2w ago

I finished my 3D Printable screw design! Excellent print ability, reliable for simple use, though the head is prone to strip... can't win them all.

If you wanna print it yourself, the model is here: https://www.printables.com/model/1348194-4n-threaded-flat-head-screws-and-nuts-10-32-equiva

Design to be a drop in replacement for 10-32 screws with a much, much higher pitch. These screws are extremely easy to print, is reliable enough that it can hold some weight.

If you wanna print this yourself, you need to make sure that the screw is sideways, so if it breaks it's no on the layer lines.

Using them in my own prints which had metal screws and they are holding quiet well.

47 comments

Torx or Robertson would probably be a better choice for head.
- I mean, theorerically, but a phillips screw is much less complicated and easier to print bc less overhang
  
  Robertson is a square. Rotate it 45 degrees and it has no overhangs at all.
  
  And given the lack of strength in these screws, cam-out is likely a positive (which is what the Phillips what's designed to do).
Have you considered a slotted head? They will scale better than a cross head, which works best with a machined fit.
Do different fills have different effects on the sturdiness? I would think that a solid fill is the best, but my kid insists different fills are more rigid.
- fuck slotted.
  fuck fuck fuck fuck slotted.
  fuck slotted all the way back to 1513 and the first clock maker who released those accursed damnable screws on humanity.
  I would rather gouge out my fingernails and toenails with a dull spoon than have to suffer through another slotted screw that's somehow worse than normal slotted screws.
  fuck. slotted. screws.
  
  Wish I could upvote you more.
  Any time I take out a slotted screw, I throw the fucker away and replace with (preferably) torx, but I'll settle for Phillips.
  
  Meh, slotted screws get used a lot. What most people don't get is the reason they have problems with slotted screw heads is because they don't use a proper fitted driver that fits the slot correctly. You always try to use an undersized driver and are then shocked to discover you have problems.
  Don't blame the tool for your poor choices. Use the properly fitted tool for the job.
- Slotted is the way to go. I've messed with a lot of drive types on 3D printed screws and I always come back to slotted, because it's the most resistant to being reamed out. Phillips, Torx, Roberson (square), and especially Allen (hex) really don't work very well when printed in plastic.
- Still testing the head designs. Should consider a slot instead since the hole doesn't take well to a metal screw driver.
  As for the fills. Wall thickness has more impact than infill due to it's size. With that said, from my testing with stylus, 100% infill bends better than 25% which I am using. But as this is not suppose to be used crooked it shouldn't make too much of a difference.
Just an FYI if you're not familiar with American screw sizes, calling this a 10-32 equivalent is probably going to confuse come people.
The naming convention used for screws in America includes the shank diameter and the pitch of the thread in threads per inch (TPI)
So a 10-32 in a #10 diameter screw with 32 threads per inch
Below about ¼ inch diameter, the American system usually uses that numbered system, a #10 screw is .190 inches or roughly 3/16
For larger diameter screws they usually just use the nearest fractional equivalent instead of the screw number, so a ¼-20 is roughly ¼ inch (actually .242in/ or #14) diameter and has 20 TPI
Most sizes have a standard coarse and fine thread, for #10 32TPI is the fine thread, and 20TPI is the coarse thread
Little back-of-the-envelope math that I'm not super confident in, this would be something like a 10-16 screw. You might want to rename it or add a note to that effect, or maybe call it something like a #10 extra coarse thread.
- I don't know much, other than I have a box of them at home which I used for my older designs, and kept with since I could buy them cheap enough in bulk.
  Wanted to make them out of plastic because I wanted to save some costs and this is what I got. I also find it funny that I got the thread which has a 15.875TPI at the end of the day. Half of what I am use to.
- Number series machine screws get their designations from wire gage sizes that they were made from. Which traces back to several different British thread systems waaayyy back in the day. While there is no hard and fast rule, the number series screws are sometimes referred to as 'Machine screws' while fractional series fasteners are referred to as 'Bolts'. No one knows why, and Whitworth is dead. (Toolmaker humor).
  All actual diameters of screws are a bit under nominal for external and a bit over size nominal for internal. This is the clearance needed to make a screw and nut fit together. This all applies to metric thread fasteners also. Thread fit classes are a story for another day.
  
  I'm not sure if the wire gauge thing is right, unless you're talking about a different system than I'm familiar with, because with wire gauge smaller number=bigger wire, and with screw sizes smaller number=smaller screw
  Also just my 2¢ on "machine screw" vs "bolt" as a casual tinkerer with various things held together by different types of threaded fasteners.
  Generally speaking if it's got a hex head or nut that I'm using a wrench to tighten, it's a bolt
  If it's got some sort of hole (or God forbid a slot) that I'm going to use some sort of a driver (for the purposes of this, an Allen "wrench" is a driver) to tighten, it's a screw.
  And of course everything gets really murky when we start talking about things like sheet metal screws, lag bolts/screws, masonry screws, etc.
Printing like this is a fun start on the path of thinking about what is possible.
In FreeCAD, there is Mark's Thread Design workbench. That includes a thread profile called the buttressed thread. The profile has a print orientation where, if you print it vertical, the thread will not produce any overhangs, like if a normal thread profile is > a buttress thread is 7. Mark has a YT upload on how to use his workbench. It is pretty easy to follow and a simple one to use.
At the stage I'm at in design, built in clips can replace most hardware. If I'm using printed threads it is usually a very large thread with some thin sleeve like clearance. I like to build splines into my threads to also create locking elements in the same space.
- Having single pointed 3"-4 Buttress threads on a manual lathe for punch presses more than once and hand grinding the tool bit from 1/2" x 1" HSS, I still have some PTSD over that thread form. But hey, if you are looking to make breaching for a 16" Navel Rifle, it's really the only game in town......
  As a toolmaker, I have seen Buttress threads used in only 3 places. Large artillery breaching, punch press ball screws, and VERY high end, (read expensive), machinist clamps. I own a pair I picked up at an auction 30 years ago in a bucket of "junk". I think they were made by another machinist for personal use.
  Edit to add: Buttress threads are directional. They can be either left or right. The choice being totally dependent the direction of the force applied to the threads. The perpendicular edge is the strong part. And the direction must always be called out.
  
  Thanks so much. I did not know the original purpose of the profile. In my experience with 3d printing, the buttress profile strength is in the opposite orientation when printed vertically. The additional length of the tapered profile creates a better distributed load across more layers of the stretching member/fastener. Still, I will prioritize overall printability without supports over thread directional orientation in most cases. I'm usually using a very large custom sized thread where the thread strength is irrelevant.
  Like here in my laptop GPU water cooler project, I am using a buttress thread and spline to retain the cooling block and pump.
  
  I worked at a shop that made large hydraulic cylinders and some of them used buttress threads.
Ackschually, you don't need to print them entirely sideways for structural integrity. Printing at an angle should also work and you don't get so many artifacts from support material.
- It's still a lot weaker than horizontal. Probably strong enough, but still significantly weaker. Personally i print horizontal without support, cutting bottom face through the head so it lies flat, though that won't work well if you need the head fully intact
Thanks for sharing! Possible to do it with a torx head?
- Anything with sharp angles will be likely strip it. Spanner Drive heads might do it, but it's an uncommon type of screw bit.
  
  why limit yourself to current bits? design and print your own bit head that's strong for printed parts.
I wonder if 3d printed driver tips would be better for matching the material strength and reduce stripping. Or maybe they wouldn't have enough strength at all and would just twist and fall apart.
I was also wondering how well printing drywall anchors would work.
- Actually a 3D printed driver might work brilliantly for this - the matching material hardness would distribute torque more evenly and prevent stripping, plus you could design a custom bit shape that's optimzed specifically for printed fasteners.
  
  Yeah, those were my original thoughts. I think it would take some experimentation to see what works.
- I print my own drywall anchors all the time and they work really well, at least for my applications!
  I use this model: https://www.thingiverse.com/thing:1288793
Is that the LTT Noctua screwdriver?
- Maybe, it was a mystery
It's hard enough getting metal screws that don't tear out from the screwdriver, what is even the point of these?
Who cares if the head strips even metal screws do that every time
- These are the size of 10-32's... if their head strips you are putting a ton of torque on it.
  
  You underestimate my power.

47 comments