LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

FFT analysis of a spectrometer signal

Hi Lynn,

 

I have a thin PET film with a thin layer of Dielectric coating.

My aim is to find the peaks along with thickness values for the dielectric material and the PET film using FFT.

Thickness computation is being combined with the fft output to get the peaks. 

I am trying to find out the FFT for the Intensity values.

I am able to do the FFT analysis with two peaks for Dielectric material and PET Film, but the resolution is very high.

The problem is the input data set is being obtained by calling dll provided by Ocean optics. I am afraid if the input data set range is defined in the dll being called. I am unable to increase the size of the data set for high resolution. 

 

 

 

 

Regards,
KM
0 Kudos
Message 11 of 13
(307 Views)

Kamasani,

 

i think it is not clear yet how you will get the information about the thickness of the film by a "simple" FFT.

 

Big questions:

- How can you make sure that your light source has the same intensity for each frequency?

- How can you make sure that the sensor detects the correct intensity for each frequency? is it "linear"? Do you have a calibration?

- The result of the measurement is a "frequency/intensity" table. Obviously, the intensity is a "function" of "frequency". But how can you extract the information of "thickness" from that function?

- How does a FFT fit into the determination of thickness, aka the parameters of your f(frequency) = parameter * intensity?

 

Keep in mind that the film can have non-linear absorption of frequencies (which are of course additionally affected by the film's thickness!).

All these parts come together to only one option (imho):

- Use a set of "golden samples" (films where you KNOW the thickness) and create a set of "calibration/reference information".

- Compare all future measurements with these samples.

 

Norbert

Norbert
----------------------------------------------------------------------------------------------------
CEO: What exactly is stopping us from doing this?
Expert: Geometry
Marketing Manager: Just ignore it.
Message 12 of 13
(303 Views)

Kamasani,

 

and it must be noted that regardless which way your procedure works, in order to increase the resolution of an FFT you have to increase the record length of the input. No way around that.

 

Cheers

Edgar

Message 13 of 13
(299 Views)