09-29-2010 09:40 AM
Greetings,
I need to measure a variety of analog signals and RPM signals and save the data to a file. I had intended to use counters for the RPM signals. The problem is that the counters use implicit timing which is totally asynchronous to the ai sample clock. Therefore, there is no time or sample correlation from the counter data to the analog data. I would like to merge the counter data with the analog data into a waveform array. Does anyone know of a method that would allow such a merger?
10-26-2010 08:58 AM
A possible solution for merging signals with varying time axes is an X-Y plot:
The XY graph accepts an array of plots, where a plot is a cluster that contains an x array and a y array. Refer to the (X and Y arrays) Multi Plot graph in the XY Graph VI in the labview\examples\general\graphs\gengraph.llb for an example of a graph that accepts this data type.
10-27-2010 05:27 AM
Thank you for your response. I understand what you are suggesting and have used that technique before for displaying data. The first, and foremost, problem is that there is inherently no timestamp and no regularity from most frequency/period measurements. Thus, there is no time relationship to other channels. I was hoping for a technique that would synthesize a waveform from asynchronous, non-regular data that could be related to other waveform data. This may be an insurmountable problem and I may just have to revert to frequency-to-voltage converters.
10-27-2010 05:54 AM
Hi,
Extrapolate the analog signal to the rate of the counter data and merge togather. By this way analog signal is "constructed" as sychnronus to counter data.
10-27-2010 06:19 AM
Thank you for that. It is a direction that I had not considered. Unfortunately, that may not be a good solution. I am sweeping a drivetrain system from idle to max speed, so the counter sample rate could be all over the place. I am also doing doing frequency domain analysis, so I need fast, regular sample rates. Again, I may have to consider the analog route.