Drive Trains & All that they encompass

[ksafin]

Hello all.

I think most teams recognize that no robot is good without a decent drivetrain. We even had the mistake last year of working on our entire shooting and picking up mechanism only to find out 3 days from packing day that the robot was incapable of moving because the drivetrain was so horrendous.

I've read many threads about various drivetrains, I've done a good deal of research, but I would really like to spawn a comprehensive discussion about several aspects of drivetrains, what makes them good, when to use what, etc.

A few personal points I would like addressed:
(It's not vital that you address every single one, but these are some things I would love to hear answered for the benefit of knowing more about drivetrains as a whole)

1) What wheels do you prefer
1a) What kind of challenge/obstacle prompts a certain wheel choice
1b) What kind of wheel would you use if you had a low budget (aka can't afford $1,000+ on swerves)
1c) How do you personally choose what size wheel you prefer?

2) What do you make your drive train out of? (Base train, custom, 80/20, etc)

3) What gearboxes do you use for your motors, and why?
3a) Do you use shifters? Why?

4) What kind of wheel setup do you choose? (4 wheels, 6 w/ 2 dropped, etc etc)

5) How much time do you normally spend building your drivetrain?

6) Do you prefer to use belt or chain? Why?
6a) What would prompt using one over another?

7) Generally, long robot, wide robot, or it depends on the game?

[MichaelBick]

1) What wheels do you prefer

We are going colsons for this next year, because they are less work. I recomend this if you do not have a lot of resources(people, sponsors, tools). If you have the resources I really like AM performance wheels or custom aluminum wheels.

1a) What kind of challenge/obstacle prompts a certain wheel choice

Resources and game.

1b) What kind of wheel would you use if you had a low budget (aka can't afford $1,000+ on swerves)

Colsons with WCP hubs.

1c) How do you personally choose what size wheel you prefer?

Game dependent, but the smaller the better.

2) What do you make your drive train out of? (Base train, custom, 80/20, etc)

Custom

3) What gearboxes do you use for your motors, and why?

We will either be going custom shifting gearboxes or AM supershifters for this next year.

3a) Do you use shifters? Why?

Yes. They allow us to play defense(good for alliance picking), get past offense, gear faster, and are useful for getting over obstacles.

4) What kind of wheel setup do you choose? (4 wheels, 6 w/ 2 dropped, etc etc)

6 with 2 dropped in general. Depending on the game we could go 8wd with center 4 dropped.

5) How much time do you normally spend building your drivetrain?

It took us a long time last year because we went the stupid route and did not prototype a WCD(the drive we went with last year) in the offseason. If you do it right it should take your the first 2-3 weeks if you are going custom, 1 week if you are using the KOP.

6) Do you prefer to use belt or chain? Why?

Chain because we have worked with it before, it is forgiving, and it is reliable.

6a) What would prompt using one over another?

You should use chain unless you have lot's of resources and/or have prototyped it over the offseason.

7) Generally, long robot, wide robot, or it depends on the game?

Depends on the game, but we default to long because it is easier to weave between robots.

[Littleboy]

http://recordings.talkshoe.com/TC-98466/TS-681568.mp3
That is a recording of a recent EWCP cast about drivetrains.

There is probably many threads if you would do a search on it.

That probably covers most topics.

[Billfred]

1) What wheels do you prefer After experience with the HiGrip wheels AndyMark puts in the kit, we're inclined to keep using them!
1a) What kind of challenge/obstacle prompts a certain wheel choice A specific need for some size or traction property. If it were a game that required every last bit of traction, we may go back to roughtop tread. If we had another game like Breakaway with a lot of really hard whacks, we may go pneumatic.
1b) What kind of wheel would you use if you had a low budget (aka can't afford $1,000+ on swerves)The AndyMark HiGrip wheels. They're in the kit!
1c) How do you personally choose what size wheel you prefer? Depends on the kit's offerings, desired speed, and the obstacles on the field. Cost is also a factor--if a 6" wheel is in the kit and gets us in our speed range, we're not going to drop big money on 4" wheels.

2) What do you make your drive train out of? (Base train, custom, 80/20, etc) Three of our four drivetrains are C-Base derivatives (we never quite build them to spec). Our 2010 robot was totally custom on the frame, but our expertise there graduated.

3) What gearboxes do you use for your motors, and why? We've used the AndyMark CIMple Box the past two seasons (and the Toughbox in 2009, but that was Lunacy and doesn't usually count).
3a) Do you use shifters? Why? We used shifters on 1618 a couple of times, and in 2010 (which, in hindsight, was a touch overkill). Recent years haven't made their necessity as obvious.

4) What kind of wheel setup do you choose? (4 wheels, 6 w/ 2 dropped, etc etc) All years except 2009 (again, Lunacy), we used a 6WD with dropped center.

5) How much time do you normally spend building your drivetrain?We normally get it together in a week or less mechanically. Electrical, programming, and other systems may influence when we turn it on and drive around.

6) Do you prefer to use belt or chain? Why? Chain every single year. Belt requires a lot more precision of fabrication, sometimes more than we have, and it makes it trickier to adjust gearing if we need it.
6a) What would prompt using one over another? The only thing that may change us over is the coming switch in the kitbot. Even then, we may go chain if we don't like the speeds the kitbot offers.

7) Generally, long robot, wide robot, or it depends on the game? Depends on the game. When a ball intake is needed, generally wide. When it's an arm game, long. But there are no hard and fast rules, especially if the field offers some obstacle.

Originally Posted by ksafin

1)What kind of wheels do you prefer
1a) What kind of challenge/obstacle prompts a certain wheel choice
1b) What kind of wheel would you use if you had a low budget (aka can't afford $1,000+ on swerves)
1c) How do you personally choose what size wheel you prefer?

2) What do you make your drive train out of? (Base train, custom, 80/20, etc)

3) What gearboxes do you use for your motors, and why?
3a) Do you use shifters? Why?

4) What kind of wheel setup do you choose? (4 wheels, 6 w/ 2 dropped, etc etc)

5) How much time do you normally spend building your drivetrain?

6) Do you prefer to use belt or chain? Why?
6a) What would prompt using one over another?

7) Generally, long robot, wide robot, or it depends on the game?

1) I (as in me myself and I) prefer high grip wheels (Colsons are wonderful). Smallish wheels (4"/5"/6" wheels are preferable, so the bot has a low center of gravity.

1a) Bumps (like this year's game and 2010) are the only reason I see for larger wheels.

1b) I think any team (large budget or not) should use a standard 6 wheel drop center tank drive. As for the wheel, traction wheels (2012 KoP wheels, treaded wheels, colsons) are the best economically, and usually the best performance wise. I'd say the 2012 KoP wheels are amazing for anyone, and a definite should-use for a low-resource team.

1c) I personally choose wheels based on my team's resources and our requirements. I suggest reading 1114's paper on drivetrains and wheels (http://www.simbotics.org/files/pdf/d...aindesign.pdf). I'm looking into Colsons, since they match my team's resources (middle) and requirements (high grip).

2) We make our drivetrain out of square steel tubing (.75" diameter, 1/16" thick wall), welded. However, steel is heavier than aluminum. It is stronger (one reason we use it), but we carefully plan out and control our weight so that our drivetrain is heavier, giving us a lower CoG, and the rest of our bot fits in the weight limit. I do not suggest this for a rookie/newer team.

3) AndyMark CIMple Boxes. I've got to say, those are some beautiful transmissions. I love them. They are small, light, cheap, come in the KoP (not in 2013), and allow for really nice driving speeds.

3a) We're yet to use shifters, however we only plan on using them when a strategy of ours requires a low gear (for example, pushing other bots up the bridge this year. No need for one in 2011).

4) For a long bot, we like the standard 6 wheel drive with the center wheel dropped 1/8", and our robots center of gravity between the middle and back wheels (better turning than directly in the center). This year we went wide, and did a 4 wheel drive. It was still pretty stable, but we designed with a really long wheelbase, which made turning a bit slow.

5) We design many different Drivetrains in the fall, so that by the time kickoff comes, and we have a general idea of what we want in a drivetrain, we'll have ideas to work off of. Designing a drivetrain shouldn't take any more than a day. If you have a smart design and know what to do, building a drive can be done easily in a few days (4-6).

6) While I hear belt is better in terms of efficiency, I personally will stick with #35 chain. I don't like designing drivetrains around the length of the belt. I'd rather be able to use chain and have it any length I want.

6a) I'd love to experiment with belt, but we've seen nothing wrong with chain, and have no reason to switch.

7) It all depends on the game. A wide bot was advantageous this year due to the bridge (and sometimes a wider ball pickup). My rule of thumb is this: If you need to fit in a specific dimension or pick something up using the wide side of the drive, go wide. Anything else, long.


I hope this helps. You should really look unto 1114's Kitbot on Steroids. It's one of the best drivetrains out there (top 95% of all drivetrains). http://www.simbotics.org/resources/kitbot Just because it's in the kit, doesn't mean it's bad. The kitbot was designed by engineers who know FRC. The KoS was designed by the team who has beaten those engineers.

[ttldomination]

1) What wheels do you prefer

I suppose I'd like to have three wheels in my back pocket.
Roughtop - If the field is all carpet, bam.
HiGrip - If the field is part carpet and part slick surface (like this year), mix these with roughtops.
Pneumatic - If a game is especially rough on the DT, then perhaps Pneumo would be something to consider.

1a) What kind of challenge/obstacle prompts a certain wheel choice

Kind of went into with 1.

1b) What kind of wheel would you use if you had a low budget (aka can't afford $1,000+ on
swerves)

I'd go with HiGrip (AM KOP) or Plaction Roughtop. Either way, wheels can often be used and serve as a good investment.

1c) How do you personally choose what size wheel you prefer?

We try to stay in the 6" range, but this is very game related.

2) What do you make your drive train out of? (Base train, custom, 80/20, etc)

We're using 8020 but I can see that its days are numbered. Most likely we will see a shift from 8020 to tubing in the next couple of years, provided we gain the required resources.

3) What gearboxes do you use for your motors, and why?

Toughboxes, and probably going to Toughbox minis this year.

3a) Do you use shifters? Why?

I think shifters do not make or break a robot, and I don't think anyone on the team is willing to drop the money to purchase one. If we work up the courage/budget to build/buy one, then we'll consider it, but for now, no one seems into it.

4) What kind of wheel setup do you choose? (4 wheels, 6 w/ 2 dropped, etc etc)

I just moved onto a new team, and they've had a ride when it comes to wheels (mecanum, 6wd, etc.). However, I think we've settled on your average 6wd-8wd DT hoping that the game isn't anything terrible.

5) How much time do you normally spend building your drivetrain?

This depends on how prepared you are going into the season. A well prepared team (as I hope we can be) will often get it done in a week. Others (as my team has been in the past), might take upwards of 2-3 weeks to get done.

6) Do you prefer to use belt or chain? Why?

We use chain, however, I've heard that belting is better (more efficient). Up until now, there isn't a terribly easy way to do belting, at least a way that doesn't require some form of machining. However, with the inclusion of belts in the KOP, we'll have to see what kind of hardware AM offers.

6a) What would prompt using one over another?

I'd say that weight, belting profile, and available hardware. It all depends on if we like what we see coming from AM.

7) Generally, long robot, wide robot, or it depends on the game?

I'd say it depends on the game, but one method won't kill you. As Billfred said, anything that has to do with massive ball intake, wider is better, but I've seen long orientation teams roll with the best of them.

- Sunny G.

[Gregor]

Originally Posted by SuperNerd256

3) AndyMark CIMple Boxes. I've got to say, those are some beautiful transmissions. I love them. They are small, light, cheap, come in the KoP (not in 2013), and allow for really nice driving speeds.

How do you know they aren't in the 2013 KOP?

[DampRobot]

Originally Posted by Gregor

How do you know they aren't in the 2013 KOP?

Because it was announced that they would be Toughbox Minis (currently sold on AM).

[EricH]

Originally Posted by Gregor

How do you know they aren't in the 2013 KOP?

Because the 2013 Kitbot details were released in the FRC Blog back in September, at http://www.usfirst.org/roboticsprogr.../blog-09-07-12. Specifically the following section:

If your team decides to receive the 2013 KOP Drive System, you will receive an AndyMark C-Base chassis system similar to the last few years. It will again be a six wheel drive robot, with similar wheels to 2012, and with enough material to drive all six wheels. The 2013 gearbox is a ToughBox Mini with 10.71:1 reduction. The estimated final drive speed is about 10 feet per second.

By the way, to the OP: You seem to have missed the three most important elements of a good robot last year. Those three are drivetrain, drivetrain, drivetrain (says Mr. Bill Beatty, who has 4 FRC Championship wins and 1 FRC Championship Finalist under his belt--I'm inclined to figure he knows what he's talking about). The KOP drivetrain is pretty close to bulletproof as far as drives go. The KOP on Steroids drivetrain (look it up in CD-Media) is good if you have a little extra money for extra parts.

Now, some answers for the questions, from past experience:
1) Depends. KOP wheels, KOP with treads attached, modified COTS all have their uses. If I had a low budget, KOP or AndyMark; typically a 6"-8" wheel with 6" being preferred; larger wheels only if terrain required and I couldn't find another method.

2) 2"x1"x1/8" wall box aluminum extrusion, welded.

3) I've seen single speed (ToughBox equivalent), multi-speed (AM Gen2 and SuperShifter, as well as drill transmissions that came in the KOP before 2005).

4) See "other method" referred to in 1 above. 6WD or 8WD with dropped centers most likely, but terrain and other maneuverability requirements may call for something else.

5) Are you including the time for designing? KOP is 0 design, about half a day to a day building; custom drives can be 3-4 weeks designing and go together in two days or be 1 week designing and 1 day assembling mechanical stuff.

6) Chain, #35 to be exact, despite the weight. It's pretty good at staying on the sprockets and staying aligned. #25 as well for some applications off drive.

7) Long, but depends on the game as to what accommodations are made for acquisition.

[Gregor]

Originally Posted by DampRobot

Because it was announced that they would be Toughbox Minis (currently sold on AM).

Oh I missed that, where was it announced?

[F22Rapture]

Originally Posted by Gregor

Oh I missed that, where was it announced?

Here

[Andrew Schreiber]

Originally Posted by Littleboy

http://recordings.talkshoe.com/TC-98466/TS-681568.mp3
That is a recording of a recent EWCP cast about drivetrains.

There is probably many threads if you would do a search on it.

That probably covers most topics.

Actually, that is the second cast on drive trains we did. It focused on skid steer systems. We have a previous one that is more general - http://recordings.talkshoe.com/TC-98466/TS-563546.mp3


Our full list of casts is available at http://www.talkshoe.com/talkshoe/web...d=98466&cmd=tc or in iTunes at https://itunes.apple.com/us/podcast/...477122221?mt=2

[ksafin]

I think somebody mentioned that many teams use Mecanum's or Swerve's (or at least explore these wheels) for the experience and knowledge of what these wheels offer, as well the advantages, disadvantages, and challenges of using them.

Part of why I brought this thread up is because next season by team was touching on the possibility of using Mecanum wheels. Now, if the game turns out to be something where Mecanum's would be a horrible choice, we obviously wouldn't use them. But if the game lends itself to Mecanum's decently, we were hoping to invest in some and hook them up.

Barring any complications you could foresee us having, does it make sense, in your opinion, for us to explore other options of wheels for our drivetrain? As much as you all agree that a standard drivetrain w/ 6 wheels like the kitbot or KoS is the best, I would still think that any team should explore different wheels, drivetrains, etc.

For some reason having a team that uses the same drivetrain for their entire career seems a bit shaking to me, as it takes away that aspect of Engineering in which we explore other possibilities, options, and then assess what we thought was best and how we move forward.

In essence, given what I've said, would you think it's ok for us to use Mecanum's next season if the game allows, and why or why not?

[ehfeinberg]

To start, a little background of my team. My team has minimal resources and limited access to advanced machining techniques. Since our founding in 2000, we have had many really creative designs and strategies, but it has always been hard for us to execute them to their full potential.

Because of this, from the years 2007 to 2011, we used Mecanum wheels. The team has always loved Mecanum wheels because they are not too difficult to put together, and they have given us omni-directional motion. What I feel might be the downfall of Mecanum is not the drive system itself, but knowing when to use Mecanum. I've found that our team automatically defaulted to using Mecanum even if it wasn't right for the game. This year when we used a 6 wheel drop center, even though it is basically the Kit Bot, we had a very rough time and most of our problems were with the drive.

This off-season, we decided to challenge ourselves and build a swerve drive! To make it easier on our team, we decided to use the Revolution Swerve Modules. We also decided to use belts. I'll let you guys know how it goes.

That being said, these are my personal thoughts on your questions.

1) I prefer either AndyMark Performance wheels or Colstons. Although I have not used either wheel, my team did not have a very good experience with the AndyMark Plaction wheels. Colstons and Performance seem to hold up better and require less maintenance.

1a) If there were bumps like 2010, I would probably consider pneumatic wheels. My team used Mecanum wheels in 2010, and even though they worked OK, they didn't hold up to well.

1b) For a low budget team, I would either use the Kit-Wheels, or Colstons.

1c) I normally like 4 or 6 in wheels. We used 8 in wheels this year making it much more difficult to get the gearing that we wanted.

2) Since our existence we have been using 80-20 for the drive frame. If we found that we were going to be overweight, we would later switch 80-20 out to quick frame. This offseason we are experimenting with 2x1 1/8 in box.

3) We normally use whichever gearbox is in the KOP. This offseason we are testing out some Tough Box Minis.

3a) No, I haven't really seen the need for shifters. In the competitions we've played at I haven't seen Shifting give teams that much of an advantage. Until we get to the level of consistently making it to nationals, I don't feel that shifters alone will give us that much of an advantage.

4) Like I said, we used Mecanum from 2007 to 2011. 2012 we used 6 wheel, and 2009 we used something which we called backer-man(3wd with the back wheel a swerve module).

5) Too much. When we used Mecanum, we normally had a driving robot week 2. When we switched to 6 wheel last year, we got it done week 5.

6) In the past we have used Chain, just because we have always used Chain. However, this offseason with the KOP switching to belts, we decided to test out some belts.

6a) Belts are lighter and quieter. I'll tell you what we think of belts when we finish our off-season drive system.

7) We normally go long, but if the game requires it we will go wide. I prefer flop bots though.

I hope you find this information, regarding a team with less resources than the normal teams that post on CD, useful.

[MichaelBick]

Use the offseason to experiment, not the season. If you haven't tried mecanums yet then you shouldn, use it during the season.