Todd Kohne responded to Jeff Smith's inquiry about the DesCAF system article
in Machine Design (May 20):
> Dr. Efrum Fudim presented this technology, coined Design-Controlled
> Automated Fabrication (DesCAF), at the National Conference on Rapid
> Prototyping in Dayton, Ohio in 1990. If memory serves me, I believe this
> conference had several presentations claiming the rights to "the first RP
> technology." I thought the concept was worth pursuing, but was unable to
> ever get samples fabricated.
> Several innovations peeked my interest:
> (please remember, SLA was still curing honeycomb structures full of resin
> at the time)
> 1) I haven't see the article you mention, but the original concept used
> conventional lamps to cure material (no lasers).
> 2) If LCD technology produced high enough resolution and power, it could
> be used to replace the photo mask creation.
> 3) Advanced designs showed curing within a vat - eliminating
> recoating/layer deposition (still a major factor in layer cycle time
> 4) Masks produced with +/- .00025 inch accuracy gave hope that parts may
> hold that also.
> 5) Irradiation time of 10 seconds per layer for any size area.
> Is the current article on-line? I am not a Machine Design subscriber.
> - Todd Kohne
The MD article describes the further evolution of Dr. Efrem Fudim's
Light Sculpting process, which he presented at the 1990 Dayton Conference.
The process generates a series of masks using a writing process, bearing
some similarity to the Cubital process. The article notes the capabilities
of laser printers, using them as the dimensional standard for the system's
ultimate resolution capability (1200 dpi in current technology). There are
several cautions that need to be exercised. 1200 dpi should be taken to
correspond to the fineness of the edge location in laser based systems, not
the dimension of the smallest feature. The mask is separated from the layer
being exposed by a layer of separating material. Potentially the
illuminating source might diffuse around the mask boundary generating an
unsharpness that will make the edge definition somewhat less than one might
presume from a 1200 dpi standard. This will need to be tested. One last
point I should mention comes from my experience making masks for printed
circuit board applications: the mask material can change dimension with
temperature and humidity. Maintaining registration to better than 25
microns (0.001") across 250mm (10") requires special material and
With regard to LCD technology - I do not conceive of LCD systems
approaching the 9600 x 9600 pixels corrsponding to 8" at 1200 dpi
represents. An array of LEDs might be more reasonable - see the article by
Professor Nakagawa and others in Rapid Protoyping Journal (I may be jumping
the gun as I do not remember if it has yet come out in print). With the
recent commercial availability of blue LEDs, this is appearing to be a
> Allan J. Lightman, Ph.D.
> Senior Research Scientist
> University of Dayton Research Institute
> 300 College Park
> Dayton, OH 45469-0150, USA
> Voice: +1-937-229-3966
> FAX: +1-937-229-3433
> E-mail: firstname.lastname@example.org
> WWW: http://www.udri.udayton.edu/rpdl
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:51:46 EEST