HELP Arm Gearbox Continues to break

Chiming in with a +1 for the following:

  • Segments of polycarbonate to take some of the whack (you could even portaband your current arm and splice it in)
  • Chain/sprocket or belt/pulley runs in the future
  • Quit using PG motors and embrace the 57 Sport life (or the VP life, but gah-lee we’ve been happy with Sports).
1 Like

We had a “C” shaped cargo intake on the end of a pink arm. Imagine ramming that into a wall with 20+ inches before you got to the bumpers.

Driver control only gets you so far. The new drive teams during the off season didn’t have the experience driving the bot. We really needed some sensors and/or software to add another level of safety.

It’s a good learning experience for the team.

Build mechanisms that are quickly and easily repaired or replaced.
Build mechanisms that are less likely to be damaged during normal game play.
Add sensor and software controls to reduce possibility of damage.

These should all be part of the design criteria for next year.


It may also have to do with the overall design concept. If the structure holding the game piece protrudes beyond the frame perimeter by any significant degree, there is a higher chance of it being damaged than if it is retracted as @Jon_Stratis suggested in his post.

When the mechanism is designed to function as Jon suggested, the forces on it are mainly the forces generated by the actuator as it is deployed. When it protrudes from the robot, the forces come from the momentum of the whole robot it is mounted on or the momentum of another robot.

We used two different pistons hooked in series for our hatch grabber. The first piston was a 3/4" bore which produces around 25 lbs of force @ 60 psi. The second piston was a 5/16" bore which produces 5 lbs @ 60 psi. This made for a very soft landing when we went in to place the hatch.

We had both of them extend and retract at the same time (on the same solenoid channel) but we had talked about moving them to different channels so that we could fire the small piston after we had made our approach to “plant” the hatch rather than using the driving forward speed to plant it, but we never did that.

Anyway, for this problem, if you removed the gearbox from this arm and had a second arm driven by the same gearbox located behind the first arm (well inside the frame perimeter) and then used a piston to connect the first arm to the second arm to create a parallelogram linkage between the first arm, then the impact forces would be absorbed by the piston and would not backdrive the gearbox. You would not technically even need an active pneumatic system. You could just connect one side of the piston to a tank and then pump it up to some moderate pressure and then seal it off. This arrangement would allow the piston to absorb the impact and then push back out to full extension afterward (the tank would absorb the working air and then return it to the piston).


We also busted up our polycarb intake this year. Had plastic 1x1 versaframe break at three different points during AZ West, early on we replaced it but it got to the point where we were adding aluminum “splints” to hold it back together until the event ended. Changed it over to 1x1 1/16" Al versaframe for Worlds and that held up fine.

We had good success with plastic versaframe for our arm in 2018 so I don’t want to write it off, I think the combination of our design and the defense in this game did the intake in. We used a pneumatic cylinder to pivot our intake but the placement of it meant it could only absorb vertical impacts, not horizontal. We had wanted to do a linear slide to absorb horizontal forces but ran out of time with our build capabilities.

Did the polycarb tubes break where holes were drilled in it or at some other point(s)?

The 188 has four 3.7:1 reduction stages. The only plastic gears are in the motor-end stage. Source: I took a PG apart, see link below.

And I’m agreeing with all that you need to have some shock loading at a minimum and likely some additional reduction or a stouter gearbox as well.

Our team had a very similar issue with the PG188 gear motor. on our First(bagged) robot we discovered no issues with the motor before bagging. however on our practice robot we destroyed a PG188 while the robot was being programmed on blocks. we proceeded to return the gear motor, and get a replacement, which lasted us about two afternoons of practice before breaking. we proceeded to return this one as well. however, at competition we opted to leave the PG188 in place instead of replacing it with a backup Versa-planetary. interestingly that motor(PG188) is still running on the robot today without issue. we have concluded within our team that a batch of PG188s received by Andymark sometime after the first week of build was entirely defective. I have include a photo of where ours is used. in conclusion I believe they work great, if you get a good one. we however wont be using them on future robots.

As Ryan as mentioned, polycarb is the savior of everything. We used 1/4" polycarb on a 4 bar intake this year, and you could have that thing twisted at a 90* angle almost, and would still twist back to shape and be fine. We weren’t using gearboxes in that case, but it would have certainly helped that if we did. We were using pneumatic cylinders, and never ran into an issue of bending cylinders, even with running into things.

Hey @GoalkeeperBoss. What @wgorgen is describing is a shock absorber system. I have seen some teams use thick rubber bands or surgical tubing to get a similar effect. Polycarb can only bend and deflect so much before failing.