I am trying to make custom parts that fit directly onto a servo. Doing this has proved more difficult than I've expected so far.

I was hoping to avoid incorporating the provided servo horns into the 3D printed part, so I've been trying this method out. Below are images of my current test - a 3D printed attachment to the servo, with an indentation for an M3 nut (the servo accepts an M3 bolt) for attachment to the servo. The plastic ring doesn't have the spline (I can't print that level of detail I think) but is tight around it. The top piece attaches to a 3/8" nut for use with the 3/8" threaded rod I had laying around.

The nut casing

The servo attachment piece - note the M3 nut indent

So far, I'm having difficulty of this setup working with any level for torque and not just spinning in place.

So... is this the correct approach? Am I going to have to design a piece with the servo horn inside of it to get the servo to connect? Are there better approaches I haven't considered?

  • $\begingroup$ Why do you not want to integrate servo horns into the print? $\endgroup$ Feb 14, 2016 at 13:11

2 Answers 2


Sounds more like a material/mechanics of materials problem to me. I'm assuming this is an FDM part, and the M3 nut is metal? The nut will have no trouble stripping this. FDM parts are far from 'full strength' parts, nearly all 3D printing is. There are too many 'spaces' inside of the print itself. This gets compounded by your design itself, since this part leaves quite a bit of room for leverage. See how the holes that allow the two piece of mate together are almost the same distance as the diameter of the nut? That is too far for a plastic part interfacing with a metal part. You need to picture an imaginary lever from the center of each of those six pins to the center of the nut, then another six levers from the center of the nut to each vertex of the nut. The plastic part is going to be able to exert far greater force than the nut, but the plastic will deform and the nut will not. The plastic is set up to destroy itself right now, and the nut will make itself a nice hole to spin inside of. Either figure out a way to get it made from metal, or decrease the size of the printed part. I would recommend metal if you can afford the weight and plan on applying any kind of torque to this system. Get quotes from some rapid prototyping services or local machine shops to get a few of these made from steel, it will probably be cheaper than you think. Just be willing to work with them, and take their advice if they suggest a design change to aid in manufacturability.

All that said, you would be better off interfacing with the servo horn if you can fit it into your design IMO. It would be better able to handle any torque in the system, regardless of material.

  • $\begingroup$ Spot on - this is a standard hobby grade FDM printer. I'm printing out servo horn attachments now and I'll try that route. I was hoping to avoid requiring metal but you might be right about it. I live in a city apartment so am limited on some material selection. I will also try the same design in nylon - I hear that is quite a bit tougher, but I'm to date inexperienced with it so I'm hoping that it isn't too hard to work with. $\endgroup$
    – HlfShell
    Feb 14, 2016 at 3:01
  • $\begingroup$ Nylon should be tougher, but know that it may still not be tough enough depending on the usage. If you find that Nylon still isn't enough, a rapid proto service would be no more difficult to call upon from a city apartment than Amazon would be. $\endgroup$ Feb 14, 2016 at 15:08

Much as stated above the power of a servo is limited. You may need to instead use a light interrupted motor as are found in printers combined with a series of reduction gears to allow the force needed to suspend your robot without adding much weight.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.