What type of metal for drivetrain?

Our team is looking into making a custom drivetrain, as we have obtained the funds and means to do so, after a year of using a kitbot. Our initial plan was to use 6061 T6 Aluminum, but after some research, it appears that 6061 T6 is extremely hard to bend without cracking it. What type of metal would you guys suggest to make a custom drivetrain out of, that could be shaped more easily?

1 Like

2102 typically uses 6061 aluminum, but we don’t do any bending with it for the drivetrain, just gussets and riveting.


I’d say the standard way to make drivetrains is with 6061 tube, so no bending is usually required.

If bending is a thing you want to get into, I’d suggest 5052.

1 Like

The tube idea sounds like a good idea, but if we were to go down the 5052 route, would you expect any significant strength issues? I know this will largely depend on the game, as the amount of defense usually varies some amount per season, but on the whole, do you think that 5052 would not be a noticeable decrease in strength over 6061?

Any time we’ve needed to bend aluminum, we’ve used 5052-H32, mostly because it was free to us. It bends well, but gums up cutting tools mercilessly.

I recommend you look in to REV’s ION Structure, specifically their MAXTube and 90 degree brackets.


Unfortunately that wisdom is not with me, I’ve only ever used 6061-T6 in FRC. I will say that 5052 is annoying to cut because it is ultimately softer and gums up cutters. If you’re new to drivetrains I’d suggest buying components from the REV Ion ecosystem and assembling them to your liking.

I might be missing something, but why do you have the requirement of needing to bend your metal? If this is your first year using something other than the kitbot, I would highly recommend using a standard tube, gusset, and bearing block setup with 6061 1/16" wall 2x1 tube. You need to make sure your chassis the most bomb-proof part of your robot, so I would recommend against trying new construction styles so close to kickoff, and if you do try something new to you, stick to things that have been tried and true in the FRC community.


Do you plan on welding?

My team uses 2x1x1/8” 6061 T6 Aluminum Tubing with 1/8” custom sheet metal gussets. We may move down to 1/16” tubing to reduce weight, but if our robot is not looking like it’s going to have weight issues we’ll stick with 1/8” to keep our CG low and have a rigid protected drivetrain.

I’d recommend a similar setup with 2x1 tubing and gussets. REV has a guide to create a WCD drivetrain using their MAX tubing, not requiring any custom parts to create a drivetrain. Thou who shall not be named (VexPro) sells their own pre-drilled versaframe and gussets that can be used to make a simple chassis without the need for custom parts, although I would not recommend that due to QC issues and other issues currently surrounding the company.

TLDR: COTS Pre-drilled rectangular aluminum tubing and gussets can be used to make an easy drivetrain. REV sells parts to make one.

So, there’s some interesting stuff here.

  1. 6061 aluminum is not inherently unbendable. That said, you do NOT want T6 (Temper 6) if you’re going to bend 6061, you want T0 (untempered) and then you’ll need to temper it properly after bending. If you happen to have a friction stir welder you can cheat this… but if you even have access to one, I’d be surprised.*
  2. There are different ways to make custom drivetrains.
    –Aluminum (or other) tube stock is probably THE most common, typically 6061-T6. Weld it, use gussets with rivets or bolts, or possibly a couple other options.
    –Second most common: bent sheet metal. Typically aluminum 5052, flanges top and bottom.
    –Plate-and-standoff is pretty rare, especially these days, but it can be done. Works like bent sheet metal but without needing to put flanges in. Also heavy as [censored].
  3. With respect to relative strengths of 5052 and 6061: Properly designed, a bent 5052 sheet can be about as strong as tube 6061. Key word: “Properly”. To me, it sounds like you’re pretty late in the offseason to be learning the proper methods here–I could be wrong, but that’s what I’m seeing.
  4. The REV Ion extrusion has been brought up. if I were looking to do a custom drivetrain for the first time, I would go with that. Actually, let me rephrase that. I would take the Kitbot frame (ONLY the frame) and build a custom drivetrain that works within that. Different wheels, different gearbox, etc. The exception here would be if I were going swerve, and in that case I’d be seriously reconsidering doing that in time for next season.

From experience. I’ve worked with a group that cracked, yes cracked aluminum plate by bending it around the minimum bend radius mandrel. An identical plate run through the nearby stir welder (in a “process” mode not a “weld” mode) was bent around a far smaller mandrel with no issues. On the other hand, FSW machines are big expensive beasts and relatively rare.


Yeah, sorry if I wasn’t that clear, we weren’t really set or anything on bending metal, but our team has done it in the past, so we wanted to see if that option made any sense. Our team is pretty well established, but we recently had a pretty big loss in workspace due to Covid and external factors, so our last season was pretty rough, and we had to end up using a kitbot. In the past though, our team always made custom drivetrains, and since we had seen some drivetrains that were bent metal, we were just exploring options, as we now have a good amount of money and workspace options. If tubing makes sense though, as seems to be the general consensus within this post and others, that is most likely what we will go towards, as though some on the team have knowledge on bending metals, it seems like it is unnecessary

1 Like

There’s nothing saying that you can’t use bent metal elsewhere on the robot, either. Example, you could do a sheet-type superstructure (if the game warrants it), and tie that onto a tube base. Or, for some extra strength: bent bellypan (either for connecting to the tube or for extra strength on its own).


Eric, isn’t that a different recommendation, rather than a rephrased one?

Or maybe I missed a very important change – will the AM14U5 be the KitBot for 2023?

I am thinking that, if a team wanted to customize, it might be better to go with your un-rephrased suggestion and build around REV Ion structure. However, if the same team only wanted to experiment with different wheels, gear ratios, motors, etc., then they might be better off with building around the AM14U5.

Many teams will have already made their choice by now, because the AM14U5 opt-out date was several weeks ago.

Custom drivetrain for the first time: KOS*–upgrade everything except the frame–would be my recommendation. Rev ION would be if I was upgrading the frame. (There is their competitor, but right now ain’t nobody going for that one.)

That said, this team has past experience (and hopefully some of it still around) building custom frames, so that part doesn’t apply.

*For those that aren’t old-timers: Kitbot On Steroids. A few hardware swaps–wheels, chains, shifting transmission, as I recall–took the Kitbot from a utility drive that didn’t suck but could be improved on to a really good drivebase, which would presumably be followed by a custom base in later years.

It’s worth noting that we have a new iteration of the KoP on steroids courtesy of 7461.

We use 6061 tube or angle for most frame components.

We did have a bent sheet metal frame using 5052 one year and won’t do it again. (I wasn’t a fan to begin with, but the team wanted to do it.)

My word of advice is that anything that is bent into shape, can also easily get bent by the impact of another robot. That year I had to straighten it so many times that the structural integrity of the material eventually snapped @ Worlds, dropping an entire swerve module. Had to rebuild the entire corner of the frame with angle to make it drivable again between matches.

This year we’re doing machined 6061 tube and using a bent sheet metal pan for mounting electronics and the compressor. But keeping the bent sheet away from structural areas.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.