Help with Plugin for Higher Fidelity Data Collection

[BHABES]

I am looking for help to develop a plugin for FlexLogger to continuously save all configured high frequency data signal channels, based on a user defined loop time averaging interval, without needing to create multiple derived channels. A second output to the continuous save file would be the RMS of each channel over the specified loop time. The plugin would also allow for intermittent full resolution high frequency data to be recorded for a user specified period of time. Notes functionality should work with the continuous save and should populate some indexing into the continuous save. Also, when a full resolution data set was taken, a note/index should be populated into the continuous save file. The files should be in TDMS and work with the CSV export plugin.

Some Background:

For everyone in the FlexLogger community to be able to gather higher fidelity data, the aforementioned plugin or built-in functionality to FlexLogger would be a significant benefit. While taking and recording high frequency data is great, doing so for a longer periods of time is not usually necessary. Shorter snippets of high frequency data at specific user defined intervals is the exactly purpose of flex logger. However, where you may have multiple data acquisition modules, and multiple instruments configured, possibly at multiple sample rates (with differing modules), and are looking to collect data for hours to days, collecting data at full frequency resolution creates a massive amount of unnecessary data. Changing/lowering the sample rate reduces the fidelity of the data collected by randomly grabbing instrument data scatter/noise. A statistically significant way of reducing the amount of data collected is to take an average of higher frequency data over a specified period and recording the rms of that signal for that period's average. Thereby, you can have a statistically meaningful long duration data set without recording an excessive amount of data. I have LabView code for a very large program that has a lot of this functionality built it, but it is old and messy. Feel free to reach out to discuss more.

[Vincent_Carpentier]

Hi BHABES,

I tried sending you a DM but it is not enabled on your profile.

We could work with you on elaborating that plugging.

You can reach me at vcarpentier (at) neosoft (dot) ca to discuss further.

We are based in Canada ( Neosoft Technologies) but we also have another company in the US (PVI Systems) as well. We've done projects in France, Australia, Poland, and more.

Let me know what you think.

All the best,

[Devin_K]

Hi BHABES,

It's not exactly what you describe with logging RMS data, but recent versions of FlexLogger do have the ability to log data at a a background rate between triggers, then at full fidelity when a trigger is observed. It does this by resampling any faster channels down to this slower background rate.  It might be worth taking a look at if you haven't already.

[BHABES]

Hello Devin_K,

I am aware of this triggering function. This was to my point, that if one was to say sample even slower at 1 HZ or 0.25Hz nominally to reduce data size, for example, to my knowledge, this is a single pole every second or every four seconds of say 1000 Hz data, rather than an average of 1000 or 4000 samples. Depending on the instrument and DAQ module, this could be really bad data, with lots of scatter. Please correct me if I am wrong in the way that this is feature is functioning.