I just started using ThunderSTORM to fit and analyze data produced with a Zeiss Elyra P.1 PALM Microscope and to compare the results to the ones the Zen Black software from Zeiss produces (The Microscope does not allow imaging with a different software to my knowledge). The Microscope is equipped with an Andor DU-897D EMCCD Camera. Since I needed the electrons per A/D count together with other values, I consulted Andor, which supplied me with the Performance Sheet for our camera, which can run at 10, 5, 3 and 1 MHz and different Preamp settings.
I managed to find the settings that Zen uses, which state that the camera is used at 10 MHz with a preamp of 1. I assume it is in EM amplifier mode (refering to the performance sheet which only shows EM amplifier results for 10 MHz mode).
According to the performance sheet, I thus have a CCD Sensitivity (electrons per A/D count) of 56.4, and a Single Pixel Noise of 83.47 electrons. Furthermore, I read in a different thread that the base level is 100. The pixel size is 100 nm. In the experiment, I used a gain of 230. This is how my camera setup looks like:
However, when I run the fitting, the Photon Number is overestimated by a factor of around 20, while everything else looks fine and comparable to Zen. What am I doing wrong? Of course I played around with the Numbers, and when I divide the A/D count by 20, I get comparable results. However, I can not make sense of this. What am I missing / doing wrong?
Thanks in advance, any comment is appreciated.