>Subject: Re: How do I turn my DeltaCad design into reality? Machine cutt
>Sent: 22/9/20 9:05 am
>Received: 19/9/00 1:35 am
>From: Bill Gottesman, [EMAIL PROTECTED]
>To: [EMAIL PROTECTED]
> Sundial Mail List, [email protected]
Steve +
Photo-etching has been my preferred method of translating ideas into
metal for some thirty years now with the addition of CNC engraving in the
last three years. Based on that experience the following thoughts may be
of use to others.
Photoetching is WYSIWIG ('What you see is what you get') and can handle
small dials up to about 12 inches with very fine detail using 'pie-dish'
methods but it's not a process which works 'straight out of the box' and
you may have to invest some time any money perfecting your skills.
Beyond that size it may be wise to use companies equipped for the
purpose. Take care to discuss etching depth or you may get something
which lloks good on completion but will weather away in just a few years
outdoors.
Your chosen photoetching company may be able to work directly from
computer files but, if not, almost any graphic arts agency will convert
Adobe Illustrator files to film masters.
Check to see if your photoetcher wants 'positive' or 'negative' film
masters and clarify what they mean by this. i.e. white lines on clear or
the other way round. As Bill Gottesman has already indicated,
Illustrator copes flawlessly with DeltaCad files although it was Mac
Oglesby who first revealed that possibility to me.
CNC engraving became a necessity for me because my one-at-a-time
photoetching process takes up to 12 hours and I had a contract for a
large number of brass dials in an impossibly short time. Its *apparent*
advantage was that computer files would become brass sundials within
minutes rather than hours and the files could be sent as email
attachments rather than taken or sent on floppies.
Believe me, it is neither WYSIWIG nor foolproof as six spoiled plates of
expensive 3/8" thick engraving brass can testify. The human element
enters the process in various ways beginning with the choice of engraving
tool size. Even a very small tool will have problems with serifs on Roman
numerals and the software may just dismiss them as impossible. Very
small tool tips break frequently and software set to operate in small
increments takes much longer to complete the process so the pressure is
on a commercial operator to use larger tools and shorter run times. Then
there is just plain human error like setting the datum to the bottom of
the plate instead of the top for a 'pierced-right-through' sundial!
I would NEVER trust a large piece of expensive material to CNC engraving
unless I'd seen a practice piece cut in cheap MDF etc first of all.
In future I will continue to use photoetching for small dials with fine
detail but for a large vertical dial set high up and necessitating thick
hourlines and blocky numerals then CNC would be my preferred alternative.
Although photoetching can reach half millimetre depth - which looks more
than it sounds - CNC can theoretically go well beyond that, so for a
large dial subject to heavy wear, e.g. at ground level with people
walking on the plate, then CNC again comes into its own.
Tony Moss
"Fools rush in where angels fear to tread"
P.S. I recently circulated a close-up JPEG illustrating etching depth to
my JPEG sub-list. If anyone not on the list would like a copy just let
me know.
P.P.S. Having sorted out the human error problem all of the spoiled six
plates were eventually CNC engraved satisfactorily.
