Phase accuracy for 1x operating speed vibration

[Coryman]

Hello,

I'm attempting to finish a dynamic balancing routine which will utilize signals from accelerometers and proximity probes.  The vi will then perform a least squares fit on on signals.  The prox probe outputs are displacement, mils p-p.  In order to get mils p-p from the acceleration sensors, I will need to double integrate the signals.  My question very basic and I don't see any issues, but it can't be this easy.....

Can I calculate the 1x phase in the acceleration fft and then flip it 180 degrees and output that as the displacement 1x phase?  

Regards,

Ron Brook

[dsb@NI]

Your proposed solution will give you the phase of the 1X component. Just like the phase of the 1X acceleration component, this will be the phase relative to the start of the analysis block. For dynamic balancing, you probably want to know phase relative to a physical angle reference on the shaft/rotor/flywheel. 

I encourage you to evaluate the Sound and Vibration Toolkit which ships with a sample project: Single-Plane Balancing (DAQmx). At the least, you could compare the (balanced) results between the two implementations.