FACULTY OF ENGINEERING

DEPARTMENT OF CHEMICAL AND ENVIRONMENTAL

ENGINEERING
___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

ECH 3112
PROCESS CONTROL & INSTRUMENTATION
LABORATORY
MODUL 2
FLOW CONTROL
(SHORT REPORT)
Submitted to:
DR. SYAFIE - LECTURER

Prepared by:
(GROUP 6)

SIOW ZHI HOONG (

HO CHAI LING (
SAIFUL BAHRI BIN ABD AZIZ (140515)

Experiment Date: 20th AUGUST 2009

Date Submitted: 3rd SEPT 2009
1.0 INTRODUCTION
 A proportional–integral–derivative controller (PID controller) is a
generic control loop feedback mechanism (controller) widely used in
industrial control systems. A PID controller attempts to correct the error
between a measured process variable and a desired set point by calculating
and then outputting a corrective action that can adjust the process
accordingly and rapidly, to keep the error minimal.

The PID controller calculation (algorithm) involves three separate

parameters; the proportional, the integral and derivative values. The
proportional value determines the reaction to the current error, the integral
value determines the reaction based on the sum of recent errors, and the
derivative value determines the reaction based on the rate at which the error
has been changing. The weighted sum of these three actions is used to
adjust the process via a control element such as the position of a control
valve or the power supply of a heating element.

By tuning the three constants in the PID controller algorithm, the

controller can provide control action designed for specific process
requirements. The response of the controller can be described in terms of
the responsiveness of the controller to an error, the degree to which the
controller overshoots the set point and the degree of system oscillation. Note
that the use of the PID algorithm for control does not guarantee optimal
control of the system or system stability.

Some applications may require using only one or two modes to provide
the appropriate system control. This is achieved by setting the gain of
undesired control outputs to zero. A PID controller will be called a PI, PD, P or
I controller in the absence of the respective control actions. PI controllers are
particularly common, since derivative action is very sensitive to
measurement noise, and the absence of an integral value may prevent the
system from reaching its target value due to the control action.
2.0 RESULT AND DISCUSSION

a. raw experimental data

FIC11 & FIC12

PB2 : 150%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 at FIC11 : 5m3/hr
SV2 at FIC12 : 4.5 m3/hr

FIC11 & FIC12

PB2 : 150%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 at FIC11 : 3m3/hr
SV2 at FIC12 : 2.5 m3/hr

FIC11 & FIC12

PB2 : 100%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 at FIC11 : 5m3/hr
SV2 at FIC12 : 4.5 m3/hr

No.4
FIC11 & FIC12
PB2 : 100%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 at FIC11 : 3m3/hr
SV2 at FIC12 : 2.5 m3/hr

No.5 (apply disturbance)

SV2 at FIC11 : 5m3/hr
SV2 at FIC12 : 4.5 m3/hr
FIC12 left alone at,
PB2 : 150%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 : 2.5 m3/hr

FIC11
PB2 : 150%
TI2 : 6 seconds
TD2 : 0 seconds
SV2 : 3 m3/hr,

(6a)= 3.5 m3/hr,

(6b)= open manual by-
pass valve for 5 min.
(6c)= increased MV by
10%

(Test disturbance)
No.7
PB2 : 100%

No.8
PB2 : 200%

PB2 : 150%
TI2 : 9 seconds
TD2 : 0 seconds

PB2 : 2%
 PB2 : 20%

PB2 : 40%

PB2 : 60%

Table: raw experimental data.

To study the responses from different PID settings in the flow control,
two systems have been used. The inflow system is using LIC/FIC11 and the
outflow system is using TIC11/FIC12. First at all, the tank T12 is filled with
water till it reaches the overflow Drain pipe outlet (D).

The experiment results were shown in the graph of figure 1. When the
experiment started, four difference color lines were shown on the paper by
the chart recorder. First, the red color line indicates the water level in the
tank T12, green color line indicates inflow, blue color line indicates the
outflow and the purple color line indicates the temperature. By referring the
purple color line in the graph printed by the chart recorder, the line is linear;
therefore the temperature in the system remains constant through the whole
process.

The Proportional (P) and Integral (I) control is set into FIC11 and FIC12
by using the trial PID values for both controllers, that is PB2= 150%, TI2= 6s
and TD2= 0s. (PI action controller.) The first disturbance is observed when
the inflow set point was 5m3/hr which is specified at point 1.The disturbance
that presence can be detected by moderate oscillation and the damping
pattern of green line.

Meanwhile, when the outlet flow rate which was changed to 4.5m3/hr,
the blue line was increased steadily to its steady state in a monotonic
pattern, which was no any oscillation. This is shown at point 2. As a result,
the respond of inflow was a sinusoidal response whereas outflow was a first
order step response. Both inflow and outflow increased gradually which then
led to the rise of water level in tank T12. The red line moved upward as
shown in the graph indicates that the water level in T12 is rise due to the
increased of water inflow. Gradually, the response would reach a new steady
point. However, it would be take some time to reach the new steady point.

When the inflow and outflow are nearly steady with slight oscillations
about the set points, the inflow and outflow were set to 3.0m3/hr and
2.5m3/hr respectively (point 3) in the auto mode. This disturbance caused
the inflow and outflow decreased to the values of the new set points in a
negative step change response. This is because the controller had detected
that the values of the controlled/process variables are much higher than the
new set points.

As a result, the response of inflow was decreased in an over damp

oscillation pattern whereas outflow decreased again in a monotonic pattern
in order to reach its steady state.

As the system was achieved steady state, the PID control was change
value of PB2 from 150% to 100% with the inflow and the outflow were
increased to 5.0m3/hr and 4.5m3/hr respectively (indicated point 4). There
was an increment due to the increased of flow with the outflow shown a
slower response. When the inflow and the outflow were adjusted to 3.0m3/hr
and 2.5m3/hr, the chart showed a decrement pattern with an over damp
oscillation.
 The pattern of the curve of PB2 = 100% is almost similar to the pattern
of the curve of PB2 = 150%. This verify that PB = 150% has no significant
difference with PB = 100%. This slight difference is due to the slightly
difference in Kc.

At this section of recording, the level in the tank increased to its

maximum value and became stable at this value. Changes in the inflow and
outflow did not affect the level because the inflow is still larger than the
outflow. From point 1 to point 4, the temperature is remain constant without
any changes and the water level is also remain constant at the maximum
value as the inflow is always set to be larger than outflow to avoid the tank
to drain out.

When the inflow and outflow were increased to 5.0m3/hr and 4.5m3/hr
respectively (indicated point 5), the green and blue lines increased to new
set points but the outflow line shown a slower response. After a period of
time the inflow will approach the new set point. When the inflow and outflow
were reduce to 3.0m3/hr and 2.5m3/hr, the lines decreased until it stabled
and reached the new set point. The level in the tank increased to its
maximum value and became stable at this value.

After the system achieved steady state, the outflow was set at
2.5m3/hr, PB2 = 150%, TI2 = 6 seconds and TD2 = 0 seconds. Therefore, the
blue line remained at a constant value. At point 6(a), the inflow was
increased to 3.5m3/hr and decreased to 3.0m3/hr. The result indicated that
the changes were due to a step input since the graph didn’t show an
oscillation pattern.
 Next, the by-pass valve was opened for 5 seconds (indicated point
6(b)), the inflow was increased in a sudden and dropped drastically when the
valve was closed.

Also, the response for the water level decreased severely due to the
shutting of valve and gradually increased after a while. At point 6(c), the MV
was step increased by 10% manually and the result showed the inflow
increased and formed a small overshoot.

When the three steps 1(a), 1(b) and 1(c) are repeated with different
values of PB2 = 100% and 200% in the disturbance test for the inflow
loop(indicated point 7 & 8), we observed The chart recorder shows the same
pattern of responses with small oscillation and no significant oscillation
amplitude difference for both PB2 values.

Theoretically, a larger PB will give a smaller steady gain due to their

inverse proportional relationship and hence they are in reciprocal. A smaller
steady gain will give damper oscillation in process controlling. However, the
chart shows no obvious difference in the oscillation amplitude because the
flow rate that we use is too small due to machine error.

Once the system became stable, the PB2 was set back to 150% and
TD2 at 0 second but the T12 was set at 9 seconds. The same results were
observed instead of the third step (indicated point 9). The system could not
go back to the previous steady set point at 3.0m3/hr when MV is increased by
10%. Instead, an offset is observed. Proportional controller cannot eliminate
the error completely, but there will be an offset between the output and the
set point. In this case, the derivative does not take place, since TD2 = 0.
Therefore the offset persists.

The step from point 10 to 13 was done to determine the Kc of the

controller. From point 10 up to 13, the PB2 was increased from 2%, 20%,
40% and 60%. TI2 was adjusted to 6 seconds once again. The TD2 was
remained at 0 seconds. The value of Kc increased drastically. These values
were too large and over the stability limit. Large dampen oscillations were
observed because of larger Kc will increase instability and cause larger
oscillation.

When a disturbance is applied to the set point, the flow rate changed
drastically. But when PB2 was set higher and higher until 60%, the oscillation
was become smaller (indicated point 13).

Thus, the critical value for PB2 is around 60%. Here, the value of Kc is
the limiting proportionality of the proportional controller. Any K c value over
this limiting value will results an unstable response. The time to reach steady
state will also increase as Kc increase.

Finally, we re-adjusted the PB2 value to 150% (indicated point 14), TI2
at 6 seconds and TD2 at 0 second. This setting brought the system to PID
system. Vigorous oscillation was obtained as the inflow was disturbed by
opening/closing the by-pass valve of pump P12. The derivative action gives
the controller the capability to anticipate where the process is heading. The
amount of anticipation is decided by the value of TD. Right now, this
controller not only can integrate the error, but also calculate the rate of
error. Thus, it is a more efficient controller. The oscillation dampened a little
when the test at point 6 was applied but became stabilize in a quick time. In
reality, time should be given more to observe the reduction of the oscillation.
3.0 CONCLUSION AND RECOMMENDATIONS

From the experiment, we can prove that Proportional Integral

Derivative Controller (PID) is a useful controller as it can deal with different
disturbances. The derivative action gives the controller the capability to
anticipate where the process is heading by calculating the error.

Different disturbances test are done purposely in order to observe PID

controller to eliminate the error. The PID controller will only take the control
action in the auto mode. That is why this controller is more efficient than
others. However, there still have some disadvantages of this controller such
as not suitable for the fast and noisy system like the flow system. Fast
process is easily susceptible to process noise. Typical of this fast process are
flow loop such as what we did in this experiment. The application of the
derivative mode on the flow process will only result in the amplification of
the noise because the derivative of the fast changing noise is a large value.

The other disadvantage of PID control is the control action will not take
place when there is a steady error. As for a steady error, de/dt is 0, thus
control action will not take place. On the other hand, the proportional
controller (P) is the simplest because the only tuning parameter is the PB but
they carry an offset in the controlled variable and do not provide the
required control. A large value of TI is not suitable in process control as the
time for the integral control to take action is too long. An error could be
larger and larger until it saturated if given enough time. Since there is
defection for proportional-integral controller, the presence of derivative in PI
controller will result in a better control. Though causing more oscillations, the
derivative can handle a drastic change (de/dt is large).

Several precaution steps should be taken to reduce error happened

during experiment. First at all, parallax error should be avoided while taking
measurements. Besides, before starting experiment, the strainer valve
should leave open for a while to strain eddies and residues. A better data can
be obtained by spending more time observing the trend of the response to
each disturbance applied. The experiment can also be repeated for a few
times to obtain a constant and more accurate result.