Our hatch mechanism is an aluminum arm driven by a pg188 gear motor. We have already had to replace the gearbox several times due to gears chewing themselves up breaking. We notice this problem if we run into the loading station, cargo ship or rocket with the arm up. We can obviously prevent this by being more conservative with our driving near those field elements, however we were wondering if there was any way we could limit it with software or mechanically to limit or prevent this damage.
Can you please provide a few photos of your setup for the arm?
What are the ratios of chain/sprocket or anything else going from the gearbox to the arm’s rotation axis?
We need more details to help
This robot? And if so, that arm holding the HP? It would be nice to see what’s down in there…but even so, I’d suggest maybe a hinge with a stiff spring, so the arm can “give” when you whack it into something.
Unfortunately the only photo I have right now is the chewed up gears but I will certainly take photos at this afternoons meeting. The arm is driven directly from the gearbox, it has a coupling to a 1/2 in hex that goes into a hex bore on the bottom of our arm. The reason for lack of chain and sprockets is we don’t really have enough space to do that unfortunately. If I’m correct the chain would be able to absorb some of the shock right?
Thats the one alright, ill be sure to take some good pictures of what it looks like down there tonight,
andymark lists the gear material as steel and plastic
Yeah off the top of my head I’m going to hazard a guess that you might have tried to direct drive the pivot of your arm without a chain reduction from the gearbox? Chain allows for loading like you might experience after a hit or running into something on the field.
This gearbox might also just not be the best solution for that arm, I suggest looking into the other planetary gearboxes AM and VEX have and running additional reduction off of that. I’ll be honest I haven’t used a PG gearbox since like 2012. There’s better solutions out there now.
You can also add a hinge/spring as mentioned above to take those loads from hitting field components.
So you’re using a 57 Sport planetary gearbox to move the arm? Just want to verify that first. Any chance you know which ratio you’re using?
Copy and paste straight from andymark
Gearbox Reduction: 188:1
No Load Free Speed: 28 RPM
Stall Torque: 33 ft-lbf
Minimum Backdrive Torque: 3.5 ft-lbs
Nah pg is the old gen that I think attempted to replace the old globes.
Ah thanks, I can’t read today apparently.
Yeah, seems like the issue is probably from the loads being transferred into the gearbox when you ram into stuff.
One idea may be to splice in a couple of polycarbonate plates. If you run into something the plates will deflect. You need some elasticity in your setup that will deflect before all of that force gets transferred to your gearbox. It will probably take some experimenting to get right as you don’t want it to be too floppy or so stiff that it doesn’t deflect at all.
There are many solutions to handle it mechanically. Often the easiest is to add an external chain/sprocket reduction to help absorb some of the shock load. Switching to a slower motor would allow you to use a gearbox with a smaller reduction a lot easier as well - a 2:1 stage is more durable than a 10:1 stage (I don’t know what the different ratios are in the PG gearboxes). The AndyMark Sport series of gearboxes has an optional “HD” final stage reduction - the carrier plate is strengthened, and the gears are twice the face width. That makes the stage more robust than “normal” planetary gearboxes. We’re using a CIM Sport with an HD stage, along with a gear reduction, on our arm and it’s handled shock loads of running into things just fine all season - it’s the main reason we chose that gearbox over the VersaPlanetary for this application. Nothing against the VP, we’ve used them in other solutions for years without issue, and actually have two of them on the robot right now for other mechanisms.
From a software perspective, you can create a “resting” location for the arm that has it retracted inwards at a safe angle. Then a timed sequence when the driver hits a button to come up and delver a hatch panel (or retrieve one) and retract. It’s essentially an autonomous sequence that’s triggered by the driver, and only exposes the arm for a couple of seconds. It’s “safer” than just telling the driver to be careful, but is not an ideal fix. But depending on time or monetary constraints, it might be enough of a fix to get you through the fall!
Interesting software implementation, ill show that to our programming team tonight to see if we can put that in before this weekends competition. Ill also be looking into seeing if we can make a new bracket with chain and sprocket
Another solution is to do what we did with our HP arm, which is to actuate it using pneumatics. The pneumatic cylinder acts like a spring, when we run into things accidentally. Although it might be that since your arm has to retract all the way into the chassis for the ramp to work, this might not be feasible on your robot.
For next year’s game I’m going to make sure to add a weighted design criteria for scoring mechanisms to NOT be able to be damaged by field elements.
We had to fix ours during the season and when we used new drivers in the off season we had to rebuild it several times.
Polycarb polycarb polycarb.
You can get it fun 1 x 1 or 2 x 2 shapes now too on McMaster https://www.mcmaster.com/3161t31
We milled out a slot in the top & bottom to get extra flexibility. Eventually we just went to polycarbonate plates & standoffs, but there’s lots of options.
My team also ran into a host of issues with our 4 bar arm that was mounted to our elevator. Failures we had were:
- Destroying the internal gears of VersaPlanetary
- Destroying 12t pinions
- Bending gear boxplate that all of the components were mounted to probably because there was only one plate so they were in single shear
- 775Pro burning out
In the off-season we have redesigned our arm to make it much more robust. Improvements were:
- Switching to a NEO
- Switching to CIMSport
- Thickening gear box plates
- Supporting all shafts on both sides
- Replacing pinion + spur gear reduction with a different gear set with 2 spur gears
I can go into more detail later if need be. But I learned a lot about custom gearboxes and the limitations of VPs with regards to shock loads.
It was polycarb. Polycarb can only take so much and our design went over that line.
More polycarb? Pics of any failures? Breaking polycarbonate is impressive.
Go look at the jvn blog about uppercuts intake.