Hamamatsu Mini-Spectrometer

[0marta0]

Hi to everyone,

I have been trying to make a programme for the mini-spectrometer, and my main problem is when I want to change the integration time value during the adquisition of data. My idea is to change the integration time automatically when the light ammount is too high. 

When I do this, an error "22" appears, that means "it was not possible to read new data". 

Somebody knows why this happen?

Thanks

[Lightinst]

A return of "22" means the spectrometer is telling you that it has not acquired new spectral data.  As you are changing the integration time, it probably needs more time.  Create a loop to check the return until something besides "22" is returned.  A return of "0" means the spectrometer has new spectral data.

[0marta0]

You are right, that was the problem. I realized shortly after making the question. Thank you very much!

[mcdjl]

I'm trying to do some signal maths based on the spectal response butu as the wavelengths overlap slightly/get blinded by other signals i need to do some signal decomposition. Using the example vis I can't find a way to do this. Is it possible using the dll outputs as shown?

[0marta0]

You can use labview fitting tools, like "Gaussian Peak fit.vi" in LABVIEW 8.5.

[mcdjl]

I need to apply the deconvolution vi first to separate the peaks as they overlap, or use multiple gaussian fits which also need to overlap.  The problem with the first method is that i can't get either of the deconvolution Vis to work with the example outputs (which are an xy graph/array of values) and the number of of overlapping guassian curves/wavelengths present changes randomly and constantly.

[0marta0]

Deconvolution vis work with "Dynamic Data" input, then convert 1D array (example output) to Dynamic Data, that can be found in: Functions: signal manipulation: "To DDT".  (I am not sure this is your problem)

[mcdjl]

It helps a bit, at least i can now connect to it. However the signal i have coming in is the XY, i still don't have anything to connect to the Y, so no result.

[0marta0]

So you are trying to obtain the system efficiency (Y). But, can you explain how are you doing it, in more detail?

[mcdjl]

The more i look at it the less convinved i am that a straight deconvolution will work. My problem is explained better in this post http://forums.ni.com/t5/LabVIEW/fitting-multiple-gaussians-to-curve/m-p/1886139/highlight/false#M636... (which i made before i found this topic). Essentially i need to remove the effects of neighbouring wavelengths on each other without knowing what they are.

Thanks for the efforts and time!