Re: reprap
Perry E. Metzger
> http://www.reprap.org/
>
> This looks interesting!
>
> Version 2 (Mendel) is supposed to have the ability to make electronic
> circuits and motors.
>
> So what are the ultimate limits of this device?
If you're squirting goo out a nozzle, the limits are objects made of
materials that solidify and that have very loose mechanical tolerances
(certainly nothing in .1mm range is possible).
> Will future versions lead to Nanotechnology?
No, though it might introduce people to the general concept.
> Will my next notebook computer be made by my home RepRap?
No. The thing can't print ICs -- it can't even print things that require
high macroscopic tolerances.
> What do you guys think about this stuff?
Fun, maybe even economically significant, but not a path to
nanotechnology.
Perry
Tim Tyler
Perry E. Metzger wrote:
> If you're squirting goo out a nozzle, the limits are objects made of
> materials that solidify and that have very loose mechanical tolerances
> (certainly nothing in .1mm range is possible).
Printers print 2400 dpi fairly routinely. That's about 0.01mm.
--
__________
|im |yler http://timtyler.org/ t...@tt1lock.org Remove lock to reply.
William R. Cousert
"Perry E. Metzger" <pe...@piermont.com> wrote in message
news:mJ-dnVTJ_q5PzMLX...@supernews.com...
>> Will my next notebook computer be made by my home RepRap?
>
> No. The thing can't print ICs -- it can't even print things that require
> high macroscopic tolerances.
>
That's true of the current version, but will it always remain so?
Couldn't today's version print out a slightly better version, which in turn
would print out an even better version, repeating the process until we have
something that can print IC's?
I read an article a few years ago that predicted just that. 3D printers that
could print out notebook computers. It said the first copy would probably
cost $1 billion, but the second only $15.
Tim Tyler
William R. Cousert wrote:
> Couldn't today's version print out a slightly better version, which in turn
> would print out an even better version, repeating the process until we have
> something that can print IC's?
The proof-of-concept of self-replication that we have works along
radically different lines.
The current plan seems to be not to copy itself, but to copy its
parts - in a way that allows them to be relatively easily assembled
by a human.
It would be surprising if a machine could print out better versions
of itself. Where would the information that allowed the improvements
come from? It happens in biology - but *very* slowly. You can't
just say "like this but smaller" - things change as you get smaller.
It seems as though you would need some extra R & D as well.
Ahem A Rivet's Shot
>
> William R. Cousert wrote:
>
> > Couldn't today's version print out a slightly better version, which in
> > turn would print out an even better version, repeating the process
> > until we have something that can print IC's?
>
> The proof-of-concept of self-replication that we have works along
> radically different lines.
>
> The current plan seems to be not to copy itself, but to copy its
> parts - in a way that allows them to be relatively easily assembled
> by a human.
That sounds like a good plan - if you can replace the human with an
assembley robot, and make the parts for that two then self replication is
achieved.
> It would be surprising if a machine could print out better versions
> of itself.
I'm pretty sure we don't want self evolving machines around, the
ones we have around cause enough trouble. Machines that can replicate
themselves and build other things reliably are much more desirable. Reprap
seems a long way from that though, the present version can make all of it's
plastic parts but not the metal, electrical or electronic ones.
--
Steve O'Hara-Smith | Directable Mirror Arrays
C:>WIN | A better way to focus the sun
The computer obeys and wins. | licences available see
You lose and Bill collects. | http://www.sohara.org/
William R. Cousert
"Tim Tyler" <seem...@googlemail.com> wrote in message
news:DsOdnZChJ_WWYfzX...@supernews.com...
> It would be surprising if a machine could print out better versions
> of itself. Where would the information that allowed the improvements
> come from? It happens in biology - but *very* slowly.
The designers would provide the information. I'm not suggesting that the
units would evolve on their own.
Could a RepRap print head print a second generation print head that was
capable of printing slightly small parts, and so on?
Ahem A Rivet's Shot
On Sun, 19 Jul 2009 13:20:33 -0500
"William R. Cousert" <wrco...@NOSPAMyahoo.com> wrote:
> Could a RepRap print head print a second generation print head that was
> capable of printing slightly small parts, and so on?
A RepRap print head cannot print a RepRap print head.
Tim Tyler
Ahem A Rivet's Shot wrote:
> Tim Tyler <seem...@googlemail.com> wrote:
>> The proof-of-concept of self-replication that we have works along
>> radically different lines.
>>
>> The current plan seems to be not to copy itself, but to copy its
>> parts - in a way that allows them to be relatively easily assembled
>> by a human.
>
> That sounds like a good plan - if you can replace the human with an
> assembley robot, and make the parts for that two then self replication is
> achieved.
Kind of - if you provided an environment with free power and enough
processed raw materials of the right kind. However building
autonomous robots seems quite a bit trickier than building a 3D
printer. More moving parts, more sensors, more complex software
needed, etc.
Tim Tyler
William R. Cousert wrote:
> Could a RepRap print head print a second generation print head that was
> capable of printing slightly small parts, and so on?
The Waldo plan - e.g. see: http://www.foresight.org/nanodot/?p=3152
It seems like a tricky plan - *normally* components have a resolution
which is worse than the components that produced them.
For example, if you use a straight edge to make another straight edge,
the second straight edge is *ususally* worse.
That is not to say that you can't use the straight edges you have to make
better and better straight edges - just that it's not as simple as all that.
Tim Tyler
Perry E. Metzger wrote or quoted:
>> What do you guys think about this stuff?
>=20
> Fun, maybe even economically significant, but not a path to
> nanotechnology.
J. Storrs Hall gives RepRap a mention here:
"Feynman=92s Path to Nanotech (part 3)"
- http://www.foresight.org/nanodot/?p=3D3160
--=20
Ahem A Rivet's Shot
However building assembly robots is a well established art
sophisticated examples of which can be seen in any car factory.
William R. Cousert
"Ahem A Rivet's Shot" <ste...@eircom.net> wrote in message
news:BsqdnRsjc_P_A_nX...@supernews.com...
>
> On Sun, 19 Jul 2009 13:20:33 -0500
> "William R. Cousert" <wrco...@NOSPAMyahoo.com> wrote:
>
>> Could a RepRap print head print a second generation print head that was
>> capable of printing slightly small parts, and so on?
>
> A RepRap print head cannot print a RepRap print head.
This might be a great subject for the next X-Prize. The goal - to build a
next generation rapid prototyping machine that is capable of producing all
of its parts and assembling them. I wonder if Paul Allen might be interested
in donating a few bucks to something like this?
Ahem A Rivet's Shot
On Thu, 23 Jul 2009 11:48:46 -0500
"William R. Cousert" <wrco...@NOSPAMyahoo.com> wrote:
>
>
>
> "Ahem A Rivet's Shot" <ste...@eircom.net> wrote in message
> news:BsqdnRsjc_P_A_nX...@supernews.com...
> >
> > On Sun, 19 Jul 2009 13:20:33 -0500
> > "William R. Cousert" <wrco...@NOSPAMyahoo.com> wrote:
> >
> >> Could a RepRap print head print a second generation print head that was
> >> capable of printing slightly small parts, and so on?
> >
> > A RepRap print head cannot print a RepRap print head.
>
> This might be a great subject for the next X-Prize. The goal - to build a
> next generation rapid prototyping machine that is capable of producing
> all of its parts and assembling them. I wonder if Paul Allen might be
> interested in donating a few bucks to something like this?
I'd settle for a set of machines that could create all the parts,
and another set machines that could do all the assembly provided that
between them they could produce all of the parts for all of the machines
and assemble them all *and* (this is important) have a proven ability to
make other things as well.
The approach is in principle simple - first make a machine that can
make one of it's components, then make a second machine (using as many of
the same components as the first as possible) to make another of the
components, continue until there are machines to make all of the components
required by all of the machines (keeping the total parts list as small as
possible will help). Then make machines using the same kit of parts to
assemble the parts makers from the parts - big benefits here to making
machines flexible enough to be used for several assembly jobs.
I've thought for a long time that this is a much more feasible
approach to nano scale fabrication than trying to design a self replicating
thing that can also do all manner of useful work.
In the days when news of this kind of work came out of Zyvex they
seemed to be headed in this general direction, but it has been a long time
indeed since I saw any news of this kind from there.