How long did it take for you to build your drivetrain?

[jonathan lall]

Building a drivetrain is not so much a matter of how long, but rather of team coordination. I would contend that you cannot have a consistently good robot without equally good teambuilding. In order to make an original gearbox for example, you really should be streaming or subteaming your human capital. I thought of it sort of like an economic model while I was a student with a building and leadership role. If you don't specialize, you can still make a great robot but you can also say goodbye to a really good drivetrain for the simple reason that multiple tasks performed by the majority of the group in order put too much pressure on the group. There will of course be an opportunity cost of specialization (a perfect function might be forgone by allocating 30% of your students to making a good drivetrain), but the final yield will be much greater.

Let's go by Billfred's example. Even though most teams don't take quite that long to make a drivetrain, if you were to syphon off 30% of your students to a hierarchical subteam for mobility (or whatever the optimal percentage you might find to be), robot building gets a lot easier. New Canadian teams for example are encouraged by 188 to use our old team model of dividing the majority of the team into two subteams of Function and Mobility (the rest might be part of a publicity or fundraising group). We have since heavily tweaked this model, but it's a good starting point.

As for some tips about how to speed up your drietrain's construction, a look into the rules will help you out (disclaimer: they can and do change at the start of every season). For example, you can order certain materials and parts ahead of time and design your gearbox on CAD software before the build period even starts. Most veteran teams have this pre-build process mastered. In our case, the actual construction of our tranny/gearbox took nearly five weeks. Considering we hand-machined all our parts, that's rather good, I think. Still, the real gains you can make in terms of time to build your drivetrain are more in adminstration than technics and tools.

[greencactus3]

well as a rookie team we used the kit pieces to make a 4wd robot and it took maybe a week to finish it, and then later on about 2 days more to add the system to switch between 4 and 2 wheel drive

[NoodleKnight]

good point, Jonathan.

But for some teams, like us, organizing all of the team members isn't as easy as we hope it to be. Mainly, we have a mentor who likes to challenge and argue about every single design created, unless he created it himself or he helped create it. So right now I'm trying to search for some hard-evidence to show that certain types of drivetrains can be built within the build period.

[Steve W]

Originally Posted by NoodleKnight

good point, Jonathan.

Mainly, we have a mentor who likes to challenge and argue about every single design created, unless he created it himself or he helped create it.

Team 188 is no different. Just ask any of them. Hmmmmm I wonder who that mentor is.

Challenging is good because if you can't explain the reasoning then perhaps not enough went into it. Remember that the mentor has gone through the processes himself and know his/her reasoning. He/she doesn't know yours and needs to build confidence in your thought process. It would be good to question their ideas to learn from them or to see if they really thought their process through.

Don't spend too long on building any part of your robot. This can cause problems on the team and there won't be time to test before ship.

[Arefin Bari]

Just like Jonathan said earlier It doesnt matter how long does it take to make the drive train. I personally believe in efficiency of the product, not how much time it took to me to build it. Last year it took us a good 3 weeks to get done with two of the transmission. but I have learned a lot of things. you should concentrate on every single aspect of the robot when you have only 6 weeks to finish it. concentrating on one aspect (drive train) and not the other (mechanisms, others) may give you a lot of problems at the end of 6 weeks. it was proven to me last year. Drive train is one important part of the robot as well as other parts of the robot are very important too. team 108 is ready to roll and we like challenges (ask any team who were at Midwest and Nationals (in galilio division), they will tell you) for 2005 season. Now only if dave would give us the game hint...

[the_mayor]

We like to get the drive-system out of the way first and it would be done in about 2-3 weeks. But the transmision took us FOREVER last year. The problem we were hung up on was trying to figure out how to make all 3 gears shift accurately. I think we spent several days trying to make the shifting mechenism work.

[Macdaddy549]

We use the same drive train as past years. I hope it will be the same this year. If you have something that works use it. You don't have time at this point to re-invent the wheel.

[rmmlg]

Last year, we spent most of the 6 weeks designing the drive train. This didn't work out so well since we didn't have any working special features. Granted, we had a very good drive train, but special features are nice to have. Things we have decided to do differently this year:
-we've done more brainstorming during the off-season
-we plan on doing the brainstorming differently and decide on a drive train and design it during the first week, so that it can be built with time to spare in testing and practicing driving for the last week of the build season
-we weren't organized enough for brainstorming last year and didn't meet oftne enough during the first week

So that's what we plan to do differently to hopefully spend less time on a drive train and more time on special features.

[Bill Gold]

Here has been my history with drivetrains in FIRST…

In 2000, as a junior on Team #258, we completed our drivetrain in three weeks. It was a simple 4wd chain and belt driven system run by the ole 3/8” drill motors. The reason it wasn’t completed earlier was because we had very few members of the team, and were working simultaneously on the ball mechanism and drivetrain.

In 2001, as a senior on Team #258, we built pretty much the same 4wd drivetrain as 2000 (but completely with chain and sprockets) in 2 weeks.

In 2002, as a college student on Team #258, we built a 2-speed shift on the fly sprocketbox drivetrain powered by just the CIM motors in 3 weeks, and then spent another week lightening it up significantly.

In 2003, as a college student on Team #258, we spent the first week and a half designing our swerve drive system powered by both 1/2” drill motors and the CIM motors (and rotated by the Globe motors), and then we didn’t complete it until about a week before ship date. We took our time on it because, since it was a simple game, we didn’t really have any other mechanism on the robot. We ghetto-rigged some wings on Thursday of SVR. If we had really tried we could have had the swerve designed in half a week, and then had it completed somewhere along the third week.

In 2004, as a college student on Team #258, we spent about a week and a half designing our “stretcher-drive” system powered by the 1/2” drill motors, and finished it about a week from shipdate. We would have finished it earlier had we had more than 3 people machining parts along with parts we outsourced to a machine shop. There was a crapton of machining on that robot for a team of just a handful of us. It was a lot of fun to work on, though, looking back on it.

[NoodleKnight]

Originally Posted by Steve W

Team 188 is no different. Just ask any of them. Hmmmmm I wonder who that mentor is.

Challenging is good because if you can't explain the reasoning then perhaps not enough went into it. Remember that the mentor has gone through the processes himself and know his/her reasoning. He/she doesn't know yours and needs to build confidence in your thought process. It would be good to question their ideas to learn from them or to see if they really thought their process through.

Don't spend too long on building any part of your robot. This can cause problems on the team and there won't be time to test before ship.

I agree that challenging designs is a good thing, but not when the arguing starts to span over many days into our limited build period. Thats why I'm trying to get as much information as I can to help limit time spent arguing over designs.

[Steve W]

Originally Posted by NoodleKnight

I agree that challenging designs is a good thing, but not when the arguing starts to span over many days into our limited build period. Thats why I'm trying to get as much information as I can to help limit time spent arguing over designs.

Arguing is never a good idea. Challenging and discussing is OK but arguing doesn't do anything but cause hard feelings.

[Veselin Kolev]

Originally Posted by Steve W

Arguing is never a good idea. Challenging and discussing is OK but arguing doesn't do anything but cause hard feelings.

A strategy we're using this year is to incorporate drive train discussions into our trainings with rookies. For example, most people that start to go to trainings have no clue about what’s going on with FIRST robotics, never really read CD, and must be taught everything. That lets us teach them from ground zero, and a little thing we do is give them mock brainstorming sessions, where we "reveal" to them old FIRST games, and they use their knowledge of drive trains (that we gave them) to determine not only the ways a drive train could go about doing the tasks of the game, but which drive train could do it best. In simple words, we let them decide what the best drive train would have been. Of course, as a veteran member, I have to sometimes step in and remind the rookies about our limited money and machining capabilities. It really helps the brainstorming process if you've done it before, but not for real. It's real enough to make a lasting impression on rookies on how to go about making rational decisions during build. And keeping the brainstorming process short.

Anyway, back to the real question. Last year my team did swivel drive, and I was managing that system. The decision process was rather quick, and the design took three days to finalize. Took another three days to distribute parts amongst our two machinists, and then took two weeks to machine. Then... the horrible week of assembly... So in the end it was mounted on our chassis by end of week 5. Well, our school has finals the first week, so we never do anything then. So it took a good 4 weeks to do it all.


The main reason why it took so long to machine and assemble was that it was not an efficient design. It was good and strong, but could have been a lot easier to make. You gotta remember that you can design anything, and build anything, but the building part will always take the most time. It helps to know how to use a mill and lathe so you can design something that is easy to make.

I have a lot to say about this actually... just because there have been too many bad experiences. A few more things that will help cut time: once you finish your design, get accurate quotes for everything you will have to buy. For example, if your design needs four bearings that cost $150 each, don't do it. Redesign your butt off. Also, once you finally do finish the design, order the parts immediately. Who knows how long it will take to ship, and its nice for the machinist to have bearings while he machines the part they go into.

Also, once you finish a design, print out all the parts drawings and give them to a machinist. Don’t worry that the machinist will be discouraged with a big pile of parts to make. They like it when they don’t have to call you and ask if there’s any more. Giving them drawings every few days seems like a good idea, but it just keeps the machinist in the dark and kinda ticked off at you… well at least for me. Anyway… enough blab. Hope that helped.

[NoodleKnight]

Originally Posted by Steve W

Arguing is never a good idea. Challenging and discussing is OK but arguing doesn't do anything but cause hard feelings.

Well, maybe what I said sounded a bit too harsh. We hold very lengthy discussions, but we never break into a full fledged arguement, maybe small ones...

Veselin, I wish I could give my drawings to a machinist, but all of our experienced machinists graduated, I'm the only student machinist left. Add to that, we used to do a bunch of machining at the Stanford Linear Accelerator Center, but now after a bunch of problems (we didn't cause them), we're not allowed back in there, which also makes it hard to teach new students how to use machine tools, since we don't have any. We do have access to private machine shops, but we can't go to the shops when ever we choose. Mainly, the majority of the team is concerned about time consumption when it comes to new designs (due to past experiences).

[BillyGoats]

Team 61's drive train is always done by the second week. We use the same drive train every year.. We make simple easy changes, becuase every year our drive train has been fail proof. In order to aviod problems we do not do any drastic changes unless we have too.

As always check out our website:

http://www.valleytech.k12.ma.us/robotics/

tons of pictures from last year and the past 9 years. You can see how our drive train has not changed as much over this time!!

[AJ Quick]

We don't have the money to do FIRST anymore, but we do BattleBots. Luckily there are lots of companies that make all in one gearbox motor solutions.. so our drive train only takes a little bit of time to get mounted!

I don't recall from the FIRST rules, but if you can use pre-built gearboxes, I do believe someone makes a gearing set for the CIM, all done with alum mounting blocks.