Sorry for the click bait title
We used pneumatics for…
… lifting our intake up and down (2x)
… shooting (2x)
ejecting (4x)
we used up so much air that we were essentially always compressing air and adding more tanks would just run the compressor longer. We ended up sucking so much battery we sometimes couldn’t climb. How can we fix this issue? Is the solution just use less pneumatics?
Use one battery to charge tanks before a match and a second battery for the match.
24 Likes
See if you can use a smaller piston and consider spring-return pistons so you only use air for one direction.
Edit: if you can’t buy new pistons, you can still get both of these effects. You can create a hard stop to reduce the effective length of a piston, you can use surgical rubber or springs to retract a piston.
9 Likes
after 1.5 cycles our compressor runs.
I wish there were spring assisted pistons rather than spring return. But yea we typically want total control so spring pistons are usually not in our best interest.
Add more tanks and/or try using less/smaller pistons.
1 Like
McMaster sells a whole assortment of air-retract (AKA spring extend) cylinders. Just use the filter tool.
1 Like
It’s a compromise, though. If you need the efficiency, sacrifice a little control.
the smaller bore is going to be something we try next year.
If you don’t need 60psi for what you’re doing, you can always turn the regulator down. Less air from the 120psi side will be required to replenish the working side.
9 Likes
Pneumatics provide fast, strong linear actuation in a compact package, however they are also heavy, expensive and require a compressor. While you can use batteries to charge beforehand, and sprung cylinders will help, at the end of the day you need to consider that all of the pneumatics on your robot are being powered by one rather small motor. Do you need four cylinders for ejecting? 2702 likes using servo powered four bar mechanism for this sort of thing. Could you have used a roller for ejecting?
While there is no hard limit in the rules for number of pneumatic cylinders, you may want to consider the limit of an 80cfm compressor.
On the next robot you design, look at the power available from the compressor. It’s not much. Only use pneumatics for functions that require little power. Don’t try to lift weight, for example. Pneumatics is great for moving things back and forth quickly, between two positions. We used three pneumatic actuators on our robot this year. One retracts the Hatch panel arm. Another grabs the panel center hole. The third flips the cargo out. We have a little storage tank, and have not had any issues with running out of air in normal game play. And we move a lot of game pieces each match.
2 Likes
Given that you have already been to an event or 2 you probably have a good idea of how many cycles you do and therefore how many piston actuations. Based on this you can generate a function of your estimated pressure, and only have the compressor turn on if the pressure drops below this curve.
2 Likes
Oh, also - if you are using default setup for the pneumatics, then the compressor is going to be running whenever the storage side falls below ~90PSI. You may want to create your own control loop so that the compressor kicks in later.
(Edited for accuracy regarding the standard behavior of pneumatics control system)
2 Likes
In the larger sizes, some spring-equipped cylinders still have ports on both ends, so you can have the spring helping you. Or you could just use your own external spring.
If you’re able to function without pneumatics that will simplify your control system wiring greatly - almost by half if you consider all the solenoid wiring and tubing runs you won’t have to do.
This year we use a piston to bring our cargo intake arm up and down (it swings out of the way when doing hatches) and an ejector on a hatch grabber which also doubles as the cargo ejector by pushing the balls out.
We also attempted to build a climber that relies on two giant pistons to lift the back of the robot - which uses a ton of air. It was close to working on our practice robot but we just don’t have the time to finish it before DCMP.
Add more tanks. We try to have 4 clippard tanks at least on the robot. We fill them all before the match and the compressor barely runs during most matches.
Pneumatics are fantastic once you get some of their nuances worked out. I’d much prefer to deal with an on / off state to move something as it’s basically purely mechanical - if I use a gearbox or a motor, often times I have to bring sensors into play which means more work all around.
The issue, as you’ve raised here, is supply of air and your compressor running all the time. I use Paul Copioli’s old Useful Calculations spreadsheet during planning so you can plug in your actuator bores, lengths, etc. and get a ballpark on how many tanks you’ll need. Check it out here Useful Calculations.xlsm (53.0 KB)
Make sure you have enough tanks and maybe add an extra just for good measure. They don’t weigh much but take up a good amount of space, but you can run tubing just about anywhere.
It’s too late now for this season, but whenever we’re using pneumatics I go through an exercise with the students to figure out how much air is used each time each cylinder fires, how many times per match we expect to be firing each cylinder, and then figure out how much the compressor needs to run to keep us topped off. For nice, round numbers, we assume the compressor turns on when it’s below 90PSI, and off when it hits 120PSI. We use the flow rate published for the compressor at 120PSI (as the rate changes based on the current pressure in the system, but we want the published worst case). We assume everything acts like an ideal gas at a constant temperature, and that the flow into the cylinders is instantaneous - this means the right side of PV=nRT is a constant for each operation.
Obviously, a lot of inaccurate assumptions made there (like, every assumption made there), but it really simplifies the math and gets the students thinking about how the air is used, and how that will impact our robot.
1 Like
Regulate each system only to the working pressure you need it to actually be at to achieve the force needed for your mechanism. For example our hatch gripper ran at maybe 30psi while our climbers ran at the maximum allowed 60psi.