WCD questions
 

My search skills must be weak, I could not find an answer to my questions, please help us out.

Fusion is contemplating switching to WCD for this upcoming season, mainly to reduce machining time and increase drivetrain serviceability. We are starting to look at designs and plan to make a couple of chassis's this summer/fall.

My question: Why do teams put their chains on the inside of the rail instead of the outside with the wheels? It looks to really complicate the build (requires: smaller sprockets, all live axles, transmission standoffs, tensioning issues etc).

FRC teams are smart, so their must be a reason other than appearance. Show me the light. Thanks

Re: WCD questions
 

One really good reason to put the chains and sprockets on the inside of the rails is to get the rails as wide as possible. This gives you a larger frame which allows you more room to put in other components. If the sprockets are on the outside of the frame rails, it forces the frame rails closer together reducing internal space.

Re: WCD questions
 

Thanks, but would not the space savings issue be a moot point?

You still can't use the extra space because it is occupied by the chain and sprockets? If you build on top of the chain and sprockets you reduce serviceability.

Re: WCD questions
 

One reason to put the chains on the inside is to allow for some uprights attached to the outside of the frame rail to create a frame perimeter outside of the wheels and a place to mount the bumper. Depending on your wheel width you can get away with 1x2 or 1x3 for those uprights.

Another benefit is that it reduces the cantilever of the axle by putting the wheels as close to the bearing as possible rather than spacing it out for two sprockets and any spacer required to prevent the chain from rubbing on the frame or wheels.

Re: WCD questions
 

In the interest of disclosure, this is all from my browsing - 3946 has not built a chain-in-tube drive, even as an off-season project.

Putting the chain inside the tube provides a built-in chain guard. This protects the environment in case a chain breaks, but probably more importantly excludes stray carpet and bumper threads and other foreign materials from getting caught in the chain.

Placing the sprockets between the two supports reduces stress all over, especially given that there is already plenty of stress on those cantilevered axles. (already noted by Mr. V)

Finally, many of the chain-in-tube designs appear to go for the largest sprocket such that the link plates will just barely fit inside the tube. This offers "insurance" against chain stretch - even if the chain stretches, it cannot come off the sprockets, because there is not enough room for the plates and rollers to fit in the space between the sprocket teeth and the tube.

Re: WCD questions
 

GeeTwo,

We have zero interest in placing the chain in the tube.

We are thinking of switching from having 2 plates with the wheels, chain, and sprockets sandwiched between them; to using the West Cost Drive system of a rectangular tube with internal transmission and exposed wheels.

Re: WCD questions
 

Putting the chain inside of the chassis give you a safeguard for the chain as GeeTwo said, even though it's not inside the tube. WCD was developed before bumpers and other robots could get tangled/break the chain. Also, there wasn't OTS FRC gearboxes then so just about everything in the gearbox was custom. Teams that developed WCD also were fairly strong in their machining capabilities, so live shafts weren't as big of an issue.

Early WCDs (with six wheels), didn't have as small sprockets. There were cutouts in the electrical boards for the chain runs. You can make it out here. Tensioning was also solved with eye shaped bearing blocks and cams.

Re: WCD questions
 

Having the wheel on one side and the chain on the other also makes maintenance a bit easier, since you can remove either without bothering the other.

Re: WCD questions
 

Putting the chain on one side of the tube and the wheels on the other has always had two reasons for me: Firstly, to balance out the loads between wheel and sprocket by making the bearings like a "pivot" for them to cancel each others' forces out, and secondly to reduce the cantilever and the torque arm length on those bearings. I would not run sprockets on the same side as the wheels mainly for the second reason.
It also increases serviceability by letting you access the chains without removing wheels or bumpers, although in practice bumper removal at least has to happen anyway.

Re: WCD questions
 

Chain-in-Tube is also safer, especially if the bot is going to be used for demos.

Re: WCD questions
 

It's also much easier to swap wheels. By having chain/belts on the other side not connected to the wheels it's usually just a matter of getting rid of a shaft collar or loosening a screw to take off a wheel. This isn't a problem if your using colons but some of the softer grippier wheels require swapping everyone once in a while.

Re: WCD questions
 

There is zero reason to use 2 tubes at all. It will actually hurt your efficiency (more misalignment) and you are just throwing weight away. If you want chain on the outside of the frame just put it there, between the frame and the wheels. If you use live axles this won't interfere at all with your ability to change wheels, and it's a perfectly valid solution for a WCD to do this. Teams have run WCDs like this many times before with no problems.

Re: WCD questions
 

In a WCD setup, if the chain is on the inside of the frame rail and the wheel is on the outside, wheel/axle flex is (almost) completely isolated from the chain alignment. Therefore there is a greatly-reduced chance the chain will come off. Sure, changing wheels is nice, but the big reason is minimizing risk on the most critical thing a robot needs to do during a match. It is why many COTS WCD products are manufactured the way they are. The details are easy to recognize if you know what to pay attention to. Here is the explanation:

During matches, drive train shafts flex under high stress (robot-robot impacts, high-torque driving, 2016 defenses, etc). Any CAD simulation shows that 1/2" steel round/hex shaft flexes at least a little. In the pits, the shaft will appear straight. Yet at the moment of impact, every shaft flexes. I forget the term that relates amount of stress to whether a material will return to its original form, but this type of thing is measurable.

It's important to analyze where a shaft flexes. In a WCD setup it flexes from the bearing block outward. If the chain sprockets are on the wheel, the sprockets will become misaligned during an impact. With #25 chain, this will nearly always cause the chain to jump the sprocket, and perhaps cause a sitting (or twitchy) robot. When a chain comes off like this, it is usually still attached at the master link.

There is also a major element of misalignment in assemlby of a sprocket to a dead-axle setup. If the bearings are not perfectly seated, or the bearing holes flex (like the do on ALL plastic VEX/AM/Banebot plastic wheels...), or the sprockets are offset from the wheel using plastic spacers, or if one of the attachment bolts is overtightened.... The list of "if x happens we are dead for the match" gets pretty large.

WCD Variants all have the same fundamental desirable feature: sliding bearing blocks that, when manufactured correctly (i.e., buy them...), come with some 99.9% guarantees. The two sides of the sliding bearing block are machined together to guarantee the bearing holes are perfectly concentric. Usually these bearing blocks have a flat face that sits flush with the frame rail as well. All of this means that all axles are parallel to each other, and are perpendicular to the frame rail. If the sprocket is isolated from any wheel impacts, then the chain is nearly guaranteed to not pop off, even under the highest of stresses - allowing us to use light & efficient #25 chain with nearly 100% reliability.

Now - in 2016 some well-known teams went to #35 chain due to the stresses of defenses on the chain itself. This is due to the fact that a wheel at full speed in the air, which then comes to a halt due to a 150lb robot 'landing' on top of it, will transfer all of the stresses directly to the chain. The #25 chain would literally break due to this stress, causing the robot to sit. This is not the same thing as a misalignment causing the robot to sit.

Re: WCD questions
 

While I think flex is an important concern, I don't think the outcome is as dire as you indicate. Even 1/8" of flex isn't enough to instantly throw a chain or anything like that. Teams have run chain on the outside for years - look at the old Team221 chassis setups for one example.

Re: WCD questions
 

I disagree with you Chris. Sure, some teams did it and had success in the seasons they did it. But that says nothing about how they made it work, the level of maintenance involved, or what their current preference is.

It's also not about length of deflection so much as it is angle of deflection.

I admit I'm also probably biased. Back in the day when my team did dead-axle wheels (pre-2013), there was hardly a first event where we didn't struggle with constant chain issues - even when the robot drove well before ship day. After the switch to 'standard' WCD, it's like night and day - we can build a drive train and do only minor maintenance for all 70-something matches in a year (including offseason).

Re: WCD questions
 

This, seconded.

Re: WCD questions
 

Thirded. Saint-Venant's Principle is useful. Additionally, using 2x2 box tubing rather than 1x2 would help things out. In normal cantilevered bearing applications, it's preferable to keep the bearings 3-5 characteristic distances (read diameters) apart, so with a 0.5" shaft, having the bearings 2" apart (4 characteristic distances) would really be ideal for long bearing life. But...space is at a premium and R8 bearings can be found rather cheaply.