Differential Swerve drivetrain attempt

[<Charlie_root>]

FC_differential_swerve-01.png (99.9 KiB) Viewed 4090 times

FC_differential_swerve-02.png (117.29 KiB) Viewed 4090 times

[FC_differential_swerve-03.png (96.34 KiB) Viewed 4090 times

So this is the project I've been working on over the past month... it's supposed to be a differential swerve drivetrain for an FTC robot (you would use 2 of these on a robot, mirrored with respect to oneanaother, in the middle of a H configuration, with 4 idler omniwheels on the tips of the H for stability and odometry purposes)

So the way it's supposed to work is you have 2 DC motors, by keeping both at the same speed the bevel gears go in counter-revolution and the wheel spins. if one motor is rotating at a different speed with respect to the other, the bevel gears will be out of sync and the entire carriage (which is suspended on 4 pulley-profile rollers) will begin to rotate, hence the differential and the swerve.

It's not done yet ... I haven't placed all the bearings and such because I had to go ahead and scrap the entire thing altogether and start from scratch because as it turns out my idea to use a small crown gear to drive a large spur gear was fundamentally flawed ... because it's really supposed to be the other way around
even attempting to specify the "other_teeth" property with the actual value of 72 results in degenerate topology. the crown gear is the only part that wouldn't work: it's there just for the looks now, as i've specificaly used a smaller than actual value for the aforementioned other_teeth parameter in the crown gear object

other thing I haven't done yet is use a freely-rotating joint on one of the core bevel gears. one gear recieves all the power, but the other is there just for stabillity and is supposed to be able to rotate freely, because otherwise it wouldn't move at all

file here: https://drive.google.com/file/d/1EaZ6yy ... sp=sharing

edit: totally forgot to mention, This was done using Assembly3 using the appimage build of FC and so

OS: (K)Ubuntu 19.10 (KDE/plasma)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.19.21300 +2418 (Git) AppImage
Build type: Release
Branch: LinkStage3
Hash: 1bf376e6abb6824991a2f8f6da1f7eefd1df9ce4
Python version: 3.7.6
Qt version: 5.12.1
Coin version: 4.0.0
OCC version: 7.4.0
Locale: English/United States (en_US)

[bitacovir]

cool!! And nice screenshots!

[thschrader]

Wow! A lot of work.
Keep up the spirit!

[OficineRobotica]

Love them screenshots and complexity of the model. Beutifull

[<Charlie_root>]

Yeah, I mean I've already started working on the next version ... the plan is to hopefully finish it AND build+test before the FTC season registration (oct20)
I haven't got a lot of time
And now since I'm in grade 12 I also have to worry about exams and what not

[100th man]

Pretty cool, especially once it is working. Got a couple like that, or part way to like that.
So if I understand
the two motors are stationary in a vehicle, one drives the top other drives the bottom large bevel gears,
they drive in opposite equal rotation speed to drive the wheel straight,
a difference of drive velocity rotates the wheel axle direction,
the gears on the edge don't serve a purpose, yet.
I don't know how this is generally done but would expect one motor to drive the wheel and another to drive the axle rotation.
This seems a compact solution that utilizes the full power of two equal motors, rather than having a smaller motor for steering that is idle dead weight when not steering. And neither motor axis rotates with the wheel axis, nice.
Yeah I would probably use two or more units in conjunction for a robot but one could get around pretty good using a single pivoting drive wheel.
You designing a controller unit too?

[mhalil]

it looks awesome.