RMS calculation of FFT

[Ronman]

I am trying to calculate the RMS of an FFT and display the result. The statistics calculates the RMS of time only. I assumed that I need to use the Formula express VI, but haven't had much success.
Thanks in advance.

[ChrisJ]

Hello Ronman,
You may want to look at the FFT Spectrum.vi. It will compute the FFT and do RMS averaging for you. Also, you mentioned not having much success with the Formula Express VI. Are you getting a specific error or are you not seeing the results you expect? If you are getting an error, please post the error and error number so we can look into it. If you are not getting the result you expect what data are you passing in and what are the expected and actual values? Also, what is the formula you are using? This information will allow everyone in the forums to provide you with better answers. I hope this helps.

Best Regards,
Chris J

[Ronman]

I am not trying to do RMS averaging of the FFT, I am trying to calculate the RMS overall amplitude of the FFT. I found an RMS vi, but the values do not reflect the expected level.

[ChrisJ]

Ronman,
It is difficult for forum members to help unless we know what you are passing in, what you actually get out, and what you expect to get out. If you would rather use the Formula Express VI, again we have to know what formula you are using, what data you pass in, and what you expect to get out. Thanks!

Regards,
Chris J

[Ronman]

For the moment, let's say I have a pure sine wave, single tone, 1 volt in amplitude. I window the waveform with a hanning window, therefore the resulting peak in the fft will consist of approximately 3 lines wide. I am using the RMS vi from probability and statistics. This vi is supposed to return the rms calculation from any number of elements. It squares each element, adds them together and then takes the square root. For an fft with a single tone 3 lines wide, the calculated RMS should be approximately equal to the peak value. My RMS answer is 0.0271 on a 1 volt peak in the fft!

[Ronman]

Looks like I figured it out. I don't know why, but I connected the RMS vi to the time domain, just before the FFT express vi, multiplied it by 1.414 (I want my overall answer in PEAK)and the answer looks very believable. I am in the process of confirming the results.
Thanks