DoN. Nichols wrote:

snip

Hmm ... at what angle do you get total internal reflection?
IIRC it is a function of the index of refraction. And while binocular
prisms get TIR at 45 degrees, the light enters and exits at close to 90
degrees, thanks to the design of the prism. If you have the light
entering at 45 degrees on one face, it will be bent towards the other
face so the angle will be steeper than 45 degrees.

And -- *if* you got TIR with the glass (say you made a full
45-45-90 prism to get the entrance angles right) you would get nothing
through to your eyepiece.

So I think that the semi-silvered is the better choice.

Hmm ... the light is not focused where it passes through the
mirror, so you might be able to take a first-surface mirror and scribe
through the silvering in a pattern to get something like 30-45%
reflection, and then you might be able to use it. I've seen scribed
patterns in mirrors in SLR cameras, with the CDS cell for metering
behind the scribed section.

41.8 degrees for light travelling from glass into air. This would mean
that the mirror would need to be at an angle of less than 41.8 degrees
to the horizontal.

The only disadvantage as far as I can see is that if the light emerging
from the lamphouse is horizontal, it won't be travelling quite
vertically downwards after being reflected. But you could correct this
by tilting the lamphouse and its lenses down slightly.

Best wishes,

Chris