Are 3 position, closed center solenoid valves FRC legal?
Students are looking to use 3-position, closed center solenoid valves, to allow holding cylinders midstroke.
There is a thread in CD about this, so I understand how to do it ( use check valves to comply with rules about venting all compressed air through valve). It seems to me that there could be some confusion regarding the rule about not connecting outputs of 2 solenoids together ( although functionally, check valves don’t really “connect” the outputs, since nothing can flow from one to the other through two check valves in series with opposite polarity), since it might visually appear that they are connected…
Is anyone aware if this is legal? Has anyone seen this done?
Seems to me that it doesn’t violate the spirit nor letter of any rules, but team would feel better if we know for certain!
Assuming they don’t have too large a port and are rated for the pressure and are self-releiving, no one has found any rules against them. That said, it won’t do a very good job of “holding position” of a pneumatic cylinder. 3946 made our own version of this using several solenoid valves in 2014 for our intake arm. Air has too much give, so even if no air is flowing, the arm or whatever still has significant latitude of motion as the forces change.
Added: adding some sort of brake to this setup might achieve what you’re looking for.
Obligatory: I am not the Q&A, my answer means nothing.
After quickly rereading this years pneumatic rules, I think what you are suggesting is legal.
My understanding of the intent of R87 is that it is to keep teams from putting solenoids in parallel to achieve higher flow rates. It states that “outputs from multiple solenoid valves must not be plumbed together”, It does not seem to ban plumbing an output to an input (to simulate a closed center solenoid for example). Although, I could imagine doing that draws a lot more attention from most inspectors. Even if it does ban that, it is referring to solenoid valves, and not check valves.
Check valves are explicitly legal this year, and I see no rule saying they can’t be used the way you stated. The fact that you understand that usage means you probably know a fair bit more than most about how to plumb it to ensure it complies with R86.
I say they violate R86. When in the center position, the relief valve can not vent pressure stored in the cylinder. We tried to use them in 2014, and Al did not like it.
Thank you very much for your replies. I understand that these are not official Q A endorsements, but it seems like it should pass inspection. I haven’t been able to get the code to submit on the Q A yet, so this does help. The students will be testing this soon, so we will see how it works, and whether the legality is moot or not! Thanks again!
Countering this point, we used something very much like this to make our pneumatic level 3 climb work on level 2 last year.
Blue alliance, far platform.
This is true if you do nothing to mitigate R86. But with some thought you can.
Due to some design decisions we probably won’t repeat, 6662 used closed center solenoids on our 2019 robot (switched in after bag day). We used check valves to make sure the pneumatic system complied with R86. You need to feed the outputs back to inputs to make sure pressure in the cylinder is relieved when you open the pressure vent plug. Here is a diagram:
It’s also nice to have check valves from the atmosphere to the solenoid outputs. Otherwise, the cylinder won’t be free moving when the pressure is relieved, even though the pressure in the output branches is <= atmospheric pressure. We overlooked this last year, but figured it was okay because pressure was relieved, even if it didn’t seem like it, and a vacuum effect was preventing the cylinder from moving. Our LRI agreed after some thought, but YMMV.
There was a discussion on legality last year on CD from which we devised our setup. Here’s LRI suggesting this setup (without the additional optional valves)
I agree with @GeeTwo that it is unlikely a closed solenoid is right for your design, and you might want to reconsider using them. If you share your application you may get some advice.
Did you have the cylinder hard against one end, or pretty close to it? Was the force stable (in this case just weight hanging)? In either (or both) of those cases, this can work decently. Trying to passively hold position near mid-stroke under variable load conditions won’t work with air.
It was about 1/3 stroke, but because we were extending it under the same load it needed to hold, it held position. Being +/- an inch or two wasn’t a big deal for that application either.
Our team is using 2 cylinders for climbing. So, with the robot weight on it, we think it will be fairly stable. Yes, it may bounce a bit, but we think it will come in handy for matching speed during climb, or even possibly intentionally lifting one side higher, to help balance the switch. With 2 position solenoids, this kind of control is not possible. Matching speeds is somewhat feasible with 2 position valves, but even that is very wasteful of precious compressed air.
I am really impatient to see how it works. The students have the parts made / bought, and are assembling. They are all super excited.
As I noted above, there are three key points I would look for as your robot inspector*:
Does the solenoid have legally small ports? (documentation or inspection, R77C)
Is the solenoid valve rated for at least 70psi? (documentation, likely a label, R73)
Does it vent pressure when the plug is opened? (demonstration, R86a)
To demonstrate venting, having a section of plastic tube in the circuits in question would be helpful - it is easy to tell whether a plastic tube is holding pressure by first squeezing it (a full tube won’t compress easily) and then releasing it from a fitting (a pressurized tube will hiss when disconnected).
* Disclosure: I have inspected robots at one event so far (Arkansas 2019). I volunteered again this year, and they want me back.
I think the extra check valves venting to atmosphere, might be very worthwhile.
Imagine lowering a climbed robot, by gradually opening a manual system pressure dump valve.
Without the extra check valves, as you relieve the pressure, the robot may not fully lower (pistons not fully extend), due to the resulting partial vacuum created on the “extend” side.
So, the extra check valves to atmosphere, will also make it easier and quicker to remove the robot at the end of the match. As well as making it more obvious to an inspector, that there is no stored air pressure in the system after activating the relief valve. (Nor any partial vacuum - which may otherwise remain in the system, without the extra check valves).
