Pneumatics: Double Solenoid

Posted by Suneet Upadhyay at 1/31/2001 8:22 PM EST

Student on team #599, Robo Dox, from Granada Hills High School and C.S.U.N. and NASA/JPL and various.

Now that that we’re thinking about controlling the systems that are being built, we could use some general advice about to use a double-solenoid pneumatics valve to control the movements of a pneumatic cylinder so it can be stopped in mid-movement, i.e. to allow the joint of an arm to stop at any angle. Thanks in advance.

-Suneet
“RoboDox” Team 599

Posted by Joe Ross at 1/31/2001 8:34 PM EST

Engineer on team #330, Beach Bot, from Hope Chapel Academy and NASA/JPL , J&F Machine, and Raytheon.

In Reply to: Pneumatics: Double Solenoid
Posted by Suneet Upadhyay on 1/31/2001 8:22 PM EST:

You aren’t the only one with this question. A few days ago someone else asked this and there were quite a few messages posted about it. You can see the link to the discussion below.

I don’t beleive that anyone has figured out a way to make the cylinders go only part way (or they aren’t telling).

Posted by Brian at 1/31/2001 9:17 PM EST

Student on team #56, Robbe Xtreme, from Bound Brook High School and Ethicon Inc…

In Reply to: Pneumatics: Double Solenoid
Posted by Suneet Upadhyay on 1/31/2001 8:22 PM EST:

Actually we have figured a way out to have the solenoids stop in the middle, but its all a trade secret so you guys will have to wait until the Competitions to hear how we did it…

Posted by Shawn McMahon at 2/2/2001 8:11 PM EST

Student on team #263, Aftershock, from Sachem H.S. and Symbol, Citycorp., Apoge, CA …

In Reply to: We go it to go half way…
Posted by Brian on 1/31/2001 9:17 PM EST:

It’s really quite simple to get the piston to stop half way. The valves last year worked because there was a neutral position. You can make a similar setup using 2 valves, one single and one double. Hook the double up to a cylinder like you normally would. Now plug both exhaust ports together and send that tube to the single switch. Sine it is spring loaded it will stay in its default position. Put a plug in the port where that air would normally come out. Now when you apply pressure to one side of the piston it will not move because you have effectively “plugged” the exhaust. To allow the piston to move, simply activate the single switch and the air is vented. Get it? The advantage to this system is that when the piston is stopped it is at full pressure, not atmosphere. A quick diagram of this setup can be seen by clicking on the link below. Have fun :slight_smile:

President of Team 263,
Shawn McMahon

P.S. Some people probably are not too happy that I disclosed this information. The way I see it, if I end up being allied with a robot that would benefit from a stoppable piston I have not only helped that team, but have also helped myself. I’m sure there are many teams who have already discovered a way to stop the piston. There are many ways. This is only 1 of 2 that I came up with.

Posted by Joe Johnson at 2/2/2001 10:15 PM EST

Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.

In Reply to: We got it to go half way too … and here’s how :slight_smile:
Posted by Shawn McMahon on 2/2/2001 8:11 PM EST:

Nicely done.

I have a possible problem.

I believe that the piston will not be as stiff in the
direction that the double solenoid valve is open
because the regulator will allow the pressure to bleed
off in this direction, but the plugged exit valve will
allow the pressure to get higher, thus allowing the
piston to be stiffer in this direction.

By the way, this is all just theorizing on my part, I
have not actually tried the idea.

Would this fix it?

Put a check valve on the input to the upper vavle (the
one marked “from compressor” in Shawn’s Drawing.

What do you pneumatics folks think?

Joe J.

Posted by Shawn McMahon at 2/3/2001 3:27 PM EST

Student on team #263, Aftershock, from Sachem H.S. and Symbol, Citycorp., Apoge, CA …

In Reply to: Good idea…
Posted by Joe Johnson on 2/2/2001 10:15 PM EST:

The piston has equal pressure in both directions. Trust me. We hooked this system up to the large 2" piston (puts out 160 lbs) and we pushed and pulled on it. It doesn’t go anywhere. :slight_smile:

Posted by Shawn McMahon at 2/3/2001 3:47 PM EST

Student on team #263, Aftershock, from Sachem H.S. and Symbol, Citycorp., Apoge, CA …

In Reply to: Re: Good idea…
Posted by Shawn McMahon on 2/3/2001 3:27 PM EST:

We did use the check valve when we tested it. So I assume if we took it out you would be correct, one side wouldn’t be as stiff, but both are as strong. In our case you would have to push back against the piston with a force greater than 160 lbs to create enough backpressure (greater than 60 psi) to force air out of the regulator. Even without the check valve its not really a problem. You have full strength of the piston in both directions. If in the rare occation you would have a force strong enough to force the piston backwards it would move (only in one direction as you stated before) This would happen in a regular pneumatics setup anyway.

Posted by Joe Johnson at 2/3/2001 5:24 PM EST

Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.

In Reply to: Thinking about it again …
Posted by Shawn McMahon on 2/3/2001 3:47 PM EST:

When I played with an air cylinder an the regulator the
other day, it was very easy to push on the piston and
make the regulator bleed off the excess pressure.

Also, in your analysis, you are forgetting that you
have put 60psi on BOTH sides of the cylinder piston.
When you net out the forces, I believe you will find
that the piston will act like a spring is holding it in
place. The higher the pressure, the stiffer the spring
but it still acts like a spring (remember PV=NRT from
physics and/or chem class?).

Joe J.

Posted by Shawn McMahon at 2/4/2001 12:46 PM EST

Student on team #263, Aftershock, from Sachem H.S. and Symbol, Citycorp., Apoge, CA …

In Reply to: Pushing on both sides of the piston…
Posted by Joe Johnson on 2/3/2001 5:24 PM EST:

You are right. The piston does act like a spring, but a very strong one (In our case). What size piston were you testing the system with? On smaller pistons it would act more like a spring, being easier to push because of less surface area inside the piston. The optimal system would have the shortest hoses. Normaly longer hoses would be better, acting as a resevoir. In this case using smaller hoses would leave less air to be compressed giving you a stronger spring.

When we tested it we had someone try to push or pull the spring. The most they could move it was about an inch. (with the check valve in the system to prevent leakage.) That was good enought for us. We don’t need to “lock” the piston in place. We just needed to stop the piston partly through its motion so we could position our arm to allow us to move under the barrier.

It would be physicaly impossible to completely lock the piston. To create a stronger spring you could use both spring loaded valves conected directly to the piston (using the shortest tubes possible). The only compressable air would be the air within the piston.

This would also allow you to use the check valve somewhere else. In our case we want to use it as a quick refill port. This gives us the ability to charge the tanks before the match with an outside compressor, eliminating the need to turn on the robot before a match.

Posted by Elliot Johnson at 2/1/2001 1:15 PM EST

Student on team #218, HP Robotics Team, from Highland Park High School and Double-E Inc…

In Reply to: Pneumatics: Double Solenoid
Posted by Suneet Upadhyay on 1/31/2001 8:22 PM EST:

We had a gripper last year on our robot powered by a cylinder that could be stopped at any position. The double solenoid was hooked up with a flow control valve after each of the outputs, and then that went to the cylinder inputs.

Solenoid Port A -> Flow Control Valve #1 -> Top of Cylinder
Solenoid Port B -> Flow Control Valve #2 -> Bottom of Cylinder

The Solenoid is then hooked up to a relay using the two diode scheme shown in the robot manual. To move the cylinder in one direction, turn the relay on in one direction. To move it the other way, reverse the relay. To stop the cylinder while its moving, just turn off the relay. When the solenoid is stopped, the pressure in the cylinders is locked in.
The valves are used to slow down the speed of the cylinder so that you can stop it in time. If the cylinder is fired at full pressure, it reaches the other state too quickly.
This set up worked for us reliably in all of our matches, but we didn’t have a huge load on the cylinders, just the force of the ball pushing back.

hope that helps,
Elliot Johnson

Posted by Suneet Upadhyay at 2/1/2001 3:15 PM EST

Student on team #599, Robo Dox, from Granada Hills High School and C.S.U.N. and NASA/JPL and various.

In Reply to: Re: Pneumatics: Double Solenoid
Posted by Elliot Johnson on 2/1/2001 1:15 PM EST:

: We had a gripper last year on our robot powered by a cylinder that could be stopped at any position. The double solenoid was hooked up with a flow control valve after each of the outputs, and then that went to the cylinder inputs.

: Solenoid Port A -> Flow Control Valve #1 -> Top of Cylinder
: Solenoid Port B -> Flow Control Valve #2 -> Bottom of Cylinder

: The Solenoid is then hooked up to a relay using the two diode scheme shown in the robot manual. To move the cylinder in one direction, turn the relay on in one direction. To move it the other way, reverse the relay. To stop the cylinder while its moving, just turn off the relay. When the solenoid is stopped, the pressure in the cylinders is locked in.
: The valves are used to slow down the speed of the cylinder so that you can stop it in time. If the cylinder is fired at full pressure, it reaches the other state too quickly.
: This set up worked for us reliably in all of our matches, but we didn’t have a huge load on the cylinders, just the force of the ball pushing back.

: hope that helps,
: Elliot Johnson

I tried it out on a cylinder, and it seems to work fine, even though I didn’t apply a the full projected load.

Thanks.

-Suneet

Posted by Joe Johnson at 2/1/2001 6:08 PM EST

Engineer on team #47, Chief Delphi, from Pontiac Central High School and Delphi Automotive Systems.

In Reply to: Tired it
Posted by Suneet Upadhyay on 2/1/2001 3:15 PM EST:

Does this actually work?

Perhaps I don’t understand double solenoid valves.

I thought that the double solenoid valve takes a pulse on one coil to move the valve one way and a pulse on the other coil to move the valve back. Put another way, the valve “stays put” until actively kicked the other way. I think of them as a kind of “latching relay” only for air.

If my understanding is correct, then I don’t think what you described above will work. The valve will always be in state A or state B (except for a few milliseconds as it transistions). So, the cylinder will always be driven to one state or the other.

Perhaps what you intend to say was you used 2 single solenoid valves. In this case, you can activate one solenoid or the other and get the behavior you describe.

By the way, if I understand you correctly, when neither solenoid is activated, the cylinder stops putting force on the output shaft. BUT the shaft is not locked. Because the both sides of the cylinder piston are open to the atmosphere.

Am I missing something?

Joe J.

P.S. Would it be legal to get a sort of “squishy” lock condition by using the check valves to keep the air from escaping? Perhaps not if the pressure regulator was not allowed to bleed off any excess pressure generated. Thoughts?

Posted by Elliot Johnson at 2/1/2001 8:25 PM EST

Student on team #218, HP Robotics Team, from Highland Park High School and Double-E Inc…

In Reply to: Does this work?
Posted by Joe Johnson on 2/1/2001 6:08 PM EST:

You’re absolutely right. I hooked it up with the pnuematics kit and a pulse would send it all the way to the other state. The valve’s in last year’s kit were a different model, they must work differently.
Next time I get a chance I’ll take another look at how we set it up last year. Maybe I just forgot something.

sorry for the confusion,
Elliot Johnson

Posted by Gary Bonner at 2/1/2001 10:42 PM EST

Other on team #433, Firebirds, from Mount Saint Joseph Academy and SCT Corp., FMC Corp…

In Reply to: Does this work?
Posted by Joe Johnson on 2/1/2001 6:08 PM EST:

: P.S. Would it be legal to get a sort of “squishy” lock condition by using the check valves to keep the air from escaping? Perhaps not if the pressure regulator was not allowed to bleed off any excess pressure generated. Thoughts?

From Team Update 4:

M21. The SMC regulator fixed at 60 psi must be connected to the output of the pump and/or volume tanks in order to limit the maximum “working pressure”. Working pressure is the air pressure that may be used to actuate the valves and air cylinders.

M22. It is acceptable to directly couple one end of a cylinder to a suction cup and drive the piston in order to create and release vacuum. Further, it is acceptable to connect a valve (connected to the “working pressure”) in parallel with the suction cup and cylinder in order to introduce air into the system. It is not acceptable to intentionally generate pressure by using a cylinder as a pump, but the normal back-pressure from a load applied to a cylinder piston is ok. Cylinders may not be used to make reciprocating pumps (i.e. pumps designed to cycle a cylinder and valves continuously to create stronger and stronger pressure or vacuum).

It looks to that you can use the check valve to prevent the regulator from bleeding off the back-pressure from the load on the cylinder.