crazy idea for stopping ABS warping

There is no new idea under the sun, it seems, so this has probably already been discussed.

It's somewhat inconvenient but it occurred to me if I could immerse the bulk of my ABS part in a high-temp bath of a suitable liquid (silicone oil?) while am printing then there should be no warp at all.
The main issues are trapped liquid in parts with internal cavities, material compatibility, adhesion loss if any of the liquid gets on printable surfaces, and maintaining the level of liquid to just below the active printing level. Accidental immersion of the extruder in the liquid is another concern.
Hmm, this would require quite a bit of redesign, although completely warp-free parts might be worth it.
I recently tried to print a largeish open-topped ABS box with a tall skirt and it warped so badly it is now on my list of holy-grails.

I have been experimenting with DAP contact cement thinned with xylene. It had really promising results even on a cold bed. Still more testing but it would be nice to have others validate either the idea or my insanity

Exactly how does immersing the print in a 70c vat of silicone oil reduce warp more than printing in a 70c heated chamber? Why would printing into a liquid reduce warping?

Quote
691175002
Exactly how does immersing the print in a 70c vat of silicone oil reduce warp more than printing in a 70c heated chamber? Why would printing into a liquid reduce warping?

Yeah, I couldn't wrap my head around that either. At some point you have to address the deficit in "effort vs outcome" on printing a yoda head with that setup..

Edited 1 time(s). Last edit at 06/22/2015 03:26PM by thetazzbot.

I have found that 45C build chamber temperature prevents delamination for at least medium sized prints. Just enclosing the machine gets it up to that temperature using the heat of the bed and extruder nozzle only. I moved all the electronics into a drawer in the bottom of my printer to keep it out of the heat, and kept the Y axis motor outside the build chamber because it runs pretty hot.

In this photo, the larger object was printed inside a closed chamber at about 45C, the smaller object without an enclosure. Both are ABS.



If you were to use a liquid, you have to worry about the drag of the liquid on the extruder and about whether the liquid would prevent the plastic layers from sticking to each other.

I already stated the obvious issues that DO make this a fairly complex and probably impractical approach.
That red print looks pretty good, especially for a non-enclosed job. What parameters of your machine allows that level of success?
BTW, FYI I have a delta.
..
As for the liquid-drag comment; I imagine the head is above the liquid and the liquid level would be maintained to some mm below the nozzle. A submerged hot-end wouldn't stay hot very long!
..
Anyway, here are the facts/my experiences to consider in the 'pro' direction:
-I payed to have a part printed in ABS in a commercial printer (I can't remember brand, but it was large, enclosed, used cartridges and probably in the $3k-$5 range). It STILL HAD some slight warp. It had a height/width radio of approximately one, and maybe 40 cm diameter. That was disappointing. I expected a commercial machine to have eliminated all these issues. So the current solutions are not so easy to maintain or everybody would be doing it without issues.
-Some designs are more susceptible to lift than others. I imagine that a wide box with thin walls is one of the worse cases (such as my colossal failure). (tell me if I am wrong)
-A liquid has a much higher heat-capacity and conductivity than air and would provide a much more gradient-free thermal environment for the part.
-Mechanical/robot parts benefit from high accuracies so even a slight curl may ruin a part. Artistic forms such as the images displayed in a previous reply can tolerate it.
-Keeping the soaking region small would limit exposure of the rest of the printer to high temps as in the enclosed approach.

I admit this is probably an idea that would be more considered on an expensive FF machine designed for abnormally-high-accuracy parts (.001" ?) where even the 2nd and 3rd order effects become important, plus assuming other practical issues don't make it an irrelevant pursuit. I'm not even sure there are FF machines with that level of accuracy.

The fortus machines have zero warp regardless of build size. Here is an example of a print I did: [imgur.com]

There are examples of even larger prints with zero warp: [www.3ders.org]

Quote
shadowphile
A liquid has a much higher heat-capacity and conductivity than air and would provide a much more gradient-free thermal environment for the part.

Thermal mass isn't a big deal because we are only interested in coarse temperature control. Temperature gradients can be fixed with a fan in the build chamber.

Quote
shadowphile
Keeping the soaking region small would limit exposure of the rest of the printer to high temps as in the enclosed approach.

There are better ways to keep the heat contained. Stratasys keeps all electronics, motors and even the extruders at ambient temperature. ( [forums.reprap.org] )

A heated chamber has already been shown to be sufficient. At best switching from air to a liquid will keep your heated chamber at +-1c instead of +-5c. That isn't going to have an effect on print quality.

I have a feeling that any system built carefully enough to hold and heat a vat of hot oil would perform just as well with air.

Edited 2 time(s). Last edit at 06/22/2015 04:49PM by 691175002.

Can't argue with most of your points, although that Fortus machine is WAY above the cost and performance I expect from my table-top Kossel.
There seem's to be little advantage to a liquid-bath. (I supposed I was inspired by liquid-cooled computers favored by over-clocking enthusiasts.)

Ehhh.... Regarding the Original Post..

The reason ABS warps is because the three different plastic polymers that make up ABS expand and contract at different rates, and a liquid bath doesn't necessarily prevent this.

In fact, because of the increased heat capacity of a liquid, relative to air, this actually might become more of an issue.

It rather depends on the coefficient of thermal conductivity of each of the polymers involved (Acrylonitrile, Butadiene Rubber, and PolyStyrene), rather then the bulk thermal coefficient that can be found online.
The bulk coefficient for ABS does not work because ABS is not a homogeneous polymer; rather it's more like a micro emulsion (it's like milk or lotion), with partial solvation of different constituents occurring at slightly different concentrations through out the mixture. Sometimes the solvent is styrene, sometimes it's butadiene, and other times it's Acrylonitrile. It depends on the manufacturer, and about 20 other factors that can be as specific as the temperature and speed at which you print your stuff.

To really stop the warping, you'd need a crosslinking agent that would activate once extruded, but, that could be tricky. It'd be better to simply print things on a proper insulated heated bed, and treat them with a crosslinking agent post printing.

ABS is a versatile material, but it's not designed to be printed. It was originally designed to be used in injection molding, where the shaping and plastic deposition occur simultaneously, and as a means of using up unused plastic that could be purchased cheaply. The fact that ABS works as well as it does in our printers is really kinda fortuitous for us.

If someone wanted to put a crosslinking agent in an ABS filament, that would activate when it's molten or when it's in air (or something like that...), that would go a long way to improving it as a printing material.

A slight modification on your idea of immersing in liquid is to use a heavy gas^*. The chamber then would not need a top, only sealing around the bottom and sides. As far as heavy gasses go, carbon dioxide has a density about 1.7 times that of air falling to about 1.25 at 100C. Early very simple trials show that it sort of works. A Sainsbury's flimsy carrier bag fed with carbon dioxide was able to maintain 60C but used about 1/2kg of dry ice per hour - fed from a second carrier bag. Heating was by a PCB heated build stage at 110 degrees. All swings and roundabouts, I guess a deeper bag would be better as would a double walled bag while stirring by actually printing something would accelerate losses.

I may return to this when I get some spare time.

*The best gasses have other problems - e.g. Uranium Hexaflouride

Try a PEI (Polyetherimide) Sheet for your printing surface....Crazy Idea! But it just might be what the Dr. ordered....No?!

PEI: [www.amazon.com]

Adhesive to use: [www.amazon.com]

Need some pointers on applying?....just ask; done it twice already--expert!

My NEW PEI print bed

Presently... waiting on new Y carriage platform, before install...
[www.ebay.com]

This stuff has the uncanny ability to hold tight to your project, while bed is hot, and releasing said project when cooled off.

Edited 6 time(s). Last edit at 06/26/2015 07:20PM by 55rebel.

Quote
Qcks_
The bulk coefficient for ABS does not work because ABS is not a homogeneous polymer; rather it's more like a micro emulsion (it's like milk or lotion), with partial solvation of different constituents occurring at slightly different concentrations through out the mixture.

Wandering a bit off-topic, this is inline with what I have observed with containers of old ABS slurry; it separates into layers. It's as if the acetone melts the ABS but doesn't dissolve it.

Does acetone break down the polymers and only dissolve part of what we refer to as ABS?

Quote
DonaldJ

Quote
Qcks_
The bulk coefficient for ABS does not work because ABS is not a homogeneous polymer; rather it's more like a micro emulsion (it's like milk or lotion), with partial solvation of different constituents occurring at slightly different concentrations through out the mixture.

Wandering a bit off-topic, this is inline with what I have observed with containers of old ABS slurry; it separates into layers. It's as if the acetone melts the ABS but doesn't dissolve it.

Does acetone break down the polymers and only dissolve part of what we refer to as ABS?

Ehh... the short answer is acetone doesn't really break down any polymer (nothing ionizes or really changes states). It just allows them to slide past one another more and more til the Vander Wal forces can't hold the stuff together. If you had a high power microscope, you could look at a blob of 'dissolved' ABS and see bits of the different polymers.

That said, Acetone does attack different parts of ABS preferentially. Polystyrene is much more easily 'lubricated' than the Butadiene Rubber or the Acetonitrile. The Acetonitrile is more polar (and thus acetone doesn't penetrate into it as easily) and the Butadiene Rubber is more sterically hindered.

Using the milk analogy, the adding acetone to polymers is like adding whey to milk. You're not breaking down the polymer you're just providing it with the ability to move farther from the other constituents, but some parts like to move faster and farther than others.