3175 Flipped 3 Falcon 500 Gearbox

This is a flipped 3 falcon gearbox that we plan to use as a backup to our swerve module. The free speed is 14.8fps with an estimated real world speed of 12fps. The gearing can be found in the picture below. This gearbox is designed with 4in wheels in mind, but it does have a 6in wheel counterpart that I will not include in this post. This gearbox is designed to be used with a chain in tube WCD. In their tests, team 1678 found that 6 falcons caused excessive slippage of the wheels due to high power (their full post is here). To counter this we plan to use a traction control system that finds the rpm of the wheels and compares it to the actual robot speed. Thanks to team 5511 for the inspiration. We are open to any questions or suggestions.



Full CAD: 3175FalconGearbox.STEP (33.9 MB)


What do you plan to make the gearbox plates out of? It looks like clear plastic of some sort in the photos, which would be very interesting.

Is there any reason to use 3 falcons? Seems excessive for the specs they run.


The plates are made of 1/4in thick polycarbonate. We chose this to reduce weight, and because our in house CNC router can cut polycarbonate, but not aluminum. Using polycarbonate allows us to make the whole gearbox in house.

Awesome. We have used quite a few polycarbonate gearboxes similar to this in the past and have been very happy with how they perform. I hope it works the same way for y’all.

1 Like

A whole lot of power. It makes basically impervious to defense, and it allows us to play really good defense if we need to. With 6 falcons total, our robot is pretty much guaranteed to be the most powerful robot on the field.

1 Like

But how will that extra power matter. With more than 2 Falcons, your wheels would slip before you could use the extra power from the 3rd Falcon. If you are going to use a traction control system, you are limiting the amount of power from each motor, effectively running with only 2 Falcons.


The traction control system does not limit total motor power, it only limits how quickly that power can be applied. Wheel slip occurs when too much torque is applied in to small a period of time. The traction control system simply increases the time in which the torque is applied, it does not reduce total motor output.

You’ll probably want to put a 50ish amp current limit on each motor to prevent brownouts on the initial acceleration. But if you have any mediocre batteries (around a 0.02 ohm internal resistance) then you’ll want to put a 45A current limit on it (assuming you are near full weight). However, a 3x Falcon gb at 45A gets the exact same pushing force as a 2x Falcon gb at 65A. But you won’t be able to draw 65A for an entire 5 sec pushing match (but I’m of the belief that if you’re in a pushing match as an offensive robot for more than 1-2 seconds then you’ve already messed up).

If you haven’t used JesseK’s calculator, I highly recommend it: https://github.com/flybotix/drivetrainsim


This is going to be one of the defining questions of the 2020 game, and I suspect we will know the answer very early in the season.

1 Like

Gearing 3 neos to 15 fps is gonna give you a whole lotta torque; Is the polycarb going to hold up in high load situations?

1 Like

Polycarbonate is surprisingly strong, so it should hold up. The motors also go through tight toleranced holes on outside plate, this is to prevent them from moving around and widening the mounting holes for the motors. Polycarbonate is our preferred material, but if it ends up not holding up we will switch to aluminum.

With a current limit of 50A(6 Falcons) would mean a peak draw of 300A. That’s without any other draw from other mechanism. Have you thought this through? What will happen to the snap actions, to the main breaker, to other systems with required voltage?


There is absolutely no point to having “the most powerful robot on the field”. Scoring functions are just as important if not more and imo, there should be more resources focused in that. Having the most powerful robot is useless if you can’t score well and consistently. Also then there is the issue of current draw and all that fun stuff. To give an example, if you look at 254’s robot this year, they only had 2 NEOs on either side, and this year there was very rough defense. They made it to Einstein Finals and it wasn’t JUST because of their drivetrain.

By ramping your power like this, I believe you will effectively have a 2 Falcon gearbox since I doubt you will be able to use all that power.

Also, for the gearbox, the output shaft is over constrained if you run chain-in-tube or even if the chain isn’t in the tube. If you’re intent on using plastic since your router can’t handle aluminum, I’d say look at Delrin instead as it is still lightweight and I’m pretty sure it’s better for gearboxes as it can handle greater loads.

It’s not about power, you could have all the power you want, it’s about how you use it. For example, engineers at Bugatti said they could make a faster car if tires existed that would let them maintain grip. If you can’t put the power down, the power is wasteful and useless.


Nothing about having power prevents us from having effective scoring mechanisms. We are doing this in the off season so we can work on good scoring mechanisms in season when we know the game. Ramping power does not prevent us from using the full power, it only slows down the acceleration, we still have the same amount of power in a pushing match.

This will have the same pushing power as a 2 falcon gearbox geared the same as both are traction limited. traction is the maximum force the DT can apply to the carpet and adding more motors keeping everything else the same only changes how long you can do that for, not the pushing force itself.

1 Like

Or you could have higher acceleration and shorten your sprint distance. I mean if you’re goal is to be a defense bot then there might be some point to a 6 Falcon drivetrain.

And like @BitSentinel said, you’re traction limited so that’s going to be the maximum force your drivetrain can apply.

You do you, but a 6 Falcon drivetrain is 100% overkill no matter how much software stuff you do and no matter what game it is. The software restrictions will essentially result in a 4 Falcon drivetrain.


Please don’t mock my students like that.

1 Like

Thank you for explaining it that way. That makes sense so after talking with my mentors we are going to fabricate the gearboxes as they are, but only use two falcons per gearbox for the 2020 season. I am confident that we can get a 6 falcon drive train to be advantageous and controllable, but that is a project we are going to leave for the off season.

I don’t want to be a jerk to kids, Anthony. Sorry.

1 Like