spectrometer uncertainty and intergration time

Hello,

I have two questions and was wondering if any of you guys had some input on them.

I am doing some experiments with the Jobin Yvon 750S spectrometer and measuring some spectral lines produced by different isotopes of hydrogen. The resolution on this machine is not that great, so the instrument itself is causing most of the broadening in the spectral peaks. I am trying to determine the best way to figure out the uncertainty in the spectral peak locations. What I was going to do was use the HWHM of the spectral peaks as the uncertainity. Why? Well I figured since the spectral peaks kind of follow the trend of a Gaussian distribution that the HWHM would be a good estimate since the HWHM falls roughly around the 68% confidence limit. Is this "logic" correct? What is your guys input?


Also, I was kind of wondering how integration time effects things. Does integration time effect each "bin" according the rules of the poisson statistics (ie the error each bin has is the square root of the value of the bin). Is this thinking correct?


Thanks guys.

Carlos R.
.