Any methods to find the envelope of this waveform?

[Rudi_N]

Franziskus,

Have you been able to dissect multiple oscillations within your signal? Please let me know what you have been able to find.

Regards,

Rudi N.

[chrisger]

hi there

is it possible just to pass the envelope to the "Exponential Fit.vi" to get the damping of the curve?

[FReich]

Hi Chris,

thank you for your reply.

Yeah it is possible to derive the damping rate out of the envelope. I will try to figure out how to.  Maybe someone knows it already and could post it here, or tell us how to get some peak points.

Thanks a million

[FReich]

Thank you Chris,

you gave me the right idea. I didn’t think about it, cos I had already a finished formula. I just thought maybe here I could find a way to get to my points and as it seems I did in a different way however. Thanks.
Its not necessary to know the peaks its much better to know the envelope.

From the envelope I select two points. The difference in my x-value is my time difference and I generate the quotient of the y values.

The damping rate is then the natural logarithm of this quotient divided through the time difference.

This applyes to all vibrations which are viscous damped.

[chrisger]

hi there

please take a look at the attachment. the example uses the "Exponential Fit.vi" to get the damping factor. the fit will lead to a more reliable value than just two data points. the example uses only a subset of the data.

[FReich]

Hi Chris,

thank you very much. Thats really a great Programm.

[laplace_transform]

Hello:

In this example, what does fArray mean? How I will determine it?

thanks

AK

[Cory_K]

---

laplace transform wrote:

Hello:

In this example, what does fArray mean? How I will determine it?

thanks

AK

---

Where in the code are you referring to?
I dont see 'fArray'.

[Mathan]

Dear Chrisger,

I tried with your example in the second post of this thread. But not able to get a clear envelope. Can anybody tell me what is the value to be given for "new value for dt" in the block diagram? Is it always constant 0.02?

I attached my waveform and the result.

Mathan

[DavidJCrawford]

Mathan,

I cant help with your hilbert transform question.

However, I have taken a very simple approach to get the upper and lower envelopes. I take an interval, say every 50 points, and find the max and min in that interval. This gives the envelopes in XY format as the timestamps are non-uniform. I extracted the first 2500 points from your example data.

I realise this might not be ideal because you may want to stay in waveform format. I hope this is better than nothing at this point.

David