WCD vs Standard

[Walter Deitzler]

What is the advantage of a WCD over a standard drive? Why would you choose to do a WCD 8 wheeled or WCD 6 wheeled over a standard 8 or 6 wheeled drive?

[Akash Rastogi]

Hey Walter,

There are several threads that discuss the pros and cons of each style of drive. I search should yield you the results you are looking for.

Let us know if you can't find them.

-Akash

[Walter Deitzler]

I tried finding them, but only found WCD vs. Swerve and many pictures of a WCD.

[Andrew Lawrence]

Some things I've found on my way asking the same questions you do:
http://www.chiefdelphi.com/forums/showthread.php?t=106243&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=88272&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=105821&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=98064&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=94288&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=96817&highlight=WCD
http://www.chiefdelphi.com/forums/showthread.php?t=95670&highlight=WCD

All in all, WCDs are lighter, allow for direct drive (more efficient), and the wheels a bit farther out for better turning. 6WD vs. 8WD is mainly based off of whether or not the robot has to traverse an object (climbing over the bumps in '12 and '10). 8WD also allows a smaller wheelbase, so that a robot can be more maneuverable, but the turning difference is often nothing a 6WD can't do, and the extra weight isn't worth the little bit of turning help on a flat field.

[Walter Deitzler]

Thank for that Andrew, some of those links will be helpful, but as for the 6 vs 8 wd, I phrased the question poorly, that is not what I am looking for. The standard vs. WCD ones are useful though!

Thanks!

[avanboekel]

SuperNerd's last link is probably what you are looking for (If I am reading your question right, you want to know the pros and cons of a WCD vs a non WC 6WD, not 6wd vs 8wd).

http://www.chiefdelphi.com/forums/showthread.php?t=95670&highlight=WCD

[MattC9]

All in all, WCDs are lighter, allow for direct drive (more efficient), and the wheels a bit farther out for better turning. 6WD vs. 8WD is mainly based off of whether or not the robot has to traverse an object (climbing over the bumps in '12 and '10). 8WD also allows a smaller wheelbase, so that a robot can be more maneuverable, but the turning difference is often nothing a 6WD can't do, and the extra weight isn't worth the little bit of turning help on a flat field.

Do you have solid proof that WCD are lighter than any other drive?

[Walter Deitzler]

Okay, so in the 4th link, I saw that WCD's use a live axle, vs. a Standard using a dead axle.
First off, could some elaborate on the difference between the two? I found the posts kind of confusing as to the difference (the live axle is spun by chain or gears or motor while dead axle is just sitting there, not powered?)

Next: Why would you choose a live axle over a dead axle, if the center wheel is lowered, thus making your robot rock between the front and back wheels? It seems that live axles (if I have their definition correct) would make the 6 wheeled work like a 4 wheeled when it turns, porblematic. But, when observing many WCD's, I do not see slick wheels or omniwheels to offset the traction/turning problems associated with a 4 wheeled.

[JosephC]

To my understanding, A live axle is one that supports part of the weight of a robot and drives the wheel(s) connected to it. A dead axle is one that carries part of the weight of a robot but does not drive the wheel(s). The wheel(s) rotate on the end of the dead axle.

[smistthegreat]

Okay, so in the 4th link, I saw that WCD's use a live axle, vs. a Standard using a dead axle.
First off, could some elaborate on the difference between the two? I found the posts kind of confusing as to the difference (the live axle is spun by chain or gears or motor while dead axle is just sitting there, not powered?)

Next: Why would you choose a live axle over a dead axle, if the center wheel is lowered, thus making your robot rock between the front and back wheels? It seems that live axles (if I have their definition correct) would make the 6 wheeled work like a 4 wheeled when it turns, porblematic. But, when observing many WCD's, I do not see slick wheels or omniwheels to offset the traction/turning problems associated with a 4 wheeled.

All of your information is correct, but you make a couple of incorrect conclusions. Yes, live axle is when the axle itself is powered, and dead axle is when only the wheel is powered. Also, you are correct about the drop center making it turn like a 4 wheeled drive. However, this 4 wheeled drive now has a shorter wheelbase, making it easier to turn. The whole point of dropping the center wheel is to shorten the wheelbase enough to where it turns smoothly without adding slick or omniwheels

[Andrew Lawrence]

Do you have solid proof that WCD are lighter than any other drive?

No outer framing. Basically eliminates about 86" of outer railing (assuming a 37" long by 27" wide bot). Center wheel is direct drive, so less chain (6 chains per average 6WD, 4 for your average 6WD WCD).

[BrendanB]

WCD's are known to be lighter but they aren't always lighter. ;)

[EricH]

All of your information is correct, but you make a couple of incorrect conclusions. Yes, live axle is when the axle itself is powered, and dead axle is when only the wheel is powered. Also, you are correct about the drop center making it turn like a 4 wheeled drive. However, this 4 wheeled drive now has a shorter wheelbase, making it easier to turn. The whole point of dropping the center wheel is to shorten the wheelbase enough to where it turns smoothly without adding slick or omniwheels

And the other implied conclusion--dead axles don't need a drop--is also an incorrect one. A 6WD, dead or live axle, without the drop will act almost exactly like a 4WD with all traction wheels (there are exceptions, like FRC25); so will an 8WD without any drop.

[Garret]

No outer framing. Basically eliminates about 86" of outer railing (assuming a 37" long by 27" wide bot). Center wheel is direct drive, so less chain (6 chains per average 6WD, 4 for your average 6WD WCD).

How are you getting 86" I would think that you are only eliminating 74" of railing, though I have never built a WCD so I may be off. Also I have seen plenty of plain non-cantilevered 6WD robots with direct drive on the center wheels.

However I do know that with the WCD it allows for wheels to be removed much easier because the shafts do not need to be removed.

[Walter Deitzler]

And the other implied conclusion--dead axles don't need a drop--is also an incorrect one. A 6WD, dead or live axle, without the drop will act almost exactly like a 4WD with all traction wheels (there are exceptions, like FRC25); so will an 8WD without any drop.

What does 25 do? why are they an exception?

[EricH]

What does 25 do? why are they an exception?

6WD flat. I don't know how they do it, but they don't bounce when turning, even with high-traction wheels. I think it has something to do with how they groove their tread, but I'm not sure.

At least, they used to do that. I'm not sure if they still do or not.

[MichaelBick]

For weight it honestly depends: if you are running tube then WCD is a pretty good choice to save weight. If you are running sheet there is really no need to run a WCD. However wether you run sheet or tube you always should drop the center wheels and should direct drive the center wheel.

[Walter Deitzler]

And I am assuming that the standard KOP C-Channel will not work for a WCD, or, at least, that C-Channel is not a good choice? I have never heard of a C-Channel WCD.

[MichaelBick]

The KOP is not made for running a WCD. WCD is overall a better drivetrain, but for teams that don't have the capabilities to build a WCD the KOP is a find alternative. I believe they even make supershifter mounts for the KOP

[Walter Deitzler]

The KOP is not made for running a WCD. WCD is overall a better drivetrain, but for teams that don't have the capabilities to build a WCD the KOP is a find alternative. I believe they even make supershifter mounts for the KOP

Why can you not run a WCD on a KoP chassis?

[MichaelBick]

The definition of WCD usually means that you are running a cantilevered, live axle, direct driven a dropped center drivetrain. While the KOP is drop center, it is not any of the other things. The C Channel of the KOP is definitely not strong enough to cantilever something off of it. because you are not cantilevering you might as well run a dead axle set up.

[Walter Deitzler]

The definition of WCD usually means that you are running a cantilevered, live axle, direct driven a dropped center drivetrain. While the KOP is drop center, it is not any of the other things. The C Channel of the KOP is definitely not strong enough to cantilever something off of it. because you are not cantilevering you might as well run a dead axle set up.

That makes sense, thanks.

[BrendanB]

Why can you not run a WCD on a KoP chassis?

Technically you can make a WCD using the KOP but it isn't the most efficient way. The team who made it escapes me but I know there is a photo here on CD.

How they did it was they took the outer most rails in the standard KOP frame and moved them inward so they ran flush with the inside rails. This meant all they had to do was put their KOP bolts through the holes with chain and sprockets and they had a WCD. Cantilevered in all.

This method could make for a great fall project for any team that doesn't want to spend money yet still wants to work towards custom drivebases. Who knows, we might do it this fall just as a learning experience! :)

[R.C.]

Do you have solid proof that WCD are lighter than any other drive?

Matt,

I can't think of another drive that would be lighter than a WCD other than a super pocketed sheet drive. I'm curious to see the weight comparison and sheet drive.

But WCD is decently light due to the fact:

-No outer Railing
-Minimal amount of parts.
-Smaller parts: Wheels/Sprockets/Gearboxes etc....

We switched over to WCD in the fall of 2010. Never looked back (thanks 973!), its such a nice system that 254/968 has perfected over the years.

-RC

[sdcantrell56]

While not the traditional wcd, doing one with fixed wheels and belts would without a doubt result in the lightest drivesystem in FIRST.

That said there is really no right or wrong answer. Do what works best for you. If your main manufacturing resources are sheetmetal then WCD might not be the best choice. It is possible and we have actually run a sheet WCD before but it's not the best use of resources.

[Kristian Calhoun]

What does 25 do? why are they an exception?

6WD flat. I don't know how they do it, but they don't bounce when turning, even with high-traction wheels. I think it has something to do with how they groove their tread, but I'm not sure.

At least, they used to do that. I'm not sure if they still do or not.

Jared from 341 made a really good post about our drive train about two years ago: http://www.chiefdelphi.com/forums/showpost.php?p=961835&postcount=31.

[Tom Line]

No outer framing. Basically eliminates about 86" of outer railing (assuming a 37" long by 27" wide bot). Center wheel is direct drive, so less chain (6 chains per average 6WD, 4 for your average 6WD WCD).

Could you show the mathematics for that? West coast drives have to carry the same amount of force that a standard chassis chassis does. Keeping in mind that both have the same load requirements, I see no reason that one can't be engineered to the same weight as the other. West coast style introduces a rotational component that needs to be accounted for that astandard chassis does not require.

Nothing stops a team from direct driving on a standard chassis, and a standard chassis does not require bearing blocks.

[Walter Deitzler]

Okay, new questions to ask.

In my WCD design I have designed the wheels to be far enough from the outside to allow for interchangeable wheel sizes of 4in, 6in, and 8in. Is this a good idea? I have read somewhere that you use 4in wheels to save weight, go faster, and keep your wheels as far to the outside as possible. Is this true? Should I design my WCD to use one type of wheel?

Also, can you weld the AM Flanged bearings (http://www.andymark.com/product-p/am-0279.htm) to standard aluminum wall? (6061 I think).

Thanks!

[Mk.32]

Generally speaking the bearings are pressed into the alum, or sometimes held in place with super glue if needed.

[Gregor]

I have read somewhere that you use 4in wheels to save weight, go faster, and keep your wheels as far to the outside as possible. Is this true? Should I design my WCD to use one type of wheel?

Also, can you weld the AM Flanged bearings (http://www.andymark.com/product-p/am-0279.htm) to standard aluminum wall? (6061 I think).

Thanks!

Smaller wheels will in fact make the robot go slower at the same gear ratio. Wheel size shouldn't be considered in terms of speed, as you should be adjusting your gear ratio to account for the wheel size. However the 4" wheels will save weight and widen your wheelbase, as opposed to 6" and 8" wheels.

Typically, teams press fit bearing, as opposed to welding them.

[Garret]

Also, can you weld the AM Flanged bearings to standard aluminum wall? (6061 I think).

I am pretty sure that those bearings are steel not aluminum, so welding them to aluminum isn't really possible (As far as I know). but anyways, there are far easier ways to mount bearings than welding them, as Mark already said they can be pressed in or super-glued. On our robots we usually hold the bearings in with a screw and washer that covers part of the flange.

The smaller wheels allow for smaller gear ratios and sometimes fewer reductions in the gearbox, this usually results in a more lightweight gearbox. I do not know about keeping the wheels as far out as possible as a motivation for smaller wheels, I would usually try to do that anyways regardless of wheel size (just my preference no real reasoning on my part behind it).

[Walter Deitzler]

Smaller wheels will in fact make the robot go slower at the same gear ratio. Wheel size shouldn't be considered in terms of speed, as you should be adjusting your gear ratio to account for the wheel size. However the 4" wheels will save weight and widen your wheelbase, as opposed to 6" and 8" wheels.

Typically, teams press fit bearing, as opposed to welding them.

So you do want to put your smaller wheels farther to the outside, to widen your wheelbase? Putting them in the same spot you would put an 8in wheel would be pointless, I am assuming, because the wheel base would be the same, but you wouldn't get better speed?

And what is "press fitting"? Could you elaborate.

[Garret]

And what is "press fitting"? Could you elaborate.

Press fitting usually means (at least how I have interpreted it in FRC) the hole is slightly undersized such that the bearing needs to be pressed in quite hard, this will result in a tight fit that will hold the bearing in place.

[Gray Adams]

Don't weld or weld near bearings. The grease has a bad habit of lighting on fire.


So you do want to put your smaller wheels farther to the outside, to widen your wheelbase? Putting them in the same spot you would put an 8in wheel would be pointless, I am assuming, because the wheel base would be the same, but you wouldn't get better speed?

And what is "press fitting"? Could you elaborate.
You need less gearing for 4" wheels though, and they have a lower mass (and therefore moment), so it can save you weight even if you don't put them further out on the frame.

Press fitting refers to having a hole that's just slightly smaller than the diameter of the bearing, so when you press it in, it doesn't come out. It's kind of tough to figure out what size hole to make and then make it accurately (within a few tenths).

[Walter Deitzler]

Don't weld or weld near bearings. The grease has a bad habit of lighting on fire.

Well, good thing I asked, thanks!

[Adrian Clark]

Ok, multiple things:

1. If you need to hold in a bearing that is not a press fit the standard approach is to use bearing loctite, not glue or welding. Although glue is a much better idea than welding. But why are so keen on holding in your bearings?

2. A press fit is not necessarily undersized, it could also be exact size. The difference between press and slip fit depends on surface finish as well as hole size. If you have a hole that you made exact size via a rat tail file chances are that's not going to be a slip fit. I caution you if you press in your bearings, too tight a fit can make a bearing seize up.

3. I would not suggest having a drivetrain with swap-out wheels with the size variation that you're talking about. If you have clearance for a 8" wheel but you use a 4" you're just not utilizing your frame space, the idea is to get the wheels close to the ends of the frame. There is no way you can swap out a 8" wheel with a 4" without changing the gear ratio. The fact that you want to do this makes me think that you didn't fully plan out your ratios and wheel sizes to your desired torque/speed. I suggest you calculate your drivetrains speeds using your different wheel sizes and you'll see how big of a difference wheel size makes.

4. The advantage of small wheels is that they're light, they lower your CG and they require less reduction. Don't use them unless you calculate your robots speed using them, or you might end up with a very slow bot.

[jspatz1]

To my understanding, A live axle is one that supports part of the weight of a robot and drives the wheel(s) connected to it. A dead axle is one that carries part of the weight of a robot but does not drive the wheel(s). The wheel(s) rotate on the end of the dead axle.

Live axle = Axle turns, wheel is keyed to the axle.
Dead axle = Axle is fixed, wheel turns on the axle.

[Walter Deitzler]

So, doing more research on CD, I have come across bearing mounts called "Sliding Bearing Mounts". What are these exactly. It seems as if they are used for chain tightening, but I have no details about how they do this.

[Aren Siekmeier]

So, doing more research on CD, I have come across bearing mounts called "Sliding Bearing Mounts". What are these exactly. It seems as if they are used for chain tightening, but I have no details about how they do this.

Many teams build bearing blocks for their drive train which hold the bearing and axle. This bearing block can slide forward and backward in the frame (moving the wheel with it) in order to adjust chain or belt tension.

Unfortunately, 254's website is under construction, so I can't pull up any of the great pics I know they had there. I'm sure someone else here can dig up some instructive photos.

[Joey Milia]

So, doing more research on CD, I have come across bearing mounts called "Sliding Bearing Mounts". What are these exactly. It seems as if they are used for chain tightening, but I have no details about how they do this.

The "Sliding Bearing Mounts" are blocks that house the bearings for the live axle and they slide in slots in the frame. This allows you to move the axles slightly further appart tensioning the chain.

[BJC]

It's interesting to hear so many people say that canilevered, direct driven systems are the lightest and best when some of the most successful teams in the world do neither of those things (most notably 67 and 1114).

I know Jim Z has done annalysis on 254, 1114, and 67's frames. I *think* 1114's frame was lighter than 254's by about two pounds.

We used .06 sheet metal this year and will probably go down to .05 sheet metal next year. I am pretty sure that our frame weight beats out 254's by a pound or two. What get's 254's weight so far down is there use of tiny wheels which takes weight out of both their gearboxes (which are custom and very light) and wheels (also custom and very light). Direct diving helps too I'm sure.

Perhaps, some 254 people could chime in and correct me here if I'm wrong on any of these points?

In any case, I'm just stirring the pot.
Regards, Bryan

[Adrian Clark]

Many teams build bearing blocks for their drive train which hold the bearing and axle. This bearing block can slide forward and backward in the frame (moving the wheel with it) in order to adjust chain or belt tension.

Unfortunately, 254's website is under construction, so I can't pull up any of the great pics I know they had there. I'm sure someone else here can dig up some instructive photos.

For the best example of a well executed sliding bearing block I would strongly suggest downloading some of 973's CAD models.

[Gray Adams]

So, doing more research on CD, I have come across bearing mounts called "Sliding Bearing Mounts". What are these exactly. It seems as if they are used for chain tightening, but I have no details about how they do this.

The idea is that the wheel, axles, sprocket, and bearings are all part of a block that slides back and forth along the chain path to adjust tension.

Here (http://i.imgur.com/ZKW8a.png) is a section view from our offseason project last fall. If you look carefully inside the tube, you can see the space the bearing block has to slide back towards the gearbox and loosen tension (to the right).

Also, here's a view of how we kept the bearing block in place. http://i.imgur.com/VrDEJ.png
It worked, but it really wasn't the greatest method. It should get you thinking though, and we came up with it halfway through building it.

[AlecMataloni]

I will side with the live axle cantilevered drivetrain not on the grounds that it "weighs less", but by the fact that maintaining one can be easier. When the wheel is supported by live axles on both sides by the frame, working on the drivetrain becomes easily more painful and irritating when frame members are in the way. With the average west coast drive, replacing components like wheels, axles, and chain is a snap because there's only one frame member to deal with.

[Pat Fairbank]

Unfortunately, 254's website is under construction, so I can't pull up any of the great pics I know they had there. I'm sure someone else here can dig up some instructive photos.
I can't find a good picture of the bearing blocks themselves, but here's a shot (https://picasaweb.google.com/115829622106274402945/CheesyPoofs2012BuildSeason#5706692307221465746) of the rectangular holes in the frame where they go. You might get a better idea from browsing the rest of the 2012 build season album (https://picasaweb.google.com/115829622106274402945/CheesyPoofs2012BuildSeason). Basically, there are two halves that come in from either side of the rail and fit inside each other, each having a bearing pressed into it. Long bolts through the rail are used to hold the two halves together, and there's a cam mounted to the rail that is torqued once the chain is installed to push the block which tensions the chain.

It's interesting to hear so many people say that canilevered, direct driven systems are the lightest and best when some of the most successful teams in the world do neither of those things (most notably 67 and 1114).
I wouldn't say that weight is the top factor in 254's choice of using a cantilevered drive year after year. It probably only comes in fourth after ease of maintenance, robustness, and aesthetics (you wouldn't believe how many design decisions come down to aesthetics).

[Adam Freeman]

I wouldn't say that weight is the top factor in 254's choice of using a cantilevered drive year after year. It probably only comes in fourth after ease of maintenance, robustness, and aesthetics (you wouldn't believe how many design decisions come down to aesthetics).


I can't say for sure, but my notes say that I weighed our competition chassis (0.06" sheet AL) at 8lbs, before we started assembling any of the components on to it.

Weight is a factor in our decision to use this type of drivetrain...but, it's probably #3 on the list. #1 would be resources available (in-shop waterjet) and ease of construction (no welding) and #2 would be robustness.

Our design is a nightmare for ease of maintenance. Changing wheels or drivetrain components is not something we could do easily/quickly.

[Chris is me]

I think the main reason anyone does any type of drivetrain the way they do is because they have access to resources that make their design optimal. A sheet metal chassis can be done excellently. So can a welded tube chassis. Rather than sketch an exact copy of the best, drive design should play to the resources your team has. What use is a drive that's 5 pounds lighter of it takes you 2 weeks longer to build it?

We use a Poof inspired but *highly* derivative drivetrain that plays into our resources and goes together relatively quickly. It's light enough. It's definitely strong enough. Probably overbuilt. And it works because it uses what we have and what we were able to get in an offseason.

The way the Poofs do it, when you get down to the details of their bearing blocks, custom wheels, etc. takes a lot of resources that not a lot of teams have as readily available as 254 has worked hard to have. Play to your strengths.

[Walter Deitzler]

I think the main reason anyone does any type of drivetrain the way they do is because they have access to resources that make their design optimal. A sheet metal chassis can be done excellently. So can a welded tube chassis. Rather than sketch an exact copy of the best, drive design should play to the resources your team has. What use is a drive that's 5 pounds lighter of it takes you 2 weeks longer to build it?

We use a Poof inspired but *highly* derivative drivetrain that plays into our resources and goes together relatively quickly. It's light enough. It's definitely strong enough. Probably overbuilt. And it works because it uses what we have and what we were able to get in an offseason.

The way the Poofs do it, when you get down to the details of their bearing blocks, custom wheels, etc. takes a lot of resources that not a lot of teams have as readily available as 254 has worked hard to have. Play to your strengths.

That makes sense, most of my parts will probably be bought from andymark. But I would still like to know how many teams mount these bearing blocks, and allow them to slide, then lock them into place when they are done with them. Thanks!

[Cory]

That makes sense, most of my parts will probably be bought from andymark. But I would still like to know how many teams mount these bearing blocks, and allow them to slide, then lock them into place when they are done with them. Thanks!

Pat's post explains how we do it. The combination of his post, the picture of our frame rail he linked, and Gray Adam's cross section view of their 3d model should be a perfect explanation of how it works.