Pneumatic valve

[Jnelson]

Does anyone have a good source for a 4-20ma pnuematic valve to control a double acting cylinder (4way 3pos). I need the valve to modulate a air cylinder to control flow of product through the gate (Infinite variable positioner). I am using a linear feedback trasducer,and a pot input to the plc for position and sp, but cannot get the thing to work right with solenoid type valves even with a deadband. Already tried festo and parker. Can't seem to find what i am looking for.

Thanks
John

[AGreavette]

You may want to try a CKD valve you can get them in AC or DC you can also look at the www.smcusa.com web site for some help. there is a 1-800 number that you can call and they will help you with what you need.

[john paley]

I'm assuming you want a 4-20 mA with something like a 0-60 or 0-100 PSI output. If so, try www.marshbellofram.com. They got some neat stuff. Or Mac Valves does I/P's, too.

I really don't know what you mean by 4 way, 3 pos. It sounds like a valve configuration, but not any analog type I've come across. I don't know what a double acting cylinder is, either. Unless you mean non spring return, with air connections on top and bottom.

If the cylinder is the "normal" type (I'm not pneumatically inclined) with air inlets on top and bottom, and you want to position it using a 4-20 type I/P, try applying a fixed pressure on one end, then modulating the opposing end with the I/P and see what happens. I never tried it but it might work, or it will go to one extreme ot the other with a slight imbalance--only slower.

[godfrey]

Try www.dev-air.com (http://www.dev-air.com)

[Steve Bailey]

Bimba has a line of linear and rotary pneumatic positioners that might do what you need.

Regardless of the actuator, traditional pneumatic spool valves are designed to be ON/OFF devices, rather than modulating controls.

John's method of modulating the back pressure can work, but unless you can tolerate a pretty large deadband, the stick/slip inherent in pneumatic cylinders will drive you crazy. By that, I mean you'll need to develop a significantly higher pressure differential to get the mechanism moving than you'll need to keep it moving. The link to Bellofram will help. They have a 'rolling diaphragm' design with lower internal friction.

If you wind up using Bellofram cylinders, be sure to design external stops at the end points of the mechanism's travel. Don't depend on the cylinder to limit the travel. The attachment of the rod to the rolling diaphragm is the weak link in the chain.

[drewcrew6]

Sounds like you're trying to control position with the wrong type of valve. As far as i know a 4 way 3 pos valve will only give full pressure to either port, or no pressure to both (when pressure is not applied that port is vented).

What you need is a proportional valve. That will accept a varing signal(4-20ma ? ,0-10v) and the pressure output will vary accordingly.

Appling pressure to one side and using a electric adjustable pressure regulator should work (never thought of that approach), as long as the air applied isn't full pressure (at least not on the back side) so that the other side will have enough force to move cylinder quick enough . All depends on desired response times.

Drewcrew6

[Dave Williams]

Proprotional control on a cylinder using back pressure regulation? I dont think so, first sign of the smallest air or seal leak will spell certain disaster & a trouble-shooting nightmare. I believe the cylinder described is being mistaken for a standard double action cylinder, the cylinder described is basically 2 in one, we have some of these on equipment in our plant. In our applications, they are simply operated by 2 mac valves. The cylinder will extend to center, full extend or full retract. If this is the application & you are simply looking for 3 positions from the cylinder, then using 2 standard type spool valves will suffice. If you are looking for infinate control, first I would not use a 3 position cylinder, i would simply use a standard cylinder of suitable length to cover what you need. This will simplify the application greatly, but as far as reliable infinate control, I/P is your best hope, but take care to use a high quality cylinder, not Bimba as their shaft seals just dont make it. But I feel if precision is important you need to look at an electric linear actuator that will accept an analog signal, this will work with the most reliability & accuracy in my opinion. Good luck.

[Jnelson]

Thanks for the responses.
First the valve that i have been using is a 4 way 3 posititon spring return to center work ports blocked.

This is not the problem, this valve configuration works fine it just overshoots the sp when i make adjustments on the fly. The lag time in energizing and denergizing the coil is enough to make the gate overshoot almost an inch.

This is not acceptable for what i am using this setup for.
What i mean by double acting is that the cylander can be powered in both directions (in and out).

I have seen proportional pneumatic spool valves in the past that do what i want but cannot remember who manufactures them. The deadband is definetely a problem since i am going to maintain batch weights to within +/- 10 pounds through a 6 inch valve. To keep this tolerance in product flow i need to keep a tolerance of 1/16 on the gate, so i dont believe the backpressure method will work.

Drew, Exactly what i need is a proportional valve 4ma is closed 20ma is open and if i send the valve 19ma it holds at 85% open "thereabouts".

Steve, thanks for the bimba link, i will check them out. Also i have hade great success with shrader bellows cylanders if that helps you any.

Thanks
John

[Dave Williams]

Well this certainly helps to clear things up for me, sorta. I've never seen a 4-20 spool type valve. You sure youre not thinking of an I/P converter, 4-20 to usually 3-15 PSI, but they do come in higher pressure ranges. These will certainly do what you want with the air pressure, but I just cant see a standard air cylinder being reliable in this application. If you were to use a standard cylinder, then a spring return would be the best as the I/P will maintain the pressure based on the analog it receives, thus you are just over powering the spring to maintain postion, this is standard application for mixing type valves on proportional blenders. Now that I type that, it occurs to me it should work fine, again barring the dreaded air leak. The advantage of the valves I mention are they use diaphrams, such as the Bellofram, sorta, & are not generally susceptible to things such as shaft leaks.

[Jnelson]

I have found a link to a valve that does exactly what i want to do check it out at http://www.svortex.com/infinite.htm if you are interested or have any comments. If i knew about this earlier i would of bought the canned deal.
So, can anyone tell me what kind of valve is on this thing?
Or where to get one like it?

Thanks

John

[Terry Woods]

Sure does look like a MAC Valve to me.

Do a search on "MAC Valves"

[Dave Williams]

Certainly does, but looks can be deceiving, or at the least misrepresented. Cant help but believe that this is still some sort of I/P type setup, in the typical sense, regardless of looks. Bottom line seems to be, if this will fit the application, why search further, seems youve found the device. I am curious about the dynamics of this device just to see how exactly it operates, but my gut tells me its still a I/P or its not very precise. I mean really, think about how 4-20 works & how air cylinders work, does anyone really think that 4-20 on a spool type valve with an air cylinder will be controlable, I just cant believe it.

[Jnelson]

Dave,
I have seen this done in a plastic processing plant a couple of years ago and it worked great.
The spool operates in either direction of center and either adds or exausts air out of both sides of the cylander to maintain the position.
Position feedback is through the transducer and the setpoint as i remember it was through a 4k pot.
All of this tied into the plc through a analog card, employing pid.

Thanks
John
http://www.theunholytrinity.org/cracks_smileys/contrib/ed/shitflagsniper.gif

[Dave Williams]

Well beats me, never seen anything like it. I have some friends in a blow molding plant, I'll check with them, maybe they know where you can find it. So it sounds like it might hunt for its position, air on, overshoot, positioner says were not where we need to be, so it goes back the other way. Just seems the way it would work to me, but then I've been made a believer before. I'll be sure to pass on any info if my friends know where these things exist.

[Eric Nelson]

"Cousin" John, (Wow, how many Nelsons are here?) :D

I'm almost positive that's a MAC 6200 series valve. If the picture was color, the valve would be BLUE. Though I'm not a big fan of MAC valves, the DO have very fast response times, which are ideal for this application. No matter what brand valve you choose, the heart of the system will be what CONTROLS the valve...

Trying to position the cylinder (accurately) without feedback will be nearly impossible. Also remember that the cylinder will not remain in the same position indefinitely. Since the rod takes up space (i.e. piston surface area), there is a lower force on that side. It will always "creep" towards the end with the rod. Without constantly correcting for this creep, the cylinder will eventually bottom out at that end. They do make shaft brakes that will hold the rod without air, but then you have the problem of what the cylinder will do when you release the brake.

In my opinion, any time you want to vary stopping position, you DON'T use a pneumatic cylinder. You just don't have enough control over it. I'd go with a servo/stepper motor, or even hydraulic, but NOT pneumatic.... Air is compressable.... Good for some application, bad for others.

But that's just my opinion :D

-Eric

[Jnelson]

"Cousin" Eric,
Thanks for the info on the mac 6200, i am going to go to their site and check it out.
By the way i am using linear feedback from a Baluff transducer.

Thanks
John

http://www.colute.net/cwm/contrib/anym/behead.gif

[Peter Nachtwey]

Originally posted by Jnelson

This is not the problem, this valve configuration works fine it just overshoots the sp when i make adjustments on the fly. The lag time in energizing and denergizing the coil is enough to make the gate overshoot almost an inch.

Are you sure that slow valve response that is causing the overshoot? Most valves spools can move move in millseconds. Unless you are going very fast the piston rod will not overshoot an inch in a few milliseconds.

Why 4-20ma?
Why not poppet valves?
Why not stepper valves?
Why not +/- 10 volt valves?

[rsdoran]

Why 4-20ma?
Why not +/- 10 volt valves?

Technically whats the difference?
The device used to control and the device used to DO the application control what is used.

Personally I think some on this site want/use things that dont make sense. Simple fact is you cant control a plain double acting air/hydarulic cylinder with solenoid valves for positioning. The time effort and extral devices used make this endeavor impractical if not impossible. That excludes the cost which overall makes it impractical.

Anyway many years ago I learned of acutators, these can be hydarulic or pneumatic. Many companies make Electroncially Controlled pneulmatic/hydraulic systems which can be controlled by plc etc.

Earlier I made a reply at work showing different things to use from different manufacturers, Bimba, Parker, Festo,Numatics etc etc. The post didnt work because I was on wrong laptop...ie didnt log in so lost the post.

Some have stated Bimba is bad but they do have an ECS series which means Electronically Controlled...kind of like an actutator. Bimba works as do others.

[rsdoran]

This is not the problem, this valve configuration works fine it just overshoots the sp when i make adjustments on the fly. The lag time in energizing and denergizing the coil is enough to make the gate overshoot almost an inch.

--------------------------------------------------------------------------------

BTW It does NOT work FINE when it does something you dont want it too.

[Peter Nachtwey]

Originally posted by rsdoran
Simple fact is you cant control a plain double acting air/hydarulic cylinder with solenoid valves for positioning.

What? I hope you meant that they can't be controlled by a PLC alone. The people that make the hydraulic/pneumatic controllers that fit in PLC racks or communicate over a field bus do this all the time.

My comment about the overshoot was meant to make one think about what is really happening. Think spring. More than a PID is required if there is much weight involved.

Note, pneumatic motion control will probably be the fastest growing means of motion for the next few years. Mechanically, pneumatics is cheap compared to other solutions. This is offset by the extra complexity of control. As stated above, a simple PID is not enough. The algorithms just require smart engineers and the processing power is getting cheaper every day.

[Eric Nelson]

Originally posted by Peter Nachtwey

Note, pneumatic motion control will probably be the fastest growing means of motion for the next few years. Mechanically, pneumatics is cheap compared to other solutions. This is offset by the extra complexity of control. As stated above, a simple PID is not enough. The algorithms just require smart engineers and the processing power is getting cheaper every day.
"Fastest growing"???? http://www.peyups.com/images/phpbb/icons/rotflmao.gif <---ROTFLMAO!

No, really, though I'm sure there are companies diligently at work on this, and someone will inevitably come up with a better mousetrap, I don't see this happening anytime soon. The idea is great, but c'mon, air has been around a LONG time (even longer than Terry!) http://www.peyups.com/images/phpbb/icons/wink.gif and no one has impressed me with a decent positioning system yet. I think it's mainly for the same reason you wrote "Think spring". Pneumatics have their place (I use 'em all the time), just not for variable positioning.

I used Festo's ServoPneumatic positioning system (ONCE!!!) on a 3-axis gantry system for Nabisco and was embarassed to put our name on it. Speeds and loads were all well within the capacity of the system, yet repeatability was basically non-existent (position OR speed). And it used gobs of CFM just maintaining position ("gobs" is a technical term, right?) :D
I would think that Festo would be at (or very near) the forefront when it comes to pneumatics.... Everything else they make is top of the line (Good 'ol German engineering!), so why do their servo-pneumatics suck? Maybe because they shouldn't have bothered pursueing it? :rolleyes:

OK, enough of my ranting... I'll shut up now shutit

beerchug

-Eric

[Jnelson]

Thanks to all that replied!

I'll let you know how it turns out.

Thanks

John

[Steve Etter]

John,

Another company to check with is Conoflow. I use to work for a company that used thier pneumatic positioners for just exactly what I think you are describing. We were metering material through an infinitely positionable gate via a 4-20ma control signal. I have not looked for a web site for them but I do know that Conoflow was the manufacturer.

Steve

[rsdoran]

quote:
--------------------------------------------------------------------------------
Originally posted by rsdoran
Simple fact is you cant control a plain double acting air/hydarulic cylinder with solenoid valves for positioning.

--------------------------------------------------------------------------------


What? I hope you meant that they can't be controlled by a PLC alone. The people that make the hydraulic/pneumatic controllers that fit in PLC racks or communicate over a field bus do this all the time.

Technically I did mean that Peter but not the way you stated. A plain double acting cylinder is meant to either extend fully or retract fully by its design. Yes you can use things to make it act differently than that but it wont be easy and never works exactly as needed, not counting the factors that if screwed with at all (flow or pressure settings) than timing issues can be a problem. Thats why actuators, electronically controlled etc type cylinders were created.

I agree with your assessment of AIR products being used more and more though. Yes air/pneumatics have been around along time but they are getting very inexpensive, more products, more features, easier to maintain with less mess.

[Peter Nachtwey]

Originally posted by Eric Nelson

"Fastest growing"???? http://www.peyups.com/images/phpbb/icons/rotflmao.gif <---ROTFLMAO!

-Eric
Don't judge the whole pneumatic world by one experience.

Attached are two zipped bitmaps file show a motion profile of a valve that is used to control the flow of natural gas or some steam. The spool, stem and piston on this valve weigh 500 lbs. The pressure used to control this was about 80 psi. There are two moves. One move goes to 10 inches out of a 12 inch stroke. The other move goes to 3 inches. The closed position is a 0 inches. Yes, the spool moves 12 inches on this 8 foot valve.
Notice that openning is much slower because the weight of the valve spool must be lifted and the surface area on the piston is less on the bottom. Notice that in both cases the main valve spool is positioned to a few thousandths of the command position. Do you see overshoot?


This pneumatic controller is positioning a 500 lb load to within a few thousands of an inch moving over 10 inches per second with 80 psi air.


This application appears to be very similar to JNelson's.

[Eric Nelson]

Originally posted by Peter Nachtwey

The spool, stem and piston on this valve weigh 500 lbs

This pneumatic controller is positioning a 500 lb load to within a few thousands of an inch moving over 10 inches per second with 80 psi air.

This application appears to be very similar to JNelson's.

OK, I think we're getting off the subject here...

Peter, you HAVE impressed me with the speed (AND accuracy!) of positioning such a heavy load... Can I ask who makes such a beast? And how much it costs!!! I am guessing that you're using air in this application because the application REQUIRES using air, no? When your options are limited, you find a way to accomplish the task within these parameters, period (else you don't make money).

But... I still stand by my opinion that using pneumatics for motion control (or, more specifically, accurate positioning) will NOT be "one of the fastest growing" means of motion (which, by the way, is the point you did not address). Yes, I agree that it makes a lot of sense to continue pursuing pneumatics for motion control, but I just don't feel that it's going to flood the market anytime soon.

Sorry Peter, but I really don't see how "this application appears to be very similar to JNelson's"... Am I missing something? He's just trying to position a small air cylinder, which in turn varies the opening of a gate. He started out by simply energizing a solenoid and quickly found out that, by the time you realize you're in position and turn it off, you've gone too far. A proportional valve is the solution, and others have pointed him in the right direction I think...

Ron said it perfectly:
"Simple fact is you cant control a plain double acting air/hydarulic cylinder with solenoid valves for positioning"

I agree http://www.peyups.com/images/phpbb/icons/smoke.gif

-Eric

P.S. Hey John, any use for a 500 pound valve? :rolleyes:

[Peter Nachtwey]

Eric,
Companies like Fisher make valves like this. This big valve is for controlling the flow of natural gas, steam or some other gas. Think of power plant and piple lines across the country. This big valve is like a gate. The big main valve is positioned using air. A smaller valve controls the flow of air into the each end of the cylinder that move the big valve. This is a three stage system.

I don't know why pneumatics are used instead of hydraulics. Motor are not used because the valve must close when power fails.

I don't know how much the valve cost. I am sure it is upwards of $50K US. Maybe as much as $100K US. The controller is experimental. It was used to test some concepts. It costs too much for a simple application but someday better technology will be far smaller and cheaper package.

Yes, we are off the topic. I just couldn't stand by while there were statements about pneumatics not being able to do position control.

I would give a stab at answering the the original post, but we were not given enough information to give an precise answer. I would want to know the require CV rating, mass of the load/gate, stroke, inside diameter, piston diameter, speed, amd accel/decel rates . I am surprised that no one asked these questions. I know. I am slicing pickles again, but I can slice pickles with precision.

[cgolding]

try tbhl@tbaytel.net They deal with hydraulics and know what they are doing

[Jnelson]

Peter,
I never asked anyone to do the project for me!

All the original post implied was that i needed a good source for this valve.

As far as the information that you request, it is irrelevant to what i am trying to accomplish(find the valve).

In any case if interested i did get the valveing to work using a 4 way 3 pos spring to center work ports open in neutral.I installed pnuematic speed controls on the working lines going into the cylinder and by slightly slowing it down, achived the goal that i wanted to.

BTW, i did find the valve from mac list price $375.00 US. What do you think i have wrapped up in the two speed control valves?
You guessed it $37.00........

Sometimes it's not about the automation, but about accomplishing the goal. Isn't the first rule of engineering "keep it simple"?

Eric,

I dont have any use for a 500 pound valve,but i could use a 500 gallon beer mug.

Thanks

John

[Steve Bailey]

John,

I've reread the posts in this thread and I'm not sure what you wound up using. Was it a proportional valve or did you find a way to use the 4-way, spring-centered valve?

[Jnelson]

Steve,

I ended up useing a 4 way, 3 position, spring centered, double solenoid valve from smc.

The valve sits in a nuetral position when no power is applied to either solenoid, with the inlet blocked and the work ports open to exaust.

When one side is energized the spool shifts position to that side and the opposite side is open to atmosphere.

When power is removed from this same side it spring returns to center.

What was happening was the valve was opening so quickly that it was overshooting the sp i have set up in the plc. This is a 12 bit analog card so my resolution was pretty tight on the transducer. So what i did was scale the resolution down from 0-4095 to 0-100 over a 6 inch stroke,so if you do the math on that it comes out to somewhere around 16 counts per inch or 1/16" tolerence. Good enough for the girls i go out with!

Also, as i stated previosly, i installed pnuematic speed controls on the work ports of the cylinder. These little beauties are great!

Thanks to all though, this was a fun little project!

Thanks

John






I was also able to get the gate to position more accurately by useing immediate instructions in the plc.

[ste1171]

You can always try FESTO. They have a proportional valve which would probably do what you want. It is an MPYE analog valve 5/2way.

[fareast]

If your set on using pneumatics , then Honeywell and Spirax make closed loop positioners (EP4/5 if i remember correctly) these will give you a fairly accurate response to an analogue signal . If however you want to do the job a lot cheaper then many companies make damper actuators which will also give you what you want (mostly 0-10V but a 500ohm resistor will give you that )The downside may be response speed (30 seconds or more)If you want a real high torque actuator then use a Honeywell modutrol valve actuator.

[iant]

The biggest problem with Pnuematic positioning is the ability to 'lock' the cylinder.
this can be done with a special cylinder brake or an external brake device.
Normally the positioning of thei type is Hydraulic.

An option is hydraulic dampening with valves fitted to use the HYD. cylinders as the brakes.

the existing system you are using relies on NO AIR LEAKAGE.
Either in the valves or cylinder or piping.
Normal Air braking is achieved by Pilot operated ckeckvalves as close to the Cylinder as possible.

Examine the use of the valving you have, To me it is wrong.
I have done this type of system before.