I tried last night 4 second exposures at ISO 400, 300mm focal length and f/5. I took 10 shots tried a few ways of stacking them including deepskystacker, rotn’n’stack, layers in paint shop pro, and a home-brew python/opencv program that just added the pixel values. The best result I got was rot’n’stack. Four seconds is clearly pushing it, the stars show definite trailing but it’s not awful. I certainly see stars that didn’t show up in the ISO 6400 1/2 second images. The two below the red circle where the pinwheel should be are magnitude 8 and 9!
Unfortunately, the 4 seconds at ISO 400 is sort of a sweet-spot. The slowest speed at ISO 800 is 2 seconds which would be the same. I can try ISO 6400 1/2 second which is nominally better but that’s what i had tried the other night.
With the Canon the longest focal length is 85mm equivalent so I could go to say 15 seconds for equivalent trailing and i could use ISO 800 or more but at 80mm focal length equivalent the smudge would be pretty small and at f/5.6 the light gathering is lessened. I may try it on the next clear night along with 1/2 sec ISO 6400 on the nikon. Playing around with telescopius I think that, at 85mm on the canon, M101 would cover about 1% of the sensor which would translate to a bunch of pixels.
Also from telescopius I get the following charts for pinwheel which shows surface brightness as 23.8 mag/arcsecond. I think that’s very dim. Andromeda at 22.2 is just a bit better. Bode’s galaxy at 21.7 a bit better still.
Somebody on reddit just told me i’d need to stack hundreds of exposures to see it – that’s not going to happen!