I was experimenting with changes in the number of sweeps per cycle for a regular spectrum analysis. I ran successive sequences of the same solution, changing the number of sweeps only. I performed repeats in a randomised order.
As you can see from the data below, there is a relationship between intensity and sweeps, which I did not expect.
The only explanation I can think of for this is that the integration time does not accurately reflect how long each element is counted and the change in sweeps alters the real counting time on each element. The software would then take the total counts and divide it by the inaccurate integration time set in the software leading to differences in CPS with different sweeps.
Has anyone else found this and have an explanation?