Eypiece projection - How much ?!

[Mraster2]

Until now I have been playing (with mixed success, see the dandelion pollen topic * ) with a (Canon) DSLR body suspended over an objective, but I would like a bit more mag. !

Is there a good ref. anywhere about the measurements involved ie. size/type of eyepiece (x) and projection distance. How large can the projection be before "empty magnification" intrudes ?

I am just 'tinkering' and learning before I contemplate upgrading my microscope.

* edit to add :
viewtopic.php?f=6&t=9189&start=30#p79716

another edit :
So far I have found others tinkering :
viewtopic.php?f=9&t=2704&sid=af8e0be54a ... b84e862849
and
viewtopic.php?f=9&t=2702
which are a good start.

[c-krebs]

with a (Canon) DSLR body suspended over an objective, but I would like a bit more mag. !

Please be a little more specific about how you are currently set up. I know you said you are contemplating an upgrade, but what do you have "in hand" now, and how are you using it?

Generally there are a few ways of setting up such a camera on a microscope trinocular tube. All can provide excellent results. The goal is usually to record in the camera close to the the view seen through the eyepieces.

Afocal. You use a normal eyepiece (generally a 10X) in the trinocular tube. It is positioned so that if you focus using the viewing eyepieces, and then look through the eyepiece in the trinocular tube the focus should be the same. Your camera, with a lens attached, is then positioned above the trinocular eyepiece. The lens on the camera should be such that the recorded image roughly matched the view through the eyepieces. For an APS-C sized sensor that means a focal length of about 40-45mm. For a full-frame camera a focal length of about 55-60 is appropriate (with a 10X eyepiece). A smaller prime lens is best, try not to use a larger zoom. The entrance pupil of the lens and the exit pupil of the eyepiece should ideally coincide. (In the Canon system, the 40mm "Pancake" lens is good for APS-C. Small, thin, and optically excellent)

Projection photo-eyepiece. Here a lens is placed in the trinocular tube and the camera is positioned above it with no camera lens attached. This can be a "dedicated" photo-eyepiece such as an Olympus NFK or a Nikon CF-PL. Typically around a 1.5X-1.67X is good for APS-C and a 2.5X for full frame. You can actually use a "regular" eyepiece as a projection eyepiece by "elevating" slightly in the trinocular tube. To get the ideal magnification you will need to play around with the eyepiece elevation (maybe in the range of 5-10mm) as well as the distance to the camera body.

Direct projection. In this case the intermediate image formed by the objective (or objective+tube lens for infinity systems) is placed directly onto the camera sensor. This is often physically difficult or impossible due to the bulk and depth of a camera body, but with certain microscopes is is easy to accomplish. This can give superbly sharp images, but you need to remember that best quality in the intermediate image is typically an 18-26.5mm circle (depends on the objective). An APS-C sensor has a diagonal of roughly 27-28mm. So this technique is usually not great for full-frame sensors, works well with APS-C (if your objectives have a wide enough field), and is quite nice with MFT (micro 4/3).

Except for "direct projection" you are already increasing the magnification on sensor with an APC-C or full-frame sensor. For imaging there is little if anything, to be gained by magnifying it more. For viewing, if you have a quality high NA Apo or fluorite you can often use with 12.5X or 15X eyepieces (maybe!) and discern more detail. But at these high magnifications (40X objective and up) you are already oversampling a considerable amount, and there is little to be gained trying to further increase the magnification on the camera sensor.

[Mraster2]

Thanks for your interesting reply, lots to think over there,

"Direct projection",is the description I should have used of my present method, instead of my rather poor "DSLR body suspended over an objective"

It is the "Projection photo-eyepiece" method that I am wondering about.
Approximate ball-park projection scales** (the closer or further away the eyepiece is from the primary image (and consequently its distance to the camera) and effect upon the image quality in the camera is the sort of thing I hope to get my brain cell round.

** I think you are telling me that I should aim to put the whole of the (18-26.5mm circle ) objective image on my sensor - and not to enlarge just a portion of it onto the sensor ??

Now to google "oversampling" !! >