NEED HELP FINDING A FAST ACTING AND EASY WAY TO AUTOMATE VALVE

[bob1111]

We are looking to automate a butterfly valve and it looks like we may want a type of servo control. The motor only needs to rotate 90* but needs to go from close to open in .1sec as well as be positioned anywhere in between very quickly. We do not "need" to know the actual position but just be able to precisely control the water we are flowing.

The valve will have a 2in butterfly, and I have not done any numbers yet but should not take to much torque. Water pressure will be 30-50psi. We were hoping to finding some we could apply a variable voltage to for various positions similar to an RC servo. Many servo motors have encoders and we really do not need that as long as we can control things well with just voltage.

So what type of motor or actuator fits this application? We had also looked at using an air cylinder to actuate the valve through a pwm control of the air supply to the cylinder but unsure how we would reduce pressure in the system. A little hard to get my head around I guess. If would could work this out, the air system might work too.

 

[waross]

Use a pneumatic valve positioner. The air control/modulating problems are already solved. Feed the output of your controller to the valve positioner.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[bob1111]

Would this offer about the fastest response of anything out there? Do you happen to know of a schematic of how these work? Just curious of the mechanics of one.

Do you basically supply an air line to the positioner and send a frequency depending on what position you want the valve or does the PWM directly control the air valve open and close?

 

[waross]

Standard industrial instrument systems usually work on a 4-20ma control signal. 4ma is zero and 20ma is 100%.
There are a great many off the shelf controllers that will accept an input from a flow element and output a control signal. The valve positioner accepts a 4-20ma signal and plant air pressure.
When the controller changes it's output signal the valve positioner changes the position of the valve to match.
Many standard devices may be converted to a voltage control scheme.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

http://www.emersonprocess.com/fisher/products/valvepositioners/index.html

They are often called "I to Ps".

That's "current to pressure". "(4mA to 20mA) to pressure".

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Moylan62]

It sounds like system is not that big you might want to look at using a VFD drive and keep your valve simply off an on. It will pay for its self and if you aren't fimliar with control loops it will save you a lot of head aches might be your simplist solution VFD are very simple an kinda cheap an can be set up very easy to simple control functions

 

[bob1111]

What would we run the VFD on?

 

[waross]

In the days of transition from pneumatic to electronic instrumentation we differentiated between An I to P transducer and a valve positioner.
Many pneumatic valves had a calibrated spring that would start to move at 3 PSI and be at full travel at 15 PSI.
(3-15 PSI was the pneumatic equivalent of 4-20 ma)
When an application required more power than was available from a 15 PSI actuator, a valve positioner was used.Typically the positioner would accept the 4-20ma signal and convert the signal into the actual valve position. A valve positioner has a mechanical link to the valve trim so it knows exactly what the valve position is. The valve positioner uses plant air typically at 100+ PSI for operating power.
I/Ps were common in HVAC applications where many valves and dampers were directly operated by 3-15 psi signals.
I have used I/Ps in industrial plants with valves and dampers operated by diaphragms with 3-15 PSI springs but if you need faster or stronger action, use a positioner.
If varying process pressures are high enough to partially counteract the force of a 3-15 PSI actuator, again the positioner was the answer.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[roydm]

Now you have the actuator sorted out, where will you get a 2" butterfly? I expect you will end up with a ball valve
"We do not "need" to know the actual position but just be able to precisely control the water we are flowing", for that you will need a flow-meter.
Make sure you have the valve sized correctly, too large and it will be hard to control.
A piston or vane type actuator will stroke faster than a diaphragm type or electric valve.
Let us know how you get on.
Roy

 

[mccla5214]

Someone is asking for a lot of conflicting requirements.

Snap for open to close in .1 second
Be able to regulate the flow in between ( is this also in the .1 sec requirement also?)

I dont' think you are going to find a flow meter that can respond that fast.

As for your torque requirements you have to have a much high torque to get the valve moved that quickly.

Why are you needing to snap open and closed that fast? It is going to cause a huge water hammer and burst pipes.

What is the primary requirement, being able to turn the flow on and off or to regulate the flow in a precise manner.

Not knowing what you are trying to do is making this complicated.

Here is a suggestion:

Use on open close 2 position ball valve from USA bluebook with the spring return down stream of the flow meter. This will satisfy the instant on/off requirement.

Put a USABlueBook air operated flow control ball valve on the line (modulating valve) downstream of the flow meter but before the 2 pos valve.

Modulate the flow control valve to the desired flow rate and use the 2 pos valve to turn on and off the flow - leaving the modulated valve in the same position so when you restart the flow it will still be at the same rate as last time.

Put surge controllers from Graingers upstream and downstream of the valve assemblies so try and limit the violence of the water hammer when you turn on/off the flow.

Hope this helps,

Mike

 

[TurbineGen]

viper, I'm going to have to agree that regulating a valve that fast is going to cause serious surge problems, but if that is truly your requirement a Hitec HS5925MG servo might be what you need. At 6 volts the transition time is .1 seconds for 90 degrees. It tops out at about 100 inch-pounds of torque, which might be enough.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[bob1111]

The specification is NOT set in stone but only represents a relative speed. NO, closing the valve quickly will never happen, we may, under emergency, fully open the valve that fast to flood our device to protect it from overheat.

I did look at this servo and it is precisely what I had in mind because it should be pretty accurate in angle changes. I do not want nor require R/C performance so we may opt to bypass parts o the circuit if possible unless something like this can be bought without R/C compatibility.

I think this is the strongest idea we have found yet. Metal gears would be a must as reliability is a large concern.

 

[bob1111]

I did notice the spec was 100 oz/in of torque, that is about .5lb/ft. Not very much but I have not done the math yet. Maybe someone can help on the force required to close or hold a butterfly valve with 50psi water and 2in bore. my 5lb/ft was only a guesstimate.

 

[roydm]

As I said previously a pneumatic piston or vane actuator will be faster than electric with no gears to break.
A butterfly is fairly well ballanced so doesn't require a lot of torque. Once it's past 70% open you will see very little change in flow.
2" is small for a butterfly, you might be better off with a ball or globe valve.

 

[itsmoked]

Roy; Do they have controllable needle valves that size?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[itsmoked]

These are the people that do this. They have all types of valves and controls.

I think you should be considering a ball valve perhaps followed by a surge tank but whatever.

Here you go...

http://www.dynaquip.com/automated.asp

Note, they do buttery flys too.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[sreid]

Look at what Exlar has to offer.

http://www.exlar.com/index.html

Fast and powerful.