Layer-by-layer annealing?

I'm still working on building my first 3D printer, but I am a little perplexed by the heated build platform and cooling fan issues.

It seems that the common way to prevent warping is to use a heated table (or a heated build chamber) to keep the part in a constantly relaxed state just near the softening point of the plastic. It strikes me that this is a little at odds with wanting to cool the extruded filament quickly to get sharp prints. I'm thinking of something where after each layer is printed, there is an annealing step where the part is scanned underneath a hot wire or heat lamp to anneal the top layer to eliminate the stresses between it and the layers below. Because you are only heating & cooling the top layer, you should be able to do it pretty quickly. Otherwise the bulk of the part is kept cool rather than hot during printing. The idea is that doing the annealing in a separate step might be more forgiving than trying to find just the right set parameters to get filament extruded and cooled at just the right temps and rates.

I know that the heated table is also used for adhesion, but adhesion would be less of a problem if the part weren't trying to warp.

Has anyone tried anything like this, or is warping pretty much a solved problem with the heated table?

Heated beds are to help ABS to stay stuck down to the build platform. True this does also help prevent warping. Cooling with a fan is for PLA not ABS.

I have problems (ABS, heated bed, heated chamber) with tall thin items cracking (delaminating) in the middle of a print.

I suppose strength and quality might be better if a heat source was to follow the printing head somehow. This might also increase strength but it would have to be controlled as if too hot it could cause sagging in the entire print.

I don't have any warping anymore, just delaminating on tall thin walled objects. Even though my machine is surrounded in plastic sheeting and it gets rather warm inside while printing. Maybe it's still no enough heat though.

I'd guess that delamination in the middle of a print is resulting from the same internal stresses that cause warping, or that also cause parts to want to peel off the table. So yeah, maybe you don't a have a high enough temp. in your build chamber. (Also could be that the layers aren't stuck together well enough to begin with.)

I Guess my idea is is that rather than trying to keep everything hot enough all the time, after each layer is printed, you turn on a linear heat source, like a hot wire (that moves with the print head), and scan it over the entire layer. Then let it cool slowly for a few seconds to let the material relax in place, and then do the next layer.

The problem is that as that layer cools it contracts, pulling on the layers below and causing warping. To reduce the stress the entire part would have to be warmed and cooled all at once.

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

I run a fan on abs as well, the bed is basically just to make it stick, it really has nothing to do with warping. the fan really helps with making sharp corners and overhangs. without a fan on the print the over hangs never come out correctly, and the corners actually start warping upwards without the fan. I suppose a heated build chamber would "anneal" the abs to make the whole piece the same temperature and reduce warping, but you would still need a fan to blow some air on the fresh plastic to make the features sharp.

NewPerfection Wrote:
-------------------------------------------------------
> The problem is that as that layer cools it
> contracts, pulling on the layers below and causing
> warping. To reduce the stress the entire part
> would have to be warmed and cooled all at once.

I'm not entirely sure about that. Let's say you've managed to build N layers that are stress relieved. You then add one layer which has cooled quickly (to get sharp features) so it ends up pulling inward on the lower layers. If you then just anneal that top layer, it should relax to conform to the lower layers. If this is done layer-by-layer, you never have to heat the whole part up at once to anneal it. The trick would be to anneal the top layer without it getting droopy.

Keep in mind, I'm just blowing smoke because I've never actually printed anything (yet).

I think polymers have some unique challenges and are very different than metals. I was doing an experiment recently with some hot melt glue that is used on polyethylene. I heated some antifreeze up to the melting point of the glue. It melted and spread in a layer over the surface. I wondered if that temperature would melt polyethylene so I tossed a piece of LDPE plastic tubing 1/4" diameter x 3" long into the hot antifreeze.

The result was completely unexpected. The tube shrank in length to just 3/4" with a diameter of about 1/2" and a wall thickness of about 1/8"! I suspect the polymer molecules are stretched out when extruded, and heating them to just below the melting point lets them snap back. So shrinkage of a printed part could potentially be far greater than one might expect. You might actually have to hold the material in place during cooling to prevent the shrinkage.

Gary H. Lucas