If no light falls onto the detector then there is no blooming. And of course there will be diffraction from the shutter edges as well, so engineering it will be

If no light falls onto the detector then there is no blooming. And of course there will be diffraction from the shutter edges as well, so engineering it will be quite a challenge (one can take hints from coronographs). If the satellites does end up on the sensor then the dimension of the streak depends on the optics, how bright it is, the used detector, plate scale, etc. (if you are extremely unlucky you might have internal reflections too and they might end up who knows where on the detector - optics is hard).