Paul's right about not reproducing.  I have tied a quarter to a string 
and lowered it to lay flat on the the center of the lower collimating 
mirror in an Optibeam 7550.  Then I printed a 5/5 board as usual.  After 
develop there was no evidence of the shadow from the coin.  Spots on the 
mirror do not reproduce.  Dirty mirrors just reduce the overall intensity.
I have found that in collimated systems the closer to the lamp a 
reflector is the more critical is it's cleanliness.  Just a slight haze on
your collector will cut the output as much as 20%.  It also tends to 
cause the lamp output to degrade faster.  My theory on that is the 
contaminant does not allow heat to be feflected away from the lamp as well 
and causes higher lamp temps.

---------------------------------------------
While cruising the Internet, be sure to visit
our home page at  http://www.automata.com
---------------------------------------------

On 27 Apr 1996, Paul Waldner wrote:

> The optics of non-collimated exposure machines are much less sensitive than the
> optics of collimated machines, though it is not true that scratches on the
> optics of collimated machines will reproduce on the panel.  Damage on reflectors
> in general (on both collimated and non-collimated systems) normally result in
> loss of energy at the exposure plane.  Our technicians are taught to clean
> optics in a three-stage method.  First try blowing the optics clean with
> nitrogen gas under some pressure.  If the contamination remains, then try a
> rinse comprised of 50% isopropyl alcohol and 50% de-ionized water.  If the
> reflector requires still more cleaning then soak a lint free, soft cloth with
> the above mixture and wipe the optics with a very gentle pressure, making sure
> that the cloth/reflector contact is well "lubricated" with the liquid mixture.
> Any further attempts to clean will probably damage the coating on the optics and
> defeat the purpose of the cleaning.
> 
> Paul Waldner
> Multiline International Europa
> [log in to unmask]
> 
>