This is a render of our 2016 offseason bot. The robot had one particularly large stress point which we were unable to fix for its one and only competition, but we are well underway fixing it now and we learned a ton in the process. All for under $150 in extra parts. Posting updated render with the stress point fixed sometime this weekend.
What was the stress point?
From looking back at previous posts I believe they had problems with the cantilevering of the CIM+what looks to be a 2-3 stage VersaPlanetary from a plastic clamping gearbox. When going over defenses the Clamping Gearbox would fail.
That thing looks pretty familiar… 
But seriously, we love that picture. That was a fun match.
Alastair sorta explained it but I’ll elaborate.
Essentially the output shaft of the tilt assembly (the one with the small sprocket on it) was cantilevered. Whenever the arm had a fair amount of velocity and stopped moving the chain would tug the shaft backward quickly causing the clamping gearbox to crack open.
They was the robot I really took inspiration from. Until today I actually never saw 4607s robot. When designing the bot I found it to be a cross between 225 and 987.
By far my favorite match of champs. Probably the most vivid FRC moment for me (I was watching as an inspector from the side of the field). You guys were awesome to compete against, hopefully we get a chance to compete in the future!
Had you considered putting a pneumatic spring on the intake/lift arm for stability, smoothness of operation, and reduced force necessary to lift it?
Assuming a pneumatic spring and a gas shock are the same thing, it was considered although it was a bit to late when it was. If I were to do a similar thing again I’d use a powerful normally open gas shock and a winch.
I love how glossy the boulder is. It looks like a bowling ball that could actually destroy the castle!
Yup, same thing!
4564 used a normally opened gas shock as well as a winch for their defense manipulator/intake arm this year. One drawback was that the arm would fall when crossing defenses like the Cheval De Frise or other rough defenses as the shocks would compress. This required them to lift the arm as they crossed the CDF and others slowing down crossing times. However overall this is a better solution than a motor at the hinge point as it puts significantly less strain on the system.
In this situation a linear actuator is also an option, allowing for very precise control with little backdrive. A good example of this is 1690, as their arm had very little flex and allowed for extremely high accuracy shooting.
A linear actuater would have even awesome. Unfortunately I’ve heard nothing but awful things about darts and with a price as high as theirs I really decided not to consider them an option.
I haven’t yet heard anything bad about DARTs; do you have some discussions to link to on that? Would be good to know for the future.
As you’ve probably learned, rotational arm joints are sometimes tricky and require some care to do with the standard sprocket at the pivot point method of rotation. You could probably have attempted something with pneumatics, which simplifies control and is easier to make strong. Stack two pistons of different throw length on each other and now you have a four position actuator.
The DARTs have several eccentricities you need to work around – they aren’t “plug and play” and require a level of hardware and software understanding. If anyone is interested in details, you can PM me and I’ll detail our experience. We used 2 on our arm last year.
It may change for next year, I know AM is working to make them better based on team feedback.
Also, thanks for using us as an inspiration! 
What gearbox are you guys using for manipulating your arm and what is its reduction?
The Versaplanetary has 2 10:1 stages for 100:1. It then goes into a 24t gear which goes into a 60t gear which brings the reduction to around 250:1. A chain is then driven by a 12t Sprocket which goes to a 36t sprocket making our end reduction about 750:1.
The reduction was pretty much perfect for the arm (It originally had a giant cylinder mounted to the back). The biggest issue that we cantilevered a shaft which had to deal with loads that big on a plastic gearbox.
How did you guys get the motors and gearboxes in with that tubing in the way? It looks like it would take quite a while to replace anything should a motor go out or something similar.