Exhaust in Solenoids

[FRC Team CC]

Hey Teams!

Is there any way to block the exhaust holes on the sides of the solenoids (some screw or contraption)? We are trying to program a piston to move out and in and found the exhaust hole leaking out air.

Also, if we want to control both the A and the B holes, do we need a double solenoid? Or can we make do with a single solenoid?

Sorry about these simple questions; we are a new team and are trying to figure out all of these new components. We greatly appreciate your help.

Thanks,
Charging Champions

[Mark McLeod]

If you plug the exhaust ports, then the solenoid will cease to work. It only works by letting air out of the cylinder through the exhaust ports.
If air is leaking, then it is for other reasons.

Both double and single solenoids control all ports.
The difference between the two types is that a single solenoid has a home position it always returns to when power is cut or the robot is Disabled, while the double solenoid stays in the last position you put it when power is cut or the robot is Disabled.


What model solenoid are you trying to use?

[ctt956]

To maybe fix the leak, try checking the ends of the air tubes, and see if they're cut straight. If they've been cut at an angle, they can leak. Cutting them straight should fix it. Note that this may not be the issue, but it's a simple thing to check before you disassemble the whole pneumatics system. Just make sure to vent any air in the tank before working on anything! ::safety::

[headlight]

A single solenoid should be enough to control both the A and B ports. The important thing is that it is a 4/5 port solenoid.

If there is air leaking out of the exhaust port it is usually either an internal leak in the seal of the air cylinder or a leak in the gasket between the solenoid and the manifold block (if you are using a manifold setup).

[FRC Team CC]

What model solenoid are you trying to use?

We are using the solenoids from this link: http://www.wcproducts.net/pneumatics.

If you plug the exhaust ports, then the solenoid will cease to work. It only works by letting air out of the cylinder through the exhaust ports.

So we are trying to control a piston that is plugged in to one of the A ports (the solenoid is connected to port 0). It seems that if we don't block off the exhaust, the air tanks don't build up in pressure as all of the air leaks out of the solenoid exhaust. In our current setup, we have our compressor connected to our compressor and a regulator. The regulator is connected to the solenoids. Is something wrong with our setup that would cause this issue?

To maybe fix the leak, try checking the ends of the air tubes, and see if they're cut straight. If they've been cut at an angle, they can leak. Cutting them straight should fix it. Note that this may not be the issue, but it's a simple thing to check before you disassemble the whole pneumatics system. Just make sure to vent any air in the tank before working on anything! ::safety::

Our air tubes seem to be cut straight, so that shouldn't be an issue, but thanks for letting us know for future reference!

A single solenoid should be enough to control both the A and B ports. The important thing is that it is a 4/5 port solenoid.

If there is air leaking out of the exhaust port it is usually either an internal leak in the seal of the air cylinder or a leak in the gasket between the solenoid and the manifold block (if you are using a manifold setup).

Our solenoids are 4 port solenoids. We have a picture attached for what our setup looks like.

Thanks,
Charging Champions

[Owen Busler]

Arent pneumatics not allowed in ftc?

[headlight]

We are using the solenoids from this link: http://www.wcproducts.net/pneumatics.

...

Our solenoids are 4 port solenoids. We have a picture attached for what our setup looks like.

Thanks,
Charging Champions


Looks good. We use that exact same setup on our FRC robot. How bad is the leak? I'm curious about what you have plugged into the B ports on the left side, and why there aren't corresponding tubes in the A ports. Also, are you feeding air into the threaded "P" port on the side of the manifold not shown in that picture?

Can you tell which exhaust the air is leaking out of? EA or EB? If you manually trigger each solenoid, does it switch which port it is leaking out of? This would be a way to narrow down which part of the circuit has the problem.

If I were trying to fix this I would unscrew each solenoid and wipe off the gasket and the mounting surface, then carefully re-mount them, alternately tightening the two screws to reduce the chance of the solenoid sitting crooked. We have also had the rare instance where the solenoid was leaking internally.

[DaveL]

If you plug the exhaust ports, then the solenoid will cease to work. It only works by letting air out of the cylinder through the exhaust ports.


I agree with what Mark has said if one input port is distributing air to 2 output ports with both output ports sending air to a cylinder (one for push and one for pull).

If you have constructed a system where no pneumatic air is used to retract the piston, then you will need a stopper on one valve port.

For example: if only one port on the cylinder is connected to a valve, when that valve is diverting air to the other output port (say port 4), all connected air stores will be dumped. In this case adding a stopper plug to port 4 and making that port the "home" port will stop air from exiting when the system is being pressurized.

This kind of set up can be used to save air when the cylinder has a return spring or other method of returning the piston to its starting position.

A stopper can be made with a round urethane belt (tube) or use a tee fitting and connect 2 of the openings together with one tube to make a dead end, then connect the tee to your valve port.

Dave

[FRC Team CC]

Arent pneumatics not allowed in ftc?

We are an FRC team. Its just that the account that we are using right now is the one we used last year when we were an FTC team :] . Sorry about the confusion.

Looks good. We use that exact same setup on our FRC robot. How bad is the leak? I'm curious about what you have plugged into the B ports on the left side, and why there aren't corresponding tubes in the A ports. Also, are you feeding air into the threaded "P" port on the side of the manifold not shown in that picture?

Can you tell which exhaust the air is leaking out of? EA or EB? If you manually trigger each solenoid, does it switch which port it is leaking out of? This would be a way to narrow down which part of the circuit has the problem.

If I were trying to fix this I would unscrew each solenoid and wipe off the gasket and the mounting surface, then carefully re-mount them, alternately tightening the two screws to reduce the chance of the solenoid sitting crooked. We have also had the rare instance where the solenoid was leaking internally.

I'm not exactly sure if there is a leak. It seems that we are only able to control the A ports on the solenoid. We are not able to control the B ports, so the air keeps leaking out of there and makes the A ports useless.

Here is our code. Let us know where we are making a mistake:

package org.usfirst.frc.team6560.robot;

import edu.wpi.first.wpilibj.Compressor;
import edu.wpi.first.wpilibj.IterativeRobot;
import edu.wpi.first.wpilibj.Timer;

import edu.wpi.first.wpilibj.*;
import edu.wpi.first.wpilibj.smartdashboard.SendableChoos er;
import edu.wpi.first.wpilibj.smartdashboard.SmartDashboar d;

/**
* The VM is configured to automatically run this class, and to call the
* functions corresponding to each mode, as described in the IterativeRobot
* documentation. If you change the name of this class or the package after
* creating this project, you must also update the manifest file in the resource
* directory.
*/
public class Robot extends IterativeRobot {
static final long TIME = 10000; // ms
static final int SAMPLE_COMPRESSOR = 0;

Compressor sample_compressor;
long m_autoStartTime;
long m_autoStartTime1;
long m_autoStartTime2;

Solenoid solenoid_0;
Solenoid solenoid_1;
Solenoid solenoid_2;
Solenoid solenoid_3;

public void robotInit() {
sample_compressor = new Compressor(SAMPLE_COMPRESSOR);
solenoid_0 = new Solenoid(0);
solenoid_1 = new Solenoid(1);
solenoid_2 = new Solenoid(2);
solenoid_3 = new Solenoid(3);
}

public void autonomousInit() {
m_autoStartTime = System.currentTimeMillis();

}

public void autonomousPeriodic() {

solenoid_0.set(true);
solenoid_1.set(true);
solenoid_2.set(true);
solenoid_3.set(true);

}

}

Thanks,
Charging Champions

[Mark McLeod]

For a single solenoid
if
solenoid_0.set(true); - seems to be activating your A port (B port off)
then
solenoid_0.set(false); - would switch to activate the B port (A port off)

P.S.
You want to watch what the individual indicator lights are doing on the Pneumatic Control Module (PCM).
If your software works correctly, the status light for each solenoid will be (could be the reverse of this):

lit=port A deactivated/port B activated
unlit= port A activated/port B deactivated

And you should be able to see them change as your code changes the solenoid from true to false to true.

Once you are sure your code is doing what it is supposed to be doing, then look at the plumbing on the manifold. But make sure your code is operating correctly first.

[headlight]

Each solenoid only has two states.

State 1: Port A is connected to supply Pressure (P), Port B is connected to Exhaust B (EB).

State 2: Port B is connected to supply Pressure (P), Port A is connected to Exhaust A (EA).

With the solenoids you are using, one of these states will be the default. If you are using air cylinders that are powered in both directions by air you should have tubes running to both A and B ports. If you are using air cylinders that are not powered by air in both directions, then you should plug the unused port. I'm personally a fan of these plugs, (https://www.mcmaster.com/#5111k504/=15brr5b) but some people like these better. (https://www.mcmaster.com/#5779k54/=15brrfx) You can also get them from here. (https://www.amazon.com/SMC-KQ2P-07-Push-Connect-Fitting/dp/B00L5HLDPS/ref=sr_1_5?ie=UTF8&qid=1480888458&sr=8-5&keywords=1%2F4%22+push+to+connect+plug)

[FRC Team CC]

For a single solenoid
if
solenoid_0.set(true); - seems to be activating your A port (B port off)
then
solenoid_0.set(false); - would switch to activate the B port (A port off)

P.S.
You want to watch what the individual indicator lights are doing on the Pneumatic Control Module (PCM).
If your software works correctly, the status light for each solenoid will be (could be the reverse of this):

lit=port A deactivated/port B activated
unlit= port A activated/port B deactivated

And you should be able to see them change as your code changes the solenoid from true to false to true.

Once you are sure your code is doing what it is supposed to be doing, then look at the plumbing on the manifold. But make sure your code is operating correctly first.

Does activated mean open or closed valve?

[Mark McLeod]

Does activated mean open or closed valve?
You can think of the one solenoid as two valves that always do the opposite of each other, or as a single value with two positions.
In one position (deactivated), it is trying to fill one end of a cylinder (let's call it cylinder end A) by sending pressurized air out port A, simultaneously it is trying to empty the other end of the cylinder (let's call it cylinder end B) by releasing the other pressurized end out an exhaust port.

In the opposite position (activated), it reverses the cylinder action by sending pressurized air through port B to fill cylinder end B, while simultaneously emptying cylinder end A by connecting port A to an exhaust port.

http://www.team358.org/files/pneumatic/5P-CYL.gif

[FRC Team CC]

You can think of the one solenoid as two valves that always do the opposite of each other, or as a single value with two positions.
In one position (deactivated), it is trying to fill one end of a cylinder (let's call it cylinder end A) by sending pressurized air out port A, simultaneously it is trying to empty the other end of the cylinder (let's call it cylinder end B) by releasing the other pressurized end out an exhaust port.

In the opposite position (activated), it reverses the cylinder action by sending pressurized air through port B to fill cylinder end B, while simultaneously emptying cylinder end A by connecting port A to an exhaust port.

http://www.team358.org/files/pneumatic/5P-CYL.gif

We tested it out, and we have no control over any of the valves, even though we used the same code mentioned above and the solenoids are lighting up...

[FRC Team CC]

Do we need to manually lock the solenoids to property use it (the lock mechanism is the blue screw)?

[Mark McLeod]

We tested it out, and we have no control over any of the valves, even though we used the same code mentioned above and the solenoids are lighting up...

If the PCM and the lights on the solenoids themselves are changing from on to off to on, then look again at the plumbing.

tubing from your pressure regulator into the P port and the pressure gauge on the regulator reading 60 psi (in your picture the P port is plugged, but needs a press-to-connect fitting and air pressure of at least 30psi to work properly)
tubing from port A to one end of a cylinder
tubing from port B to the other end of the same cylinder

Don't leave any of this plumbing open to the air. It must connect to something or be plugged to prevent the pressurized air from running straight out to atmosphere.

Test manually:
Does the solenoid change when you push the blue manual button shown on top of each one in your photo (with pressure applied to the input port of course)?

Can you add code to toggle the solenoid value at the press of a button?

[Mark McLeod]

Do we need to manually lock the solenoids to property use it (the lock mechanism is the blue screw)?
No, it must be unlocked to work.
You can temporarily press the blue button to test, but don't twist and lock it.

[headlight]

No, the blue screw is a way to manually activate the solenoid valve. Unless it is pushed down or set in the lock position it shouldn't stop you from controlling it. A picture of your entire setup might be helpful.

Here is a link to the series catalog for your solenoid, it might give you some helpful information. http://content2.smcetech.com/pdf/SY3.5.7.9000.pdf

[FRC Team CC]

If the PCM and the lights on the solenoids themselves are changing from on to off to on, then look again at the plumbing.

tubing from your pressure regulator into the P port and the pressure gauge on the regulator reading 60 psi (in your picture the P port is plugged, but needs a press-to-connect fitting and air pressure of at least 30psi to work properly)
tubing from port A to one end of a cylinder
tubing from port B to the other end of the same cylinder

Don't leave any of this plumbing open to the air. It must connect to something or be plugged to prevent the pressurized air from running straight out to atmosphere.

Test manually:
Does the solenoid change when you push the blue manual button shown on top of each one in your photo (with pressure applied to the input port of course)?

Can you add code to toggle the solenoid value at the press of a button?

We have closed of all the ports and connected it to the cylinder, yet when we push air through it, we can only move the piston in one direction, and to move it in the other direction with would have to switch tubing placement, instead of just being able to code it to change. We have made sure to keep the air pressure to go in at 35 psi, and have closed all other valves (except the exhaust openings). We also have the code to toggle the solenoid using our game control to change, but haven't tested it out yet due to this problem.

[chapman1]

I'm not sure whether you are wanting to power the cylinder action in just one or both directions, so I wonder if you are using a 4-way valve when you need a 3-way valve?

One of the best explanations I've seen on the different types of valves is this one, from a Halloween prop website:

https://www.frightprops.com/faq/content/2/51/en/solenoid-valves-the-different-types-explained.html.

[FRC Team CC]

I'm not sure whether you are wanting to power the cylinder action in just one or both directions, so I wonder if you are using a 4-way valve when you need a 3-way valve?

One of the best explanations I've seen on the different types of valves is this one, from a Halloween prop website:

https://www.frightprops.com/faq/content/2/51/en/solenoid-valves-the-different-types-explained.html.

We want to power the cylinder in both directions, and our using a 4-way valve. Is that what we should be doing?

[feverittm]

Please post another set of pictures from your connections. It might be helpful.

If you pressurize the system on the compressor side of the regulator to 100psi and set the regulator to 40psi. Will the system hold pressure? How long?

Pressure input in the 'P' port, leave exhaust ports open, make sure all output ports are either plugged or connected to a cylinder (ports go vertical on the manifold).

If you can get it to hold pressure. Press the manual button and the cylinder should move. If you let go of the button it should go back to the original position.

You need to make sure you keep enough pressure on the valve or it won't work. And make sure that when you activate the valve you don't lose all your pressure. If you do as others have said, something isn't hooked up right.

[FRC Team CC]

No, the blue screw is a way to manually activate the solenoid valve. Unless it is pushed down or set in the lock position it shouldn't stop you from controlling it. A picture of your entire setup might be helpful.

Here is a link to the series catalog for your solenoid, it might give you some helpful information. http://content2.smcetech.com/pdf/SY3.5.7.9000.pdf


Here is a picture of our setup:
http://imgur.com/a/Wwh4t

[GeeTwo]

Here is a picture of our setup:


First, though probably not your issue, the CAN bus is not properly terminated. There should either be CAN leading away to the next device (most likely a PDP with the terminator set) or a 120 ohm resistor on the currently empty pins.

I think I've figured out everything up to the pressure regulator. It looks like the output of the regulator is split, with one going to the manifold, another going underneath. Where does that connect?

Trying to work out the black tubing connected to the manifold outputs. I cannot tell if the two inputs to the cylinder are coming from corresponding A/B ports, or adjacent sets of ports.
In any case it appears that you have plumbed the three unused A ports together, and the three unused B ports together. If all three (or four) of the solenoids are in the same position, this shouldn't be a leak, but if any of them is different from the others, you will be in an exhaust condition - pressurized air coming in one branch of the T and right out through another.
If the cylinder is plumbed to different solenoid valves, The only way you can operate it in this setup without dumping air is to switch all four ports at the same time.

The real solution to this problem (more manifold positions than you have separate items to control) is to plug the holes individually (we have some plugs in our shop, but don't know whence they came) or to use a "Blank Station Kit" like Vex's 217-2949 (http://www.vexrobotics.com/vexpro/pneumatics/solenoids-and-manifolds.html) for each unused control. Edit: I see it is also called a blank station kit with the same part number on the WCP link you provided (http://www.wcproducts.net/pneumatics).

Edit2: Found the plugs (http://www.mscdirect.com/product/details/09400839?fromRR=Y)like we have on MSC.

[DaveL]

Your set up is hard to follow. This is a problem for others trying to help, but more important its hard for you and your team to understand, troubleshoot and fix.

To make thinks easier for everyone, establish some color coding standards.
For example, you can use white tubing for the high pressure line. Then use green tubing for air that extends the piston and blue tubing to retract the piston.

That said there are way too many connectors and air lines.
Step 1. Simplify by only using 1 air tank.

Step 2. The regulator looks like it has 2 output pipes.
I have not seen this done. My suggestion is to cap the bottom opening.

Step 3. From what I can tell, air to push the piston is being sent from the 3rd valve (top) and air to retract the piston is being sent from the 2nd valve (bottom). You want to use one valve per cylinder.

To clean this up....
Remove all the tubes to the valves system except 1 providing air, 1 pushing the piston and 1 retracting the piston. All the other ports should be capped. Use a tee fitting and a loop of tubing to build the cap.

Flow limiting connectors can hide what is happening. I would replace these connectors with plain connectors. Once you understand how things work, try one before adding a second, as it might not be needed.

After you make these changes, take another picture but this time make sure all the tubes are visible, so we can help you trace their path.

Pneumatics is fun. Like programming, keep it simple to make it easy to understand.

Dave

[Mark McLeod]

I agree with the others about not connecting port-to-port on the manifold.
You need to deadend each unused port using, either a plug you can purchase, or juryrig a simple plug like the attached.

Other than that the tubing seems to be correct for your one cylinder.
The supplied pressure of 35psi is good to activate the solenoids.

Try swapping the two end hoses of the cylinder to see if the activated side is being affected by those flow controls.

Test using the manual blue button to make sure the plumbing is right, but then look back at the electrical connections.
Make sure the solenoid you expect (the second from the bottom or right in your photo) is wired to the correct PCM output. (The wires cross behind the regulator with another pair so they can't be traced from your photo)
Are both the light on the solenoid itself and the LED below the PCM solenoid connection changing when you attempt to switch cylinder direction?The photo doesn't show the back of the regulator well, so look for the flow arrow on the regulator and make sure your high-pressure line is going in the port directly opposite the flow arrow (on the side of the tail end of the arrow).
The regulator has one input high-pressure port and three regulated low-pressure output ports and all can be used.
We use one of the output ports for the low-side pressure gauge and the other two low-side outputs can be used or plugged as desired to direct the low-pressure outputs.

[headlight]

Here is a picture of our setup:
http://imgur.com/a/Wwh4t

Here is what I see:

1. You cannot connect an A port to a B port to plug it. That is not a plug, it is a short circuit. All the air will either come out of the A port and go into the B port and out Exhaust B, or vice versa. You can plug an A port into another A port, as long as the solenoids are actuated simultaneously or not actuated at all. Connections with this problem: A4/B4/A1 and B3/A3/B1. Swap those tee systems around so they are: A4/A3/A1 and B4/B3/B1.

2. Once you fix the problem in 1, that air cylinder should work. Test it by pressing the blue button on solenoid 2. Pressing it in should cause it to change state, releasing it should cause it to return to the previous state. If that works, then your system is physically sound.


If these things don't work for you, the next thing is to take note of your gauge pressures and behaviors and write a detailed narration of what you did, where the air came out of, what the gauges were reading... etc.

[bobbysq]

The real solution to this problem (more manifold positions than you have separate items to control) is to plug the holes individually (we have some plugs in our shop, but don't know whence they came) or to use a "Blank Station Kit" like Vex's 217-2949 (http://www.vexrobotics.com/vexpro/pneumatics/solenoids-and-manifolds.html) for each unused control. Edit: I see it is also called a blank station kit with the same part number on the WCP link you provided (http://www.wcproducts.net/pneumatics).

Edit2: Found the plugs (http://www.mscdirect.com/product/details/09400839?fromRR=Y)like we have on MSC.

You can also use smooth polycord of the same diameter as the tube. I don't know if this is any better/worse, but it's what we had on hand when we used pneumatics.

[PVCMike]

Step 3. From what I can tell, air to push the piston is being sent from the 3rd valve (top) and air to retract the piston is being sent from the 2nd valve (bottom). You want to use one valve per cylinder.


Dave

These manifold blocks can be confusing to look at because the associated A and B ports are not right above each other, they're diagonal. I think they have it plumbed right (That part anyways)

All of the other points Dave made are great and I agree, clean it all up, remove the flow restriction fittings, plug the second regulator port, plug all the unused ports on the manifold.

You definitely cannot plumb A to B, I suspect that is why you're venting all your air. Plug those up or at the very least plumb all the excess A together, and all the B together.

You can make your own plugs for the push fittings in a pinch, super easy. Cut a 2" piece of tubing, heat one end with a heat gun (until it's very soft), then clamp it between some pliers. I would not recommend you use this method on the robot, but we do all the time in testing.

[FRC Team CC]

Here is a diagram I drew of our setup for the solenoids

It might not be a great drawing, but I hope it helps you understand our setup a little bit better

http://imgur.com/a/g2MT4

[dardeshna]

Here is a diagram I drew of our setup for the solenoids

It might not be a great drawing, but I hope it helps you understand our setup a little bit better

http://imgur.com/a/g2MT4

Team 8 literally had this issue earlier today with the same manifold solenoid setup. While bridging A ports and B ports to plug them is probably also contributing to your issue, the way we resolved the original problem you described was actually quite simple.

Try unplugging all of the tubes from the manifold, unscrewing the solenoids, removing the rubber gaskets, and blowing out the internals of the solenoids and manifold with compressed air. Then reassemble carefully. And while you are at it, plug the unused ports properly with the loop setup someone posted a picture of above. For us, it was likely something was either loose or improperly installed.

You could think of this as the good old reboot! Good luck :)

You can make your own plugs for the push fittings in a pinch, super easy. Cut a 2" piece of tubing, heat one end with a heat gun (until it's very soft), then clamp it between some pliers. I would not recommend you use this method on the robot, but we do all the time in testing.

Edit: ^^^ Haven't tried this myself but it does seem like a very good option for testing if you are out of connectors.

[DaveL]

These manifold blocks can be confusing to look at because the associated A and B ports are not right above each other, they're diagonal. I think they have it plumbed right (That part anyways)



Thanks for the correction Mike.

One other point I'll make is that at each connector there is a chance of an air leak and each sharp bend causes internal air drag.

When you build the system for your competition bot, I would work out how to build the system with the fewest connections and the fewest bends.

For example: instead of having 2 air paths from storage, run air to tank1, then tank2 and then to the gauge. This eliminates, 2 tee connectors, 6 connections and 4 sharp bends.

Dave

[Daniel_LaFleur]

You can make your own plugs for the push fittings in a pinch, super easy. Cut a 2" piece of tubing, heat one end with a heat gun (until it's very soft), then clamp it between some pliers. I would not recommend you use this method on the robot, but we do all the time in testing.

Mike,

Just be aware that this method reduces the burst pressure of the tubing and can allow the tube to balloon and pop.

[FRC Team CC]

We tested it out, yet air is coming out from both valves (A and B) for one of the solenoids, but the diagrams show that it should only be coming out from one valve. For all the other solenoids, they work perfectly fine. We are not sure why this is happening, and are wondering if we need to buy a new solenoid.

[Mark McLeod]

Does it happen with each (all four) of your solenoids mounted to the manifold?

[Al Skierkiewicz]

CC,
A few items you might want to address.
Please use only one output port on the regulator, plug the unused ports.
The photo you linked has one way valves. These are air piloted valves that you show and if you back feed them with pressure (as shown in the photo), they will leak. Another source of leaking with the manifold you show, is damaged gaskets. The gasket between the manifold and the valves must be placed correctly on the valve body and the valve must be tight to the manifold. If the gasket is not placed properly and if the valve is loose, it will leak through the gasket from one port to another.

[PVCMike]

Mike,

Just be aware that this method reduces the burst pressure of the tubing and can allow the tube to balloon and pop.

Totally. That's why I mentioned it was not a good idea to use on the official robot, good for a quick test if you don't have any proper on hand. We purchase off-the-shelf rated plugs to use on the real thing, although I don't think we've ever had one of our makeshift ones fail.

[FRC Team CC]

So we tested out the solenoids with all the advice you guys have given us, and have found out that one of the solenoids was just not working properly, while all others were, so we have ordered a replacement solenoid.

[Al Skierkiewicz]

CC,
Please reread my post. If you parallel these valves they will likely not work as intended.