and recorded and if there was interference I would have expected to see it on the recording of from the OC cable.  I also generated a 0V signal and would have expected to see the spikes then as well.  I've posted a zoomed in picture of the recorded signal and you can see the spikes are occuring at the zero crossings.

Again thanks for the guidanc.  New post at http://forums.ni.com/ni/board/message?board.id=250&thread.id=48142

sam

[Ben]

Two idea that come to mind are the resolution of your output. When you get to small size only the peaks actual result in a change of the output.

The other idea would be impeadance matching where if the sink impeadance does not match the output, you could be seeing reflection of each step change.

Just trying to help,

Ben