How strong can you make a 3D-printed gearbox. Would you believe strong enough to lift an anvil? [Gear Down For What?] likes testing the limits of 3D printed gearboxes. Honestly, we’re amazed.
3D printing has revolutionized DIY fabrication. But one problem normally associated with 3D printed parts is they can be quite weak unless designed and printed carefully.
Using a whole roll of filament, minus a few grams, [Gear Down For What?] printed out a big planetary gear box with a ratio of 160:1 and added some ball bearings and using a drill as a crank. Setting it up on a hoist, he started testing what it could lift. First it lifted a 70 lb truck tire and then another without any issues. It then went on to lift a 120 lb anvil. So then the truck tires were added back on, lifting a combined weight of 260 lb without the gearbox breaking a sweat.
This is pretty amazing! There have been things like functional 3D-printed car jacks made in the past, however 3D-printed gear teeth are notoriously easily broken unless designed properly. We wonder what it would take to bring this gearbox to the breaking point. If you have a spare roll of filament and some ball bearings, why not give it go yourself? STL files can be found here on Thingiverse.
I’d like to see the condition of the gear teeth after running it.
The gear teeth are very long on a cylindrically shaped plain allowing for plenty of surface area which helps keep them from wearing but I too would love to see the condition after the testing.
The gear teeth aren’t the issue. Plenty of surface area, and, it would appear, plenty of well fused mass inside to hold it together. I’d worry about the bearing balls on the plastic. The frame only has maybe 100 to 200PSI (tension- estimated from the vid with no dimensions) but where the balls contact there will be much higher contact stresses.
Remember, all of the weight is not being supported from the bearings, it’s probably impossible to calculate, but the planet gears also have 8mm linear shaft inside them and that supports some of the weight also.
>”The frame only has maybe 100 to 200PSI ”
That’s quite optimistic, or best case ideal loading with no bending stresses.
You have to remember that plastic is bendy, so it deforms out of the ideal shape the moment you put any load on it, and it creeps, soo keeping the load on it turns it into a permanent deformation.
The elastic modulus of steel is something like 200 GPa and for plastics (ABS) it’s around 2 GPa. It bends a hundred times more for the same load, so all the gear teeth get out of alignment and that makes for huge stress concentrations where the mis-aligned teeth try to mesh – the mechanism just grinds itself to pieces.
I mean. if you took the same parts and cast them out of aluminium, you still wouldn’t trust the gearbox to do anything useful. It would lift, but it wouldn’t last for very long, and with the usual standards you’d have to de-rate it down to something like 20 kilos to be man-safe to operate.
Aluminium, E = 70 GPa, Sy= 90 MPa
ABS plastic E = 2 GPa, Sy = NA, Su = 40 MPa
ABS doesn’t have a yield strength because it yields at almost all loads (creep). For a significant load, such as lifting up an engine block, it’s only a matter of time that the gearbox fails – if the load is small it fails by fracture, and if the load is large it undergoes a ductile failure.
Indeed, to make plastic lasts in applications like car bumpers, they’re intentionally made a bit soft so there wouldn’t be any stress concentrations that would lead to creep. That way they keep their shape. As soon as you try to clamp them down really tight, one year, two years, the part fractures off.
ABS isnt the only plastic, PET-G and nylon can have a much higher life and durability than aluminium for things like gears, hell nylon gears can often outlast anything but hardened steel gears.
the total load by mass is a lot lower though.
ABS was only one example. The elasic modulus for most ductile plastics is in the single digits. For nylon it’s 2-4. That’s something you just have to take into account when designing plastic gears – nylon gears can outlast other stuff, but they won’t carry the torque and power.
If you want to print plastic gears that don’t deform under load, or at least no more than aluminium, you have to make them out of stacked sheets of Aramid.
Hey! I just found out that there is an article about my video. Its pretty cool to see it here so I thought I would answer some questions.
Visibly, the gear teeth look “EXACTLY” the same. The only difference is that after the tests it will spin much easier and actually keep going for a second or two.
Why cant I log into my Hackaday.IO account here?
It’s just the way it is but I’ll link your Hackaday.io into the article if you post the link
Lots of surface area, is that why you can use plastic?
“How strong can you make a 3D-printed gearbox. Would you believe strong enough to lift an anvil?”
Coyote’s named Willie hope as well.
funny you mention this as I was trying to think of some sort of wile e coyote type pun but it wasn’t really forthcoming.
https://www.youtube.com/playlist?list=PLQ3WET7-tdyCAlKsR6ZKYF2Iv52f975d0
EMCH327 – Design of Mechanical Elements class project, 3D printed reduction gear assembly. Final lift weight was 220 lbs (99.79 kg), lifted by a 6.64 oz (188.3 g) assembly. We had to keep the design under 12 cu inch in volume so the gears are optimized.
Please turn down the volume the drill has a high pitch whine.
You, the original uploader of that video (I assume), post your video on another website. Then you disable playback of the video on external websites…
*slow clap*
Haha
So about $40 in filament? Yikes.
About $30 for the brand that I buy, so probably not worth it for the average joe, but for me it was totally worth the youtube revenue.
“Youtube revenue” aside, it’s easily worth $30 in filament to try out something new and amazing.
[GearDown], how much time do you think the design and production of the gears took? (I’m guessing that $30 in filament is nothing in comparison.)
“it’s easily worth $30 in filament to try out something new and amazing.” Granted. No one was saying otherwise. But $30 is $30.
Designing them was a 6 hour event, the most time consuming part was the printing.
Printing took all week.
Design time only 6 hours? that’s amazing, The more I hear about you the more I like you. I did take a look at your other video’s too and I really liked your 3.6 trillion to 1 gearbox, I can see myself use a few more of your video’s in future if you can keep up this sort of calibre of content.
Ill be working on my next video on Sunday, planning on doing another video on the 160:1. I plan to find out how much torque the 160:1 can handle before it fails, and that may end up being what the video is about.
So, uhh.. this is embarrassing. That guy put that video out I think more than a month ago. Which rock exactly does hackaday live under? Don’t get me Wong though, I do enjoy the quality content you bring to my life every few days. Don’t forget to repost about the company in MA that has the metal sinter 3D printer on the cheap ready to go now. Winner winner chicken dinner.
Don’t feel embarrased! Just send the tip in promptly next time.
Hackaday _is_ the community. Hackaday readers send in tips, and Hackaday contributors write them up. You are we, and we are we, and …. something something eggman.
See something cool? Send it in. tips@hackaday.com Or suffer from watching videos that were posted more than a month ago.
I want to see a Lego one tested next. Yesterday I needed a custom gearbox and I didn’t have a 3D printer, so I made it out of Legos. It turns out Lego gears are fantastic for the job because they’re more durable than your average cheap plastic and have pretty tight tolerances. The gearbox connects to my bike wheel and turns an odometer, and so far it’s survived 8 miles including fast and bumpy downhills.
Yeah I’m sure Lego use Nylon which is a great material for gears. I think without going metallic that’s the best you can go for.
Lego is made of ABS. Recently, James Bruton made his giant Lego skateboard from 5x size Lego parts 3d-printed from ABS so they were actually quite accurate.
I like how you can read the script off the laptop while he’s saying it :o) The split ring assembly works really well
Haha, I think I’m going to go back to printing the script out and hanging it on the wall.
come on, i can lift an anvil and i’m 60-70% water, just go outside and lift a car :)
So… I pay you $30 (the cost of the materials) and you’ll lift something for me anytime I want?
That’s what I’m gonna do
I pick things up and put them down.
I love that commercial
As cool as I think this is, I still have to say that it kind of loses on the news worthy front since he used metal rods through his planet gears.
Hey Guys, the New 160:1 video is done if you want to check it out! Oh yeah and Macona, I show the gear teeth at the end just for you!
https://youtu.be/Z2XkdLy9pxE