INDIAN INSTITUTE OF TECHNOLOGY KANPUR

DEPARTMENT OF CHEMICAL ENGINEERING

Unit Operation Lab (ChE-492)

INTERACTING AND NON-INTERACTING TANK SYSTEM

1. Objective:
To study the dynamic response of liquid level in Single Tank, Two Tank
Interacting and Two Tank Non-Interacting System.
2. Introduction:
The principle distinction to be made in multi-capacity processes is in how the
capacities are joined. If they are said to be isolated or non-interacting, the
capacities behave exactly as they would alone. But if coupled, they interact with
one another, in which case the contribution of each is altered by the interaction. In
non-interacting system the two tanks levels do not interact because the flow from
the first tank to the second tank is independent of the level in the second tank. In
interacting system the levels in both tanks interact because any change in the down
stream level will affect the upstream level.

The following general rules are applied to the principle of interaction;

1. The degree of interaction is proportional to the ratio of the smaller to the larger
capacity (not time constant). Where this ratio is low (<0.1), the capacities may
assumed not to interact .
2. Interaction always works towards increasing the larger time constant and
decreasing the smaller one.
3. Specifically with regard to the behavior of the system with equal time constant
and of equal capacity, the effect is a combination of one large and the rest smaller
time constants.
3. Description:
Present set-up is specially designed for studying the dynamic response of Liquid
Level in Single Tank, Two Tank Interacting and Two Tank Non-Interacting
System. Apparatus is self-contained re-circulating unit. Water from Sump Tank is
 sucked and delivered to a Constant Level Overhead Tank by means of a
Centrifugal Pump. Flow rate of water is controlled and metered with the help of
given Rotameter. Apparatus can be operated for any of the desired modes viz.:
Single Tank System, Non-interacting System or Interacting System by operating
the valves.
Water, passing through Rotameter, can be allowed to enter first in Process Tank-1
or Process Tank-2 by fixing the valve position provided for this purpose. To run
the apparatus in Single Tank System mode, the water will be allowed to enter in
Process Tank-1.The drain of Process Tank-1 is supplied to Process Tank-2 with
provided valve and then drained back to Sump Tank. Resistance will be fixed for
the drain valve of Process Tank-1 and reading will be noted as briefed further in
the this Manual.
For conducting the experiment on Two Tank Non-Interacting System, the
apparatus will be operated in the same mode as Single Tank System, but resistance
will be produced at drain of both Process Tank-1 and Process Tank-2. Reading will
be noted down as per Manual instruction.
When the system will be operated Two Tank Interacting mode, the water will be
allowed to enter in Process Tank-2 and drain valve will be fixed in close position.
Water form Process Tank-2 will be allowed to Process Tank-3 from the bottom by
opening a valve. Resistance of this valve and drain valve of Process Tank-3 will be
fixed and reading will be noted as per the instruction of this Manual.

4. Utilities Required:
1. Electricity Supply: Single Phase, 220 V AC, 50 Hz, 5-15 amp socket with earth
connection.
2. Water Supply (Initial Fill)
3. Drain Required
4. Floor Area Required: 1 m x 1 m.

5. Precaution & Maintenance Instructions:

1. Never run the apparatus if power supply is less than 180 volts & above than
230 volts.
2. Always keep apparatus free from dust.
3. To prevent clogging of moving parts, Run Pump at least once in a fortnight.
 4. Always use clean water.
5. It apparatus will not in use for more than one month, drain the apparatus
completely.

6. Troubleshooting:
1. If pump gets jam, open the back cover of pump and rotate the shaft manually.
2. If pump gets heat up, switch off the main power for 15 minutes and avoid
closing the flow control valve and by pass valve simultaneously.

EXPERIMENT NO. 1
 SINGLE TANK LIQUID LEVEL SYSTEM

1. Aim:

I. To obtain the step response of a single tank liquid level system to a step change in
input flow and to compare it with the theoretical response.
II. To determine the time constant of the system from the step response curve in a
single tank liquid level system.

2. Theory:
A single tank liquid level system is first order in nature. The transfer function relating
deviation of liquid level in the tank to the deviation in the inlet flow rate is:

where τ = AR is the time constant of the system.

For step input of the magnitude “M” in the inlet flow to the tank :

Taking inverse Laplace transform of equation (2)

This equation gives the theoretical variation of liquid level in the tank with
respect to time for a step input of magnitude “M” in the inlet flow rate.
The experimental value of τ is obtained from the step response curve in which
63.2% of the step response will be completed within a time equal to to one time
constant. A liquid level system is non-linear in nature.
The discharge from the vessel is proportional to the liquid level in the vessel.

where “C” is a constant.

In the linear region of the operation the above equation becomes qo = h/R
where “R” is the resistance of the discharge valve. The resistance “R” is the
 reciprocal of slope of the tangent line to head (h) vs discharge (Q)curve at the
first steady state value.

3. Experimental Procedure:
1. The liquid level in vessel is maintained at a steady state value around the lower
middle part of the level gauge by properly adjusting the inlet and outlet flow rates.
2. At the steady state condition note the level “hs” and discharge “qs”.
3. Now a step change in input flow is introduced by opening the cock valve at the
bypass of inlet valve.
4. Simultaneously start a stopwatch and note the rise in the level with time till the
level reaches a new steady state value.
5. At new steady state, note the level and outlet flow rate. The difference between the
initial and final flow rates gives the magnitude of step input “M”
To Determine the Resistance of the Valve:
1. The final steady state level and the discharge of the system in the above
experiment are noted.
2. Now the inlet flow to the vessel is decreased by a small amount and allows the
system to come to a new steady state value.
3. The experiment is repeated for different steady state value noting the level and
discharge each time.
4. A plot of discharge vs. head is drowned from which the resistance of the valve can
be obtained.

4. Observation &Calculations:
D = 0.108 m
Observation Table:

t, min h, m hS, m qS, LPH qfs, LPH

Plot a graph between qS vs. hS to determine the Resistance (R) of the valve

Calculations:
5. Nomenclature:

A = Cross-sectional area of process tank, m2

D = Inner diameter of tank, m
f = Subscript of final state
h = Height of liquid in process tank, m
hs = Initial Height of liquid in process tank , m
H = Change in height (h – hS) in process tank after step Change, m
M = Magnitude of step change, m3/min
qfs = Final steady state feed flow rate, m3/min
qs = Initial flow rate for the process, m3/min.
R = Resistance given to the valve
S = Subscript of steady state

τ = Time constant, min

6. References:
1. Donald R. Coughanowr, “Process System Analysis and Control”, 2nd
ed., McGraw Hill, NY, 1991, Page 64-67.
 EXPERIMENT NO. 2

TWO TANK INTERACTING LIQUID-LEVEL SYSTEM

1. Aim:
To obtain experimentally the step response and to compare the experimental
and theoretical responses.

2. Theory:
A dynamic system is the one in which there is some varying amount of
accumulation of conserved quantities with time. Consider a liquid level system
shown below in which two tanks are arranged in series such that the response of
first tank depends on the conditions in the second tank. Such a system is said to be
an interacting system and is a lumped parameter system.

This arrangement is such that the flow through the resistance R1 depends on both h1 and
h2. Such a system is known as interacting system. Selecting a macroscopic system
consisting of the entire tank, the total transient material balance for Tank 1 is:
 d (h2 − h2 s ) h1 − h1s h2 − h2 s
= −
dt A2 R1 A2 R2

In terms of deviation variables

dH1
A1 = Q − Q1 − − − − − − − − − − − −(5)
dt

dH 2
A2 = Q1 − Q2 − − − − − − − − − − − (6)
dt
H − H2
Q1 = 1 − − − − − − − − − − − − − (7 )
R1
H2
Q2 = − − − − − − − − − − − − − − − −(8)
R2
where
Q = q − qs , Q1 = q1 − q1s , Q2 = q2 − q2 s
H1 = h1 − h1s , and H 2 = h2 − h2 s
Taking the transform of equations (5) to (8)
Q( s ) − Q1 ( s) = A1 sH1 ( s) − − − − − − − − − (9)
 Q1 ( s) − Q2 ( s) = A2 sH 2 ( s) − − − − − − − − − (10)
R1Q1 ( s) = H 1 ( s) − H 2 ( s) − − − − − − − − − −(11)
R2 Q2 ( s) = H 2 ( s) − − − − − − − − − − − − − − − (12)
Combine these equations to eliminate Q1, Q2, H1

This equation represents the theoretical response of the system in terms of variation of
liquid level in tank 2 with respect to time for a step input of magnitude M in the inlet
flow of tank 1. There are two ways of solving Eq (15).
1. First obtain the time constant τ1 & τ2 of each tank (adjust the resistances R1 and R2
such that τ1 = τ2 = τ) . The response in terms of H2 (t) can be obtained from Eq. 15
and compared with the observed response.
2. The observed response, H2 (t) vs t data can be fitted to Eq. 15 and τ and R2
obtained by non–linear regression techniques. Compare the values of τ and R2 with
the observed values of τ & R2.Discrepancy in the response or in values of time
constant should be accounted for. One of the principal reasons for discrepancy
could be the assumption of linear resistance R1 & R2 that may not be valid.

3. Experimental Procedure:
1. Adjust valves V1& V2 (partially open).
2. Switch on the power supply and the pump; adjust the inlet flow rate, q, to the
Tank-1 at around 40 LPH by completely opening the feed valve, Vf and adjusting
the flow by the control valve of the rotameter.
 3. The liquid level in the two tanks will rise. Wait till steady state is reached in each
tank.
4. Note the steady state liquid level (h1s and h2s) and the inlet steady state flow rate
(qs).
5. Now increase the inlet flow rate to tank one to say 50 LPH by suddenly opening
the rotameter needle valve in a ‘single go’ and maintain the flow rate at this
level . Simultaneously start the stopwatch and record the liquid level with time in
each tank (h, h2) till next steady state is reached. Note down the final steady
state value of h1, h2 and qfs .
6. Steps (3) to (5) can be repeated for another step change in feed flow rate without
changing the valve positions V1 & V2.
7. From the steady state values of qf (s), h1(s) and h2(s) obtain the values of τ1 and
τ2 and adjust R1 and R2 such that τ1 = τ2 = τ.

4. Observation &Calculations:

Plot steady state discharges qf(s) vs level h2(s).

5. Nomenclature:
A = Cross-sectional area of process tank (1), tank (2), m2
D1 = Inner diameter of Tank (1) & Tank (2), m
D2 = Outer diameter of down comer, m
f = Subscript of final state
h1 = Height of liquid in process tank (1) after step change attaining steady
state, m
h2 = Height of liquid in process tank (2) after step change attaining steady
state, m
H1s = Change in height (h1fs - h1S) in process tank (1) after step change
H2s = Change in height (h2fs – h2s) in process tank (2) after step change
h1s = Initial height in the process tank (1)
h2s = Initial height in the process tank (2)
h1fs = Final height in the process tank (1)
h2fs = Final height in the process tank (2)
qfs = Final steady state feed flow rate, LPH
qS = Initial steady state feed flow rate, LPH
QS = Initial flow rate for the process, m3/min
 R1 = Resistance given to the valve (1), min/m2
R2 = Resistance given to the valve (2), min/m2
M = Magnitude of step change, m3/min
τ = Time constant, min

6. References:

Donald R. Coughanowr, “Process System Analysis and Control”, 2nd ed., McGraw Hill,
NY, 1991, Page 83-86.
 EXPERIMENT NO. 3

TWO TANK NON- INTERACTING LIQUID-LEVEL SYSTEM

1. Aim:

To obtain experimentally the step response and to compare the experimental &
theoretical responses.

2. Theory:

A dynamic system is the one in which there is some varying amount of accumulation of
conserved quantities with time. Consider a liquid level system shown below in which two
tanks are arranged in series such that the response of first tank does not depend on the
conditions in the second tank. Such a system is said to be a non- interacting system and is
a lumped parameter system.

Tank 1 =A1 Tank 2= A2

This arrangement is such that the flow through the resistance R1 depends only on h1. Such
a system is known as a non- interacting system.
A mass balance on Tank 1 gives :
dh1
q − q1 = A1 − − − − − − − − − (1)
dt
A mass balance on Tank 2 gives :
 dh2
q1 − q2 = A2 − − − − − − − − − (2)
dt
The flow-head relationships for linear resistances are given by

h1
q1 = − − − − − − − − − − (3)
R1
h2
q2 = − − − − − − − − − − (4)
R2
Combining eqns (1) and (3) and using deviation variables

Q1 ( s) 1
= − − − − − − − −(5)
Q( s )  1 s + 1
where Q1 = q1 − q1s , Q = q −q s , and  1 = R1 A1
Similarly combining eqs.(2) and (4)

H 2 (s) R2
= − − − − − − − (6)
Q1 ( s)  2 s + 1
where H2=h2-h2s and τ2=A2R2
From eqns.(5) and (6)
H 2 ( s) R2
= − − − − − −(7 )
Q( s ) ( 1 s + 1)( 2 s + 1)
On taking the inverse Laplace transform, we obtain the theoretical response of the system as:
   1 −t /  1 1 −t /  2 
H 2 (t ) = MR2 1 − 1 2  e − e  − − − − − (8)
 1 − 2 
 2  1 
In case τ1 =τ 2 =τ, then
 t 
H 2 (t ) = MR2 1 − e−t /  − e−t /   − − − − − − − (9)
  

Equation (8) represents the theoretical response of a non-interacting two tank liquid level system
to a step change of magnitude (M) in the feed rate to tank (1).

There are two ways of solving Eq (8).

1. First obtain the time constant τ1 & τ2 of each tank and the resistance R2 (τ2= A2R2), the
response in terms of H2(t) can be obtained from Eq. 8 and compared with the experimental
response.
2. The observed response, H2(t) vs t data can be fitted to Eq. 8 and τ1, τ2 and R2 evaluated by
non –linear regression techniques. Compare the values of τ1, τ2 and R2 with the
observed values of τ1, τ2 & R2.
Discrepancy in the response or in values of time constants should be accounted for. One of the
principal reasons for discrepancy could be the assumption of linear resistance R1 & R2 that may
not be valid.
3. Experimental Procedure:
1. Adjust valves V1& V2 (partially open).
2. Switch ON the power supply and the pump; adjust the inlet flow rate q, to the tank
(1) at around 40 LPH by completely opening the feed valve, Vf and adjusting the flow by
the control valve of the rotameter.
3. The liquid level in the two tanks will rise. Wait till steady state is reached in each tank.
4. Note the steady state liquid level (h1s and h2s) and the inlet steady state (flow rate q(s)).
5. Now increase the inlet flow rate to tank one to say 50 LPH by suddenly opening the
rotameter needle valve in ‘Single go’ and maintain the flow rate at this level
Simultaneously start the stopwatch and record the liquid level with time in each tank (h, ,
h2) till next steady state is reached. Note down the final steady state value of h1 , h2 and q
f(s).

6. Steps (3) to (5) can be repeated for another step change in feed flow rate without changing
the valve positions V1 & V2.
7. From the steady state values of qf (s), h1(s) and h2(s) obtain the values of τ1 and τ2 and
adjust R1 and R2such that τ1 = τ2 = τ.
4. Observation &Calculations:
DATA:
D1 = -------- m D2 = ---------- m
Observation:
Initial steady state value (i.e. before step change is introduced)
t = 0
qs =--------- LPH
h1s = ---------- m
h2s = ---------- m
Final steady state values after the step change
qfs = --------- LPH
h1fs = -------- m
h2fs = -------- m

Observation Table:

Plot steady state discharges qf(s) vs level h2(s).

Calculations:

Theoretical response
 H 2 (t )    1 2  1 −t /  1 1 −t /  2  
  = 1 −  e − e  − − − − − (10 )
 MR2  th   1 −  2  2 1 
5. Nomenclature:
A = Cross-sectional area of process tank (1) & tank (2), m2
D1 = Inner diameter of Tank (1) & Tank (2), m
D2 = Outer diameter of down comer, m
f = Subscript of final state
h1 = Height of liquid in process tank (1) after step change attaining steady
state, m
h2 = Height of liquid in process tank (2) after step change attaining steady
state, m
H1s = Change in height (h1 - h1S) in process tank (1) after step change
H2s = Change in height (h2 – h2s) in process tank (2) after step change
h1s = Initial height in the process tank (1)
h2s = Initial height in the process tank (2)
h1fs = Final height in the process tank (1)
h2fs = Final height in the process tank (2)
qfs = Final steady state feed flow rate, LPH
qs = Initial steady state feed flow rate, LPH
QS = Initial flow rate for the process, m3/min
R1 = Resistance given to the valve (1), min/m2
R2 = Resistance given to the valve (2), min/m2
M = Magnitude of step change, m3/min
S = Subscript of steady state
τ = Time constant, min

6. Reverence:

1. Donald R. Coughanowr, “Process System Analysis and Control”, 2nd ed., McGraw
Hill, NY, 1991, Page 80-82.