I had been thinking about buying a vice brake but I kept putting it off since it felt like at $10/inch I could just make my own and have all the fun of making one. Now I learn that maybe it makes sense to just print the brakes I need as I need them! Commercial fabricators are finding that just plain cheap PLA is working surprisingly well for brakes. Just think though layer direction to manage stresses well.
I’ve been playing with an openscad model for a vice brake today. I think that it would be possible to provide a bunch of different angles in the anvil block to facilitate different angle bends, but I’m still thinking what that would look like. In the meantime, here’s a mock-up of the assembly as I have it so far. It would sit in a vice.
There are several different sizes of fingers that you would slide into the block. If anyone wants to print and try it out, let me know and I can send you my work-in-progress to play with.
I haven’t printed this yet, and won’t have a chance at least until the weekend, but… I have now accounted for the angle of the stock in the shape of the base block holding the vice fingers, and have parameterized the included angle and allowed for stock thickness in the anvil.
If you try it out, feedback appreciated. Feel free to ask questions about how to modify the parameters to get a desired result.
Put a couple of magnets onto the angle so it wants to stay in place in your vice. That’s one of the nicest features of the pre-bought one that I have. I also like that you could custom print fingers in such a way to allow for tabs to pass through for more complex geometry. I think with thinner ductal material like steel it would work pretty well. But I have to put a heck of a lot of force when I’m bending 3 mm aluminum on my steel version. I wonder how PLA would hold up?
Yeah, I have magnetic nylon soft jaws for my vice, and was thinking about magnets as well. But I didn’t want to embed them in the blocks because of worrying about weakening them. Maybe I could put holes for magnets in the lip that sits on top of the vice. I’ll look at that.
The company quoted in the article says that 12 gauge is the highest gauge of steel they bend with PLA on a commercial basis. I think that 3mm aluminum might be a bit more than this design would work well with. But here’s a profile of a heavy-duty version with 10mm thick fingers and 25mm tall blocks:
I have been wanting to purchase an inexpensive benchtop shear/brake for awhile but now I think I might do something like this instead.
I am not sure if you have seen the Hackaday article on someone who has already done this;
Thanks @Eclsnowman — this looks easy enough:
Looks good. Don’t forget that unless you use architectural steel, you will have an inside radius on the angle iron. That will hold the top plate up slightly from the vice jaws. Perhaps just magnets on the underside of the angle would hold it up enough to mitigate the Inside corner radius.
I wasn’t thinking angle iron. I was thinking that this whole thing is 3d-printed. The angle is printed separately from the finger and anvil blocks so that it prints without supports. I was expecting to glue (or maybe tape) the blocks (green, above) into the frames (red, above) and similarly drop magnets into the holes in the frames, which are meant to have just one layer of plastic under the magnet to keep it from just dropping through.
Here it is as oriented to be printed:
Gitlab has an interactive 3d viewer for the model that will let you zoom and rotate the model.
I could add to instructions the option to instead cut off pieces of angle iron for the frame, in which case you would use double-stick tape to hold the blocks to the outside of the angle iron, and just tape magnets under the top of the angle iron to hold it in place on the vice.
Seems hackaday has picked up on this as well. Nice little video with bending parts.
I printed (all perimeters, no patterned infill) and tested today, starting with a piece of spring steel scrap.
Initially, the forces on the anvil are towards the vice, but as the metal bends, there are greater and greater forces pushing the sides of the anvil apart.
The anvil broke, and two of my magnets are somewhere in my shop. (When I run across them in five years, holding on to some random piece of metal, I’ll wonder where they came from. Yes, I was wearing eye protection while testing.) The fingers seemed to be fine, and the design for holding the fingers seems to work easily enough.
I’ll have to think about a more robust anvil design to handle the side loads.
New cross section:
The frames are now different from each other, so using different colors to help keep them separate. They are also thicker.
The anvil has more plastic at the base, and the blocks are wider.
I’m suggesting 100% or so rectilinear infill at 90⁰ for the anvil only for greater strength in the direction the first version broke, and 100% perimeters for the other parts. Haven’t printed this version yet. But I do have some 3mm aluminum set aside to test.
Also thinking about holes for bolts and retaining plates top and bottom for the anvil, but I’d rather find a design that doesn’t require that.
What are your thoughts about printing them standing up verses laying down how you currently have them, so that way the stress is always compressing into the layer lines as opposed to forcing across them.
Frames obviously (a least to me) need to print “laying down” as I have them, but that’s an interesting thought for the other parts. (I have a bias for laying things down because I’m printing on a bed-flinger until I get back to my corexy rebuild.)
Most FFF strength tests I’ve seen are under deflection loads rather than pure tension or compression loads. I have intuition here but I don’t have much faith in it, and I don’t have equipment to test reliably.
I have only skimmed the following recent paper reviewing research as of earlier this year, but I note that “stacking direction” results in it are slim at best.
This thesis, in my opinion, could have used some editing:
I think I won’t spend $40 to buy this article:
But local universities might have that journal available.
My parts here that are under the largest proportional pure compression load are the fingers. I would expect their dominant failure mode to be crushing/buckling, and I would expect highest strain to be on axes normal to the vice travel. I think I’ve talked myself into changing from printing some parts with all perimeters (for which I’ve seen recommendations for parts under compression load) to printing all the parts with high percentage rectilinear infill at 90⁰.
I think I could print two otherwise identical anvils, one printed horizontally and one vertically, put a solid metal cylindrical rod between them, put them on a hydraulic press, and see which one fails first. That would ensure that they are seeing the same force so it should be a reasonably fair test. I wish there were a gcode onebox widget to render this set of pieces to test for anvil and fingers:
…Well, it looks like printing the fingers on end is not a winner:
The magic pile of flying spaghetti where the on-end finger fell over is pretty cool though!
@Eclsnowman for the win! Here’s the setup:
The top piece was printed horizontally, as I had it, and the bottom piece was printed vertically, as @Eclsnowman suggested.
The horizontally-printed piece broke. Here’s the test:
Here’s a section of the edge:
Note that when it breaks, it breaks with lots of energy. I’m starting to feel that the 3d-printed anvil might not be a great idea. At any rate, using eye protection every time I use this seems like a really good idea. I added more explicit warnings to the source file to make it clear that this is a bit dangerous.
Now I’m wondering about inexpensive ways to contain failure. Maybe I could wrap the anvil in spectra/dyneema fishing line? I don’t expect it would add operational strength, but it might make failures slightly less exciting.
I always wondered about putting a pause in the gcode at several layers and integrating spectra as a reinforced strand.
Well, UHMWPE glass transition temperature is about 120°C and melting point is around 130-136°C, so much lower than printing nozzle temperature… I would expect it to compromise the integrity of the strands too much to trust.
Ahh, I never knew the raw material Spectra was made from. One option when printed vertically is an inserted safety wire:
The wire is an interesting idea. It wouldn’t stop the print from breaking (the removed print material would actually make it a little easier to break), but it should keep the pieces from going flying if it does.
For reinforcement, I often just put screw holes through a part. Particularly vertical ones. A lot of the parts I make are designed to be mounted to something against the top/bottom face using a screw that it longer than you might otherwise use, so that the screw will provide a compressive force on the print along its Z direction and brace it against shearing in X/Y.
While I still think that printing vertically is the way to go for this piece, laying it flat and adding holes for screws through the print’s Z direction (which might also be used for mounting it) should increase the strength significantly, because the side force would be pushing on the lower layers that reach all the way across, and not just the upper ones that are narrower (assuming the base is thick enough and the holes tight enough that the screws won’t splay outward). Even if they’re not effective enough to prevent the print from breaking across them, they would serve the same purpose as the safety wire idea (though if they do their job well enough that the failure is through the base instead, all bets are off).
I’ve been trying to avoid solutions that are a hybrid of machining and 3d printing, because as soon as I’m machining, I might as well just cut the anvil on a mill.
I’ve thought of printing the anvil a size that fits inside metal c-channel, and a frame to surround it. Probably most people with a vice have a hacksaw…
Safety wires, or threads, could go straight through instead of being curved segments. Cotton, or maybe synthetic blends with high temperature plastics (this does not include nylon or viscose/rayon), could be printed in and contain most of the force I think. Or just drill holes and thread after printing. I had thought of screws and bolts, with and without metal plates, too.
But ultimately I’m looking for a design that has a meaningful benefit over paying $10 / inch so if it ends up too much work to safely print it’s not a winner and this goes into my “journal of negative results” file, because we all know that for a lot of people, inconvenient safety features get ignored. So I keep pondering.