I'm still looking for the best chemical to finish the surface of my PLA prints - if there exist any appropriate chemical at all. I don't know as I'm not a chemist, so I have to rely on the statements of the reprap community where some people recommended a organic solvent called tetrahydrofuran (THF) that should work the same for PLA as acetone for ABS. To make it short: It doesn't. At least for the white PLA that I am using.
In my 'experiments' I found out that it behaves just like acetone (to PLA) - but only much slower. It acts hygroscopic and pulls out all the water from the plastic making it brittle and spongy.
Dipping such a pyramid in acetone for 15 or 20 seconds and I get a part with a super smooth surface.
Pyramid in THF for 30 mins makes the very outer shell a bit softer. Some (but not all) layer lines are merged giving a clean and smooth surface. Corners are strongly rounded now. But for large parts of the object the chemical made no visible difference. Strange behavior.
After 2 hour in THF the walls and layers of the test part are bending and distorting.
After 12 hours the test part looks the same as with 1 hour in acetone. The whole object is beginning to fall apart.
So, THF is NOT an alternative (for me).
Any ideas? :-)
Edited 1 time(s). Last edit at 03/21/2013 11:35AM by jk25.
I'd recommend that you simply sand and then finish using a coating like polyurethane. I don't know that there are any reasonable solvents that will work in the same way as acetone does for ABS. And even acetone isn't a very nice chemical and in larger quantities should be reclaimed using evaporators and closed environments. As individuals, we may not be subject to the laws that business has to follow, but that doesn't mean we don't have a responsibility to avoid environmental dumping using an excuse of "It's a hobby".
I haven't found anything that works either. Though there are far far far worse chemicals than acetone. I wouldn't call the ambulance if I spill some acetone or similar laquer on my skin, I worry a little more about DMSO, THF, Dichloromethane, etc.
In theory, if there is water absorption in the polymer, none of these would work consistently anyway. Maybe a water based chemical with mild heat?
@Enlightx, easy enough to build one for yourself and try it. Plans are all over the net, and it really doesn't take much in the way of hardware. The basic recipe is a coffee can, a couple bearings, a motor to turn it, and some abrasive (walnut shells, etc.) And any bits and pieces you need for the tumbler assembly can probably be printed...
probably, but I'd be worried about injecting abrasive material into the print...granted, under certain circumstances, that might not be a bad thing. Sandblasting also tends to produce very fine grained byproduct which is a dangerous inhalant (see silicosis for more info)
edit: (Plus, even if you don't breathe it in, the crushed abrasive tends to get into everything if it's not properly contained in a sealed sandblasting unit ...not saying it's a bad idea, just that you should keep the issue of unexpected "mess" in mind)
Edited 1 time(s). Last edit at 03/21/2013 03:37PM by xiando.
A friend of mine tried sandblasting his 3d prints and got bad results. The only thing that happened was an even more rough surface than before. It didn't smooth out his parts.
And because of the dust the abrasive material creates it's even more dangerous than such chemicals like acetone which you only have to deal a few seconds with. Without a proper working station including an exhaust system I wouldn't try to deal with sandblasting.
I think you'll be pleasantly surprised how well a tumbler works if you are now doing it by sanding. Get small triangular plastic abrasive media and it will smooth every little nook and cranny. It won't take two weeks either, that rock polishing from the rough. I'd bet overnight or less would do the job. If you find that you need faster, get a vibratory unit instead of tumbler. I've owned both, the tumbler did a beautiful job but was slower and more work to fill and dump. The vibratory was much faster and dumped out through a chute. However it was very noisy so we put it in another room.
I was in Harbor Freight yesterday. They sell two small vibratory finishers, and also 5 lb. quantities of various abrasives. The larger vibratory was $149 and the abrasives varied in price, but cheap enough to do some experimenting. I'm going to be machining metal parts soon, I'll probably grab a vibratory finisher and some abrasive too.
tk25, I'm looking to make a connection between two PLA prints watertight. The part is too big to be printed in one piece, so it will be multiple pieces with the junction between the two pieces keyed like a puzzle piece. The PLA part will eventually be used as an investment cast core, so it needs to be water tight. Regarding shape, it is essentially a 10" pipe (imagine two 5" pieces of pipe with a keyed connection between the two).
Based on your tests, do you think THF, if applied on the surface at the junction of the PLA printed parts, would seal the surface of the junction?
So as THF didn't work well for me so far I tried Dichloromethane (DCM, also called methylene chloride, see Wikipedia) as some people recommended to be an appropriate solvent for PLA.
Well, I'm excited about it as it truly WORKS to get smooth surfaces on PLA objects. It works similar to acetone, but I also found some differences. I want to share my experiences here.
DCM is nasty stuff. It looks like water but has a slight sweet smell. It's declared to be toxic and carcinogen, so don't use in your house but do it outdoors. Also proper care and protection (glasses, gloves) are recommend.
DCM is cheaper than THF but much more expensive than acetone (acetone is about 1/3 of the costs of DCM here in Germany).
DCM is also found in many PVC glues so it's not an 'exotic' chemical.
Here are some of my tests where I used pure (99.9%) Dichloromethane for.
DCM is a quite aggressive stuff and much more aggressive to PLA than acetone to ABS which means it dissolves more in the same time. You have to be very careful not to over-do it with smoothing when dipping an object into the chemical. A few seconds normally are fair enough to get a smooth surface. I found that it's stronger to white PLA than to golden PLA. At least to the filament I use here.
It isn't a good idea to use it on 'scientific' objects like this 3d-scanned ruin that was printed as a cool miniature. Details get lost easily and quickly. Although I dipped the object into DCM for only about 10 seconds many details of the walls and stones got lost. 1 or 2 seconds would have been between in this case.
You can apply it also to mechanical parts. Like acetone it improves the overall strength of an object by 'blending' the outer shells and layers in between. Instead of having an object that is made only out of stacked plastic layers you now have an uniformed 'monolithic' object. More or less, of course.
The object shouldn't be applied for too long with DCM. When the entire plastic absorbs the solvent an object with a rubber-like surface is the result. Also after a few days when all the chemical is evaporated the plastic remains a bit elastic which is nice for artistic objects but the worst case for mechanical parts. An interesting point in my opinion.
DCM also works on ABS! So it's very universal. Also on ABS it's more aggressive than acetone.
It's very interesting that DCM doesn't leave a shiny surface like acetone always does. Normally it leaves a 'matte' finish that feels velvety and not as smooth as with acetone although the surface is really smooth to the eye. Only on small parts and details it leaves a shiny finish. Bigger and flat features tend to be remain matte. Thin layers of DCM also tend to leave shiny surfaces instead of matte. That's typical for the areas where the transition between chemical and air was. The surface tension forces the liquid to form a shallow film there leaving a shiny finish. Sometimes little bubbles are given off the plastic when the dissolving process takes place and got easily embedded into the surface when the plastic re-solidifies.
Great post. DCM also has a low boiling temperature; what is your opinion about possibly vaporizing it and using it in vapor form, similar to how people are using acetone vapor to smooth ABS (ignoring the health implications)?
Now that we are on the subject of solvents, I just wanted to chime in to warn about the dangers of messing around with DMSO. It has the ability of bringing anything dissolved in it through gloves or skin right into your body. I work in a chemistry lab and this is one solvent that we try to avoid using and when we do, we treat it with an enormous amount of respect.
Anyway, I don't wait to rain on the parade and those are some great looking prints! Carry on.
The vaporizing setup would have a cold condenser ring setup below the opening, I would be doing this outside, and I would be wearing a respirator/glasses/etc. Placing the carcinogenic dangers and potential health issues to the side for a minute, would DCM functionally work in vapor form, similarly to acetone?
If all health hazards were to be put aside, you should be able to vaporize DCM the same way as acetone. I would however still really advise to not do so, there's a reason that DCM is a banned substance in some states in the US. A "better" way to do it would be to stand behind a fan and use a mist bottle to stray at the part but even that is a bad idea in my opinion unless you plan to not breath or be absent for the duration of time you are treating the parts.
>A "better" way
> to do it would be to stand behind a fan and use a
> mist bottle to stray at the part but even that is
> a bad idea in my opinion unless you plan to not
> breath or be absent for the duration of time you
> are treating the parts.
Misting also sounds like a decent idea, but part of my desire to vaporize it is because what I'm going to be printing is a part that is too big to be printed in one piece (so it will have puzzle-piece type joints for the separate sections), and the piece has an internal structure of snaking tubes. I'm looking to seal the where the sections of the part joins inside of the tubes (this print is likely going to be used as an investment casting pattern), and the only way to really do that would be to completely submerge the part so that the liquid flows through the internal structure, or as a vapor, in hopes that the vapor may be a little less quick acting and hopefully more homogenous in its solvency.
I understand that these chemicals are harsh, but there does exist a proper protocol of handling and protection that could work. Hence I'm trying to determine the effectiveness/feasibility, rather than writing it off wholly because of the health risk potential.