Arm Mechanism Advice

[Timothyw0]

Hi from team 4573! We plan on creating an arm that moves vertically. We would like to move the arm 30 degrees. What would be the best way to approach this movement? With a motor or pneumatics? And also, if we are using a motor, what kind of sensor should we use? Thanks!

[Bennett]

I would suggest pneumatics, as a motor could accidentally go past 30 degrees, but pneumatics wouldn't go past it. If you use a motor, a limit switch, or rotary encoder could be used to let you know when it is at that point.

[BenGrapevine]

Originally Posted by Bennett

I would suggest pneumatics, as a motor could accidentally go past 30 degrees, but pneumatics wouldn't go past it. If you use a motor, a limit switch, or rotary encoder could be used to let you know when it is at that point.

I would agree with this, I attempted a similar deal a few years ago and asked the same question. All the responses I got where overly complicated compared to a pneumatic system (our team lacked any ability to use pneumatics at that point so we scrapped the idea altogether. Shame).

[IKE]

How often do you need/want the arm to move?
How heavy is the arm and the load it will move?
Will the arm only be in 2 positions or will it need to stop in intermediate locations?

If 2 positions, then pneumatics could be a really good answer, but you need to be mindful of how many times you want to do the operations and pay close attention to your storage volume calculations. With the unlimited allowance of on-board storage (from a volume perspective), you have a lot of volume you can work with, but recharge is still limited to a specified compressor.

Motor solutions for arms allow for more felxibility, but often require a fair amount more engineering. One mistake I often see with teams is using the window motor for an arm. It has high torque and is essentially self-locking which makes it seem like a natural for an arm. However, it is low power which makes it really slow. The motors also have thermal protectors which keep the motor from overheating, but also keep them from functioning at critical times.
Instead, using a high reduction gearbox and some additional chain reduction can get you the high torque with a higher power set-up (quicker movement with higher loads). FRC 33 2005 and 2007 have some nice shoulder driven arms. Check "Behind the design" for 2007 for more details.

For arms with a small range like you are discussing, I really like the geometry of the HotBot 2007 (FRC 67) or Robo-wrangler 2011. These arms use a push rod and a lower pivot that drives the arm up. The geometry can also be set up so that the linkage will "lock out" a vertical height which can alleviate the stall concerns.

If you can answer my initial questions, we can start to work through some solutions.

[Timothyw0]

Thank you guys responding. I appreciate it. But mike: The arm will have to move quite frequently because it will provide our picking up ball mechanism. It won't be very heavy because it's made out of square aluminum and will have a motor attached for another mechanism. And lastly, our arm will need to stop in immediate locations.

[Timothyw0]

In addition to that, if we're using pneumatics, should we put the compressor on board or off board? (To put it on the robot or not). How many tanks would we need if the compressor is on board? And would you guys happen to know the best brand for a compressor as well? Thanks!

[DonRotolo]

Originally Posted by Timothyw0

And lastly, our arm will need to stop in immediate locations.

I assume you meant to write "intermediate". If so, that's a deal-killer for pneumatics.

Why? Several reasons:
1. As the load on a pneumatic cylinder varies, its position changes. So that intermediate position will not be the same each time, as it depends on load (e.g., whether you have a ball or not)
2. Controlling air for intermediate positions is difficult. Not impossible, but difficult. There are dozens of threads on this from up to 8 years ago.
3. Pneumatic cylinders prefer either of 2 positions: Fully extended or fully retracted. For accurate end positions, you need to 'stack' cylinders, adding cost and complexity.

Bottom line: Be sure to TRY what you want to do before finalizing your design. You may find yourself agreeing with the above...

[Ether]

Originally Posted by DonRotolo

...that's a deal-killer for pneumatics

^^Yes. Watch this video