Gearbox failure?

So I built the pillow block described in the last post. First I 3D printed one in about 2.5h and when I went to install the gearbox I snapped the block in half along the layers (what's commonly known as 'delaminating').

Then I designed and laser cut a replacement in about 20 minutes. So there. Ask your local make if 3D printing is right for you.

The new block was assembled and a 45cm bar was attached to the shaft. Now at 55:1 reduction on a 50 N·cm stepper motor should give me 2750 N·cm, or a holding torque of ~61.1Nm at 45cm. That's a LOT of torque for such a small motor.

The new block assembled, the gearbox stalls even with my hand on the bar at the 10cm mark. This thing should be almost unstoppable.

I don't have enough engineering experience to solve the mystery and I'm led to believe neither the machinist who made the parts nor the designer who drafted them can spot what's wrong. I've got some expensive ideas, but I'd really like to get a Mechanical Engineer to take a look and tell me what's going on.

So... two steps forward, one step back. Next step is to find the Mechanical Engineer who can help.

Discussions

TTN wrote 02/27/2015 at 08:52

Wait, so your 3d printed part delaminated? Usually, this indicates you're printing at too low temperatures.

Are you sure? yes | no

Dan Royer wrote 02/26/2015 at 20:05

@Kert - I have. I even wrote my own https://www.marginallyclever.com/other/hypocycloid-generator.html good link!

@zakqwy - I can't yet. If I get the gearbox working it'll probably be my first Kickstarter.

Are you sure? yes | no

zakqwy wrote 02/26/2015 at 20:51

Share it = we help you?

Are you sure? yes | no

Dan Royer wrote 02/26/2015 at 20:58

I have some local people helping me offline. I can't do more than that at this time. :(

Are you sure? yes | no

zakqwy wrote 02/26/2015 at 23:32

No time, or IP concerns?

Are you sure? yes | no

Dan Royer wrote 02/27/2015 at 00:06

ip.

Are you sure? yes | no

zakqwy wrote 02/27/2015 at 00:25

http://i.imgur.com/DWSqMrY.jpg

Are you sure? yes | no

Dan Royer wrote 02/27/2015 at 01:14

I've already had one company try to hire me so they could get the plans, absorb the IP, and then fire me. It was laughably transparent. So.

Are you sure? yes | no

Kert wrote 02/27/2015 at 05:36

Hm. It is indeed odd problem you are having in there. As nothing in breaking down mechanically with your new laser cut part instead of 3D printed then I can see few possible causes for that (do note that I'm not a mechanical engineer tho):

(1) Motor not putting out the rated torque value in reality.

(2) Some kind of inefficiency (for example, friction) consuming significant amount of motor torque.

(3) Something with the gearbox design process (unlikely, as with careful checking of calculations that would be found quite likely)

I personally would start with assumption (1), presuming its a generic stepper from china or if using microstepping (which is said to take the motor torque fair notch lower)

Are you sure? yes | no

zakqwy wrote 02/26/2015 at 18:28

Can you post more details? Prints, CAD files, more pictures, that sort of thing? I'd love to see more information on your gearbox. No backlash and 0.002 degrees per step sounds great, but I want to see it!

Are you sure? yes | no

Kert wrote 02/25/2015 at 05:24

Have you tried any of the online gearbox engineering calculators? Surely something like that should exist - although I have had no need for that so have none bookmarked. At least the gearbox stuff seems to be well documented for robotic uses, like http://www.societyofrobots.com/mechanics_gears.shtml for example. Then again you already seem to have all the numbers figured out so the issue might not be on the theoretical side but something related to practical application.

Are you sure? yes | no

Gearbox prototype made

Gearbox 3 Attempt 2

Discussions

Become a Hackaday.io Member