Technical: Scaling FTIR spectroscopy results I have atechnical which I hope can be answered by someone's experience.
I'm studying absorption on a sample by FTIR spectroscopy and, to catch the signal coming out from the little guy, I'm using a MCT detector connected to a lock-in amplifier which is the SR810 here https://www.ece.cmu.edu/~mems/resources/HH1212/SR830br.pdf . I made measurements with different setups and for some reasons I had to change the sensitivity on it to catch all the signal, leadiing me with a problem of scaling at the end. 
What's coming out from the FT of the interferograms is this:
Where the green one has been made with lock-in set on 200µV sensitivity and 
the blue has been made with lock-in set on 500µV sensitivity, being the signal higher than 200µV. The point is that I would like to scale properly the two absorption peaks. 
Has someone a certain algorithm to match peaks' intensity and the sensitivity on the lock-in or I have to make the measurement (of the blue) again with the same scale of the other?
Thank you very much (:
 A: Just do them both again. This is a mentality you can take to heart. You will always be happy to have the better data, and it's never as bad retaking the data as it seems. Many times I have realized: "If I had stopped thinking about how to adjust the data and just retaken it, I would have been done by now." But maybe that's not your situation, and it really is a burden to do it again. 
Here's the thing: it's actually rather difficult to get quantitative amplitude measurements with an FTIR to high (~1%) precision. The reason is that there is long term drift. You need to first think about all of the other things that may have caused your amplitude drift other than the change in lock-in sensitivity. After all, if the lock-in is working properly, the scaling factor between those two ranges should be 1.  And in my experience, the SR810 is pretty solid.
So first, think about these:


*

*Is the lamp well warmed up and stabilized? Has it been on at its current brightness for at least 30 minutes? Or it seems that you are measuring the output of a (CO2?) laser. If so, do you have an idea of its power stability? 5-10% power drift is not unheard of for some systems, especially if it's warming up or the environment is changing. 

*Is the MCT well cooled and stabilized? Is it not running out of LN2?

*Are the samples (if you're doing a transmission measurement) in exactly the same position in the sample chamber, without clipping the edge of the beam? 

*Are the experiments performed in as quick succession as possible to minimize drift over time?


Since you are performing a step scan measurement, each scan can take awhile. Have you tried taking a few identical measurements in a row to get a sense of the stability of the system?
Once you have considered these things and determined that your system is stable enough for the precision required for your experiment, repeat the experiment using the same sensitivity (and all other parameters) for the lock-in in both measurements. Then you will have good data and know more about how much you can trust your results.
Good luck! 
