Sensor tilt, adjustment

[bojan]

My modified Canon 450D suffers from mild sensor tilt problem...
On top of that, Canon 400mm f2.8 adds it's own amount, at different angle of course.
I would like to solve both problems independently... and start with camera (to eliminate one variable from the equation).

The idea is to build the optical bench for EOS camera, which would allow precise measurement of sensor corners offset relative to the sensor centre (with help of live view).
This may help with sensor tilt adjustment (by inserting thin spacers... btw, where to get them from? Are there available in various thicknesses? Did anybody tried to change the canon's method by using elastic spacers between camera frame and sensor PCB and adjusting the screws?)

I am wandering if someone tried to build something like this?

BTW, I tried CCD Inspector, but the results are inconclusive, so I gave up.

[JA]

Hello Bojan,

You could start with brass machinery shim as it is easy to cut with scissors/knife or punch and bend as required and available in 0.05mm to 0.25mm thicknesses, sometimes as as a kit or thinner & thicker individually.

Some ebay possibilities:
http://www.ebay.com.au/sch/i.html?_o...+shim&_sacat=0

You could even, if you wanted a specific, non-available size, cut the spacers/washers from a feeler gauge set, although this is often stainless steel, with some brass leaves.

What led you to be concerned with the sensor tilt- any image differences in corners? In addition to any optical comparisons, it may also be possible to also measure this mechanically using a dial indicator and finding the distance between the camera mount flange and sensor (being careful with delicate glass, of course). BTW- check that the mount flange is OK and no sign of movement/deflection. There are some excellent inside camera shots on some camera disassembly websites that may be of use to see what sort of adjustments or shims, etc... would be possible. Most sensors have a 3 point to camera chassis mounting, sometimes with thin washers. You could also carefully "machine" these washers to thickness, as required if they are between the body and sensor.

Best
JA

[JA]

Bojan,

You could also more easily shim the camera's lens mounting flange, either as temporary fix to see if it cures the issue, or maybe even permanently.

Best
JA

[bojan]

Yes, 450D sensor has 3 screws that hold it in place
And I do have a difference in focus between corners and centre of the frame - not much but enough for it to be noticeable (you know.. when you KNOW about a certain problem, it is usually perceived to be bigger that it is in reality).
Rotating the camera by 90° changes the situation significantly, but not consistently so I concluded the lens flange has it's own tilt (this came as no surprise to me because in process of conversion from FD to EOS I dismantled the back end and saw some problems... and left them for later)

[bojan]

To shim the camera flange is possible... will have a look.

I would like to avoid doing those things on the scope and in the dark... and I can't see anything suitable further away than 10 metres from where my pier is mounted...

[JA]

Originally Posted by bojan

To shim the camera flange is possible... will have a look.

It would be good as a "quick" reversible test or even possible fix

Originally Posted by bojan

I would like to avoid doing those things on the scope and in the dark... and I can't see anything suitable further away than 10 metres from where my pier is mounted...

Bojan, I'm not quite sure what you mean I wasn't referring to any outdoor tests, but possibly inside on a bench.

One thing is for sure, you need to fix it in the way you described initially- fix/adjust the camera, then fix/adjust lens separetly, BUT before ANY of that I would test the situation with different lenses to see if the issue is still present.

You mentioned your 400mm 2.8 was an FD modified to EOS mount. In modifying an FD lens to an EOS mount isn't there a lens mount flange to focal plane difference that needs to be accommodated in the modification - 1.5-2 mm or something? If that isn't perfect perhaps that causes a slight difference in the corner focus vs centre focus due to the flat film plane vs spherical surface (however slight) of the light wave, causing problems with infinity focus. Maybe that 1.5mm-2mm difference should be checked or what's the word from Bridge, ....finessed?

Best
JA

[bojan]

This lens was probably involved in a war... judging by the condition of metal parts.
The back end was wobbly, and one of the critical parts (aluminium cast, the part that had threaded holes for flange) was partly broken, but I could not make the new one for various reasons, one of them was I wanted to keep the iris mechanism in original condition and functionality). So I compromised here a bit....
Also, those lenses have back end that could be rotated (in increments of 90°), and the part that rotates was hit (or lens landed on it with it's full weight of 5.5kg), resulting in damage at the inside rim, which was visible as slight thickening at the end of the tube).
I repaired this on lathe by removing 0.3mm or so, but I am not sure the cut was done with enough precision (when I tightened the part in the chuck, I didn't remove the paint from the tube, so there was a possibility of some introduced eccentricity and/or non-orthogonality).

Originally Posted by JA

It would be good as a "quick" reversible test or even possible fix



Bojan, I'm not quite sure what you mean I wasn't referring to any outdoor tests, but possibly inside on a bench.

One thing is for sure, you need to fix it in the way you described initially- fix/adjust the camera, then fix/adjust lens separetly, BUT before ANY of that I would test the situation with different lenses to see if the issue is still present.

You mentioned your 400mm 2.8 was an FD modified to EOS mount. In modifying an FD lens to an EOS mount isn't there a lens mount flange to focal plane difference that needs to be accommodated in the modification - 1.5-2 mm or something? If that isn't perfect perhaps that causes a slight difference in the corner focus vs centre focus due to the flat film plane vs spherical surface (however slight) of the light wave, causing problems with infinity focus. Maybe that 1.5mm-2mm difference should be checked or what's the word from Bridge, ....finessed?

Best
JA