Capture Torque vs. Angle readings with multiple DAQ devices simultaneously

[Mrgoomis]

I am trying to capture a torque signal from an NI 9237 and combine it with an angle signal from a PCI 6251 with an SCC-68. It seems that I am able to capture each individually without a problem, but I seem to be getting off on time bases during aquisition. I am running a 1mhz oscillator for the aquisition clock through the SCC-68 for the angle signal and a 10khz internal clock for the torque aquisition on the 9237. I take every 100th point out of my array to size it correctly down to the 10khz collection of the torque signal. Since the signals are coming from 2 different devices I am having trouble getting them to align accurately on the time base. I am new to labview and am stumbling through this at the moment. Any suggestions would be appreciated. Also i'm guessing there is probably a better way to do this with different hardware?

My code is attached.

[ColeTrain]

Hello Mrgoomis,

Welcome to the forums. This is almost certainly an issue with the delta-sigma convertor on your 9237.  The delta-sigma will introduce an offset because of the way that it functions.  This KnowledgeBase has a great outline of how to compensate for this in code. 

A few asides for you, while this is being programmed in LabVIEW, our DAQ forum also has great users and would be a great place to post if you have any other troubles setting your system up and getting all of the timebases aligned.  Also, if you are using the 9237 in a cDAQ chassis, we can share clocks through a PFI line to ensure even tighter syncronization.  Hope this helps!

[Mrgoomis]

First off, Thank you for you first reply. I have finally gotten to the point where I have the correct equipment for the job.

I have a cDaq 9174 chassis, 9237, and 9401. I have been working with NI support to get the clocks synchronize between the modules. I initially had the 9401 set up to generate a clock signal and the 9237 would read it internally for sampling. Unfortunately this method worked when using virtual equipment but the 9237 would not obtain samples with real hardware(Times out before any samples could be acquired) So now I am running the angular encoder off of the oversample clock generated by the 9237. This way seems to work correctly as far as data acquisition goes. Now I am back to the offset between the sources, I understand the 9237 will have some offset due to the oversample clock and acquisition/filter delays. The problem I am having now is that the offset does not seem to be constant. I am running a 1hz digital waveform (using an external function generator) into both the 9401 and 9237 using a voltage divider. I confirmed there is no lag between signals using an oscilloscope. When reading these inputs into my vi there is an offset but it is not constant, given the synchronize clocks I do not understand why this is happening.

*Also the previous knowledge base article you referenced is missing links to images that should show up on the page so it is hard for me to grasp what it is describing.*

The attached image shows the offset scatter I am seeing.

As always, any help is greatly appreciated.

Thanks

Message Edited by Mrgoomis on 11-24-2009 08:27 AM

[ColeTrain]

Hello Mrgoomis,

Sorry about the KB, it should have the images now. As far as this offset, if you have the task set up to be hardware timed, it should have the same offset everytime.  How are your tasks set up and configured?  Could you show your code please?  This will help us move forward on this issue.