# Flow Control Question

T

#### Thomas Szovik

Hi All

1/ Object: Maintain of constant flow rate with a regulating valve at a
discharge gravitational.
2/ Control value: Pressure transmitter at the bottom of a cylindrical
vessel.
3/ Available controller: Honeywell UMC800, without differentiating
function.
4/ Proposition: Utilizing the D-function alone of a PID block 2 generate
a signal proportional to the variation of pressure.
5/ Question: Will this functioning? Has anyone experience or idee how to
proceed?=20
6/ Observation: Sorry for my English what I have pick up by reading, I
speak French, German, Hungarian of course and a little Malagasy

Thomas Szovik

2330 Dunaharaszti
Petofi S. ut 52.
Hungary
[email protected]

S

#### Steve McAlpin

Hi Travis,

We use a magnetic flow meter in our water treatment plant to monitor flow it
has a accumulative option associated with it they are called Tiger Mag's.
The company is called Sparling Instruments and you can find them @
http://www.sparlinginstruments.com/http://www.sparlinginstruments.com/ .
Hope this helps you out they seem to be pretty good meter. Our plant has
been online for about 6 years and we have had very little problems with
them.

J

#### jmGiraud

Constant gravitational discharge:
The velocity of a free gravitational discharge is given by the Toricelli formula.
v=SQRT(H)
v=velocity

Introduce some geometry

(1) Q=k.SQRT(H)
Q=flow rate
k=geometric discharge coeffient

Introduce the valve flow rate

(2) Q=cv.SQRT(DP)=cv.SQRT(H)
cv=Valve discharge coefficient
DP=Differential pressure across valve
=H (back pressure is negligable)

Solving for valve 'cv'
(3) cv=Q/SQRT(H)
Calculate the required 'cv'for about 8 points.
NOTE:The valve opens on decreasing signal.
Graph inversely to compare diectly with valve manufacturer chart cv:%opening.
If the two graphs do match closely then you wil have a 'Closely Constant Gravitational Flow Discharge'.
An air to close ball valve should be very close
The flow rate depends entirely upon the vavle 'cv'
and the closeness of the relationship between the valve 'cv' and the calculated "cv".
The remaining of the loop components are just to
adapt the signals to each individual.
If things would be right, this is a perfect
controller because it has nothing to do and is not in the loop.
Check with valve manufacturer what they can offer.
They should be able to shape the cam of the positioner so that it will meet the calculated "cv".