"The APSD must be approximately uniform over the half power bandwidth of each oscillator." 1

[CarForumPoster]

Hi, I am reading this paper:

http://www.vibrationdata.com/tutorials2/FDS_FDET_McNeill.pdf

In the intro it says "The APSD must be approximately uniform over the half power bandwidth of each oscillator."

What is the significance of the Acceleration PSD being "uniform" over the HPBW? What does this mean exactly and what should I do to ensure this is true? What are the effects of it not being true?

 

[GregLocock]

It means no big spikes or dips. The more you have,the less accurate the result

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[amanuensis]

You have to have in mind the plot of the FRF of an oscillator. There are three areas : spring, damper and mass.
The damper area is the most energetic and the most "narrow band".
So suppose you excite the oscillator in the whole frequency range (white noise) in a first time, next you excite the oscillator with a white noise filtered in the damper area, i.e. in the half power bandwith in a second time.

In the both cases, the vibration energy is quite the same.
So you can neglect spring and mass effects as indicated by eq. 15.

If you don't respect this assumption, then you can't writte the eq. 15.

 

[CarForumPoster]

Thank you both, that is very helpful.

I am interested in calculating fatigue damage spectrums for base excitations where this is not the case. E.G. broadband random with narrowband spikes) The powerpoint here provides a general equation that does not make this assumption.

https://vibrationdata.wordpress.com...sponse-to-base-input-in-the-frequency-domain/