Resonance Dwell - 90 degree phase shift??

[davidd31415]

I've been looking at Vibration Research's "Resonance Dwell" feature. It keeps the phase shift between control and measurement channel at +/- PI/2. How is a 90 degree phase shift related to resonance? Many times I see acceleration peaks with little or less than 90 degrees of phase shift, would these still be considered resonant frequencies?

Thanks,

David.

 

[GregLocock]

In a single degree of freedom model, with light damping, the maximum amplitude of response is at, or very close to, a phase shift of 90 degrees.

In a lightly damped linear structure with a low modal density you will often see behaviour approximating to this as well. That is, if you look at the driving point FRF the phase swings from zero through 90 to 180 or vice versa during each successive resonance.

On real engineering structures in their operating range you may see this behaviour, or you may not. I typically don't. Even if the phase stays in 0 to 180 range (hurrah) it is rare for modes to be sufficiently decoupled to manage the full swing, and to be centred on 90 degrees. Look at the Nyquist plot and things become much clearer.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[davidd31415]

I will see if I can produce Nyquist plots with the software I am using. Are graphs of acceleration vs. frequency and phase vs. frequency sufficient to determine where the resonant frequency of a part is or if a part is in resonance?

For the parts that I am vibrating, high peaks in acceleration occur as the phase begins to shift from 0. By the time the phase reaches 90 degrees the acceleration is lower than the peak. Would the acceleration peak, the 90 degree phase shift, or neither be considered a resonant frequency?

Thank you,

David

 

[GregLocock]

"Are graphs of acceleration vs. frequency and phase vs. frequency sufficient to determine where the resonant frequency of a part is or if a part is in resonance?
"

yes

"For the parts that I am vibrating, high peaks in acceleration occur as the phase begins to shift from 0. By the time the phase reaches 90 degrees the acceleration is lower than the peak. Would the acceleration peak, the 90 degree phase shift, or neither be considered a resonant frequency?"

Practically speaking, I'd use max magnitude. Theoretically that is wrong. It's a lot easier to explain on the Nyquist plot.

http://www.modalshop.com/techlibrary/Fundamentals of Modal Testing.pdf

fig 3.1

From this you can see that the resonant frequency of each mode, using the fancy-pants definition that it is the point of greatest spacing of the points on the circle, is not at 90 degrees (the imaginary axis), and is not at the point of maximum amplitude (quite). This is good data by the way, real data off cars for example is much messier than this.

If you haven't seen circle fits before you may be a bit lost.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.