G Unit To PSD conversion within given frequency range

[yusf]

Hi all;
I am given a vibration profile that is stated in G unit between some certain frequency values I wonder if anyone knows conversion and theory behind it to extract the vibration profile in terms of PSD ..


Example:
0.01g^2/Hz 5-10 Hz
2g 10-100Hz

Thanks in advance.

 

[IRstuff]

Is the second line missing its ^2/Hz following the 2g?

I don't understand what you want, since you have the PSD right there.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[yusf]

Hi;
Unfourtunately no..The vibration profile given as a combination of PSD and i think sine sweep such that;

10-40 0.15 g^/Hz
and the rest is 40 to 500 as 5.5 G limit.

I would like to convert the second part to PSD namely g^/Hz unit so i will get full PSD vibration profile
There is a formula i found from the following link but it doesnt sense me since it only converts a sine peak within df whereas i need conversion within a broad range of frequency

http://www.mpihome.com/english/sineamplitudepsd.htm

 

[yusf]

Hi Again;

Also from the reference below i found that:
"The frequently used MIL-STD-810 standard states that a 1-to-1 conversion of a sine sweep
test to a random vibration test is not possible and that a measured vibration level is the best
basis for formulation of vibration requirements."

Ref:

http://www.sebert.nl/Brochure/Sine_sweep_vs_Random_vibration_testing.pdf

So is there an ambiquity on that given profile?

 

[IRstuff]

You should ask your customer, or whoever provided you the data. The units of your original data are specifically for PSD. Sine sweeps are generally not specified in that fashion, since a sine sweep does not have a "density" which requires a g^2/Hz density unit. Your second set of data does not have full definition of units. Is the 5.5g "5.5 g-RMS? Is it supposed to be flat? Why are the two sets of data different?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[BrianE22]

Your sine sweep calls for a constant peak acceleration (5.5 g's) for each frequency. If you have a random waveform with all frequencies present at the same time and each frequency at its peak acceleration then the the peak acceleration of the random waveform will be considerably higher than the 5.5 g's for most of the time. Is that acceptable?

 

[yusf]

Hi All
I have returned to my data provider to give me full PSD
thanks for your support