To swerve or not to swerve

[Caleb Sykes]

I think that this sums up pretty well many team's experience with swerve.

...Tried swerve. Didn't move for 1.5 regionals.

http://www.chiefdelphi.com/forums/sh...1&postcount=28

Seriously, the team that posted this has 70+ members, a couple of good years under their belt, and this still happened to them. Don't do swerve if you haven't already prototyped it over the pre-season. I don't really care how big your team is or how much FRC experience you have. Just don't do it.

[AdamHeard]

Originally Posted by aldaeron

#SwerveSoHard

I haven't built one, but I have always wanted to (ask my team). I looked into it quite a bit in the offseason. One of the items that is required (for a Revolution Swerve Style) is a bevel gear set per wheel. I haven't seen these at FRC type stores (AM & VEX). The cheapest place I found them is on Amazon

http://www.amazon.com/dp/B005AVUCJ8/ (straight bore)
http://www.amazon.com/dp/B005AVU6T4 (keyed)

There are a lot of sizes and bores there.

My cautions after doing some research. Compared with other drive styles, Swerve :
- Is more expensive (count up all the miter gears, bearings and sprockets)
- Is heavier
- Is not in the standard WPI library (probably not in LabVIEW either)
- Takes up quite a bit of room in your chassis corners (you have to be able to rotate the whole wheel and all the hardware with it 360 degrees) (it's tall too!)
- Forces a smaller wheel track

A swerve is a significant undertaking and is a potential season breaker (imagine if you can't drive at all!). Make sure you have a backup plan.

Consider your budget and number of team members before committing. Unless you are a huge team it is likely that the effort expended on learning how to build, program and drive a swerve will take away from your other game mechanisms. Essentially the drive train is your main mechanism. Luckily there are some examples (BuildBlitz & Ri3d) to help you build a quick shooter and pickup.

Also not a bad idea to read this from 1114. They address the swerve in some detail http://simbotics.org/files/pdf/drivetraindesign.pdf

I am basing this on the experience of a friend of mine from 1296 who described his experience making a swerve during the build season. It essentially took most of the season's effort. In following years it was easy to do, but the first year and all the lessons learned were tough.

Good luck

-matto-

Vexpro has 3/8" hex broached 15T 12DP bevel gears that are kind of the standard for FRC swerves (starting with 118, then spreading to 1625, 1717, 973, etc...) for a great price.

[gurellia53]

Originally Posted by inkling16

I think that this sums up pretty well many team's experience with swerve.

http://www.chiefdelphi.com/forums/sh...1&postcount=28

Seriously, the team that posted this has 70+ members, a couple of good years under their belt, and this still happened to them. Don't do swerve if you haven't already prototyped it over the pre-season. I don't really care how big your team is or how much FRC experience you have. Just don't do it.

We (1736) had a similar situation in 2012. I wasn't on the team but we ended up locking the swerve modules in place on the Thursday of the regional and driving it tank style.

[lynca]

Originally Posted by Tyler2517

We are running it.

I would imagine you are not the only one, my guess would be the teams with experience are strongly considering swerve. (i.e. 111, 118 , 973 , 1625, 1717)

If they are not doing swerve, many of the top teams are considering butterfly , octocanum and beyond.

Swerve modules are available from 221 if people are interested.
http://www.team221.com/viewproduct.php?id=89

[Popmaster288]

A swerve drive gives faster strafing and more traction than Mecanum. Depending on what your priorities are, it could be a very good option. If you don't have experience with building swerve drives, then it's a bit too late to start developing one for this season. The only thing that swerve falls short is with potential pushing power. A 6 CIM-6 wheel drive with a ball shifter in low gear will out push a swerve. If the Pro's of a swerve drive outweigh the difficulty of making one, then a swerve drive, in my opinion, is a very good choice.

[AdamHeard]

We won't be doing swerve, as fun as it would be. I can guarantee (off assumptions) that 1717 will be doing swerve though because they're crazy and dance like penguins.

Originally Posted by Popmaster288

A swerve drive gives faster strafing and more traction than Mecanum. Depending on what your priorities are, it could be a very good option. If you don't have experience with building swerve drives, then it's a bit too late to start developing one for this season. The only thing that swerve falls short is with potential pushing power. A 6 CIM-6 wheel drive with a ball shifter in low gear will out push a swerve. If the Pro's of a swerve drive outweigh the difficulty of making one, then a swerve drive, in my opinion, is a very good choice.

All things equal (motors, rollout ratio, tread, weight, etc...) swerve and 6wd are the same in pushing power.

[runneals]

We built swerve our first year (Rebound Rumble) and didn't really use it since our drivers didn't get much time to learn it. Plus, our drive was kinda glitchy (which could be improved), but I would suggest to pick something with less parts (ie. Butterfly).

[jeremylee]

Originally Posted by gurellia53

We (1736) had a similar situation in 2012. I wasn't on the team but we ended up locking the swerve modules in place on the Thursday of the regional and driving it tank style.

I was on team. Disaster. Mentors trying to save it, students disengaged. Even tank drive was disappointing.

[Jefferson]

Originally Posted by aldaeron

#SwerveSoHard

I haven't built one, but I have always wanted to (ask my team). I looked into it quite a bit in the offseason. One of the items that is required (for a Revolution Swerve Style) is a bevel gear set per wheel.

http://en.wikipedia.org/wiki/Slip_ring

[cadandcookies]

Originally Posted by inkling16

I think that this sums up pretty well many team's experience with swerve.

http://www.chiefdelphi.com/forums/sh...1&postcount=28

Seriously, the team that posted this has 70+ members, a couple of good years under their belt, and this still happened to them. Don't do swerve if you haven't already prototyped it over the pre-season. I don't really care how big your team is or how much FRC experience you have. Just don't do it.

As that team, I can tell you that our performance that year was tough for a lot of us to swallow-- especially the juniors and seniors who spent countless hours not only during the season, but also prototyping and designing before hand. Your team is the only ones who can realistically assess whether you can pull off a swerve drive. We made that assessment and thought we could pull it off, but we were wrong. Be prepared for that if you do end up going with swerve having never tried it before.

If your priority isn't to win competitions and you have little to no experience actually building and programming a swerve drive, it's still a completely valid option for this year's game. Just understand that what you might be gaining in knowledge you will most likely lose in terms of being competitive on the field.

In summary I guess, I can say "Been there, done that, got burned." But sometimes it's necessary to get a bit burned in order to learn a lesson.

[Tyler2517]

The year before i joined my team they tried running swerve or more of a "crab drive". It ended in a way that those member of the team then are still scared to death of trying to run it. We ended up with it falling apart on the field...
Running a swerve is probably the most risky thing you can do but it has great rewards if done right. If you don't have 1000 hours of meeting time to make program and test it in the build season it will be a nightmare to run. If you do try to pull it off have a back up plan and look at the teams who have mastered it and see how they pulled it off some of them even release cad. Even if they do give out cad some times copying it will not work you will not understand why little things were decided apon and may end up costing you dearly in the end.
They also cost a lot of money to buy and the parts can take weeks to arrive even if you built it, but did not have the belts to run it you may not have the programming to make it the best possible.

=No matter what your team is the only one who can decide if you can run it.

[FIMAlumni]

The best swerves (1717 and 16) are the best drive trains in theory I've seen in first. The 4 downsides to these are reliability, time, weight, driver comfort. If you think you can finish a reliable swerve that fits your weight constraints in enough time to get your drivers used to driving it, I don't see why you wouldn't.

[NickTosta]

I doubt you'll see many top teams using a swerve (or mecanum) this year. With a wide-open field, symmetrical game pieces, and many different spots to score from, the additional control and motions you gain from swerve really won't be a significant advantage.

Swerve is a major engineering task, especially for teams who aren't intimately familiar with it (at least one or two successful implementations). The way I see it, you have to be either extremely comfortable with building/programming/maintaining/driving a swerve robot, or you have to be 100% sure that it's going to be a major advantage to have it on your robot. Preferably both.

Originally Posted by shades23

I see the advantages with a 2 speed being able to our run defense bots and push around others but if you get someone who can operate swerve drive you can practically get out of all sorts of situations as seen with 195 in 2012 and 1717 D'Penguineers... in almost every year recently

<3

Glad to see people from other teams like it and remember it! Swerve is pretty crazy to drive.

[Bryce2471]

Team 2471 made the decision to go with a swerve drive system this year.
In short, we decided that in this years game it's important to put the robot in the right place on the field facing the right direction, as quickly and efficiently as possible.

It will be a tough rode to go down, but we did have a successful off-season bot with swerve drive this year. we also did a somewhat successful crab drive in 2012.

Some of the disadvantages people have mentioned earlier in this thread not a big issue with the design we are using. We also did a drop center drive bot this off-season. The drop center drive base weighed in at about 36 lbs and the swerve drive base was about 37 lbs. The modules weigh 7.5 lbs each, only take up a 6" by 6" square, the wheels touch the carpet 2" in from the edge of the robot, and they only cost 222$ per corner.