Is there a stroke limit on pneumatics?


#1

We did not read any restrictions on the length of stroke, we would just like to verify.


#2

No.


#3

What size can you fit into the robot size retracted? That will be your stroke limit.


#4

There is a practical limit to how much air you can have available to fill the cylinders…

We’ve found that using pneumatics to power little things is a great thing to do. But using it to power big things is not a good idea. Look at how much power the motor on the compressor draws, compared to a motor like the CIM. Think about how much power it takes to do the job that you want to do with pneumatics.

If you don’t understand what I mean, ask.


#5

I’m open to suggestions. We are thinking a lamp arm design. Spec a 15" stroke cylinder. It’s swinging about 160 degrees. Swinging arm is around 37" long.


#6

What bore cylinder would you need to support the load? Keep in mind that you want to have about twice as much force available, as you really need, so that it will actually work. Also, how often would this cylinder be used? How many times per match?


#7

That is well within workable. We were using 24" cylinders last year. Just figure out how much force you will need to move your arm and use an appropriately sized (bore) cylinder. Also, think about how many times you will move it per match (and any other pneumatics) to size your accumulators (tanks). I can search/provide more details if needed.

^^


#8

QFT. I have used pneumatics in FRC applications a number of ways, and I’ve found that anything bigger than about a twenty or thirty cubic inches per cylinder would have better as a motor driven device (e.g. lead screw, rack and pinion, arm) because of the inefficiency of the compressor (most of the energy applied to the air becomes heat before it can become work) and the limited power allowed by the legal solenoid valves. Also, I have found that long thin cylinders are much easier to bend or otherwise damage than short, stout ones.


#9

This is key. Last year, we had a 36" stroke cylinder to raise our hook to the bar. This worked fine for use because (a) the diameter was tiny (like pencil width) and (b) we only used it once per match.


#10

Dial 7/8 bore x 15". 18ci total. They should be pretty protected during game play.

We will ad flow controls but are concerned about stopping at the mid height position. The upper and lower position will be dead stops. Any suggestions?

BTW, thank you for the information thus far.


#11

practically no.
factually, yes.

You’ll find that in bore, cylinder manufacturers limit their own stroke length. Probably so they don’t sell cylinders that easily break. Also, spring return cylinders are usually much shorter.

Usually you’ll find “DXP” mounting for Bimba will give you the longest stroke. Here’s their link to their old fashioned catalog…it has a pretty good listing of all the stroke lengths.

Bimba round line original cylinders

5/16 max is 4"
7/16 max is 12"
9/16 max is 12"
3/4, 7/8, 1-1/16, 1-1/2, 1-3/4, 2, 2-1/2 max is 32"

If you want to order any (new) cylinder longer than 24 stroke, it takes longer. Might be too long for robot build.

I have not ordered the Automation Direct cylinders, but they look good and mostly are identical to the Bimba.


#12

That is not easy to do. I suggest that if you need to control something with intermediate stops, that you use a motor to do it, and encoder feedback of some type.

Or, if there is only one intermediate stop, figure out how to use two shorter pneumatic cylinders, so that one will be used to go half way, and both to go all the way.


#13

This works. If you want to stop exactly half way, an easy and compact way to do this is to mount the two cylinders to each other, with the piston rods extending in opposite directions. If you use cylinders with mounting threads at both ends (Bimba D style), this is easily done with two pieces of bar with two holes each and four nuts.


#14

You will have to be very careful with the design to get 160 degrees of swing on your arm when using a linear actuator like a pneumatic cylinder. The forces at the end of the arm (torque around the pivot) will not be constant either so just the weight of the arm might make it stop moving if the geometry is not correct. You need to do a detailed analysis and prototype it to be sure that it can really do what you want.


#15

Very good point. A general rule of thumb is 90 degrees arc using a pneumatic cylinder, although we have fudged this rule a bit, if the forces weren’t too great, or gravity was assisting.


#16

Ok, I just checked it. 125 degrees. I think it is ok design wise. Just concerned about getting the heights correct at mid stroke or anything that is not to the stops.


#17

Just because you get the range of motion desired does not mean that your design will work. You really need to draw out the free-body diagrams at several positions through the desired range of motion and work out the forces to ensure that the cylinder can actually push your arm through the whole range of motion. It is very likely that the equivalent torque that the cylinder can exert around your pivot will be the greatest in the middle of the range of motion and a small fraction of the maximum at the two ends of motion. Thus, if you are actuating an arm that starts out horizontal and swings up and ends up 35 degrees past vertical, it may have trouble starting that swing.


#18

To do this, there are cylinders that are designed for three position, or what we’ve done a couple times is fasten two cylinders together with the rods going out both ends. All in, lowest, one out, mid, both out, highest. Note, in this arrangement, your cylinders are ‘traveling,’, not fixed on the robot. Telescoping pipes help with this.

But know that it is against the rules to close the exhaust with a valve to stop a cylinder from moving because you’re trapping potential energy. Not allowed.


#19

To clarify, R88 states that the working pressure must be no greater than 60psi, and specifies a relieving regulator to vent in the case that pressure is later raised through heating or back forces. Also, R93 specifies that opening the vent plug must relieve all stored pressure in the system promptly. Trapping air where it can be compressed and not relieved is going to be ruled against one if not both of these, and likely the generic robot safety rule S1.