LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

how to find time delay between two input channels in ni pci 6036 DAQ card

Hello ,

I have read many related posts about simultaneous measurments of two analoge voltage input channels in  daq card .I know the best hardware is " simultaneous measurments Series of DAQ cards " but I only have pci 6036 DAQ card and I am trying to figure out what is the time delay between reading of the two channels .Is this time always constant ?(does it depend on  input voltage (frequency,amplitude,waveform ,..).I am sending sine wave(s) to both channels and reading off the V values, if they read the same value the difference should be always zero but I am getting -0.002 to 0.002 volts difference (I must find a way to convert this to time) .Attached is a screenshot of my VI.I am wondering how I can measure exactly the time delay between the channel .

I am open to any suggestion , my final goal to read the two channels exactly at the same time ((or find out the exact time delay so I can match the corresponding data considering the time delay))

0 Kudos
Message 1 of 8
(4,184 Views)
Unfortunately there is no simple answer to your basic question. Converting the voltage to time for your current set up is not hard knowing the frequency and amplitude of the sinewave, and assuming you are measuring this voltage difference at the zero-crossing point.

What is it exactly that you are trying to do?

Mike...

Certified Professional Instructor
Certified LabVIEW Architect
LabVIEW Champion

"... after all, He's not a tame lion..."

For help with grief and grieving.
Message 2 of 8
(4,045 Views)

@mikeporter wrote:
Unfortunately there is no simple answer to your basic question. Converting the voltage to time for your current set up is not hard knowing the frequency and amplitude of the sinewave, and assuming you are measuring this voltage difference at the zero-crossing point.

What is it exactly that you are trying to do?

Mike...

Even doing this will be troublesome as there is no way to ensure that each input will be perfectly accurate (or that their individual inaccuracies are the exact same).

 

If your final goal is to make sure that you, "read the two channels exactly at the same time" we should figure out what offset would be acceptable for your experiment.  The two inputs will never be "perfectly" synchronized, but at what point is it good enough?

Matt J | National Instruments | CLA
0 Kudos
Message 3 of 8
(4,032 Views)
I think the time should be constant, there is a sample clock and a multiplex clock on the daq card (i'm not exactly sure these are the right names) the time between channel samples depends on the MUX clock rate which can be read from a property node of the task. You can set it too, but be careful as it must sample all channels in enough time to meet the sample clock rate.
0 Kudos
Message 4 of 8
(4,022 Views)

Thanks for the reply , as long as there is constant time delay between the two  channels (for instance ch1 reads always 200us before ch2) I do not worry and I can safely find one to one corresponding data using the time delay .As for my goal , I want to be able to read the two channels (ch1 and ch2) and plot V1 vs V2 .So it is obvious that (V1,V2) has to be syncronized .

Thanks again

0 Kudos
Message 5 of 8
(3,900 Views)

The maximum sampling rate of the 6036 is 200KHz, so the time between channels has to be less than 5 microseconds.  What sampling rate are you planning for your signals?  If you really seriously want to measure the time difference between the channels, you could do something like put the same sinusoid into all your channels, sample at your desired sampling rate (the sinusoid frequency should be about ten times lower than the sampling rate), then determine the relative phase difference beween the sampled channels.  Translate phase difference into time difference and you'll have your answer.

 

Bob Schor

0 Kudos
Message 6 of 8
(3,824 Views)
Yes, but the problem is that you will only find an answer for that particular set of conditions. There is no guarantee that it will hold true in all conditions. To do this in a s wy that guarantees a known skew between channels requires specialized hardware. So the question really us how much skew can you tolerate (hunt: 0 is not an option).

Mike...

Certified Professional Instructor
Certified LabVIEW Architect
LabVIEW Champion

"... after all, He's not a tame lion..."

For help with grief and grieving.
0 Kudos
Message 7 of 8
(3,803 Views)
The convert clock rate or the inter-channel delay does vary with the sample rate but you can read what it is and set it with a property node.

https://decibel.ni.com/content/docs/DOC-14548

http://www.ni.com/example/30797/en/

0 Kudos
Message 8 of 8
(3,771 Views)