Valve controller selection 1

[thrusterpump]

I need to select a device to control a pneumatic valve using the 4-20mA input on the actuator. The control must be manual, by that I mean the user will vary the 4-20mA current to set the valve in the desired position by whatever means the device allows. We would prefer that the control allow for very fine adjustments. Any recommendations on what type of device can be easily incorporated and used on this type of application? It is in an industrial setting that will be regularly used. Any advice would be greatly appreciated.

 

[ScottyUK]

Something like

http://www.beka.co.uk/setpoint_safearea.html

perhaps?



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

If we learn from our mistakes I'm getting a great education!

 

[thrusterpump]

Yes, thanks for the tip. For reference, any other brands or types, maybe with more features? This is a good start, thanks again

 

[itsmoked]

Perhaps:

http://www.omega.com/ppt/pptsc.asp?ref=DPF30&Nav=greg09

Keith Cress
kcress -

http://www.flaminsystems.com

 

[roydm]

How about any cheap PID controller locked in manual.
You can feed the output back to PV to give a nice large display.
I assume you arn't using a PLC or DCS
Regards
Roy

 

[PNachtwey]

A cheap PID will not work for controlling pneumatic systems if the load is a big valve and the valve must be positioned rapidly and accurately. A pneumatic system postion system is a third order system. This means that the controller should have 3 gains to place the poles. This means there should be a proportional, derivative and second derivative gains. Clearly a PID does not have a second derivative gain. If the integrator is added it will only make the steady state error zero. The integrator has its own pole so that can be counted as one of the three required poles.
In addition, the system is not linear which causes problems for cheap controller. These systems typically require a position feedback, duh. The trick is to get the second derivative for the second derivative gain. This can be done by differentiating the position twice. Good luck with out using an observer. One can use an accelerometer. This works well in the lab but will not survive in many industrial environments. The best method is to put pressure sensors on either end of the cylinder. The pressure on each end of the cylinder is mulitplied by the area on each side of the piston and a differential force is computed. The differential force is roughly proporitonal to the acceleration of the load or the second derivative.

So how big is the valve and how fast and accurate must the control be?

Peter Nachtwey

 

[JLSeagull]

A manual station is open loop -- no feedback. Manually adjust the setpoint to 100 percent, the output is 20 mA, a 3-15 psig I/P converter establishes 15 psig output - or a digital valve controller establishes 100% open (or closed) as appropriate. Scotty locked in the firing solution. Other manufacturers or models for a 4-20 mA manual station could be helpful.

 

[roydm]

PNachtwey,
What the heck are you smoking?
A PID controller locked in Manual mode is nothing but a Manual loading station.
Set the output at 50% and you get 12mA, right now, no delay nothing!
I suggested feeding the output back to PV is just for display purposes.
Other advantages of using a controller-
Can be the same hardware as other loops for operator familiarity.
Allows for future automatic possibilities.
Probably cheaper than a dedicated manual loader.
The original post said the valve has "4-20 mA input" & to "very fine adjustments" so presumably it has a positioner rather than just an I/P. This will be the speed limiting factor not the 4-20 mA source.
Regards
Roy

 

[roydm]

BTW, I have nothing against using a Manual loading station as ScottyUK, Itsmoked & JLSeagull suggest, I was just offering an alternative.
Regards
Roy

 

[thrusterpump]

Valves are 1.5" and 4" v-port ball valves with positioners. Control is not required to be fast but accuracy would be nice, or at least accurate enough to set a flow rate (gpm) based on what is read from a differential pressure transmitter. We want to be able to set say 900 gpm on the 4" line by opening/closing the valve with the controller in question. Could any of you explain what I/P and PV is? Newbie here!

Thanks again to all who have posted

 

[ScottyUK]

I/P is an abbreviation for current-to-pressure transducer, the device which interfaces between the electrical world of 4-20mA and the pneumatic world of 3-15 PSI

PV is an abbreviation for process variable, usually the quantity being controlled by a control loop. The term PV is often used in conjunction with SP, the abbreviated version of Set Point. SP is the value to which the controller tries to regulate the PV.


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

If we learn from our mistakes I'm getting a great education!

 

[roydm]

Since you seem to have a flow meter, why don't you close the loop by wiring it to a controller, that way you can set the flow where you want and the controller will take care of any variation in supply pressure.
A flow loop is probably the easiest loop to set up.
As I said before, a controller will be about the same cost as a manual loader and the operator will be happy with the results. I recently purchased 1/4 DIN controllers for < $250 ea.
Regards
Roy

 

[thrusterpump]

Thanks again to everyone who responded.

Roy, do you have any specific controller references. We are using Rosemount 3051 dP transmitters

 

[roydm]

Thrusterpump,
I have used a bunch of Honeywell UDC2500 lately, they have a NEMA 4 front which makes them nice for a wet area.

http://honeywell.silverw.com/docs/51-52-25-127.pdf

A UDC2500-C0-0A00-200-00000-00-0 would do it for you, no special features, they sell for just under $500. Make sure you get the -0A00- option otherwise the Auto/Manual buttons are dissabled.
If you want something cheaper check out the

http://www.Omega.com

Regards
Roy

 

[PNachtwey]

PNachtwey, What the heck are you smoking?

oops, I just got a little ahead of my self. I am used to controlling the spool and the actuator that it controls.

 

[roydm]

PNachtwey,
Ah, now I see, you are using cylinders for high speed robotics or servo positioning. No wonder I couldn't understand it.
Sorry if I came back a bit strong.
Regards
Roy