INTERACTING AND NON-INTERACTING SYSTEMS

Objectives:
To study the dynamic response of single and multi capacity processes when connected in
interacting and non- interacting mode.

Block Diagram:

(a) Non-interacting system

(b) Interacting system

Specifications:
Product

:Interacting and Non interacting system

Product code

:328

Rota meter

:Make Eureka, Model MG 11, Range 10-100 lph, Connection

BSP back, screwed, Packing PTFE +Silicon Make

scale

Process tank Acrylic, Cylindrical, Inside Diameter 92mm With graduated

in mm. (3 Nos)

Supply tank

:SS304

Pump

: Make U.P. National Mfrs.Ltd., Model THS 3000, Type submersible,

Head 3 m, 1200 lph discharge, Watts 35, Volts 240 AC, 50Hz Fractional
horse power, type submersible

Overall dimensions

:410Wx350Dx705H mm.

Theory:
Non-interacting system:
In non-interacting system, we assume that the tanks have uniform cross sectional area and the
flow resistance is linear. Writing a mass balance around the tanks gives the transfer function of
the system that relates h2 to q.
Writing mass balance for tank1
q- q1 = A1(dh1 / dt) (1)
Mass balance for tank 2
q1- q2 = A2(dh2 / dt) ..(2)
The flow head relationships for the two linear resistances in non-interacting system are given
by the expressions
q1= (h1 / R1) .(3)
q2= (h2 / R2) .(4)
From (1) and (3)
..(5)
Where Q1 = q1 - q1s, Q = q - qs and

= A1 R1

From (2) and (4)

(6)
Where
H2 = h2 - h2s and

= A2 R2

Overall transfer function is calculated as follows

..(7)
For a step change of magnitude A in Q,
Q(s) = A / s
...(8)
H2 at time t is given by
{

}) .(9)

For an impulse change of magnitude V (volume added to the system)

Q(t) = V TM(t); So, Q (s) = V
..(10)
For impulse change, H2 at time t is given by
.(11)
Considering non-linear resistance at outlet valve of the tank R2 can be calculated as
R2 = 2 dH2 /dQ
Where dH2 is change in level of tank2 and dQ is change of flow from initial to final state.
Put the values in equation (11) to find out H (t) Predicted and plot the graph of H (t) Predicted
and H (t) Observed Vs time.

Interacting system:
Assuming the tanks of uniform cross sectional area and valves with linear flow
resistance the transfer function of interacting system can be written as:

.(12)
Let
..(13)

(14)

For a step change of magnitude A

.(15)
In terms of transient response the interacting system is more sluggish than the non-interacting
system.
For an impulse change of magnitude V (volume added to the system)
Q(t) = V TM (t); So, Q (s) = V
..(16)
For impulse change H2 at time t is given by
(17)
Considering non- linear valve resistance, the resistance at outlet of tanks can be calculated as
R1 = 2

(18)

R2 = 2

(19)

Experiment - 1
Aim:
To study the step response of two first order systems arranged in non-interacting mode.

Procedure:

Start up the set up.

A flexible pipe is provided at the rotameter outlet. Insert the pipe in to the cover of
the top Tank 1. Keep the outlet valves (R1 & R2) of both Tank 1 & Tank 2
slightly closed. Ensure that the valve (R3) between Tank 2 and Tank 3 is fully
closed.

Switch on the pump and adjust the flow to @90 LPH. Allow the level of both the
tanks (Tank 1 & tank 2) to reach at steady state and record the initial flow and
steady state levels of both tanks.

Apply the step change with increasing the rotameter flow by @ 10 LPH.

Record the level of Tank 2 at the interval of 30 sec, until the level reaches at
steady state.

Record final flow and steady state level of Tank1

Carry out the calculations as mentioned in calculation part and compare the
predicted and observed values of the tank level.

Repeat the experiment by throttling outlet valve (R1) to change resistance.

Observations:
Diameter of tanks ID : 92mm
Initial flow rate (LPH):
Initial steady state level of Tank 1 (mm):
Initial steady state level of Tank 2 (mm):
Final flow rate (LPH):
Final steady state level of Tank 1 (mm):

Final steady state level of Tank 2 (mm):

S. No.

TimeGf Level of tank 2

(sec)

1
2
3
4

H(t) observed

(mm)

(mm)

H(t) predicted
(mm)

()
--

Take readings till the steady state reached in level of tank 2, fill up columns H(t) observed and
H(t) predicted after calculations.

Calculations:
H 2(t) observed = (Level at time t - level at time 0) x 10-3
{

H 2(t) Predicted =

})

Where
H 2(t) Predicted is level in Tank 2 predicted at time t in mm.
A = magnitude of step change
= Flow after step input - Initial flow rate in m3/sec.
According to Harriots plot method:
From the graph, find the time taken to reach 73% of maximum observed height
t73/(

1+

2)

=1.3

Calculate 0.5( 1+

2)

, find the height corresponding for the same time from graph

find H/km ,where km is the maximum observed height

From Harriots chart shown below estimate
Where
tank1.

1
2

1, 2

is time constant of tank1, A1 is area of tank1 and R1 is resistance of outlet valve of

is time constant of tank2, A2 is area of tank2 and R2 is resistance of outlet valve of

tank2
Area of tank 1 = /4 (d12) in m2
Area of tank 2 =

/4 (d22) in m2

Considering non-linear resistance at outlet valve of both tanks, it can be calculated

as R1 = 2
R2 = 2
Where dH1 is change in level of tank1 and dQ is change flow of from initial to final state and dH2
is change in level of tank2 at initial and final state. Put the values in the above equation to find
out H2 (t) Predicted and plot the graph of H2 (t) Predicted and H2 (t) Observed Vs time.

Model Graph:

Summary and Conclusions:

1. The step responses of two first order systems arranged in non-interacting mode is
studied and a graph is plotted between Time versus H2 (both observed and calculated).
2. It is noted that the theoretical value of height matches with the experimental values and
the response is found to be over damped.

Experiment-2
Aim:
To study the step response of two first order systems arranged in Interacting mode.

Procedure:

Start up the set up.

A flexible pipe is provided at the rotameter outlet. Insert the pipe in to the cover of the
Tank 3. Keep the outlet valve (R2) of Tank 2 slightly closed. Ensure that the valve (R3)
between Tank 2 and Tank 3 is also slightly closed.

Switch on the pump and adjust the flow to @90 LPH. Allow the level of both Tank2 and
Tank 3, to reach the steady state and record the initial flow and steady state levels of both
tanks.

Apply the step change with increasing the rotameter flow by @ 10 LPH.

Record the level of the Tank 2 at the interval of 30 sec, until the level reaches at steady
state.

Record final steady state flow and level of Tank 3

Carry out the calculations as mentioned in calculation part and compare the predicted and
observed values of the tank level.

Repeat the experiment by throttling outlet valve (R1) to change resistance.

Observations:
Diameter of tanks: ID 92mm
Initial flow rate (LPH):
Initial steady state level of Tank 1 (mm):
Initial steady state level of Tank 2 (mm):
Final flow rate (LPH):
Final steady state level of Tank 1 (mm):

S. No.

Time

Level of tank 2

(sec)

(mm)

1
2
3
4

H(t) observed
(mm)

H(t) predicted
(mm)

()
--

Final steady state level of Tank 2 (mm):

Take readings till the steady state reached in level of tank 2, fill up columns H(t) observed and
H(t) predicted after calculations.

Calculations:
H (t) Observed = (Level at time t - level at time 0 ) x 10-3 m
H2(t) Predicted =
Where
A

= magnitude of step change

= Flow after step input - Initial flow rate in m3/sec

Use the Harriots method and find

Where
tank1.

1
2

2.

is time constant of tank1, A1 is area of tank1 and R1 is resistance of outlet valve of

is time constant of tank2, A2 is area of tank2 and R2 is resistance of outlet valve of

tank2 . Considering non-linear resistance at outlet valve of both tanks, it can calculated as
R1 = dH1 /dQ and R2 = dH2 /dQ
Where dH is change in tank height for change in flow dQ. Calculate values of b,

and from

equations given in theory part. Put the values in the above equation to find out H (t) Predicted
and plot the graph of H (t) Predicted and H (t) Observed Vs time.

10

Model Graph:

Summary and Conclusions:

1. The step responses of two first order systems arranged in interacting mode is studied
and a graph is plotted between Time versus H2 (both observed and calculated).
2. It is noted that the theoretical value of height matches with the experimental values and
the response is found to be over damped.

11

Experiment-3
Aim:
To study the impulse response of two first order systems arranged in Non-Interacting mode.

Procedure:

Start up the set up.

A flexible pipe is provided at the rotameter outlet. Insert the pipe in to the cover of Tank
3. Keep the outlet valve (R2) of Tank 2 slightly closed. Ensure that the valve (R3)
between both Tank 2 and Tank 3 is slightly closedSwitch on the pump, adjust the flow
to @90 LPH. Allow the level of both thetanks to reach at steady state and record the
initial flow and steady state levels.

Apply impulse input by adding 0.5 lit of water in Tank 3.

Record the level of the Tank 2 at the interval of 30 sec, until the level reaches to steady
state.

Record final steady state level of Tank3.

Carry out the calculations as mentioned in calculation part and compare the predicted
and observed values of the tank level.

Repeat the experiment by throttling outlet valve (R1) to change

resistance.

Observations:
Diameter of tanks: ID 92mm
Initial flow rate (LPH):
Initial steady state tank 3 level (mm):
Initial steady state tank 2 level (mm):
Volume added (lit.):

12

Final steady state tank 3 level (mm):

Final steady state tank 2 level (mm):

S. No.

TimeGf Level of tank 2

(sec)

1
2
3
4

(mm)

H(t) observed
(mm)

H(t) predicted
(mm)

()
--

Take readings till the steady state reached in level of tank 2, fill up columns H(t) observed and
H(t) predicted after calculations.

Calculations:
H (t) observed = (Level at time t - level at time 0) x 10-3

V = Volume of liquid added as an impulse input (in m3)

(For 1, 2 and R2 refer values obtained in experiment 2)
Put the values in above equation to find out H (t) Predicted and plot the graph of H (t)
Predicted and H (t) Observed Vs time.

13

Model Graph:

Summary and Conclusions:

1. The impulse response of two first order systems arranged in non- interacting mode is
studied and a graph is plotted between Time versus H2 (both observed and calculated).
2. It is noted that the theoretical value of height matches with the experimental values.

14

Experiment-4
Aim:
To study the impulse response of two first order systems arranged in Interacting mode.

Procedure:

Start up the set up.

A flexible pipe is provided at the rotameter outlet. Insert the pipe in to the cover of Tank
3. Keep the outlet valve (R2) of Tank 2 slightly closed. Ensure that the valve (R3)
between both Tank 2 and Tank 3 is slightly closedSwitch on the pump,adjust the flow to
@90 LPH. Allow the level of both the tanks to reach at steady state and record the initial
flow and steady state levels.

Apply impulse input by adding 0.5 lit of water in Tank 3.

Record the level of the Tank 2 at the interval of 30 sec, until the level reaches to steady
state.

Record final steady state level of Tank3.

Carry out the calculations as mentioned in calculation part and compare the predicted
and observed values of the tank level.

Repeat the experiment by throttling outlet valve (R1) to change

resistance.

Observations:
Diameter of tanks: ID 92mm
Initial flow rate (LPH):
Initial steady state tank 3 level (mm):
Initial steady state tank 2 level (mm):
Volume added (lit.):
Final steady state tank 3 level (mm):

15

Final steady state tank 2 level (mm):

Sr. No. Time Gf level of tank 2

(sec)
1
2
3
4

(mm)

H(t) observed
(mm)

H(t) predicted
(mm)

()
--

Take readings till the steady state is reached in level of tank 2, fill up columns H(t) observed and
H(t) predicted after calculations.

Calculations:
H (t) observed = (Level at time t - level at time 0) x 10 -3

V = Volume of liquid added as an impulse input (in m3)

(For calculating 1, 2, , and R2 refer experiment 3)
Put the values in above equation to find out H (t) Predicted and plot the graph of H (t)
Predicted and H (t) Observed Vs time.

16

Model Graph:

Summary and Conclusions:

1. The impulse response of two first order systems arranged in interacting mode is studied
and a graph is plotted between Time versus H2 (both observed and calculated).
2. It is noted that the theoretical value of height matches with the experimental values.

17