Drive Train War

Everyone knows that almost every robot has to have a drive train (I say “almost every robot” because my team some how pulled of a decent drive trainless robot in 2013). Of course all drive trains are not equal, some are more manuverable, others are better a pushing. Arial Assist I my opinion was one of the most drive train centric games, if not the most drive trains centric game FIRST has had. So I have a few questions I would like to ask:

First off, what is your favorite drive train?

Second, what do you think is the most important trait a drive train should have? Examples would be, Highspeeds, pushing power, manuverablity, ect…

Third, why do you believe this thing about drive trains?

And finally, it has been brought my attention by this article that FRC may be in the midst of a “Drive Train War”, I’m curious about everybody’s thoughts on that, is FRC in a metaphorical “Drive Train War”?

Swerve drive (field-centric).

This year, we used swerve, and it worked out pretty nicely.
We had been prototyping for the past couple years, and we managed do deal with the complexities of the drivetrain decently well.
However, it was difficult to get working correctly. During our week 0 event, we could hardly move. But by our first real event, it worked nicely.
Our main problem throughout the season was dealing with gyro drift in our field-centric system. It was not until this offseason that we managed to alleviate these problems.

This gyro drift was the only major issue with our drivetrain, and in my opinion, the swerve was ultimately advantageous.
We were pretty quick, and obviously had exceptional maneuverability.
We can’t win pushing matches versus powerful tank drives, but we can still play effective defense. And if defense is being played on us, we can just slide around them.

Next year, if we do swerve again (which is likely), it will be even better now that we have all the issues worked out.

If you can figure out how to build a swerve and make it work, I think it is very beneficial to have.

On #2. Yes.

My favorite drivetrain is the one that gets the most practice time, because a drivetrain’s single most critical component is the driver. If your team has not fielded a complex drivetrain before, you had better be practicing with a prototype of that drivetrain NOW if you expect to use it for the 2015 season. This is especially true for swerve drivetrains.

My least favorite part of inspecting robots has been repeated visits to the pit of a team that is still trying to get its complex drivetrain working on Friday morning. This has happened far too often, and it is extremely frustrating both for that team and for their alliance partners in early qualifying matches.

  1. West Coast Drive, particularly those with octagonal or hexagonal bumpers. Simple, easy to repair, and very upgradable.

  2. A drive train should be, first and foremost, reliable. If the drivetrain breaks, it is USELESS. In addition to reliability, the drive train should support the robot that is using it to drive. A robot like the Cheesy Poofs this year benefited from a very fast and maneuverable drive train, and wouldn’t have been as capable if it had the drive train of say, 610. On the other hand, a defensive robot, like my team’s, benefits heavily from the ability to push. Shifters make it more possible to have the best of both worlds, but what you focus on more should be based on your robot’s design.

Note: This post is not intended to discredit complex swerve or H drives, but to state that they’re not always that beneficial in terms of robot performance. The experience gained by creating one of these drives, and the skill required and gained by to create a drive like this is incredible, and can teach students a LOT about design. My team, and several others, have found a simple drive train works best, but teams like 16 have been very successful with swerves. Don’t let other teams discourage you from building a complex drive train if you feel that the experience will help your students, but recognize the difficulty in creating a drive train like this.

Swerve drive seems like it would be obivous but in my opinoin there are more exotic drive trains that are wold give swerve a run for its money. Check out 624’s drive train.

  1. Gear Drive - It’s durable, easy to maintain, and, when done correctly, can push basically any robot on the field.

  2. Control of Position - In my opinion, the fundamental objective of a drive system is to allow you to get to where you need to go on the playing field and stay there as long as you need to. Depending on the game, this could mean that your drive might be fast and maneuverable so you can avoid object or other robots, it might mean your drive is slower but strong enough to allow you to defend opponents or prevent opponents from defending you, or it might mean your drive needs to do a little of both depending on the situation during the match and you use a two speed shifter or some other mechanism to achieve this goal.

  3. A robot is only effective if it can get to where it needs to be on a playing field. You could have the most accurate shooter, or the most effective arm, but if your opponents can hold you on the other side of the field from the goal, they probably won’t do you any good. I’ve seen matches, and even entire regionals that were won simply by having a robot on an alliance that could stay between an opposing robot and the goal, even if blocking was the only thing that robot could do.

  4. The “Drive Train War” was mostly the effect of this years field being so open (no obstacles), by comparison, the games of the past several years have had field obstacles and defense rules that made such drive systems impractical/unnecessary.
    Strong drive systems are nothing new, 6CIM drives were possible as far back as 2006 (maybe even before that) before bumpers were mandated or even common, they’re simply easier to build in recent years due to the availability and relative low cost of high quality COTS gearboxes. Drivetrains will of course, continuously improve, and teams need to be able to adapt to handle the different play styles that come with better systems each year just as they would with any other part of an FRC challenge.

  1. I am partial to swerve drives. 2451’s recent coaxial shifting swerve drive weighs in at a mere 7lbs, rivaling “normal” drivetrains.

  2. Although you sacrifice the ability to easily have 6 cims by going swerve, you gain so much maneuverability that it doesn’t matter. Maneuverability and a good driver are the most important parts of a robot IMO. Defense can easily be played by swerve by just putting the wheels perpendicular to the other team, but the ability to snake around defenders makes it a very powerful force on the field.

  3. Most games can only be played by moving. 2013 was an exception, where you could just climb or shoot form teh safe zone, but I can’t recall a champs-winner that couldn’t move.

Motor limits were in place to prevent 6 CIMs every year besides 2013 and 2014. The MiniBike motors (things that looked like large CIMs) used to be allowed (2006, not sure about 07) but you couldn’t have 6 CIMs on any robot until 2013 as far as I know. We also didn’t get MiniCIMs until 2013 as well. Before that adding drive power normally meant adding either Fisher Price motors for a long time or eventually BB775s (when they weren’t case shorting) when they were introduced both of those are normally worse on drive trains do to their lower thermal mass and need for air cooling. 6 CIMs and the addition of MiniCIMs has made the process much easier, COTS gearboxes or not.

It’s worth noting that 6WD and 8WD drive trains really are the gold standard in FRC. I say this so that teams who have never done a exotic drive train before don’t think they have to come up with a swerve drive or some such to be competitive. BeyondInspection has a nice breakdown of the drive trains that made it to Einstein this year and previously. You’ll see the vast majority aren’t exotic, just very well executed 6WD or 8WD.

One also has to consider the ratio of teams that use a drive versus the number that actually make it to Einstein or elims.
That being said, 6WD still makes it Enstein more often than other types of drivetrains.

The best drivetrain is the one that gets you where you want as fast as possible as consistently as possible.
Tankdrives have the pushing power, speed and simplicity that are required to accomplish this while being the most reliable and, usually, lightest drivetrain.
You usually can’t win a match that you don’t attend because of mechanical problems.
I have yet to see a mecanum or omni based robot that is more maneuverable than 254s 6 cim tankdrive.

I have yet to see a mecanum or omni based robot that is more maneuverable than 254s 6 cim tankdrive.

Agreed. I don’t think anyone who saw 254 or 1114 on the Einstein field last season would say “Too bad they don’t have swerve drive … they’d really be competitive then.” So much of it is how well your driver is able to leverage the drive train you have. If your exotic drive train is fragile, took too long to get working, or it takes too much practice to learn to drive well, it’s not a good tradeoff IMO.

What makes a gear drive more durable, easier to maintain, or better at pushing than chain or belt drive?

To answer your questions in order…

What makes a gear drive more durable?

  • It completely eliminates the problem of breaking a chain or belt, and, in my experience, we’ve never had a gear break on a gear drive.
  • Throughout 3 districts this year the only maintenance our geardrive required was replacing worn wheel tread, which was typically done between events. Also after our last competition, we re-examined the drive to find that aside from a bit of dirt, the gears had virtually no wear.

What makes a gear drive easier to maintain?

  • Changing out wheels (to replace worn tread or otherwise) is substantially easier as there is no need to fight with tensioners, just pull the shaft out (which in our case requires pulling one pin) and the wheel drops out the bottom of the drive.

What makes a gear drive better at pushing?

  • Better pushing depends more on the particular implementation of a geardrive. In our case we utilize gears to allow us to have a lot of wheels (12) on the drive system, this combined with the arc pattern of the wheels and a two speed gearbox with a very low low gear (~3fps) allow our drive to maintain maximum traction during a pushing match. In two iterations of using this type of drive we’ve never found a single robot that we could not block or push, and in many cases its possible to push multiple robots at once.
  • Though not as critical as the features above, a well build gear drive also has less friction than many (but probably not all) chain drive systems making it slightly more efficient.

I have thought for a few years that “fork-lift drive” would become a powerful drivetrain in FRC, but I have yet to see it take 0ff. This would be steerable rear wheels and fixed front wheels. The fixed wheels would utilize a two speed gearbox with a large spread, and the two rear swerve modules would be geared to match the faster gear.

Didn’t 67 run something similar in 2003?

Possibly. I don’t know the 2003 67 robot.

Here’s a video of it. Its an interesting drive but I’m not sure how well it would lend to an open field.

Not sure on 2003. But they did most recently in 2008.

I also believe that 16 only had one swerve module in their 2003? Robot. Hopefully John is lurking around and can confirm. I’ve got a pic of the module but I forgot to tag what year was what.