>On Thu, 30 Mar 1995, Chuck Grant wrote:
>
>> > Note that to 'appear' black doen't necessarily mean it must be painted
>>black.
>> > For instance the reflections from the lower section of tube in (3) are
>> > returned back along a 'glancing angle' and in this case a *smooth, gloss*
>> > surface offers less reflection than a roughened one. Of course the smooth
>>
>> Excuse me, but simple experiments seem to show clearly that
>> smooth gloss surface + glancing angle = really good mirror.
>> This would seem to be to be a very serious mistake.
>
>I'm not exactly sure what the original poster had in mind, but some
>baffle surfaces should be left glossy, though still as black as
>possible. It depends on where the residual light will end up. The
>advantage of a specular baffle is that if you know where the light is
>coming from, you can predict where it will go, or even better, direct it
>someplace harmless. A matte surface, on the other hand, scatters
>whatever light does reflect off it randomly.
Thanks Sam for explaining my point better than I did in the original posting.
What I was trying to say was..
The walls of the tube are being illuminated at a glancing angle
The view of the lower section of tube through the secondary is *back* along the incident direction not in the reflected direction which was Chucks concern.
This view of the tube walls thru the diagonal occupies a larger angular area than stuff on the tube end(like dust on the mirror or mirror cell) which might scatter the light from the glossy reflection towards the image.
Until you can make your roughened walls as 'dark' as the dust on your mirror you may as well leave the lower part of the tube glossy.
Hmm, I'd better check how dusty my mirror really is (:-)>
JohnH