02-23-2012 12:45 AM
Is there any LabView code that can do a swept sine test with a single output (stimulus) and read & process multiple inputs (responses) from the same DUT ? In the Sound & Vibration toolkit the Getting Started Swept Sine example does a single response channel very nicely, but modifying it to handle multiple responses is not a simple task !
02-23-2012 08:22 AM
One of the drawbacks of wrapping the generation, AO, AI, and analysis in a single monolithic API is that it made extensions (including multiple responses) challenging. We are actively working on this challenge. Can you post your requirements for a discrete frequency sweep here?
02-23-2012 08:30 AM
Depending on your test requirements, you can also use a series of One Shot, Single-Tone measurements. Step the frequency for each measurement to cover the desired bandwidth. Refer to the example <LabVIEW>\examples\Sound and Vibration\Audio Measurements\SVXMPL_One Shot Gain Phase and Distortion (DAQmx).vi as a starting point.
02-24-2012 02:47 AM
At each frequency increment we want to measure the response of a hydraulic system in terms of pressures at 3 points and accelerations at 3 points (ie 6 response signals). The stimulus signal is the force applied to the system and is common to all 6 responses. The frequency range is 20 Hz to 2 kHz. At this stage we're just interested in the magnitude responses.
02-24-2012 03:01 AM
re: Single Shot Mode: I've tried using it in a loop and incrementing the stimulus frequency with each iteration. It probably could work with some further refinement of the code, although its noticably slower that the Swept Sine example vi and the start/stop nature of the stimulus requires a longer settling time.
In modifying the Swept Sine example vi the acquisition of multiple response channels doesn't seem to be a problem, but processing them causes problems - the main loop errors out after 3 iterations, regardless of sampling rate or no. of input channels !
02-24-2012 06:39 AM
Send your modifications. We can help to look for the performance bottlenecks. That may be the quickest way to get your SIMO swept sine test running. If you post it, the LabVIEW community can help, but if you would prefer to send it to me dirctly by email, send it to firstname.lastname@example.org.
02-26-2012 06:57 PM
Thanks Doug, but we've just finished the testing by doing it one response at a time. So we are right for now, but this requirement will come up again some time. Hopefully by then the S&V toolkit will be able to do a multi-response swept sine.
08-29-2012 02:39 AM
I made some changes to the Sound and Vibration swept sine toolkit to support multiple channel response analysis.
Essentially this could be used only on cascading type systems. (Systems with single stimulus and multiple response), however, the response should be cascade thus, response1 vs stimulus, response 2 vs stimulus and so on...
However, the hardware to be used had to support multiple synchronization, thus we were force to used 4461 for generation of stimulus and 4462 for acquisition...note 4472 being of different family cannot be used.(difficult to synchronize multiple channels).
Unfortunately,I currently do no have the modified code and would not be in position to share. I believe it is possible.
08-29-2012 04:48 AM
Also, refer to this link