14/03/2021
Signal & Guideposition
How does the precise position of a star on a slit does influence the signal.
In order to improve the observation experience we raise the question: how does the precise position of a star on the slit influence the signal of the spectra on camera sensor.
In best situation this should never be a question, because we like to achieve to get all light of the star, to be observed, through the split and finaly end up on the camera sensor. In principle, this can be achieved by selecting a slit which is simply wide enough to cover the whole star. This clearly influcences the resolution of the final spectrum. The broader the slit has to be selected, the lower the resolution will be.
Sometimes, it's required to achieve a certain resolution to make some spectral features visible. This has a direct constraint on the slit to be used and might end up in the situation that the slit will be smaller than the disc of the star at focus driven by the atmoshperic seeing conditions. Therefore it's usefull to know how does error in posittion of the star relative to the slit influcences the final signal in the spectrum.
The answer to this question depends for sure on the instrumental parameter of the telescope used and in principle can be also derived theoretically modelling the star as a certain functionand caluclate the throughput defined by the slit in use. We decided to do it simply experimentally by observing a star and modify the position the telescope is guiding to.
We selected Alpha Leonis using LHiRes III with a 2400lines/mm grid, 23mue slit and ST8 camera in binningmode 4 to lower exposure time. As telescope the 20Inch Newton in primary focus has been used. So we took several spectra and track the maximum ADU at a certain position in the spectrum. Than, we varried the position the telescope guide to by single pixel steps and checkout the signal in the spectrum as function of this position. The outcome of this procudure is shown in figure 1 below.
We have, as expected, strong increase of the signal the more of the starlight is positioned on the slit. We expect, that the rate of decrease by position error is a function of the seeing. However we learn that in this particular setup an error in position of even 1 pixel only results in a 20 percent loss and corresponding longer exposure time to achieve the same S/N compared to best position.
More images