Creating 3D QR codes

General discussion topics
jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Creating 3D QR codes

Post by jsc » Thu May 08, 2014 6:46 am

I'm going try to have a friend of mine with a milling machine make something like this. The filament channel is split by bridges in an attempt to keep the heat contained to the center portion. The handle bit is for pliers. There would be two of them. The cone on the back is intended for introducing a soldering iron or a hot printer nozzle. Maybe some alignment pins. Possibly move the bridges to the mating face to try to keep them away from the heat cone. Not sure what is possible machining-wise or material-strength-wise.

Anybody have any objections why this wouldn't work?
Attachments
bottom.jpg
bottom.jpg (46.51 KiB) Viewed 11698 times
top.jpg
top.jpg (76.19 KiB) Viewed 11698 times

Toby
Posts: 330
Joined: Wed Apr 09, 2014 9:44 pm

Re: Creating 3D QR codes

Post by Toby » Thu May 08, 2014 8:01 am

First reaction- I think if the filament melts in the center section then it will likely melt into the gaps between the channels as well. I would keep the bottom part one continuous channel and split the top part into three sections- left, center and right, with just a hairsbreadth separating them.

The center takes the heat and the left and right hold the filament in place. Also, make the left and right channels longer, both to better secure the filament and also to have a wider margin of error for how much away from center the filament melts. It may take some practice but I think you'll learn by experience how long to apply the heat to make sure it doesn't melt outside the channel.

And if you want to be fancy, curve the model so that it better matches the natural curve of the filament. I just started modeling my own version of it that way for printing and casting in plaster. I think yours has a better chance of success, but it will be fun to try them both.

Are you going to use aluminum?

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Creating 3D QR codes

Post by jsc » Thu May 08, 2014 8:08 am

I think I had a drawing error, the center portion should be 5mm long. I think the channel is drawn with a 1.75mm radius, not diameter, so the melting part is longer than it appears in proportion.

Toby
Posts: 330
Joined: Wed Apr 09, 2014 9:44 pm

Re: Creating 3D QR codes

Post by Toby » Fri May 09, 2014 11:01 am

I only had time to make the bottom portion of the mold this evening. One suggestion based on this: make the channel a fraction larger than 1.75mm, say 1.77 or 1.78. The filament just barely fits at 1.75. I don't see any reason to be that tight.

Here's the PLA mold and the plaster cast from it:
splice.1.jpg
PLA mold
splice.1.jpg (44.29 KiB) Viewed 11690 times
splice.2.jpg
Plaster cast
splice.2.jpg (39.71 KiB) Viewed 11690 times
splice.3.jpg
Plaster cast with filament
splice.3.jpg (38 KiB) Viewed 11690 times

Toby
Posts: 330
Joined: Wed Apr 09, 2014 9:44 pm

Re: Creating 3D QR codes

Post by Toby » Fri May 09, 2014 4:35 pm

Alas, the heated plaster is capable of melting the PLA, but heating plaster causes it to degrade in just such a way that the melted PLA bonds to a fine layer of plaster dust, which stays with the PLA when it's removed. Not something I want to run through my hot end.

There's some heat-resistant paint (for barbecues) that might work as a coating for the plaster, but it's expensive and toxic and probably non-trivially thick. Plus the plaster is still going to degrade and eventually crumble from heating and cooling several times.

I think metal is the way to go. Theoretically it should be possible to make a plaster mold and cast a low temp alloy right in your back yard, but that's more than I can do at the moment..

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Creating 3D QR codes

Post by jsc » Sun May 11, 2014 1:32 am

Picked up some Art Plaster today, they didn't have Permastone. Can I have your mold files to try myself? Do you use two in a sandwich, or just one on the bottom?

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Creating 3D QR codes

Post by jsc » Sun May 11, 2014 1:40 am

Also, there is no really good way to make plaster casts with features on both sides, is there? Other than putting on some sort of lid with projections, I guess?

Toby
Posts: 330
Joined: Wed Apr 09, 2014 9:44 pm

Re: Creating 3D QR codes

Post by Toby » Sun May 11, 2014 2:43 am

Attached are the files for the bottom mold. There's 4 .obj files in the zip. Arrange them on the bed in S3D and you'll probably see how they go together. Any questions, just ask.

As for printing I recommend separating out the one that has the filament arc and print that in two processes, with the lower one at .2 mm layer height and the upper one, just for the filament arc, at .1 mm. I got good results that way.

Actually come to think of it I also recommend scaling everything up by 1.01 just to have that bit of extra room. I never got that far, because what happened is I tested the method with just the bottom piece and that's when I found the problem with the plaster being compromised by the heat and imparting a layer of dust to the filament.

There is a top portion- it's in three sections and fits inside the bottom mold exactly over the filament as expected. The idea being the left and right portions are taped to the bottom and hold the strands of filament in place, while the middle portion is heated up and then pressed into position and held there.

I modeled the top parts but never printed them, and at the moment the files are disorganized. As I remember there was a small modeling error due to one arc being modeled with 20 vertices and a smaller one being modeled with 10, but they didn't absolutely line up the way they should have because of the differing locations of the vertices. It probably doesn't matter but at that point the project got tabled. Let me go back and straighten that out, or at least put together what I have with the small error in it and upload those files in a bit.

In terms of casting objects with geometry on all sides, yes you can do it. Here's something I cast just to test that very process:
cc.JPG
cc.JPG (86.57 KiB) Viewed 11678 times
The underside of the model is the same as the top side, so they both have that inset in them. It's a question of the mold fitting together so the last piece can go in without needing more plaster after it. Or else you model it with two extra tubes sticking out- one to pour into and one to allow air to escape. If you get into doing something like this we can compare notes. It's a lot of fun, especially if you like geometry visualization problems.

And with this filament splice thing, don't forget you need some kind of mold release or else you'll never get the bottom piece off without damaging the plaster. Like I said I use that paste wax stuff.

I haven't given up on the splicing yet. Yesterday I got a bright idea for how to do it in a simple way using soft clay to hold the strands in position and apply heat directly on the join with a soldering iron. I've assembled the components but don't have any results yet.
Attachments
SpliceObj.zip
(11.02 KiB) Downloaded 392 times

Toby
Posts: 330
Joined: Wed Apr 09, 2014 9:44 pm

Re: Creating 3D QR codes

Post by Toby » Sun May 11, 2014 4:18 am

Here's the top portion of the mold. If you make and cast three of them, they should fit together inside the bottom mold and hug the filament just right.

I haven't printed or cast these, but they look right by every measurement I made. (Except they do have that slight error I mentioned. It's very slight. Shouldn't matter. Fingers crossed.)

And if you scaled the bottom portion by 1.01, then these need to be scaled as well.
Attachments
TopSpliceObj.zip
(6.47 KiB) Downloaded 379 times

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Creating 3D QR codes

Post by jsc » Sun May 11, 2014 10:52 pm

Toby, looks like there might be an issue with TobyCad that's causing some strangeness to occur with the retaining walls. Are you sure they're fully joined up? They get sliced like this, which is not ideal:
Attachments
sliced.jpg
sliced.jpg (79.13 KiB) Viewed 11663 times

Post Reply