Full-Frame DSLR's and Vignetting

[michael8554]

I've had a number of astro modified Canon DSLR's over the years - 1000D, 450D, 500D, and currently 600D.
Canon improves the noise performance with each new model, but also keep shoving more and more pixels into the sensors, at the theoretical expense of sensitivity, and the real affect on imaging scale - on my scope it's 0.67 arcsecs per pixel with the 600D, which has 4.3um pixels.
So when I saw a Canon 6D on sale at an affordable price I snapped it up. 
The 6D has 6.5um pixels, and is famed for its sensitivity and low noise, but of course is Full-Frame, which usually means significant optical vignetting and mechanical cut-off on many scopes.

Here's a Flat from my 600D, converted to JPG, and a 1.5 gamma stretch applied to show up the corner vignetting. I measured the centre intensity to be 141 and the corners 126 (out of 255).

 

Here's the Full Frame 6D Flat, converted to JPG, and again a 1.5 gamma stretch applied. A fair amount of vignetting in the corners, but surprising to me, no mechanical cut-off.

 

Finally here's the area of the 6D Flat that has the same centre and corner illumination as the 600D. Not surprisingly this is the same size as the 600D sensor.

 

So although the edges and corners of the 6D Flat are significantly vignetted, I think you'll agree that there seem to be a goodly amount of useable pixels outside of the central grey area.

FOV goes up from 58 x 39 arcmins to 81 x 55 arcmins.
Imaging scale is now 1.05 arcsecs per pixel.

At this stage I'm waiting for the battery eliminator to arrive, so real-world values may not be so rosey.......... ?

Michael

[StuartJPP]

I much prefer my Canon 6D to my Canon 650D and I am sure that you will as well. My imaging lens, the Canon 500mm f/4 L IS, is designed for full frame but there is still of course vignetting but to be honest I just ignore it and throw everything into the pot along with flats and just crop off the bits that I don't want. This is usually the edges caused by dithering.

Processing images from the 6D is so much easier and if you get enough subs hardly any noise reduction is necessary, depending on how hard you stretch of course...the same could be said of the 650D, but you'd need a lot more subs to begin with.

As a consequence I hardly ever use my 650D any more for imaging and if the next season is poor it will be sold off to some lucky recipient for a steal

 

[ollypenrice]

You're talking about field illumination here. That's not the only issue, though so far so good. Indeed very good. Stars are abominably exacting targets for any optics, so how will they be near the edge of field? You don't mention your scope. Mine were chosen to cover full frame (Tak FSQ106N and TEC140 with flattener.) Both give clean stars to the edge, but the Taks (note old models) are probably less evenly illuminated than your system based on your flats. However, their significantly uneven illumination corrects with flats. The TEC is as flat as a Dutch airfield and about as large...

Olly

[michael8554]

Well despite the full moon I couldn't resist a try-out with the battery,

Results were pretty much what youse guys predicted - good news and not so good news.

Flats corrected the dip in Field Illumination ((like that term, must remember it Olly), but the coma inherent in my native F6.3 SCT didn't  magically improve outside the "old" FOV.

So like Stuart said, it's the usual judgement of what to keep and what to throw away.

Michael