Recall Saved Data in SoundCheck
Additional Resources for SoundCheck
Detailed explanations of how to use SoundCheck can be found in the SoundCheck manual.
Video Transcript: Recalling Saved Data in SoundCheck
It’s often useful to bring test data back into SoundCheck for additional analysis. This is useful for troubleshooting problems remotely, for example helping a contract manufacturer figure out some questionable results or if you want to re-analyze your own data without re-running the sequence.
While you can always bring this data into the memory list manually, replacing an acquisition step in a sequence with a recall step automates this process to save time.
Let’s look at an example.
Here, I have the factory test sequence that my CM is using.
In the past, they have sent me data that I needed to open up manually in SoundCheck, but now I have the actual test sequence. It’s also important to get the waveforms, saved as WFM or TDMS files, included with the sequence – these are the formats that contain the metadata that SoundCheck needs.
I’m going to make a copy of this sequence, and append ‘recalled data’ to the sequence name. Next, I’m going to replace the stimulus and acquisition steps in the sequence . These are the steps that define the test signal, play it on the device under test, and record the response.
I’ll replace these steps with a recall step… I’ll drag it in here. It’s important when adding this step to the sequence you place it right after the acquisition step but before any other step in the sequence uses these waveforms. …. You need to select the appropriate TDMS or waveform format depending on the file format of your date. In this case I’ll use TDMS.
Now here we have two options. We can have the sequence prompt the user to pick a file. If you choose this option, it’ll prompt the user to select the file each time the sequence is run. This is useful if you have multiple files or data sets you want to select from.
Alternatively, we can specify a file path. This is what I’m going to use in this case, and I’ll specify a path to the factory test data that they sent me.
It’s important that the names of the waveforms match the exact names used in the sequence, as we’re replacing them with the recalled data.
I also recommend saving the stimulus, just in case there’s any sample rate differences between stations.
If all the necessary data is contained in one file, you can use a single recall step. Alternatively, you can create multiple recall steps for the stimulus and as many response waveforms as you have in your sequence.
My recommendation is to use one TDMS file that contains all the waveforms, so a single recall step, whether specified or prompted, replaces the acquisition and stimulus steps.
Once our recall step is in place, click OK
Since these are the exact same names as the output, we need to configure this step to overwrite data.
It already is, so all I need to do is remove these stimulus and acquisition steps.
Now, I don’t need to actually have any hardware on my station. When I press start, the sequence will run, recalling the data. All the remaining steps, the analysis, post-processing, and displays will execute as if the data were acquired in real-time.
And we can see, when I press start, it outputs all my data here.
Now we can troubleshoot in many different ways. For example, we can go in and change some sequence parameters, say if we want to look at the resonance over a different range, and re-run the sequence.
Or maybe I want to add some additional distortion analyses and examine more harmonics to find the cause of some audible artifact. I can just add these in, and now when I run this I’m able to capture all of this data and analyze it from those same exact waveforms.




