Time Domain Test Data Interpretation in Frequency Domain!

[struclearner]

When a certain value of test data repeats or reoccur during the test, in Frequency domain, how this will be stated.
Frequency is usually referred as Number of Time/Unit Time, but the value occurrence in Time domain is not per unit time it is during the test whole time.
For example, in strain time history test data a certain value of strain occurs 50 times over the whole test time, in a Frequency domain, how it will be stated, 50 hertz mean 50 cycles/second, but this value occurrence is not in one second.
Thanks for your explanation and referring to relevant material/article/book for reading and understanding.

 

[Heaviside1925]

OP,
I like MIT's open course work. May be better ones out there for specific instances but this is where I go for a broad understanding.

https://ocw.mit.edu/courses/2-161-s...ous-and-discrete-fall-2008/resources/fourier/

What you are trying to do is not a simple conversion.

a certain value of strain occurs 50 times over the whole test time

In time domain, all you know is a signal of a certain amplitude is occurring 50 time during a certain time period. Was it periodic? Did it value occur for the same duration at the same specified interval? If it is and it did, then the frequency would be whatever that repetition rate is per second at whatever amplitude. If not, without more information, there is no way of knowing the frequency or frequencies without performing a transformation like shown in the link above.

 

[GregLocock]

I suggest you synthesise data showing your 50 events, and another with say just 20 and see what the difference in the spectrum is. I think you'll be disappointed.

Cheers

Greg Locock


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

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

 

[electricpete]

For example, in strain time history test data a certain value of strain occurs 50 times over the whole test time, in a Frequency domain, how it will be stated, 50 hertz mean 50 cycles/second, but this value occurrence is not in one second.

Ok, so let's say your test duration is 10 seconds. Then you can say the event "occurred 50 times in 10 seconds, which is an average rate of 5 events per second"

I'm not trying to be a wise guy, just proposing what seems reasonable to me. But it's not clear to me what you are trying to achieve. I have a suspicion frequency domain is not what you're looking for. If average rate of some event occurring over time is not what you're looking for, then maybe you can try to build some statistical description of the signal.