Learning and Understanding different drivetrains

Now in an effort to learn more about FRC and better understand it we are currently conducting research about the pros and cons on drivetrains. We want to understand why teams use a certain drivetrain.

Currently we are stuck on West coast drive… We are absolutely stumped on what benefits it gives and why we see some of the more elite teams using it?

The live axles and cantilevered (meaning the axles are only supported on one end) wheels allow for more freedom of space and much easier maintenance.

One of the best ways to improve stability of any object is to move the contact patches as far to the outside as possible. On a robot, this involves moving the wheels as far out as possible. Because on a WCD the wheels are cantilevered, there is no need for an outer frame rail. Because of this, the wheels can be as close to the edge of the allowed perimeter. This allows for a stabler robot.
Another advantage is because the wheels are on the outside of any frame, and usually on live axles, remove a wheel only requires removing 1 clip or screw, and the entire wheel pulls off the side. This allows a wheel to be changed very quickly, usually without having to adjust chain tension, or remove alot of robot parts to get to it.

It’s a very balanced drivetrain.

For reference, a “true” WCD is 6 wheels, cantilevered outside of the frame, with the center wheel directly driven off the gearbox and the rest powered by belts or chains, with the center wheel dropped some fraction of an inch–oh, and it uses live axles. I’ll be talking about the general 6WD drop-center (as opposed to a 6WD with corner omni wheels and the 6WD scrub turn), rather than a true WCD.

First, it’s dirt simple. About the only thing simpler is a 4WD, and it’s not that hard to swap it down to that. However, it has a shorter wheelbase, due to the dropped center wheels, so it turns much more smoothly than the “Mexican jumping bean” that I’ve heard the 4WDs compared to. If one or more of the wheels are directly driven, you just about can’t lose half of your drivetrain in one mishap (or you’d have lost it anyway). It’s easy to turn when you want it to turn, but not quite as easy as a corner-omni configuration.

The true WCD adds ease of maintenance–the wheels are easy to yank off when you want to swap 'em–but also adds complexity of machining (or increases the cost as you buy the parts).

As a note, the WCD has been in use for 10 years or more. It might not be as maneuverable as a swerve or mecanum, but it’s pretty effective, and is likely to be sticking around another 10 years or more.

It offers a wider wheel base, which is more stable, and easy wheel access (as wheels on high-traction drives tend to wear a lot, this is very helpful for facilitating maintenance).

On the down side, it makes maintaining the belts or chain a bit of a pain as they’re generally run off pulleys that are mounted inside of the gearbox.

A west coast drive provides a ton of advantages. As others have said, you can maximize your wheel base, improving robot stability. It also can make servicing a drivetrain insanely fast. Our modified WCD set up allows us to switch out our entire drivetrain for entirely new components in under 10 minutes without having to temporarily remove/loosen any non drivetrain component.

I have experience with two different drivetrain styles. #1 is closed frame 6WD/8WD similar to last year’s kitbot drivetrain. #2 is West Coast Drive.

#1. Closed frame (8WD) using 1" x 1" x .125" square aluminum tubing. This is the style my team used in 2011 and 2012. We cut and welded the drivetrain frame together before anything else. This method uses dead axles with all six/eight wheels driven by chain from the gearbox.
PROS: very simple design
Low maintenence
Proven and reliable
Robust
Flexible Design

CONS: Medium to heavy weight depending on wheels
All wheels are chained- no direct drive, so it’s possible to throw a chain and lose control of one side
Takes up more space
More difficult to swap wheels/tread

#2: Standard West Coast Drive. We used this style in 2013 by necessity in order to package our wide intake under the bumper. We liked it so much that we decided to use it this year as well. It’s 1" x 2" x .125" aluminum tubing, riveted together with Vex VersaGussets with a Vex 3-CIM ballshifter. The center wheels are directly driven and the outside wheels are chained.
PROS: Simple to design
Direct Driven center wheels
Easy maintenance
Fast swap wheels/tread
Lightweight design
Robust Design
Widest possible wheelbase
Less space consumed

CONS: Bearings on cantilevered axles need to be well supported and aligned, though the VexPro VersaBlocks are awesome for this.
Requires more machining capability

I love the WCD and don’t see any reason not to use it, due to it’s low weight and super easy maintenance as well as the benefit of a maximum width wheelbase and direct driven wheels.

Just to be clear: wheelbase is the distance from the front to rear axles (length). trackwidth is the distance between the two front (or rear) wheels (width).

CONS: Medium to heavy weight depending on wheels
All wheels are chained- no direct drive, so it’s possible to throw a chain and lose control of one side
Takes up more space
More difficult to swap wheels/tread

Actually, throwing a chain does not cause a loss of control on one side. At the WPI regional in 2013, we (Team 20), lost a chain in our west coast drive during a finals match, and had no time to replace it before our next match, so we played that match without a chain, and we experienced was very little loss of control. In fact, it was impossible to tell (from the stands) that we had lost a chain.

We had a similar situation this year at the Tech Valley regional, were we smashed a drive train bearing, and had to remove the affected wheel, as we had no time to fix it. Our WCD performed just as well with 5 wheels as it did with 6, and we won that match.

Thanks, I’ll remember that.

I think you misunderstand me… Throwing the chain from the gearbox to the wheels in design 1 would result in losing power to the wheels on that side. Of course this depends on how the chain is run, but it’s assuming that you have one chain from gearbox to the center wheel and from there the wheels are chained to each other. Also, one of the benefits of WCD that I listed was that throwing a chain wouldn’t cause you to lose power on that side because the center wheels are direct driven from the gearbox (in our design at least.)

Note that one can direct-drive center wheels on a non-WCD 6WD setup (a la AM14U), so this isn’t strictly a WCD benefit.

Back when I was a student on 330 and we were using that type of setup, we never ran just to the center. The gearbox was independently chained to the center and the back wheels; center wheel chained to front wheels. That way, we’d have to jam a chain when it was thrown, or throw two chains, to lose the side.

Think of things in more atomic terms than ‘drivetrains’ and you’ll be much better at understanding why teams have arrived at certain styles of drives and what the trade-offs were that led them there.

There’s nothing intrinsically good about 6WD or swerve or anything else; good implementations are the result of well-considered details.

This sounds like it’d introduce a lot of frictional losses over just direct-driving the center wheel and chaining to the two outer wheels.

Not too much. And it helped with our weight distribution (we tended to run with the CG aft of center whenever we could). The gearbox was between the center and rear wheels, so we had two short chain runs.

This was back in the day when if you wanted to direct-drive your center wheel, you had to make your own output shaft, if not your own gearbox–we didn’t have that type of capability, then. Besides, we were a dead-axle team–that is, at the time we did NOT go live-axle if we didn’t absolutely have to. (In my time on the team, I can think of maybe 4 live axles that weren’t in a gearbox, none in drivetrain–they were all reduction stages in arms or shooters,)

What about the weight of the bumper mounts? That needs to be added into the weight of the frame as well.

We used a 6 wheel center drop dead axle (JVN axle tube style) and the outside edge of our wheels were within a .5" of the frame perimeter. I am not going to worry about that last inch of width but that is important for some teams.

WCD would have the edge in quickly replacing a wheel, though we could change a wheel pretty quick if needed.

We also used HTD timing belts between wheels which are a lot lighter than chains. We used a short #35 chain between the gearbox and the first wheel.

Part of the appropriateness of the drivetrain is what your team can build, and should be a key part of your decision. Determining that is a good offseason project.

For those of you with WCD I have several questions:

  1. How do you locate the wheel in the axial direction? Both inwards and outwards?
  2. If you are using the 3 CIM ball Shifter,
    a) Do you use the 3rd stage from Vex?
    b) Do you only use the support from the bearings in the gearbox or do you add additional support?
  3. Is there a set wheel width that you use? (narrower to be able to move the gearbox outward further? ie 1" traction vs a 2" traction?)
  4. How much drop do you use for the center wheel?

Thanks! I’m sure as we look into it, I’ll have more detailed questions!!

  1. If you custom machine wheels, you usually keep a nub on the wheel to push up against the bearing. If you don’t machine your own wheels, you can just use spacers of some type. This year we 3D printed our own. As for the outside, most teams use either e-clips, snap rings or a bolt and washer with the axle threaded on the end.
  2. We used the WCP gearboxes, so I don’t have much of an answer here, other then you usually would need the 3rd stage to actually have enough reduction for a functioning robot.
  3. The wider of a wheel you go, the more grip you can get from that wheel. Since we don’t machine our own wheels, we use whatever we can get. This year, that was VersaWheels. The general rule is to have the wheels as far to the edge of the perimeter that is allowed.
  4. This year we used 0.1". The old standard used to be 1/8", but with the smaller wheelbases that we are currently seeing, that number has to drop, and we have found 0.1" to work the best.

A well done WCD can give a team a pretty good step up from the competition for a lot of reasons that have been mentioned above and i dont think i need to delve into here but it does seem like your team is looking to take it to the next level (correct me if im wrong). Finding a consistent drive base and sticking with it isn’t a bad thing, it gives you time to perfect your manipulator(s) which can really help your gameplay. But if you are looking to upgrade here are a few options to mess with in the off season.

  1. 2-Speed gearbox: If you haven’t already started using one it can really help your game play. Our team has loved the ability to race across the field and push people around at the other end if we need to once we get there

  2. Integrated Manipulators: I actually missed it at SVR in 2013 but Team 118 (go watch their video, its brilliant) had a pick up that (this is what it looked like from their reveal video) dropped out of the bottom of their drivetrain. (Now that’s cool) and its was a intresting way to keep from getting their pick up smashed.

  3. I’m not even sure: Look to 118’s drivetrain this year. In their reveal video this year (https://www.youtube.com/watch?v=PtRewwr59d8) just at around 15 seconds. Its amazing system but what happens is that omni’s fire down using pistons and its got three great usages that i can imagine. A) Just as they do in the beginning a launch to really get you going might be helpful. B)Lets say you are getting defended and pushed across the field, you’ve got no traction at this point and even though you can get your wheels moving you cant get out of the push (ask your drivers about this, they know the situation) you fire down the omnis and just coast away, while ramping up your wheels and as soon as you coast out of their reach you sprint out of the way, freeing you from the pin. C) Turns, the ideal turning system would be high traction in the back and omni’s where you want to turn right? So… fire down just one set of omni’s (Look to 1:06 in the video) they manage to pull of these almost diming maneuvers throughout the video which i can only imagine happen because of the turning help that the omni’s provide. This must be an incredibly powerful set of pistons and some fairly complicated code to make it easy to use but its a great summer program.

I’m reminded of a quote by Adam Heard of 973 (Correct me if I’ve got the wrong person) “World Champions are build in the off-season.”

Good Luck!!