Dark thought about support materials

_______________________________________________
Developers mailing list
Developers@reprap.org
[reprap.org]

> Perhaps we should put the crosshatching down first and the perimeter last?

Good suggestion. It may also help to start the cross-hatching on known structural matererial, rather than on support material. If you think of a circular table, with a small central column and a large flat top, 90%+ to 100% of the cross-hatching will start over support material.

-Sebastien

Quoting Forrest Higgs :

> I know Vik has reported this all before with Zaphod, but this has a
> new twist. When you print if the extruded thread doesn't bond to the
> previous layer you can wind up dragging the thread all over the show
> and make a real mess of things.

Even the Strat does that occasionally, creating what they call a bird's nest.

> Think about that with respect to "support" material. If you cross a
> stretch of support material and the thread bonds to more plastic on
> the other side of the stretch no problem. If you have a trace that
> you are putting down that turns a corner on support material I think
> you probably have a problem if the thread doesn't bind to some degree
> to the support material. If the thread binds to the thread beside
> it, assuming that there is one, you are probably okay. If you are
> just crossing new territory, however, there could well be problems if
> the thread can't bind to the support material itself.
>
> If we put down the perimeter of a layer first I can see there being
> real problems where the perimeter crosses support material in that
> perimeters tend to be curvy in a lot of instances.
>
> Perhaps we should put the crosshatching down first and the perimeter
> last? Are you guys already doing things that way?

We do outline first. I think you have to do that to get any decent
sort of surface finish.

What this is telling us is we have to get it right - each layer must
bind to the one underneath, and be geometrically correct.

I'm messing about with the parameters in Ed's ARNIE machine, and also
adding new ones. What this is teaching me is that getting them right
is critical to a good result. For example, as Vik pointed out aeons
ago, cooling between layers improves things a lot. When I made it
possible to infill at a different speed to the outline that made a big
improvement in quality (as did my previously making the speeds true
regardless of direction). Also, I'm in the process of making it speed
up as it goes round corners (proportional to 1 - the scalar product of
the normalised line direction vectors). Provisional results indicate
that this makes a further improvement; I'll test some more then check
that into the repository too.

All this is consistent with experience with the Strat, which tells us:
don't re-design the machine, just make the software smarter.

Yours

Adrian

[staff.bath.ac.uk]
[reprap.org]
_______________________________________________
Developers mailing list
Developers@reprap.org
[reprap.org]

On Monday 19 February 2007 16:25, Adrian Bowyer wrote:
> Quoting Forrest Higgs :
> > The stuff melts at 40-50 degrees C? What use is that going to be?
> > HDPE is coming out of the barrel of the Mk 2.1 at about 160 degrees .
> > Wouldn't the PEG just turn into a puddle when HDPE at that
> > temparature hit it? Also, I'd have to talk somebody into making it
> > into filament which means bulk orders. I've been through that
> > Chinese fire drill too recently with CAPA to want to do it again with
> > something else. I don't think my supplier would stand for being
> > asked to do that twice.
>
> Well - I'm extruding CAPA at about 65, so it could be fine for that.
> We won't know until we try it. And something that melts that low
> shouldn't be too hard to roll into a filament if need be.
>
> Yours
>
> Adrian
>

*I'll ship you some 5-6 lb of 1450MW PEG powder by Friday or so.

Dow Chemical makes PEG under the tradename Carbowax
[www.dow.com]
They've got quite a wide range of MW, but according to page 21 of the pdf, at
[www.dow.com]
even their highest MW PEG, 8000 MW, only has a melting point of ~65C.

I'm waiting for them to get back to me regarding a sample request via their
Europe/UK distributor; we'll see but I'm not optimistic.

*It took some poking around, but people sell solid PVA granules for
specialized trades like artifact conservation:
{
1kg PVA granules - USD$23.75 - UK?? but probably available.
MW - unknown.
[www.conservationresources.com]
conservationresources.com has other pva resins as well.
Note that Conservation Resources has a UK retail operation:
[www.conservationresources.com]

Polyvinyl Acetate Resins - 1 kg $29.50
[www.museumservicescorporation.com]
This is just a US-based Conservator supply company.
}

If you go looking for it, try searching for "PVA resin", also "pellets,
granules, beads, kg, etc." I've also sent information requests to a couple
of uk adhesives companies, I'll let you know if anything positive comes of
it. I imagine there are dozens of other industries using dry pva resin, but
I was starting to get burnt out googling.

It occurs to me that PVA resin may be a very ... interesting water-soluble
support material, seeing how it turns into glue when you immerse it in the
tub of hot soapy water and start scrubbing. Try not to absent-mindedly run
your hands through your hair when you do so.

*I have been thinking that food starch may be a good inert inexpensive bulking
filler for support material, and this may also boost melting temperature.

*---wild goose chases involving Polyvinyl butyral and "gelatinized starches"
(aka pasta) as support material candidates redacted, due to inanity.---

*Adrian, do you have access to mech-e material selection software like CMS or
CES? ( [www.grantadesign.com] ) Is such software useful?
KevinZWolf suggested them some time ago.

Regards,
Sebastien Bailard
_______________________________________________
Developers mailing list
Developers@reprap.org
[reprap.org]