Sandblasting to smooth printed parts

[jd_oc]

On the Tips and Tricks section (http://wiki.makerbot.com/tips-and-tricks) there is a mention of smoothing parts via sandblasting. However, I'm having trouble finding any info on anyone who's actually done this. Does PLA sandblast well? If anyone has any links or first-hand experience that would be great.

Alternatively, I know companies like Shapeways use vibration and abrasive material to smooth their nylon parts, is there any sort of DIY equivalent that works for PLA/ABS?

[Jesse Donaldson]

I tried crushed walnut shells in a vibratory tumbler for ABS... it had
surprisingly little affect over few hours.
I haven't gotten around to any more testing, but I'm planning to try
harder abrasives, more weight, rotary tumbling, and longer periods of
time.

Let me know if you find anything interesting!


Jesse

> --
> You received this message because you are subscribed to the Google Groups
> "MakerBot Operators" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/makerbot/-/Wdq20fXhzFoJ.
> To post to this group, send email to make...@googlegroups.com.
> To unsubscribe from this group, send email to
> makerbot+u...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/makerbot?hl=en.

[Dan Newman]

On 15 Oct 2012 , at 11:09 PM, Jesse Donaldson wrote:

> I tried crushed walnut shells in a vibratory tumbler for ABS... it had
> surprisingly little affect over few hours.

You'd likely want to through a little carbide grit in as well. Maybe 800?
Plastic pellets + grit might do better even better. Of course, with the
grit you'd likely also want some water in order to get an abrasive slurry.

Dan

[PropellerScience]

This is a perfect excuse to break out my Air Eraser. Stay tuned!
http://youtu.be/YtBaG58P5gc

[jd_oc]

That looks promising. If you can take video (like the penny clip) that would be awesome. Whether or not it ends up working, either way it would be a great data point to have so people don't need to repeat this down the line. 

[jd_oc]

Hi Jesse,

Thanks for the info. I would definitely be interested in any other tests you come up with. I would be willing to spend $100-$200 for a system that could smooth PLA prints. I'm guessing with the proliferation of printers out there, a new market for a simple inexpensive system for finishing parts would emerge. Anyone want to create a Kickstarter?  ;)

[PropellerScience]

I tried sandblasting PLA and ABS with my air eraser, and all it did
was take the shine off of them. Maybe I need a bigger blaster, but I
haven' figured out how to use my bigger blaster yet, or it's just
junk, it never did anything on my birdbath last Summer. In any case,
if you were to sandblast anything I suggest a sandblast box. I'll
never get rid of the stuff in my garage!

[geneb]

You can flame polish acrylic with a Hydrogen torch. Can something similar
be done to ABS and PLA?

g.

--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.

ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://www.scarletdme.org - Get it _today_!

[hellphish]

Yes, but I bet it is tricky. I sometimes use a little pocket torch to remove whiteness on colored plastic if I have to remove support, remove boogers, or have sanded the surface.  For smoothing prints I use a big vat of acetone!

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To post to this group, send email to make...@googlegroups.com.

To unsubscribe from this group, send email to makerbot+unsubscribe@googlegroups.com.

[PropellerScience]

I have a little brush of acetone :-)
I found that Bondo is pretty awesome at filling in gaps, like the top
of a head if you're going to paint it anyway.

[Ian Johnson]

I was just experimenting with this last night.  I had been wanting to try out a tumbler, but didn't really want to spend $150 for a small one and not have it work.  Then I discovered that I already had a giant one, next to the washing machine.  I got 5lb of ceramic media from the local Harbor Freight, a 2 gallon tub and a cheap floor mat.  I cut up the floor mat and lined the tub with it so it wouldn't make too much noise.  I tossed in an ABS print with the media, closed up the tub, and put it in the dryer on "air fluff".  I packed pillows and quilts all around to hold the tub in place, and oriented it so that it would turn end over end since I've heard that will create more aggressive action.

The print I used had .1mm layers, and after 4 hours the layers were still visible.  I would estimate that they were maybe cut down by half to 2/3.  It had a nice matte feel to it, and the edges weren't rounded so much as not quite sharp.  I've been getting a bit of a moire pattern on my prints, possibly due to the voltage on my E driver being a little off.  If it weren't for that, I think the appearance would be smoother.

I often smooth my prints with acetone vapor so they are better for making molds with.  You get the best results if the print is high resolution and sanded ahead of time.  The more you melt it, the more inconsistent the surface will be.  The less melting required to eliminate the layer lines, the better.  I don't think that tumbling will ever completely remove layers lines (not even Shapeways polishing does) but it would be a good way to prepare a print for final smoothing with acetone.  You can also use it as the final step.  If you need it smooth, but don't want the gloss, you can tumble it after acetone to get a smooth, satin finish.

I haven't tried it with PLA since I didn't have something on hand to test with.  I'd also like to see what adding a polishing compound would do, but I don't know what to start with.

[jd_oc]

Thanks Ian for the detailed report, sounds like an inexpensive method anyone could try before paying money for a dedicated tumbler. I was just doing some research on rock tumblers and it looks like just to do the initial rough polishing on rocks takes 4-6 weeks in the tumbler. So maybe a few hours isn't enough for ABS/PLA. I wonder what your part would look like after 48 hours in there. 

[Lincoln K]

A vibratory tumbler will do it, and it doesn't round the corners like a standard rotary tumbler.

Depending on the final surface you want it could take up to 12 or more hours, and possibly a few different changes off "grit" like when you use different strengths of sand paper.

[Ian Johnson]

The part I tried was a ring, about 2mm thick, 70mm across and 35mm tall.  I found that the inside of the ring wasn't affected nearly as much as the outside.  A vibratory tumbler would probably have polished the inside better.  Also the ceramic pieces are maybe 15-20mm across, so they aren't getting into any tight spaces.  They were recommended for aggressive deburring of metal which sounds like it might be too much for plastic, but we are looking to remove more plastic from the surface than you would typically be looking for in a normal polishing operation.

I could run it longer, but the timer on the dryer maxes out at 1:20, so I have to be there to keep resetting it, rather than letting it run for hours while I'm at work.  It does make a good test to see if a real tumbler would be worthwhile.  Harbor freight has a decent sized vibratory tumbler for around $150, but some reviews indicate that it isn't that well made.  There are some homemade ones on youtube.  One person mounted a bowl to an orbital sander held upside down in a vice.

[Ian Johnson]

DId you use a compressor with the air eraser, or canned air?  Do you think it would do more with higher pressure?

[Robert Edwards]

Has anyone found an aggregate that works well with PLA?

We're going to try and place a bunch of broken PLA pieces into a rotary tumbler.

I'll post up how it went…

Is anyone able to find PLA powder anywhere?

[Thomas Kulvik]

I was just asking myself the same thing. Have you found out anything regarding this? Sandblasting PLA.. if it actually works, it may be a perfect way to: 

- Improving the surface overall.

- Removing rafts/support. 

Have anyone tried this? 

Mvh

Thomas

[Stan Velijev]

I was really curious about just tossing some PLA and sand into a vibrating paint shaker, but I work out of an apartment and I assumed the noise would be too bothersome to neighbors so I never gave it a try.

Would be an interesting experiment.

[Jameshs]

Sandblasting rarely smooths - it tends to lower the surface uniformly - so would 'follow' the ridges.

The only other 'polishing' systems are like this http://www.ukge.co.uk/UK/Stone-Tumblers.asp?gclid=CM3T76jwq7cCFaTItAodqFIAWQ which are low cost and come with polishing compounds to add.

I used to use similar approach on wood - sandbalst to remove the basis sawmarks and then tumbling to smooth (takes a LONG time and does produce a 'roundness' and softening of edges)

[Jay]

Hmmm....I have 3 large tumblers I use for brass casings. I think my medium would be wrong though (corn cob, pecan, and walnut shells mixed with different chemicals)...anyone got a guess at what I could use? and easily found? I think I've got a few pounds of virgin corn cob left over....anyone?

Jay

[Richard]

I think part of the problem with rock polishers for smoothing plastic
is the plastic is very light weight compared to stones in a tumbler.
As a result the surface pressure that allows the grit to polish the
plastic is very light if...you just have plastic parts and grit in
there. Its why you always have to put multiple stones in a tumbler,
not just one so that they can "mash" against each other with the grit
in between.

I would suggest putting some small stones or maybe marbles or BB's
(ball bearings) in with the plastic part. Obviously delicate parts
could get destroyed with this method if your stones were too big and
heavy. Fish tank gravel (the smooth, natural kind) might be just the
ticket. Small hard stones that can get into tight places and that are
individually light, but in volume provide some bearing mass.

Corn cob, and nut shells work with shell casings because the brass is
dense compare to plastic so it provides enough weight for the shells
to push the polishing/burnishing material against each other.

[Kristofer]

So have anyone tried sandblasting nylon?

I am looking for a matte finish on my bridge nylon.

Regards Kristofer