RMS calculation in Statistics Express

[mtan3]

Hi all,

I want to figure out whether the RMS calculation in Statitics Express will output a signal with a different "number of samples" than the raw signal. I read about the following knowledgebase:

http://digital.ni.com/public.nsf/allkb/862567530005F09C862567210074EA10

but still not quite sure if averaging will reduce the number of samples. Can anyone help me out?

Thanks in advance for any answers!

[crossrulz]

An RMS is a scaler.  So I'm not sure where you are talking about reducing the number of samples.

In the end, this is the VI that is called: RMS VI

[mtan3]

I have attatched a picture of my program. In my program, without the Statistics Express it works fine. But once I put in that block, DAQ output says number of samples does not match, and wouldn't let me output the data to the oscilloscope. That's why I suspect that the RMS calculation somehow changes the number of samples when it processes the signal.

[crossrulz]

Oh course it does.  That VI just gives you 1 scaler value.  You started off with a waveform and you are left with a single number.  That is why your number of samples do not match.

[mtan3]

Is it possible to get a continuous RMS signal instead of just points, so that the number of samples will match?

[Dennis_Knutson]

That does not make much sense.

All of your code is hidden in the express VIs so it's impossible to tell what you are doing. What do you need to write? There is no scope code in your VI so your comment about that has no meaning.

[johnsold]

How would you define "a continuous RMS signal"? The RMS process operates on an array of data to produce one number.  As an example suppose your array has 100 values. RMS is a calculation which produces 1 value representing a particular characteristic of those 100 values. If you want to plot a continuous RMS signal, what values will you use for the other 99 elements?

Lynn

[crossrulz]

It is getting away from the Express VIs, but you could use the RMS PtByPt VI inside of a FOR loop.  Then it is a question of how large of an RMS window do you want.  Same idea as a moving average.

[mtan3]

Hi Lynn,

so my this is what my signal look like. I want to represent it with a linear envelope wrapping it around. But the output RMS only gives me a value each time. Is there a way to create this evelope signal with a certain sampling rate?

[mtan3]

If I want a signal with 1000 samples and 10,000 Hz, does that mean I need to run the for loop for 1000 times?