4 replies to this topic

HI ALL ,
I need a help.Can some one tell me about the following:

To control the flow rate of a liqiud we use pneumatically actuated control valve which takes in signal from the controller.
now this controller gets the Process variable(Flow rate again in this case)and accordingly asks the control valve to open or close in order to get close to the set point flowrate of the liquid.



I have a situation

I have a small tank of liquid which is 3/4th filled.this liquid has to be sprayed over a solid at a specific flow rate.
But different flow rate for different solids .hence we use control valve /controller(PI)/flow sensor arrangement
to ensure that specific flow rate.Now my doubts are:

1>WHETHER THE FLOW SENSOR SHOULD BE PLACED BEFORE THE VALVE OR AFTER THE VALVE.(AS IN FEED FORWARD OR FEED BACKWARD)IN THE ABOVE CASE.
2>IS PI CONTROLLER THE BEST ONE (I NEED A SPECIFIC FLOWRATE).
3>WE USE COMPRESSED AIR WITH CONSTANT PRESSURE TO TRANSFER THE LIQUID FROM THE TANKS
BUT IF WE APPLY BERNAULLIS' EQUATION ,WITH VARYING HEIGHT OF THE LIQUID THE FLOW RATE WILL FLUCTUATE EVEN WHEN STEADY STATE OF CONTROL VALVE IS ATTAINED.SO I SUPPOSE COMPRESSED AIR PRESSURE SHOULD KEEP ON INCREASING TO ENSURE CONSTANT FLOW RATE.

I will be greatful to get feed backs from you all







thank you

some one please reply ......

[djack77494]

Gatty,
In answer to your questions:

1)WHETHER THE FLOW SENSOR SHOULD BE PLACED BEFORE THE VALVE OR AFTER THE VALVE.(AS IN FEED FORWARD OR FEED BACKWARD)IN THE ABOVE CASE.
You do not understand the concepts of Feed Forward and Feed Backward. Placement of the valve will have no impact on whether or not you have a Feed Forward or a Feedback control system.

I typically put the valve downstream of the flow measuring element for a very specific reason. The most common flow measuring element type is an orifice plate with attached differential pressure transducer. There are restrictions on the placement of the orifice plate in terms of straight lengths of piping before and after the plate. The restrictions are needed to achieve accuracy. Because of these restrictions, it is generally much more convenient to locate the valve downstream of the plate. BTW, these restrictions also apply to SOME other types of flow elements; other types do not have such restrictions.

2)IS PI CONTROLLER THE BEST ONE (I NEED A SPECIFIC FLOWRATE).

A PI controller is frequently used in such applications and should be acceptable. For especially critical control loops, you could go to a PID controller.

3)WE USE COMPRESSED AIR WITH CONSTANT PRESSURE TO TRANSFER THE LIQUID FROM THE TANKS
BUT IF WE APPLY BERNAULLIS' EQUATION ,WITH VARYING HEIGHT OF THE LIQUID THE FLOW RATE WILL FLUCTUATE EVEN WHEN STEADY STATE OF CONTROL VALVE IS ATTAINED.SO I SUPPOSE COMPRESSED AIR PRESSURE SHOULD KEEP ON INCREASING TO ENSURE CONSTANT FLOW RATE.

For the variations you describe to occur, you infer that there are significant level variations and that the static head portion of the total differential pressure is a large part of the total. Under these conditions, you could allow the air pressure to rise and fall to offset level changes. But, why not just let the control valve handle this situation. That's what it's there for, right?

Use of all CAPITAL letters is "equivalent" to yelling. Please talk softly.
Doug

[Radionise]

Gatt,

A feedback control is characterised by a simple concept of:

"what you control is what you measure".

So, regardless of the position of the valve, since you still measure the flowrate and control the exact same flowrate, it is still simply, a feedback control.

[sgkim]

Hi gatty12:

(1) The location of flow sensor shall usually be on the upstream the valve as djack77494 commented. Both sensor locations, either before or after the valve, can generate feed back signals for flow control.
(2) PI mode for your situaton may not give any trouble but the sensitivity of valve positions on the flow rate is quite high. So additional differential mode seems to be required.
(3) The sensitivity of liquid level on the flow rate of liquid would be greater at lower air pressure range. The flow rate would be proportional to the root of liquid level when without air pressure. Check if over pressure is applied to the vessel, which makes the control valve almost closed. Chances are that no pressure release line is provided for the vessel.

I guess the control valve seems to be ill-sized: check the "differential pressure available" and "flow coefficient of the control valve". Check the present open-posion of the valve first. At normal flow rates, the valve should preferably be open about 30~80 % of stem position.

To solve the problem I recommend the following two ways:
(1) Control the pressure of the vessel manually at suitable presure range if the control valve is properly sized. The flow controller will do the rest job.
(2) Provide an adjstable dosing pump.

sgkim