pic: Experimental VP drive gearbox



Experimental VP drive gearbox I designed (still missing things like fasteners). I would use threaded rods to connect the 4 plates together. The gears are attached to the CIM using the andymark 8mm to hex adapter.

Can you explain what benefits you were looking to get with this design over a standard single speed gearbox? I’m not sure how this would save weight (pretty sure this would weigh more than any custom single speed gearbox) and it doesn’t seem like it cuts down on custom parts much. Just wondering if there was some other objective driving this design, or if you were just doing it for exercise to see how utilizing a VP for a drive train gearbox would work.

I did it mainly as an exercise, and just to make something that was compact and looked cool. I wasn’t really looking to save weight, I was more focused on saving space. I initially had this idea looking at your planetary drive a while ago, but never really got the time to do it. I was also inspired by 1296’s 775 pro vp drive gearbox. At the end of the day this is probably that is never going to get made.

Here is another view of the gearbox:
https://gyazo.com/5bb9699132af2e6b6ffa06fb561d478a
link: https://gyazo.com/5bb9699132af2e6b6ffa06fb561d478a

Are you sure that the VP will hold up with 2 CIMs powering it?

The load ratings sheet for VPs shows very different charts for a 775 motor and a CIM motor, let alone 2. While this design may hold up okay for a 775pro or 2, I think this may put the VP out of spec with 2 CIM motors driving it.

I would be pretty cautious about 1 CIM going into a VP, let alone 2. Especially in a dynamic loading situation like a drivetrain.

Edit:

What transfers power from the initial gear reduction into the VP?

Picture did not load for me.

Yeah that is something that I definitely did not consider when designing this gearbox. Talking to a student on 1296 about their gearbox, he said that they welded the pins and bored the sun gears and added needle bearings, but considering the step up from 2 775 pros it probably would encounter problems. It would be interesting to see how a GEM would work for this application.

Interesting design challenge, but for a single speed just minimizing the amount of gears (single reduction ideally) and minimizing the machined parts tends to be the best way to go. No need to get fancy with a single speed gearbox, that’s the main benefit; simple, light-weight, and saving space.

The reason the VP load rating charts are very different is that the 775pro has a free speed of around 18730rpm and produces 0.71 N.m of torque at stall. The CIM on the over hand has a free speed of around 5330rpm and can produce 2.41 N.m of torque at stall.

The 775pro needs a greater reduction (3 or 4:1) in order to match the speed and torque of the CIM. This is the reason why with a CIM 64:1 is the maximum ratio for safe operation, and with a 775pro a 210:1 is the max.

A major benefit of having a gearbox based around a VP is its ability to be highly versatile. It enables you to fine tune the robots FPS easily. Meaning that you can find the perfect gearing for getting over that obstacle or moving to that point as fast as possible. A gearbox like this could be really good for drivetrain prototyping and testing, as it would enable quick tuning of gear ratios, the results of the testing could then be used to determine the required reduction of the conventional spur gear gearbox.

That’s his point - the max safe ratio for a box with 2 CIMs would probably be something in the ballpark of 32:1.

While you get some flexibility in gearing in a setup like this, I think in reality you’d find the ratios you get to choose would be a bit too spread out to actually be practical for this application. You can do 3:1, 4:1, 5:1, and 7:1 stages. So you’d have to jump quite a large gap in your reduction to change things out. Generally when you want to change a drivetrain gearbox ratio you don’t want to go from 16 FPS to 8 FPS, you want to raise or lower the ratio just a small amount. This kind of gearing flexibility can be designed into custom spur gearboxes fairly easily - ie a 20:54 reduction could be replaced with a 24:50 or even a 18:56 reduction in order to fine tune the precise reduction.

Another thing to consider when going through these exercises is that generally a planetary gear system will be less efficient than a spur gear system.

This may not always be the case, but there are usually more interfaces and bearing surfaces that create more friction in the planetary system. The planetary shines with its packaging for high reduction instances, where a spur gear system will take up more room for the same reduction.

A side note for the VP load cases. If you use a 3, 4, or 5:1 as the output stage is the VP, the hex shaft will fail before the stage will. Those are the three strongest stages, we did some rough solid mechanics calculations on it this year for our intake rotation. However, the gears are more brittle and if we twisted the hex shaft, the gears exploded due to the hardness, but usually the hex shaft twisted before the gear box exploded. The shaft is more ductile than the gears. I would be concerned about shock loads in a drive train if you are geared too high with the ratios. If the gearing is low enough, it should reduce that concern.

We (1296) did end up trying a swap to 9:1 from the 7:1 is was running at, but decided to stay faster.

We did TIG weld the pins into the carrier, and ream out the planet gears for needle bearings.

Overall it was 12:48 via a 3mm gt2 best with WCP pulleys, and then 7:1 and an Integrated encoder, direct driving a 6" wheel from WCP.
Gearbox was fine, rode a bit too close to the sun speed wise with the motors.

Would I recommend this to most teams? nope, but being as familiar as I am with the system the Crew and I gave it a whirl.

-Aren

Would you mind sharing some pictures/details of your gearbox especially the VP input? It would greatly appreciated.

The larger pulley is on a keyed 8mm shaft that is going through a bearing, then we have the input block bearing on an input coupler. Then we have a modified input coupler (removed the flange). This is what drives the sun gear. The VP is set up: dead stage, 7:1, Encoder. Hope this helps!