[How to Mechatronics] on YouTube endeavored to create a comprehensive guide comparing the various factors that affect the performance of 3D printed gears. Given the numerous variables involved, this is a challenging task, but it aims to shed light on the differences. The guide focuses on three types of gears: the spur gear with straight teeth parallel to the gear axis, the helical gear with teeth at an angle, and the herringbone gear, which combines two helical gear designs. Furthermore, the guide delves into how printing factors such as infill density impact strength, and it tests various materials, including PLA, carbon fiber PLA, ABS, PETG, ASA, and nylon, to determine the best options.
The spur gear is highly efficient due to the minimal contact path when the gears are engaged. However, the sudden contact mechanism, as the teeth engage, creates a high impulse load, which can negatively affect
durability and increase noise. On the other hand, helical gears have a more gradual engagement, resulting in reduced noise and smoother operation. This leads to an increased load-carrying capacity, thus improving durability and lifespan.
It’s worth noting that multiple teeth are involved in power transmission, with the gradual engagement and disengagement of the tooth being spread out over more teeth than the spur design. The downside is that there is a significant sideways force due to the inclined angle of the teeth, which must be considered in the enclosing structure and may require an additional bearing surface to handle it. Herringbone gears solve this problem by using two helical gears thrusting in opposite directions, cancelling out the force.
Some key recommendations include increasing the wall count to five passes, and using a minimum infill of 35%. After testing a few filament types, it was somewhat inconclusive as each brand of filament has different breaking strengths due to factors such as age, moisture absorption, and color. It seems that using PLA is still a safe and cost-effective option compared to other fancier materials. The conclusion regarding gear type is to use herringbone gears whenever possible, but ensure the use of bearings that can handle the increased axial force compared to spur gears.
The reason his 100% infill ones weren’t stronger was because even at the lower infill percentages the gear teeth tips were solid already, due to perimeter counts. Even at fairly low infill you get solid tips on the teeth, high infill then only serves to increase the overall strength of the gear between its teeth and whatever else is mounted upon its axis. For small gears I’d always recommend 100% infill, but it increase overall strength, not tooth strength, small teeth are already as strong as they can be at anything with 3 or more perimeters. The age of the filament was probably not the dominant factor here.
Being able to 3D print gears has saved my bum on countless occasions. I’m still on a quest how to easily generate mating Herringbone transmission gears in Fusion. Anyone?
I had to design and print a gearbox for a class recently and I used the Helical Gears Plus addon. Worked great for me. https://apps.autodesk.com/FUSION/en/Detail/Index?id=1259509007239787473&os=Win64&appLang=en
This was a rather unsatisfying conclusion, for a lot of work: All materials are roughly the same.
Test setup, using 2 to 3 inch gears. What loads do you expect when using gears that size? Is there a point to still use plastic? Was this all academic? Isn’t there a book from 1880 that shows what precision is needed for spur gears that size? Where is my coffee?!?
regarding your question “Where is my coffee?!?”, see the link below
https://hackaday.com/2024/06/12/supercon-2023-reverse-engineering-commercial-coffee-machines/
“This was a rather unsatisfying conclusion, for a lot of work: All materials are roughly the same.”
But that’s a GREAT practical conclusion! If I can print a herringbone gear in PLA or in PAHT-CF and get functionally the same performance, then not only does printing in PLA reduce filament cost, it also means I can avoid the need for a hardened 0.6mm nozzle and switch back to a standard 0.4mm nozzle, or even go for a 0.2mm nozzle for improved X/Y resolution. Which then allows either finer modulus gears, or less error per tooth for the same modulus (which means smoother operation, which itself reduces wear and heating, and increases service life and reduces operating noise).
Null results are just as useful as “this particular filament for this particular situation is the best in this particular metric” results.
Plastic and wood aren’t ideal materials for gears. Use metal instead. 🙂
I think you’ll find an abundance of plastic gears. Even the Ford model T used a bakelite-like plastic gear to drive the camshaft.
As for wood gears: https://woodgears.ca/
A fantastic example of an absolutely terrible use of a plastic gear
Sometimes you want low durability: some mechanisms use plastic gears so that if something gets stuck, the gear gets destroyed, and the rest of the mechanism survives.
Also to concentrate normal wear. For example, kitchen mixers use a worm drive, with thte screw cut into the motor shaft and one or two plastic wheels engaging with itt, which are easier to replace.
I see. I’m confident the usual plastic being used by 3D printer owners will meet this requirement.
Yes, and one invariably discovers that the plastic gear is no longer available as a spare…
Laser-cut wood seems pretty convenient for all that it wouldn’t be a like-for-like replacement gear material. You can still get the friction down with some more effort. Also, older woodworking included various things that were gear-like or screw-like, and they lasted fine even if they’re not in the same category as precision metal gears with modern oil and such.
Plastic can also be cheap of course, but it can also have low friction and low noise/vibration, and it can be more cost-effective to produce all kinds of complex little mechanisms with other materials than metal. Or you can just let it be the sacrificial wear part.
You can always use burnout casting for the gears, gotta get the form right with a printed positive first though
Depends on the application, millions of very reliable things have plastic gears in them and a lot of cars have plastic gears in gearboxes and engines, not to mention other mechanisms like wiper motors and all of them work as well or better than their metal equivalents.
Wow. Look at the comments so far. I went to bed last night in my own world but did I wake up in some strange parallel universe where all or even most people have the ability, equipment and a safe space for working metal? Is the metal shop a standard room in every house just like the bathroom, kitchen and bedroom?
I mean.. sure that would be the best material to make a gear from something like 99.999% of the time. But most people aren’t in a position to do that. And a lot can still be accomplished with a 3d printed plastic gear!
+1
PLA gears, especially for quick repairs, can get you out of trouble quickly. Watch the load and maximum temperature and you’re fine. Those gears can be sturdy as frick!
Also… I’ve heard of PLA used in the creation of “lost wax” (lost PLA?) molds. Couldn’t you, therefore, use these prints as the basis for casting aluminum gears?
Lost wax, lost foam, lost plastic … the proper name for these processes is “investment casting”.
You would need to machine them afterwards though, which kind of defeats the point.
Depending on the tolerances required, hand filing the aluminum (or zamak) gears would be fine.
I’ve 3D printed some nylon gears for the cold end extruder (bowden type) and after 3+ years they’re still working and going strong.
You’ll find nylon gears in a lot of smaller devices like 2D printers, shaving/trimming machines, fans etc. and they last a long time.
Anything high torque, yeah metal is probably the best material.
Those comments seem to have gone missing.
The are a few reasons to use plastic gears. The first is noise reduction, plastic gears are just so much quieter than metal, even just putting one plastic gear among your metal gears will help reduce noise. The second use of plastic gears I see is having a sacrificial gear, i.e. if something goes wrong this gear will get stripped and save your metal gears. My small lathe has a plastic gear so that if you crash the carriage you only need to replace the plastic gear (not that I ever crashed it carriage).
Even 3D-printed change wheels can be entirely adequate on a small lathe, where little force is applied and backlash is generally taken up before the tool is advanced to contact the workpiece.
However in the general case the hacker/maker community appears oblivious to the vast amount of effort that- over the course of a couple of hundred years- has gone into perfecting the surface finish of gears. If an application originally used wheels with polished teeth, a replacement with even microscopic layer lines or made out of a material that is basically glue-bonded sawdust is rarely an adequate replacement.
Well, of metal working is out question, other materials such as plastic, clay or cardboard can be used as alternatives, of course.
I agree with your conclusion but not necessarily the premise. Even under heavy restrictions and without much budget you can do a few things with metals, like mix their powders into resin or plate the surface of something. And there’s some alloys at soldering temperatures and below that may not be that bad depending on the purpose?
I found PLA gears quite durable. Under a light load – a gadget I built spun continuously for two years with no detectable wear (Backlash didn’t seem to increase..)
nylon has been used as a gear material in industrial machines,10hp+- electric motors on an air compressor
built into a power hammer that I have seen
but there is nylon and then there is nylon
gear trains have less vibration if they are odd toothed
ie: that the same teeth do not hit each other on every
revolution
soft starts on motors help
and torque reversals eat gears
plastic gears are OK, pulleys not so much.
Lost two teeth on the gears of my dremel screwdriver just yesterday night. Metal planetary gears ;(
Not the same as gear-to-gear but the belt-to-gear transmission of my electric longboard is printed in ABS (iirc). 200+ km and the only noticable “wear” is evened/smoothed out layer lines. Here is a picture of a pair of gears after about 100km https://github.com/HateFindingNames/MotherBoard-Drivetrain-System/blob/master/Footage/Pictures/Wheel%20pulley%20v1.jpg
ok, very informative, I like the article a lot.
BUT ( yes I know, I’m always a butt)
helical gear trains dont reverse. Its “warp factor 10 mr Zulu” or nothing.
Straight gears, yes noisy, but you can back up.
Now, being an Englishman perhaps I’m a little odd,
but I do like the chance to back up when I make a mistake.
Not often, but it does happen.