Rigidly mounting an odd combination: T2i + 85L f/1.2

[ChadS]

@DrC thanks for the heads up. I was totally unaware of that lens. See I should have asked all of you first! I've got the 85 f/1.2 on my camera now and it's performing far better than I could have ever hoped so I'm happy with things so far. However, now that i know that lens exists I will have to find a use for it...

p.s. if you only illuminate only with green light and you operate at the minimum focus distance of the 85 f/1.2 the final resolving power is not limited by the lens but by the sensor itself even wide open - at least for any current-generation EOS (the 4.3 micron pixels on my T2i are about as small as they come now). I may have to go get some crappy cell phone sensors just to find where this lens falls apart. The blue light focuses reasonably well but the red isn't close. It's at least half a pixel off! *sarcasm off*

edit: dumb question, how can a lens with a 52mm thread and an 85 mm focal length have an f/1 aperture? Ah, it can't image to infinity so it doesn't need a full 85 mm clear aperture. I'll shut up now but will leave this edit in.

edit 2: Apparently it's 0.9X to 1.1X so not a very large range. That makes the overall field of view not much more than 15mm (for my case) whereas my customer has decided 75mm is more appropriate.

[ChadS]

I have a copy of the SDK. The documented commands are ... limited. However, I know some of the telescope community especially has reversed engineered the software bundled with the camera and have discovered other capabilities. Unfortunately I don't know of anyone that has figured out how to pin the shutter open.

I think I'm having a good amount of luck on mounting my rig. Based on the links provided earlier I've created a steady-rest with two rollers under the lens, a spring loaded clamp to go over and a variable riser under the back of the lens (that also supports the camera). On top of this I've added about 12 lbs. of beanbags. I can go several hundred shots and not drift a full pixel (~40 micron object distance) but I'll need to wait until I get a finer positioning stage to see how well I'm splitting the pixel distance.