Creating a skew between AI channels

[jarjour.3]

Hi,

I have an PCI-6115 board that I am using to generate scans for, and recieve data from an electron microscope. Essentially, the DAQ generates pairs of voltages on the analog output pins, which causes the electron microscope to take data at a certain location, and then that data is reported back to the DAQ card. All of that works fine, I have the card running at 10 MS/s input, and 1 MS/s output, meaning 10 samples are taken at every location and averaged. The data coming from the microscope is so noisy, however, that it would be ideal to take even more samples and average them. Is there a way to cause a specific skew between the sampling times of the 4 analog inputs, such that they could all be connected to the same physical input and acheive an effectively 4 times higher sample rate? It isn't key what that skew is, as long as it is about 10-100 nS/channel. I understand I would have to use a clock other than the analog in sample clock for the other 3 channels, but I'm not sure what clock to use. Assuming I find a suitable clock, how would I go about triggering that clock 10-100nS after the AnalogStartTrigger?

Thanks

Alexander Jarjour

[Call.Me.Wilbur]

Hi Alexander,

While you could modify your code to sample from all channels in this way I think it may be better to address the root cause of your noise.

I recommend taking a look at the following documents to help begin this process.

http://www.ni.com/white-paper/3344/en/

http://digital.ni.com/public.nsf/allkb/4327BA58E55F172A86256A380055BC65

Additionally, depending on the behavior of the noise you could apply filters to your analog input to improve the signal. You can do this in LabVIEW using a DAQmx Channel Property Node.

Finally, your device is capable of sampling all channels simultaneously so you could easily average all 4 simultaneous channels. Depending on where the noise is being introduced this could be a valid option as well, although I believe a filter would provide a better result.