Are 3D printed gears applicable for industrial use?

I want to print some gears with ABS.

  • What will their lifespan be? How long will they last if I use them, for example, every day?
  • $\begingroup$ What sort of gears, and what size? It is possible to print gears, but the teeth need to be quite large (and tolerances will be poor). Some extruder designs use 3D-printed reduction gears. If you are planning mass-production, forget it. $\endgroup$
    – Mick
    Commented Dec 16, 2017 at 19:04
  • $\begingroup$ Worm gear. The size will be about a half sheet of paper. It is not for mass production. I want to use it for a robot. This robot will used every day for at least 10 hours. For me it is important that this arm works without damage for a few years. The size of the teeth are big. But when will it break (ABS)? $\endgroup$
    – user9350
    Commented Dec 16, 2017 at 20:16
  • $\begingroup$ How much load will the gear be under? $\endgroup$ Commented Dec 17, 2017 at 11:32
  • $\begingroup$ The steppers torque is about .5Nm. The worm gear will increase the the torque about 50-60 times. I need a Torque of 20-30Nm for the output. $\endgroup$
    – user9350
    Commented Dec 17, 2017 at 11:52
  • 2
    $\begingroup$ @user8886193 you should edit your question to add the comments informations. $\endgroup$
    – Tensibai
    Commented Dec 18, 2017 at 14:15

2 Answers 2


Survivability of parts is a very tricky topic, because a lot of factors go into it. While ABS is a common industrial plastic for molding, FDM introduces quite different challenges that can impact the time a piece lives. I can't estimate a lifetime for you, but I will illustrate why we can't estimate it for you, giving you things to think about in your design process:

Problem 1 - What's the printed internal part geometry?

FDM introduces boundries in 2 (r,z) dimensions . Not just the z layers above each other do have boundries that can and will become plane of failure, each layer consists of one filament1 that was deposited side by side to itself. These neighboring pieces (distinguishable by r in cylindrical coordinates) have a boundry that is not of the same strength as staying on the same piece and following it around (and changing ψ) a solid chunk of ABS (as you would get with molding). under stress, these boundries can crack. If you want to force your piece to have such a fate just to see how it looks: mount a 0.4 mm nozzle in a machine calibrated for 0.35 mm and run a 0.35 mm sliced print - it should be easy to crack it apart into a long snail of filament. Or declare your filament to be 3mm in a 1.75mm machine. The Horrible underextrusion and lack of pressure against already deposited filament makes it possible to unravel the whole filament at times.

Problem 2 - What is the intended use?

Use is not the same as use. Yes, it might sound unintuitive, but depending on how a piece is used, stress on the part is different. Let's take the same two gears. We put one of them in a hand mixer and a superlight drone. In the mixer it will spin rapidly against medium to tough loads (depending on dough) over medium periods (the timeframe here is usually minutes at max) of time. In the drone it will have considerably less load, but it will spin for much longer, maybe up to hours if the pilot is very capeable and the batteries last. In both cases wear and tear will be quite different.

Problem 3 - What determines strength?

Strength of the part is not only determined by the filament used, it is ALSO determined by tons of other variables. Print orientation. With enclosure or not. Humidity during print. If the surface of the part is sealed or not after the print. If it was postprocessed somehow to increase capabilities. If the piece is printed hollow or solid. How long did it cure or harden after the print... There are so many variables, that each guess would be quite wild.

Problem 4 - How to get the lifetime now?!

You can't guesstimate the reliability of a product from its design and makeup only. That is why design departments create prototypes: To rigerously test the products. This is how they learn how safe or sturdy their product is. They make prototypes and purposefully put them under various kinds of stress until they break. For gears this involves spinning them in a gearbox for hours nonstop until they break, force them against a blokaded gearbox till they break, run the gearbox dry, hot or freezing, and also under other very destructive conditions. Part of this destructive test is an accelerated life time test that, just like other tests in this stage, tries to find out the maximum parameters it is useable with. A common test for hand mixers apparently is to run them 2 minutes against some gooy substance, then stop some time before repeating.

1 - For the math inclined: the filament can be represented as a function in cylidrical coordinates, f(r,φ,z)=r(ψ)*φ(ψ)+z(ψ), where ψ the path-parameter of the filament - or in other words the length already traveled. To some degree, a G-code is generated by first creating such a function and then creating the tool path from this.


Well, to determine the time life of the gears you will need to do a test called ALT (Accelerated Life Time) but the parts should be last for a long time (not years) however this can be determined by thickness.

The torque required is not than much like a tuning up the radio volume, so if you are going to create a gear box to increase torque, just grease the parts to reduce heating on friction.


You must log in to answer this question.