durusmail: mems-talk: Re: Economical Mask-Making Technique Using Desk-Top Publishing
Re: Economical Mask-Making Technique Using Desk-Top Publishing
Re: Economical Mask-Making Technique Using Desk-Top Publishing
Christopher G. Levey
1995-06-19
...
> Typically any popular DTP uses 2540 DPI, there are machines
> that produce 3386 DPI and there is one in the market that offers
>>4000DPI resolution.
...
 Ash
We see little improvement in using 3386 vs 2540 DPI (not the 30% one might
expect). My understanding is that the laser spot size used in a typical
Linotronics is ~20 microns regardless of the DPI used. Higher DPI simply writes
on a finer grid. The increased overlap in laser spots results in slightly
sharper edges.

An additional limitation we have found with this method is that defects in the
clear portion of the film can scatter or absorb enough UV to show up on the
resulting chrome mask. (For example, yield would be low for a large
interdigitated electrode pattern with a gap less than about 50 microns.)

Also, although one can use the Lino film to print directly onto wafers, it is
important to first transfer the pattern to a glass/chrome mask if more than one
mask level is required; alignment is difficult when using the film directly.

Dartmouth students have used direct printer output from several CAD packages
and also the cif to true scale postscript converter from Ash's group (see the
original posting of this thread).

--Chris Levey
  Solid State Microengineering Lab
  Thayer School, Dartmouth College
  email: chris.levey@Dartmouth.EDU
  http://hypatia.dartmouth.edu/levey.html


reply