After learning of how many teams used 775’s and versa-planetary’s on their drive trains my confidence in this working out good is high.
Still ~18:1 overall ratio with 4" wheels. May go higher reduction to make sure we can always spin the wheels on carpet.
2x1 crossbeam added for stiffness. It is set just above the 775’s so that if you laid a flat sheet over the top they wouldn’t be touching it.
2x1 chunks added to exterior drive plate to facilitate bumper mounting, 2x2 angle for full length bumper mounting support rules.
Sitting at about 25 lbs with no chain or wheels. hoping to stay under 30 with everything except electronics.
This entire setup exists in a space 5.25" tall. With wheels it’ll be another inch.
Overall I’m happy with how it’s coming along. If anything just for the bumper mounting because almost every robot i’ve worked on has been a nightmare to take bumpers on and off
This still looks pretty weak. I highly recommend at least adding another crossbar but even then I can see this entire chassis twisting after a big hit. Drivetrain problems are not problems you want to have, If you can’t move you can’t score. Not being able to move is not fun.
Yeah, my idea is to have whatever side of the frame that has mechanisms on it, the opposing side will have another cross beam.
I left the front and back open for now for whatever scoring mechanism we need for the game, the super structure for that will add more cross supports.
When we build this for real to test over the off season it’s gonna get rammed pretty hard so we will see.
Take a look at 610’s design tutorials here. They’re a few years old, but all of the information about making custom drivetrains/gearboxes is still relevant and very helpful
Not everything has to be about winning. If it works and the students like it, no big deal. This does have the added advantage of remaining under 2" tall along most of its length, so a plate can be placed right on top of the side plates to protect electronics and motors.
That being said, the way you’re using plates over 2x1s still rubs me the wrong way. Your side plates are still way less stiff than they should be or could be with just tubes.
If you add a crossbar opposite the side of the manipulators the tweaking problem might be ok, but even if you do that I would still move your center crossbar closer to the end of your chassis.
Also, using 6 775pros DOES NOT give you more power than a CIM drive! Theoretically if you ran all the motors at 60A, you could generate more output power, but in practice you won’t be running your 775s more than half that much. That’s why most of the 775pro drivetrains this year and last year were geared so slowly (relatively of course; 15fps is fairly slow for this year in my book).
It also has the advantage of folding like a taco along most of its length. Which is why I was trying to get some of the apparent advantages out of OP so any suggestions on how to remain mobile for more than the first 5 seconds of driving wouldn’t interfere with them.
It should be equal too or slightly more than a 4cim drive.
Which the whole thing I’m going for.
Also it’s all 1/8th plate, I don’t see how it could fold like a taco with that 2x1 going across the middle. Even without more cross beams it should be pretty stiff.
How does this compare in cost to a Kitbot, or even a custom chassis with more standard COTS drive gearboxes? Also, I second Anand’s comment about using some sort of tubing. Right now, you’re just asking for misalignment on your shafts.
6 775pros may be equal to a 4-CIM drive in some respects, but keep in mind you are gearing this for something below 15fps. 1072 and 299 geared for 17fps easily this year, and 115 geared for 21+fps off of just 4 CIMs. Doing that with a non-shifting 6 775pro option is asking for trouble.
Right now your chassis is H-shaped. If you add a crossbar on one side, it’s a rectangle with two pairs of spindly plates sticking out connected with standoffs. A good hit from the side will bend the chassis inwards, not to mention ruin your bellypan.
That’s why I said “Doing that with a non-shifting 6 775pro option is asking for trouble” because unless you make it traction limited at a low current, the 775s will burn out quickly. CIMs are much more forgiving and tend to die long after a brownout has occurred.
Well we have almost all the metal already, the 775’s themselves are cheap and we have enough extra vp’s to build one of these at the moment.
The last two year we have been using 2x1’s with cantilevered wheels this year, and non cantilever wheels in 2016. The wcp gearboxes we used are like $400 each! So the cost should be realtively equal.
Yes. I’ve said multiple times now that this is going to be in a perpendicular drop drive/h-drive setup anyways.
It’ll probably only have one traction wheel on each side and the rest be Omni’s or mecanum wheels.
If I can bend 1/8" plate with my hands, then almost any robot going above 10ft/sec probably could. I would seriously consider looking into making this out of Rev extrusion.
I think you’ll be fine with 1/8th plate if you tie the 1/8" plates to the beam and if you are pretty generous with your spacers between the plates. I’d even consider making C channel out of small lengths of the same profile you use to make beam and then you can install them at the bottom between the rails or even here and there between the chains (since when have FIRSTer been scared away by nightmare assemblies?
1/8 plate is pretty strong stuff if you know how to design supports.
Also, don’t forget that you have a bumper that ou can potentially use to strengthen the chassis.
