Pneumatic Control

How about asking the control system Designer? What does it say on the "for construction" pneumatic schematic drawings? Only, they cannot be seen from here.

Thanks i thought that the pressure to the actuator is always in the range of 3-15 psi. But upon looking at some specification sheets i have come across the term "Actuator Bench Range" and its getting clear to me.

i have two further questions:

1)What i was told by one of my colleagues is that for an I/P positioners we receive a 4-20mA signal that is converted to the respectve 3-15 psi signal which is then fed into the actuator. But from what i understand the output of the I/P can be any value of psi suitable for the valve diaphragm(no standard there). Am i correct in my understanding?

2)If we have a valve whose actuator can tarvel from 0-100% within the 3-15psi range can we directly connect the output from the pneumatic controller to the actuator or will we need a positioner in that case as well?

I am working on elec control systems and am getting introduced to pneumatic systems for the first tme via your reply so i dont have a better ununderstanding of them.

Re: Pneumatic Control
Remember that air is compressible.It does not move at the speed of electrical signals.
A brand new,properly calibrated valve will operate 0-100% at the specified pressure.This pressure will vary according to the manufacturer and application.
The diaphragm would become excessively large using 3-15 psi in some cases,so a higher pressure,smaller diaphragm valve is sometimes designed.
The standard 3-15 psi valves are very rugged,and under engineered.I have seen them tolerate 400% overpressure without damage,but I would not use a 3-25 for 60 psi service routinely.
A control system is usually designed by an engineer in the particular field,and it is best to use specified materials and procedures.
The standard I/P pressure is 3-15 psi,but other values are available.It is usually easier to use a standard I/P because then you only have to stock one type for spares inventory,and use bias or multiplying relays to boost the output.
The valve positioner can be thought of as a traffic cop of sorts.It is "looking" at the valve stem position,comparing it to the input value(from the I/P) and making sure that the valve goes to that position,using whatever force is necessary to do it.This accomplishes 2 things:It overcomes hysteresis and sticking in the valve actuator,and by having a very low air volume demand from the I/P it reduces time delay in the control loop.Remember, a diaphram requires a relatively large volume of air for a full valve stroke.The actuator has internals that require very little volume,so it can be though of as an amplifier,or step up transformer of sorts.The positioner has it's own high pressure supply for providing this action.
I had a similar problem as yours many years ago when transitioning from pneumatics to electronics.There is a lot of correlation between the two.
What I still like about pneumatics is they can be totally independent of the grid and still perform very precise control,lightning and power surges have no effect on them,and if properly maintained,will last forever.Panama canal still has the original pneumatic controls .A pneumatic valve can function where electrically operated valves will fail,such as high temperature,high humidity conditions, such as inside of steam chambers.And of course, safety in explosive environments.
Hope this helps you.

Welcome to the world of pneumatic instrumentation!
Some pneumatic control valves have 3-15 PSI actuator bench ranges, while others do not. For those that are 3-15 PSI it is possible to operate them directly from an I/P output or from the output of a 3-15 PSI pneumatic controller. Valves having some other bench range *must* use a signal converter or a positioner to allow a 3-15 PSI signal to command the valve. The same may be said for double-acting (e.g. piston) valve actuators, which require *two* pneumatic pressure sources: one to drive the valve open and another to drive the valve closed. Here, a positioner is required to convert a single command signal into two pneumatic drive pressures to actuate the valve. However, even 3-15 PSI single-acting valve actuators controlled by 3-15 PSI signals benefit from the use of positioners.
Simply stated, the job of a valve positioner is to guarantee the control valve's stem position is faithful to the command signal, whether that signal be 3-15 PSI, 4-20 mA, or digital (e.g. Fieldbus). Lots of factors contribute to stem position error, chief among them being (1) force on the valve plug resulting from process fluid pressure drop across the valve, and (2) valve stem packing friction. When a positioner sees a valve stem's position deviating substantially from the equivalent value of the command signal, it throws as much or as little pressure to the valve actuator as it can to move that valve's stem to the correct position. In essense, a positioner is a "stem position controller" all its own. Pneumatic positioners function as proportional-only controllers, while some modern digital electronic positioners utilize full PID control to achieve tight response.
Positioners also have greater air flow capacities than most I/P converters and pneumatic controllers, which means a 3-15 PSI valve driven by a positioner will generally be faster-responding than the same valve driven directly by an I/P or driven directly by a pneumatic controller. If fast response is important but accurate valve positioning is not, you may find a pneumatic volume boosting relay used in lieu of a positioner. In applications where fast and accurate positioning is required on a large pneumatic actuator, you may find both a positioner and pneumatic volume-boosting relay(s) installed on a valve.
For more information on pneumatic instruments and control valves in general, feel free to peruse chapters 14 ("Pneumatic Instrumentation") and 27 ("Control Valves") in the online "Lessons In Industrial Instrumentation" textbook. Right-click on the following link and save the PDF file to your computer before opening it, as it is very large (> 50 MB):

http://www.ibiblio.org/kuphaldt/socratic/sinst/book/liii.pdf

The reason is as you indicated 3-15 PSIG for transmitter and controller, but the valve may require higher pressure to operate.

Look at the valve manufacturer data sheet.

Another way to picture a valve positioner is like an Op Amp.It tries to balance both inputs,one from the I/P(pneumatic) and one from the valve stem(mechanical),by adjusting it's output.

Imagine the positioner as having a lever with a rolling fulcrum.It tries to balance the lever by moving the fulcrum(output) in the required direction to achieve balance.
The fulcrum may be centered in some cases, but if the valve hesitates or sticks, or develops a small leak,the gain(output) will adjust as necessary to achieve equilibrium between the desired input signal and the desired output position of the valve.

Here is a general "rule of thumb" to apply when tuning a pneumatic control loop:Flows are very fast,so they require relatively low gain(P),high reset(IO) and high derivative(D).Heating loops, in large volume retorts,or tanks, are very slow,so they require high gain,low reset,low derivative.Here is a method I have used very successfully,and it can apply to electronic controls as well:Set I and D Values to Zero.with gain at mid range.(Unless it is a flow loop,then set gain VERY LOW,to avoid run away oscillations)Bump the process set point upward,till you see the process variable begin to move,and then back to the original setpoint.Observe the response time till it returns to a Stable Value.Record this value.(Graph paper will prove handy if the display is not electronic with a trending graph option).Do not worry about offset from setpoint at this time.There will be an offset.Stability is what we are looking for.You are looking for a 1/4 wave decay,that is, the process oscillates 1/4 the amplitude in each successive oscillation.Do not concern your self with the time to return to setpoint at this stage of calibration,because it will have an offset at this stage of the calibration.Once you have established 1/4 wave decay, adjust reset,(I) to the reciprocal of the time period in minutes of the above mentioned wave form.(Basically,convert from frequency to time).(sound familiar?)Bump the process again, and optimize the return to setpoint by adjusting the reset (I) value.It should return to setpoint this time, but do not be concerned with how long it takes to do so,that will be taken care of when we adjust the derivative(D).If the process oscillates a little at this point, reduce gain Slightly Bump the process once more, and adjust the Derivative (D) to 1/8 the value of the reset in minutes.Notice that the return to setpoint is much quicker. Optimise all of these adjustments in small increments for best performance.The ideal is a process that returns to setpoint quickly,with minimum oscillation and minimum overshoot or undershoot.

In summary:Gain(P) adjusts the amplitude of the output signal as a percentage of the deviation from setpoint. It does not care about offset from setpoint.

Integral,(sometimes called RESET in certain pneumatic controllers),is designed to get the process back to setpoint,by repeating the initial (P) output as many times as necessary to accomplish this.It does not care how long it takes.

Derivative (D) is designed to operate in a time-sensitive manner.(Actually, like looking forward in time to predict the next response needed) It's output depends on how fast the process deviates,and the output is a function of this time period.This is only one of many methods to tune a loop,but this is the one I chose and it worked for me.

Be aware that some pneumatic controllers refer to Gain, as Proportional Band,or simply PB.Do not confuse this with Proportional,as used in PID.

Tricky,I know,but read documentation carefully.

Don't let it worry you.It is simply the reciprocal of gain.

(Honeywell controllers come to mind here.)

Ok, that's it:Basic Loop Tuning, 101.You can dig as deep as deep as you want on this subject,and it gets very deep, but when confusion reins,(like going behind someone that did not have a clue), go back to the basics and work from there.I hope this helps you get started.Have fun.

Let me know when you have that AHA! MOMENT, the bulb lights up and it all comes together.