Hey I’m currently trying to design a custom gearbox for next season. After talking to teams about drivetrain I found that those who attempted to use 3 neo motors in a gearbox ended up with stripped gears. And had to take a motor off. Has anyone attempted this?
Concentrating three motor’s worth of power or torque on one small gear can overstress teeth, leading to this:
Try not to do that. Design with a larger gear where the motor powers come together. That may mean using an additional stage.
Picture: 3604 Goon Squad, 2016 FiM DCMP (MSC)
Richard - that picture looks very familiar!
In my experience, most gear wear that occurs in high-power drive systems can be mitigated through the use of a small “acceleration ramp” in code to prevent sudden direction changes from causing damage due to backlash. It’s not a cure-all, but it definitely helps; along with keeping your gears well lubricated.
1018 has been using the Evo-Slim for CIMs.
3 NEOs … current limiting … no issues
Almost tempted not to post!
We use 3 Neo’s per gearbox. A 12T pinion gear on each neo going to a single 84T gear for a 7:1 total reduction. Haven’t had any issues, but I will make sure our team takes a look at our gearboxes now.
195 is running a custom gearbox with 3 NEOs per side. This isn’t the best picture, but it was most handy. Excuse the gear madness that was necessary for packaging the PTO for the climb in, there are a lot of idler gears in the transmission.
Overall redux is 8:1 with 12:48, 30:60 to the wheel (seriously… this is only a 2 stage reduction). 12T pinions and 30T spur gears are steel. No issues with any gears yet on Comp Bot through 2 events of driving and climbing, or on our Practice Bot that has seen well over 100 hours of runtime.
That is a crazy drive gearbox. Not to derail the thread too much, but how are you dealing with the backlash of ~5 gear stages between the motor and wheel? We had a 2-speed gearbox with a PTO for climbing last year and had backlash issues.
It’s actually 6 to the wheel. 12:48, 60:60, 44:48, 48:44, 30:60, 60:60
By feel, the backlash actually isn’t too bad compared to any COTS transmission I’ve used. There’s an encoder on the same shaft as the wheel (hidden by flange of drive rails) so there’s no backlash that the software sees.
Haha, 60:60, 60:60, 44:48, 48:44. Awesome.
Encoder on the drive wheel is probably the answer here. We put ours early in the reduction for the added resolution, but probably could have benefited more by reading off of the wheels directly.
This is something to keep in mind for NEO design- in theory an advantage of the NEO’s is being able to use the internal encoder to reduce parts count and increase reliability. Keeping lash down can definitely be a higher priority in NEO drives than CIM or 775pro drives.
I suspect that the “teeth rounding” isn’t really caused by the NEOs, but rather by the WCP gearbox they look to have been attached to from your pictures. NEOs don’t produce that much more torque than CIMs. The WCP SS gearboxes as sold by Vex have a nasty habit of stripping gears if the motors attached to them start to loosen or if the “snowman holes” wear out when using 11 tooth pinions. We destroyed many gears last year because of this.
There are a few things you can do to mitigate the damage if you’re unable to replace the plates with a non-snowman hole version of the plate. Try spacing the motors out gently such that as you change direction, the motors are kept from moving. In addition, try wrapping a 6" clamp around the motors to keep them from walking.
These holes are your enemy for anything WCP SS or DS. We’re moving to modified EVO gearboxes next year. I’m tired of dealing with this.
We’re running NEOs this year and had problems with the internal encoders. It’s possible that the problem was either software on our end or firmware on REV’s end. Regardless, all of the gearboxes were designed so that we could drop in MiniCIMs or CIMs if the NEOs weren’t working, so they have Grayhill mounts which we ended up using instead. Those have been working reliably.
In years where we ran these gearboxes we epoxied the snowman holes and drilled out only the set we needed, worked great all season.
We just had our sponsor cut out a few plates with the snowman holes replaced with regular holes. I haven’t seen any aluminum glitter coming out of the gearboxes all season thus far.
Not having used those gearboxes before, I didn’t realize they had that feature. If a student proposed this to me as a gearbox design, I would suggest they go back to the drawing board.
The snowman holes are certainly not the issue here.
Firstly, this is your drivetrain gearbox. The motor bolts should have loctite on them. They should never get loose, and should start off torqued down.
In the ~6 years of my team using these on our final bots, practice bots, and off season bots, we have never had any issue relating to these.
We had issues with the snowman holes back when we used WCP SS gearboxes. Can confirm adding loctite (and a washer to help the screw not dig into the aluminum) worked.
Not specific to your question but this is a video (by @PatrickW of 2910 swerve fame) that I found really helpful for designing gearboxes and such. Check it out.
@pchild 401 has a pretty nice 3 neo drive train if they wanna share.