Anyone tried eyepiece projection for EAA?

[vlaiv]

In regard to longer focal length EPs and exit pupil - I suspect, but I need to think about that some more, that longer FL EP will be better in terms of correction.

It would be somewhat similar to stopping down achromat scope - if you just use central part of the lens you will be reducing aberrations from strongly curved edges. Exit pupil does not really apply in this case in traditional sense, since rays exiting eyepiece are not parallel like in normal viewing situation. I think couple of diagrams are in order to explain different configurations.

This is how focusing to a distance works. Blue rays are focus at infinity (this is how telescope works), incoming rays are parallel and they bend and reach focus at focal point of the lens. If you have an object that is closer to infinity (red lines), rays come in diverging and are again bent, but this time they converge behind actual focal point of the lens (you need to put sensor behind actual focal point of EP). If you move object even closer, you reach point where lens acts as relay lens. This point is at twice focal length of lens - green rays - and they act symmetrically coming in at the same angle as they come out and they converge at the same distance as object. Image in this case is 1:1, or just reimaged at different place.

Eyepiece for observation works in "blue" mode, but reversed (look diagram from right to left). Rays from scope are converged to focal point and continue to diverge and hit eyepiece and then are "straightened" or "parallelized" (collimated) when they exit EP. They reach your eye same as would rays from very distant object naturally.

Exit pupil is spread of blue lines on left side. It makes sense only when you have parallel rays coming out of EP.

Here is another diagram, for EP projection:

This is diagram for proper EP projection scenario. Light comes in converging from scope, and light has certain "shape" of the cone - which depends on the speed of the scope. For given green marks, F/ratio of the scope will determine ratio of vertical green line to horizontal one. We can say that "Exit pupil" in this case is area covered by diverging light beam when bent inside lens. This will depend on distance of focal plane of the scope to center of the lens and F/ratio of incoming beam. This is related to vignetting as well, above scenario shows point at optical axis being projected to sensor - exact center. When we have point that is a bit shifted (at some angle to optical axis) then central line moves (up/down) but so does whole beam. Due to its "thickness" at some point part of it will land on field stop - this is when vignetting starts, and some of light rays will be blocked - this is why vignetting is gradual and not hard stop.

If we keep beam narrow (this is for slow scopes) then we will be using only central part of the lens. Alternatively using larger lens - EP with longer FL and wider field stop  will have the same effect.

With regards to coma - coma is progressive and on second EP projected image it is compounded by other type of blur. There is coma in the first image as well, but it only starts to show.

Here is spot diagram for F/5 newtonian (look at central column). At twice the distance from optical axis - coma will be twice as strong.

There is another matter that might be at play there, since you say that you have regular focuser. This kind of setup is probably very long and can bend somewhat. If it does bend / tilt, optical axis will no longer be at the center of the sensor, and one side of the image will display field even further from the optical axis - stronger coma. It is hard to tell from the image because there are no obvious shapes in upper part - but it looks like there is shift of the optical axis - bottom part of the image looks like it's been affected by coma more.

Do you have brick wall that you can image? Something with regular pattern over whole field? That will show distortions better over whole field.