Falcon 500 Failure Mode Completely Sheared Shaft Screws

So the shaft whose screws loosened is the one with its pinion further away from the motor, correct?

Still it doesn’t seem like the motor mounting should be a factor here. Are there four bearings on that axis? Two inside the rotor cup, another on the housing cover that centers the output shaft, and a fourth in the module?

We (#1577) just found out this happened to one of our practice robot’s falcon motors.

I hope we can fix this or atleast get a replacement. Our falcons were loctited, the screws’ thread is still tight in their place.

2 Likes

Thanks for providing another example. Was this also in your drivetrain?

2 Likes

Yes, in a swerve module

Yes, it’s the one with the pinion further from the motor. The only bearings on that axis are in the motor.

You guys aren’t using MK4is, correct?

Currently we’re using mk3, but this happened in our custom self-made modules

1 Like

Did it have a similar distance from motor mount to pinion as the MK4?

I’m wondering if the driver here might be that distance… If you think about pushing the shaft to one side (as the gear does as the shaft spins, it puts stress on the bolts on one side, while removing it from the other. If we assume a perfectly rigid shaft, then the amount of stress on the bolts increases the further away you put the force (t=f*d). And as the shaft spins, this stress oscillates - you go from full stress to no stress (aside from the normal stress from tightening the bolt). Oscillating movement/stress can cause material fatigue leading to failures.

And, to be perfectly honest, I’m hoping that it’s that distance that’s the driver, because it means other gearboxes, where the pinion is close to the motor mount, are going to hold up better and my robot’s going to be fine :slight_smile:

@PatrickW have you seen these failures before with re-loctited falcon 500s?

Hmm not sure about exact distance but the distance is calculated in the same way for sure.

Yes, stress and fatigue will do that to screws - but shearing 5 steel screws is a bit extreme. This is from our practice bot, it barely had any clashes and played far less than our main robot.
I would suggest a redesign of the falcon shaft to be attached in a different way, perhaps as part of the rotor like the NEOs or cim motors to save us from these issues that keep on recurring.

We’ve had falcon issues ever since 2020, and lost a few of them to various issues. This is getting really expensive, especially considering that as an Israeli team we pay about 300$ for each motor including taxes and shipping.

9 Likes

There’s two operating modes for this shaft: simple torque output and overhanging load output. For simple torque you just need to maintain enough preload on the screws so that the mating surface does not slip back and forth. For overhanging loads the screw tensions is quite a bit more important, as they will have variable tension in them. If the variable stress is a significant fraction of the yield stress then they are likely to fail in fatigue. Its a bit fiddly to get the stresses into the right range. Too loose and they would fail quickly in fatigue. Too tight and you twist the heads off :wink: This is something where you -really- want steel screws, not stainless. Strong screws are easier to get into the right range.
A torque spec for for these screws would be smart.

4 Likes

We have not seen this type of failure on any of our Falcon 500 motors including all of the swerve modules using Falcon 500 motors, that’s all of our robots starting in 2020.

We did have one Falcon fail on a MK4i module. The shaft developed a huge amount of side movement. I was a little worried this could be a reoccurring issue until we realized that motor was totally missing the nose bearing which caused the central structural column of the motor to break. I’m pretty sure the nose bearing fell out when we were thread locking our shaft screws. 99% sure this was our fault.

We have been really happy with the reliability and performance of the falcon 500 motors overall and with the MK4i modules. We are nearing about 12 events worth of driving on our 2022 robot without issue.

It may be possible that the screws were extra stressed due to over tightening when they are reworked.

5 Likes

I was waiting a bit for this…I’ve seen various similar failures in small screws, often due to my own or the assembler’s lack of a good feel for how much torque is appropriate.

Given how difficult these are to get off, I wouldn’t be surprised if teams apply a similar torque to reinstall (inappropriately), stretching the tiny screws. Torque wrenches that size are really torque screwdrivers, and I’m pretty sure Harbor Freight only sells torque wrenches.

Because the failure mode is related to screws coming loose there prolly a misinformed bias toward over-tightening these screws. @Jon_Jack is there a torque spec for these screws?

3 Likes

Thanks to expensive bicycles, you can buy good low torque value torque wrenches. You will not find them at Harbor Freight.

5 Likes

Joining the sheared screws gang… but did yours go sideways in the tapped holes?

7 Likes

It may just be me, but I don’t think that’s supposed to happen…

I have seen some things from some “creative” students, but not quite that creative lmao

2 Likes

yikes… welcome to the club i guess? sorry about ur motor :frowning:

2 Likes

Idea I just had:

Do we think there’d be a market for a reseller that purchases falcons then conducts the QC Vex apparently isn’t? They’d verify or apply the relevant fixes and guarantee each motor to have them at a slight upmark.

4 Likes

I’d definitely pay a little more for the peace of mind that they aren’t going to grenade themselves and not having to tear them apart ourselves.

1 Like