pic: Average 2014 Chassis

[Richard Wallace]

loading photo ...

[apples000]

Originally Posted by Richard Wallace

-- 3 CIMs per side
-- 28"W x 27.5"L
-- Grayhill encoders
-- Four AM 4" Performance Wheels with Nitrile Roughtop
-- Two VexPro 4" Dual Omniwheels
-- #25 Chain on 34 tooth sprockets with simple roller tensioner
-- AM gears replaced by VexPro gears to reduce weight and friction

How do the Vex Pro gears reduce friction?

[Richard Wallace]

We measured about 17 Amperes free current using the stock AM gears, and about 10 Amperes free current after replacing with VexPro gears. Gears were well greased for both tests, and the results were repeatable.

AM steel gears appear rougher, while VexPro 7075 gears have smoother finish. Also, VexPro gears are AGMA 11 -- closer tolerances, better fit.

The free current really tells the story, though. Lower friction, less power wasted.

[JamesCH95]

I believe the VexPro gears are coated with some Teflon/ceramic mixture that reduces friction. In addition to the reasons mentioned above.

[Ian Curtis]

Originally Posted by Richard Wallace

We measured about 17 Amperes free current using the stock AM gears, and about 10 Amperes free current after replacing with VexPro gears. Gears were well greased for both tests, and the results were repeatable.

AM steel gears appear rougher, while VexPro 7075 gears have smoother finish. Also, VexPro gears are AGMA 11 -- closer tolerances, better fit.

The free current really tells the story, though. Lower friction, less power wasted.

Wow! That is a big difference. For those of us that don't know much about gears/wear-in/etc. would the AndyMark gears improve over time as the surface finish improves? I think I remember someone (971?) ran their gearboxes on a test stand prior to mounting them on their robot and found it wasn't really worth it. A 50% reduction in free current sounds like something that would be worth it though.

[xForceDee]

Is the center wheel dropped? If so, what was the thought process for using omni's? Last year, my team did the same thing sort of against my wishes because I didn't see the purpose. It seems to me that as long as you power every wheel and only 4 can physically be on the ground, then turning will never be an issue. Despite that question though, looks like a great drive train. Nice job guys!

[Richard Wallace]

Originally Posted by Ian Curtis

Wow! That is a big difference. For those of us that don't know much about gears/wear-in/etc. would the AndyMark gears improve over time as the surface finish improves? I think I remember someone (971?) ran their gearboxes on a test stand prior to mounting them on their robot and found it wasn't really worth it. A 50% reduction in free current sounds like something that would be worth it though.

Free current has two main factors: (1) internal rotating losses in the CIMs, about 2.5 Ampere per motor so 7.5 Ampere for three motors, and (2) mechanical drag in the rest of the drivetrain. So the reduction we saw using VexPro gears is really about a factor of four; i.e., 9.5 Ampere reduced to 2.5 Ampere. That is about 85 extra Watts of mechanical power per side. Not chump change.

[apples000]

Originally Posted by Richard Wallace

We measured about 17 Amperes free current using the stock AM gears, and about 10 Amperes free current after replacing with VexPro gears. Gears were well greased for both tests, and the results were repeatable.

AM steel gears appear rougher, while VexPro 7075 gears have smoother finish. Also, VexPro gears are AGMA 11 -- closer tolerances, better fit.

The free current really tells the story, though. Lower friction, less power wasted.

Wow, that's really neat. I thought that the fancy coating might help, but I never imagined it would be that much of a difference. I got to play with a few vex gears today, and they're really high quality. They're also pretty cheap (as AGMA 11 gears go...)

[Richard Wallace]

Originally Posted by xForceDee

Is the center wheel dropped?

We didn't drill any extra holes for axles in the AM14U chassis, so yes, the omniwheel axles are about 1/8" higher than the other two axles when the chassis edges are parallel to the floor. This is effectively about 1/16" center wheel drop. I agree that a custom chassis would not need a center wheel drop for this kind of omniwheel setup.