loading

Vapor polishing is nothing new. Experienced 3D printing enthusiasts are well aware that you can smooth the surface of 3D printed objects by exposing them to the right solvents. There are a few existing methods for doing this, but each of them comes with unique drawbacks.

What's needed is a better solvent application method, and that's what this instructable is about. (Be sure to check out the video in the last step!)

The current vapor polishing methods and their drawbacks include:

1. Hot treat via boiling solvent in a crock pot or similar chamber (Potentially dangerous and very hands-on process)

2. Cold treat via slow release of solvent from paper towels in an enclosed container (Very time consuming, can't observe parts during this process.)

3. Spray canned solvent aerosol onto part. (Inconsistent surface finish, must be done outside)

4. Dipping parts directly into liquid solvents (Unpredictable/inconsistent finish, Likely over exposure of part)

All I wanted is a machine that lets me quickly drop parts into a transparent container and be able to press 'go' and have the machine produce a predictable finish on its own. I do not want to have to put together a really involved setup that may be a fire hazard, fume hazard, or something that produces unpredictable surface finishes. Essentially I want something as convenient as a microwave. I'm also forgetful so I don't want my parts to be destroyed if I forget that I left them in the machine.

The Ultrasonic Misting 3D Vapor Polisher is the solution to all of these problems.

This key component of this machine comes from ultrasonic humidifier, which uses a piezoelectric transducer (like a speaker) to create a high frequency mechanical oscillation in a liquid. This vibration forms an extremely fine mist of droplets in a fog/mist. The density of the fog is controlled by varying the intensity of the vibrations via a potentiometer.

This fog mist is very dense and wont move far on its own, so I used an aquarium air pump to blow it from the misting chamber into the glass finishing chamber where it can condense on the 3D printed part. This airflow keeps the air moving inside the finishing chamber, which helps produce a consistent finish on the part. The airflow system is open by necessity but we do not want the exhaust air to become a fume hazard so there is a water bubbler on the exhaust to absorb excess solvent. (Note that this only protects you when using water miscible solvents such as acetone.)

If a spot treatment polish is desired the tube going into the finishing chamber can be disconnected and manually directed onto the part where ever you want. This a great way to apply chemicals for solvent welding (aka gluing parts plastic together).

The other cool thing about this design is that the fog creating components are separate from the fog applying components. This means that you could use a micro controller to do some very precision applications of the fog. As it is, using a simple mechanical timer as shown in the drawing will still give you a ton of control.

Step 1: ​Polishing Chemicals & Solvent Science

Obviously you want to be really careful around solvents. Read the MSDS sheets for any chemicals you use and always wear proper PPE when handling them.

While this machine was designed to allow for use indoors the only solvent I recommend doing this with is acetone. All other solvents should be treated like poison.

Why is acetone special? Actually acetone is some pretty amazing stuff.

Acetone is miscible with water (they mix well), so the bubbler does a decent job of removing it from the air. The absorption of acetone by water is about 1 to 1 by volume, so keep your water fresh to minimize the smell. In any case acetone is perfectly safe for humans under normal circumstances and normal amounts. Believe it or not, the human body actually produces acetone in small amounts!

Acetone is the only commonly available solvent other than water that isn't classified as a VOC or HAP. This means that there are no regulatory restrictions on how much of this solvent we use. No other commonly available solvent is so free of restrictions!

Acetone is the fastest evaporating and one of the strongest of all commonly available solvents and it is very dry (non-oily). So it makes an excellent cleaner and degreaser and this is how it is used in most industries.

Though it doesn’t smell good, you would have to consume a lot of this stuff to be harmed. It is just an irritant in low concentrations, all the bad stuff happens at high concentration levels. Acetone is not regarded as a carcinogen, a mutagenic chemical, or a concern for chronic neurotoxicity effects.

Cool, so what solvents work best for 3D printed plastics?

ABS: Acetone

Acrylic: Most Solvents

PLA: MEK or 'MEK Substitute'

PVA: Water

PVC: Most Solvents

Polycarbinate: Pretty solvent resistant

Nylon: Pretty solvent resistant

Polypropylene: Pretty solvent resistant

Polyethylene: Pretty solvent resistant

Final Notes:

*Always clean your chambers before switching to a new solvent. This is no joke, mixing chemicals as common as ketones & alcohols can cause an explosion. When in doubt consult a chemical compatibility chart: http://www.safety.vanderbilt.edu/chem/chem-compati...

*Solvents do not "melt" plastic, but rather they dissolve them. Melting means a physical change from solid to liquid caused by an increase in thermal energy. The reaction here is more similar to what happens when you pour water onto a sugar cube.

Step 2: Bill of Materials

(1) Air Pump & Check Valve: Aqua Culture 1/2 Gallon Betta View with Base (Same one available at Walmart)

(1) 8"x6"x3" Project Box: www.radioshack.com/project-enclosure-8x6x3-...

(1) Ultrasonic Humidifier:Crane Adorable Ultrasonic Cool Mist Humidifier - Frog

(1) Airtight Glass Jar: A 1 gallon pickle jar would be perfect.

(1) Brewing Airlock: Make your own like I did or get a fancy one here: Cylinder Airlock

(1) Empty Paint Can: 1 Quart or 1 Gal, available at any hardware store.

(3 ft) 1/4" OD polyethylene tubing: available at any hardware store.

(1) 90 deg threaded quick disconnect fitting for 1/4" OD tube: available at any hardware store.

(3) Straight threaded quick disconnect fitting for 1/4" OD tube: available at any hardware store.

Optional:

(1) 5 Minute Spring Loaded Mechanical Wall Timer: Intermatic FD5MW 5-Minute Spring Loaded Wall Timer, White

**Note: Some parts used in the photographs are different than are named here. I used a 7x5x3 project box but I think it is too small. Also, I used a 1 quart paint can but I think 1 gal would be better because it could hold an entire jug of acetone. Finally, it is critical that your finishing chamber is airtight. For this reason I do not recommend using the one I got from Walmart (seen in the main photo).

Step 3: Take Apart & Prepare the Humidifier

My wife was sad that I took apart her frog humidifier, but I insisted on dissecting it for science!

The humidifier comes with a power cord, power switch, potentiometer, float switch, and on/off light. We will use everything so there is no need to cut out any electronics. All you need to do is take apart the humidifier with your screwdriver.

When it is all apart all you have to do is cheat the float sensor into thinking it is ON all the time by ziptieing the float in the up position. Note: Its probably best not to operate the vibrator when there is no liquid, im sure theres a reason they installed this float switch.

In photo 5 i cut the power cord to the aquarium air bubbler soldered it onto the humidifier power posts. This makes it so that the air bubbler and the humidifier are powered by the same power cord, and are only on when the main power switch is on.

If you choose to install the optional 5 Minute Spring Loaded Mechanical Wall Timer that I recommend then you simply install it in series with the main power, like a switch. The installation directions will come inside the timer box.

Step 4: Prepare the Paint Can (Misting Chamber)

In this step you drill the holes in the bottom of the paint can so you can mount the vibrator. Measure and drill 4 mounting holes and one big through hole for the vibrator itself. The big hole will be leak proof because of the rubber seal but the tiny holes are trickier. Solvents will dissolve most any glue you use here so drill your mounting holes as small as possible. I used 2 part epoxy to help seal it in and I haven't noticed any leaks.

When you drill the holes in the lid for the tube fittings make sure to size them as small as possible. You want the fitting to screw itself in and seal the hole mechanically as much as possible because it will be difficult to glue over a large hole without the solvent destroying the glue soon after.

EDIT: Silicone glue is a great choice for solvent resistance.

Step 5: Assemble the Project Box

Notice from the very first photo in this instructable that I ran the air bubbler tube up the outside of the paint can, rather than hiding is on the inside. I did this to avoid having to seal up another hole on the base of the can.

Its difficult to see in the pics but I added a slice of foam around the air bubbler to prevent it from vibrating against the box and making noise.

After you jame everything in there you are done! I'd sujject letting all the glue dry for 24 hours before butting any solvents into the chamber. It'd probably be better to test the equipment out on water before you just straight to solvents. Also if you bypassed the float switch like I did then be careful not to turn the machien on when it is empty or bad things will probably happen.

Step 6: Mechanical Testing of Acetone Polished ABS Parts

I was curious what effect acetone vapor treatment would have on the mechanical strength of abs specimens. Good thing I happen to have a TestrBot Universal Test Machine to figure it out.

The testing was pretty conclusive that treating abs specimens with acetone vapor caused them to lose strength despite having given the specimens a full 24 hours to dry out. Skeptics can check out the attached pdf file for detailed test results.

This result was surprising to say the least. I expected the polishing to strengthen the parts by reducing surface stress concentrations. My highly speculative guess for the resulting loss of strength is that the exposure to the solvent caused a permanent chemical change to the surface of the parts resulting in a softening affect.

Looking at the inside of the parts and it is clear that the treatment did not penetrate very deep. It is interesting to note that the mass of the specimens was measured before and after treatment and right before they were tested. Specimens increased in mass by about 0.2g right after the treatment but they went back to their normal weight after 24 hours.

To be fair, this was not an exhaustive study. I tested a handful of specimens in a single orientation. But then again, my standard deviations were low so my setup is clearly capable of producing repeatable results.

All things considered, having an ultrasonic vapor treatment chamber could still be beneficial to anyone wanting to improve the aesthetic value of their 3D printed parts.

Check out this video showing the whole process:

UPDATE 9-2-15: I've just released a new batch of test results that show how to optimize 3D printed parts for strength.

<p>UPDATE: There is a new Instructable out with a more developed version of this device.</p><p><a href="https://www.instructables.com/id/Automated-Ultrasonic-Misting-3D-Print-Polisher-PRO/">https://www.instructables.com/id/Automated-Ultrason...</a></p>
<p>Perhaps having 2 of these machines would better serve you since PLA and ABS are the most common materials in the world of 3D printing</p>
<p>I've tried MEK on PLA, doesn't do anything. The stuff that works is Ethyl Acetate(MEK substitute), chloroform, di/trichloromethane, or tetrehydrofuran(pvc pipe cleaner).</p>
<p>Thanks for sharing your experience! There are so many different brands of PLA out there that it makes me wonder if the chemistry varies enough to make a difference? In any case all those chemicals sound pretty scary! Though the pvc pipe cleaner comes with a build in manual applicator wipe thing, perhaps that is the best option.</p>
<p>Yeah. That's quite possible. Even with the different solvents you get different results, not to mention the various additives to PLA thrown into the mix. Ethyl Acetate is commonly available in beauty/cosmetic shops as nail polish thinner(not remover, thinner).</p>
<p>I know PLA was already addressed in this thread, but would this setup work with MEK?</p>
<p>Nice! I like your method. Very innovative. Can you use it for even more drastic chemicals? Like hydrochloride acid for nylon polishing?</p>
<p>I looked for a long while for a technique to polish up PLA. I found that using TETRAHYDROFURAN (THF) works amazingly well. But it's no good for a vapor bath. At least I had no success. </p><p>I use and old coffee pot with less than 1/4 cup poured in. Throw your prints in and shake the pot enough to wet the prints. You can experiment with time, but I jumble the pieces only 5 seconds or so. Take them out quickly and put them on a non or semi-porous surface to dry. 1/2 hour drying time is enough to handle them. Make sure your hands are clean.</p><p>The yellow print had errors in the file but you can see the finish on the red print.</p><p>I think you'll like the results.</p>
<p>Was the left over THF any good to use again? Also, I'm taking it you only shake/rotate the pot gently for about 5 seconds?</p>
<p>As far as resusing the THF after the bath, I do pour the remaining liquid back into the bottle. I haven't had any problems. But maybe it will get too cloudy at some point if it starts staining. If your concerned about that. You are correct about the time. Swirl/shake/rotate for 5 seconds, even less. The key thing is covereage. That's a good rule of thumb. But experiment with the time, who knows?</p>
<p>Thanks for that. I'll give it a go at the weekend.</p>
<p>While the process would, in theory, be useable for almost any liquid the build would have to be entirely redesigned to handle nastier chemicals. Something like an acid would most likely either have to do this in a full fume hood or have some far more serious form of ventilation and filtration systems. Also most of the containers would need to be either glass or chemical safe plastics. Maybe if you glued a thin chemical proof membrane to the piezoelectric &quot;agitator&quot; it wouldn't be damaged by stronger chemicals with the tradeoff of a reduced capacity. I am interested myself to how the peizo holds up to prolonged solvent contact as it was initially designed to only be in water. I might consider having wrapping the guts in something just in case it decides to fail catastrophically at some point.</p>
<p>Couldn't have said it better. I thank you for the compliment Peter, and I would also recommend against using any stronger solvents or harsh chemicals with this particular machine; for both safety reasons and because the existing setup couldnt handle it. </p>
<p>One safety issue is not addressed! Acetone is extremely flammable, especially when misted by the humidifier. You have electrical circuits right next to the acetone. Worse, I see an illustration with a gallon can of acetone with a stopper in it leading to a hose with a flame! If that flame propagates back into the can, the whole house will go up!</p><p>The heating scheme is safer because soon the vapors become so rich in acetone, they can't explode, and the heat can be provided by hot water. Timing can be provided by having a set amount of acetone. When it boils away, there will be no more vapors fed to the chamber. In the humidifier, fresh air is being continuously fed into the mister chamber. So there will always be a region where explosion can start.</p><p>The hot water method also demands a smaller amount of acetone, as it ends with the acetone being exhausted, so the fire hazard is reduced.</p><p>This process should not be automated. It should be watched, and preferably done outside (because of fire hazard, not chemical hazard.)</p>
Since he's using a air pump already, why not use it for double duty. Why not use a nebulizer, venturi style fogger in the chamber.
<p>Maybe a slicer could be modified to make G-code that produces 2 shells, an inner and an outer one. That way you could treat the outer shell and the inner one would still supply strength. then again.. maybe that could be accomplished by just making the one shell thicker.</p>
<p>Hi Mike,</p><p>This is a really cool idea - well made. I've used piezoelectric foggers for quite a few things in various places... The ones you use are certainly quite powerful (you might want to warn people you can actually hurt yourself if you leave your fingers in the 'jet' of water - you can even cut through 3mm plastic (I've done both by mistake!). Anyway, I also thought you/other readers might also want to know that there are possibly more straightforward ways to incorporate a fogger into a can, though I admit I have not tested this with solvents, (but I'd suspect that the cable is PE and probably pretty inert). Check it out - they are quite cheap, and may save your wife feeling sad about Mr Frog getting disassembled (you are not alone on this!). <a href="http://www.amazon.com/AGPtek%C2%AE-Aluminum-Fountain-Machine-Atomizer/dp/B00P91ZFPA/ref=sr_1_1?ie=UTF8&qid=1431177828&sr=8-1&keywords=pond+fogger" rel="nofollow">http://www.amazon.com/AGPtek%C2%AE-Aluminum-Founta...</a></p><p>These foggers also have the great feature of having a 'low-level' sensor built in (that little black handle-type-thing), so it saves you running the fogger dry (or having to have a reservoir) and risking burning the piezo disk out.</p><p>Anyway - thanks again for a really clever I'ble. Great job! I'd never have though of this, and look forward to trying it out with my ABS parts.</p><p>Cheers, </p><p>Jude.</p>
<p>Thanks for pointing out this alternative source for a fogger. I was looking to find something besides a humidifier. </p>
<p>Cool - let me know how it goes, if you buy one. I have not tried it with Acetone as I mentioned. I think the design could also be much simpler as a result, as you can just drop this in a bucket, fill with acetone, and elevate/hand the piece in the mist....of at least that's what I think might work well anyway... Good Luck!</p>
<p>Did you try a mechanical test of a tall part? Results will probably be similar, but it's possible the acetone treatment would increase the interlayer bond strength?</p>
<p>Interesting... =D</p>
<p>Why is the only step I want to try marked as &quot;Don't try this!&quot;?</p>
<p>This is excellent. I've been meaning to try this with PLA, but I hear that the results aren't quite as good as using acetone on abs, and that MEK is more harmful by several orders of magnitute. For now I've resigned myself to finishing parts by coating, sanding and painting.</p>
<p>Than you! I've tried a bit of PLA with acetone but the effect was not desirable. MEK is more dangerous from my understanding, the MEK substitute is supposed to be a bit better though. </p><p> Have you tried tumble finishing? I've seen others get great results with converted rock tumblers.</p>
<p>I haven't yet. Simply because if it doesn't work well for me I don't want to have a rock tumbling setup I have no other use for. Haha! Right now I'm currently doing some work with filler primer, sanding, and painting to see how it turns out. Obviously a little more work and it only works for pieces that aren't precisely engineered to be used mechanically.</p>
<p>Brilliant. The only improvement I would make is to try and find a way to couple the transducer to a quart can of acetone as it comes from the store and a custom replacement lid. The little high strength magnets suggest themselves.</p>
Fantastic project and will written. Saving this for a late date.
<p>Thanks!</p>
<p>on strength, would adding extra shell thickness when you slice offset the loss? I could see that if you only have 2 layer shell dissolving part of the outer one and spreading it around could weaken slightly. maybe add a 3rd so lessen the loss.</p>
<p>For my particular tests Im certain that adding perimeters results in a stronger part much more so than adding more infill. The funny thing is that there are oddities in the way the parts are made that may affect the results. For instance, slicer has a bug that makes the second perimeter not bond well to the first, making the part weaker than it should be. More testing is needed to figure out things like this. </p><p> In any case the perimeter count was set at 2 for all specimens tested here and the only variable changed was the vapor treatment. </p>
silicon sealer can be used, won't be affected by the acetone.
<p>Ahh thank you!</p>
<p>Many people confuse melting with dissolving. When Dorothy throws the water onto the wicked witch and she cries, &quot;I'M MELTING!&quot;, I always scream at the tv and correct her, &quot;YOU'RE NOT MELTING, YOU'RE DISSOLVING, YOU STUPID ILLITERATE TWIT!&quot;.</p>
hardly matters, might as well just say go bye bye
<p>Definitely cool - I was trying to think of applications for work ... but there doesn't seem to be the need........ yet. We are looking at a new 3D printer though &gt;.&gt;</p>
<p>Proper clever. </p>
<p>very interesting idea!</p>
<p>Great project! This is definitely a problem... I use the usual vapor chamber and keep getting acetone all over my hands -- not good. Thanks for posting... will definitely be making this.</p>
<p>i just hang my prints above a jar of acetone and leave the door open so i dont breathin to manny fumes </p>
<p>Great work! but, still have to be done with extreme carefull. Any leakage can turn into serious health problems so it would be best to do it outside. </p>
<p>Awesome work, thanks for sharing this!</p>

About This Instructable

59,161views

613favorites

License:

Bio: Hi I'm Michael! I'm a dog owner, husband, writer, and mechanical engineer and I love getting my hands dirty building stuff. If you ... More »
More by MechEngineerMike:A Practical Guide to FDM 3D Printing Gears Easy Removable Milk Crate for a Bike Automated Ultrasonic Misting 3D Print Polisher PRO 
Add instructable to: