3D Print Finishing





Introduction: 3D Print Finishing

About: I'm an inventor / maker / designer based in the Bay Area. My background is in residential architecture, film set design, animatronics, media arts, exhibit design, and electronics. I use digital design and fa...

There's no need to live with the layer lines that show up on 3D printed parts. With a little work, you can get your parts looking smooth and glossy.

Assembly and finishing are two crucial skills to have when it comes to just about any kind of making. With 3D printed parts, there are lots of options for creating different kinds of finishes.

The image above is from a speculative architecture project of mine. It's an ABS print that's been sanded, bondo'd, and painted with matte spray paint. The wooden façade and window panels are laser cut. There are tons of options for finishing!

Remove Parts + Clean Supports

The support structures from Simplify 3D are really good in general. They come off really easily and control the shape of the objects very well.

With the bigger parts of the bike fender, I first tried printing them with the big surfaces directly on the build platform. Every time I tried to remove the parts from the bed, the first layer of the part would peel off a little bit because so much surface area was stuck to the bed. I probably could have tweaked the settings in some way to avoid this problem, but I came up with a quicker solution.

In Simplify 3D, I moved the part above the bed by a few millimeters so that the whole part was sitting on top of the support structures. Because the entire weight of the part was resting on the supports, I think the support structures had a tighter bond to the part than usual.

In any case, with the combination of my hands, the can opener from my multi-tool, and a putty knife, I got the support structures off of the part pretty quickly.

Step 1: Glue

With all of my parts cleaned up, it's time to glue them together. There are lots of different glues that work with PLA, but I find that E6000 is easy to use and makes an indestructible bond. The bond is also slightly flexible, so there's no danger of a brittle bond (like one you'd get with 2 part epoxy or cyno), and it will stick pretty much anything to anything.

I use a syringe to apply the glue into the trough on each part. The syringe keeps the glue from getting everywhere- this stuff is really sticky!

It takes 24 hours to fully cure, so I use painter's tape to keep the parts in place. I don't need to worry about clamps or jigs or any other work holding because I designed the parts with ridges and troughs that keep the parts aligned.

Step 2: Sanding

A raw printed part will almost certainly never come out with a completely uniform surface finish. Even with Simplify 3D's superior slicing algorithm, the vertical walls of a part will look different than an overhanging surface, as will a top surface. This is the nature of the technology, and it doesn't seem likely to me that any improvements in software are going to make much of a difference here.

In order to get a uniform surface for finishing, I'm going to sand the part, then use a 2-part epoxy to get a polished finish.

With sanding, always start with a coarser grit and work your way up to a finer grit. With this part, I start with 200 grit, step up to 320 grit, then wet sand with 400 grit. This gives me a nice, even surface for the epoxy finish.

Step 3: Epoxy Finish

When FDM printed parts are finished, they have small ridges on the surface. This is an unavoidable result of the building process- single layers of filament welded on top of each other. With epoxy finishing, we can smooth out these ridges while creating a shiny surface and even stronger adhesion between the layers at the same time.

I used a product called XTC-3D which is a 2-part epoxy resin to finish the piece. It's got relatively low toxicity, is easy to mix, and has about a 20 minute working time so it's easy to control. It's fully cured in two hours, which is desirable when you're going to be doing multiple passes of sanding and finishing.

First Coat:

Like proper finishing with just about any chemical and any material, multiple coats make for the best results. The epoxy fills in the gaps and uneven textures in the surface of the part, but since the part is uneven, the coated surface is uneven. For best results, at least two coats are usually needed.

Second Coat after Sanding:

After sanding down the first cured coat, I apply a second coat. I found that wet sanding wasn't really necessary, and that a soft 220 grit sanding block did a good job evening out the finished surfaces. A hard sanding block is necessary for the big flat surfaces- if you use a soft block here, the uneven surface will remain.

Third Coat: Complete

After three rounds of wet-sanding and finishing, the final result is pretty smooth. The epoxy is very viscous once it's mixed, so it's hard to get a perfectly smooth finish with it.

Step 4: Ready for a Ride!

The black color of the PLA means that with a glossy finish, the part is basically a mirror. This really makes all the little bubbles and imperfections in the surface show up. I would imagine that in white or another color, this finish would appear much more uniform.

In any case, there is no trace of any ridges, and the smooth, glossy finish will make it easy to clean and wipe off any mud.

The clamp part attaches pretty easily to the seat post. All you have to do is spread the clamp a bit to snap it on, then insert the screws and tighten them in place.

I'm happy with the finish, and it keeps the mud off of my back, but the form of the part still feels a bit boxy to me. I think for round two I might try out Fusion's Sculpt Environment to make a more organic form.

12 People Made This Project!


  • Clocks Contest

    Clocks Contest
  • Water Contest

    Water Contest
  • Oil Contest

    Oil Contest

12 Discussions

I recently discovered that XTC 2-part epoxy, found it to be wonderful stuff and I will probably use it for a lot of my 3d printed builds going forward.


1 year ago

You mention the epoxy being thick and difficult to get smooth coats with. You might try thinning it with methyl (denatured) alcohol. At $20 per gallon it's cheap, and the epoxy paint flows better. Might save you some sanding time, and it cleans paont brushes really well.

3 replies

thanks for the tip! Hadn't thought of thinning it.

Using Eepoxy and Acetone cleanup have shown to be toxic to surfboard makers over time. Something about the mix isn't right and you might get supersensitised to the epoxy as time goes on. Keep that in mind and wear a respirator/gloves if using acetone with epoxy.

Thanks for pointing out the safety concerns, I should add that warning to the Instructable. I'm not wearing a respirator in the videos because I'm in a powerful spray booth where there's no need for one. I'm spoiled, I know.

Why would you put support structures under a flat print seen like a waste of filament to me

1 reply

It does use a lot more filament, but the surface comes out so much better than to have it flipped. Orientation is super important for surface quality and sometimes it's worth it to me to sacrifice filament and time for a better result.


1 year ago

If you print in ABS, you can use the Acetone method to get a smoot and glossy surface.

Looking at Google you can find many sites that explain the method.

Basically it's necessary to put the part in a glass jar (not plastic) suspended from below. Place 2 mm of acetone in the jar and close (the part must not be in contact with acetone).

Time depends on the quality and dilution of acetone, ABS and temperature, you have to do some testing, but the results will be amazing! I personally test it various times.

You could also heat the jar with an electric warmer, it will reduce the time of exposion.

2 replies

Thanks for the tip! I've done that before with ASA prints and a rice cooker. Timing is crucial, and if you get it right the results are awesome.

With PLA apparently you can use MEK for a vapor bath with the same results, but I haven't been able to try it because it's illegal in California.

Yeah I've heard of that technique too, it'd be nice to have a hands-on report on how the results compare to what is presented here! Jon, go do experiments por favor :)

Thanks for the tip! I just looked it up and it seems like a great way to get better results. I'm all about increased strength to weight ratios.