DESIGN OF LEVEL CONTROL SYSTEM FOR

LIQUID FLOW

Submitted by:

Uzair iftikhar- F2017019011

Hassan Ali- F2017019020

Shiraz Khalid - F2017019018

Submitted to:
Dr. Shakeel Ahmed

Bachelor of Science in Electrical Engineering

January 2021
Page 0 of 13
Problem Statement
We have to design our system in such a way that when the output of the level sensor

V  0.8h  0.4
(having transfer function h ) multiplied with the designed gain reaches the value of

6.xVolts it opens the output valve of the fluid tank. And similarly when the output of the level

sensor multiplied with the designed gain reaches the value of 6.yVolts it opens the output valve

of the fluid tank. It is given that we can only control the output of the tank whereas the input of

the tank cannot be controlled.

For our group the calculation for the values of x& yare shown in the table given below.

According to the given formulation the value of x is the most significant digit (MSD) and value

of y is the least significant digit (LSD) of the average of last two digits of the IDs of our group

members.

Student ID Last 2 digits Average x y

F2017019011 11
11  18  20
F2017019018 18  16.33 1 3
3
F2017019020 20

Thus from the above table we have the value of x and y required to find the exact voltage

value at which the relay closes and opens. So, for our group the relay closes at 6.1V and relay

opens at 4.3V.

V  4.3V
relay _ open

Page 1 of 13
 V  6.1V
relay _close

Part A: Find the value of amplifier gain, K, required to open the valve when the level

reaches 1.5 m.

To find the desired amplifier gain (K) required to meet the given specification, we first

V  0.8h  0.4
have to use the given sensor equation i.e. h to find the output voltage of the sensor

when the fluid level reaches to the height of 1.5 m. Putting this value of h we get

V  0.8*1.5  0.4
h

 V  1.6V
h

From the above calculations we have found the value of the output voltage produced by

the sensor when the fluid level reaches 1.5m. Using this sensor output voltage value at 1.5m fluid

level along with the voltage value at which relay closesand the relation between these two

voltages we get

Vr  KV
h

 6.1  K *1.6

 K  3.8125

Page 2 of 13
Part B:At what level does the valve close?

To find the level at which the output valve closes again, we will use the value of the

amplifier gain calculated in the last part along with the relay open equation given above. Also we

will use the amplifier input-output equation as given in the figure below

Page 3 of 13
 Using the above stated values and equation we will get the sensor output at which the

V  4.3V
relay opens i.e. relay _ open and the valve closes again

Vr  KV
h

4.3
V 
h 3.8125

V  1.13V
h
By putting the above obtained output voltage value of the sensor in the given transfer

function of the sensor we will get the level at which the valve will closes

V  0.8* h  0.4
h

1.13  0.8* h  0.4

1.13  0.4
h
0.8

 h  0.91m

Page 4 of 13
Part C:Suppose
Q1  5m3 / min , Q2  2m3 / min and Qout  9m3 / min (when open). What isthe

period of the level oscillation?

Form the previous part we know the valve opens when the level reaches 1.5 m and closes

when the level reaches 0.9 m. So, to calculate the time in which level decreases from 1.5m to

0.9m we will first calculatethe difference between the total inflows and outflow so,

Netoutflow = Q1 + Q2 - Qout

 Net outflow  2m3 / min

This is the flow rate at which the tank will drain when the output valve is opened. Next

we will find the total volume of fluid to be drained from the tank to decrease its level from 1.5m

to 0.9m by using following equation.

Volume to drained   * r 2 *(1.5  0.9)

 Volume to drained  7.54 m3

Now we have both the volume to be drained and net out flow so the time (t1) for drainage

can be calculated as

Volume to be drained
t1 
Net outflow

t1  3.77 min

Page 5 of 13
 Next we will calculate the time required to fill the tank from 0.9m to 1.5m. Because the

valve is closed when the tank is filling so there is no out flow from the tank. The net inflow will

be

Netoutflow = Q1 + Q2

 Net outflow  7m3 / min

This is the flow rate at which the tank will be increasing its level from 0.9m to 1.5m. The

total volume of fluid to be filled in the tank to increase its level from 0.9m to 1.5m is the same as

a volume that was drained from the tank to reach from 1.5m to 0.9m and was calculated

previously, so in this case we have

Volume to be filled  7.54 m3

Now we have both the volume to be filled and net out inflow so the time (t2) for filling

can be calculated as

Volume to be filled
t1 
Net inflow

t2  1.08 min

So the period of level oscillation will be

Oscillation period  t1  t2

 Oscillation period  4.84 min

Page 6 of 13
Part D:Design a circuit to interface the sensor with a 6-bit dual-slope analog-to-digital

converter(ADC) with a 10-V reference

Dual slope ADCs are popular for their high resolutions. The basic schematic of dual

slope ADC is shown in figure below. For the operation of dual slope ADC it is required that

signal should be unipolar and its polarity should be opposite to the reference signal. Before

starting the conversion the switch S2 shown in the figure is closed to discharge the capacitor.

After discharging the capacitor the switch S2 is opened and the switch S1 is connected to input

analog signal. Suppose that we have positive reference voltage so the input analog signal must be

negative for successful operation. The negative input analog signal will cause the current to flow

away from the integrator. The output of integratorv1 will increase forfixed duration T1 during this

time a counter is enabled to count a fixed number of clock pulses. For an N-bit converter the

counter counts 2N pulses.

Page 7 of 13
 After this time T1 the next phase of conversion begins and the counter is reset. In this

phase the switch S1 is connected to the reference voltage. Because the reference voltage is

positive it will reverse the direction of the current in the integrator causing the output voltage v1

of the integrator to decrease. During this phase the counter is again enabled to count the clock

pulses until the output voltage v1 of the integrator reaches zero volts at which the counter is

stopped. The count in the counter at the end of this stage is the digital equivalent of input analog

signal.

So for our case it is given that we have to use a reference of +10V for conversion of

sensor reading by dual ADC. Since from the given sensor equation we can see that it will

produce only positive voltages, so we have to pass the output of the sensor through an inverting

amplifier before feeding it to the ADC. The gain of the inverting amplifier can be set unity

because for our system the maximumoutput of the sensorat the level1.5m is 1.6V which is less

than the specified reference voltage i.e. 10V. For maximum resolution we set the gain of the

amplifier such that the maximumoutput of the sensor i.e. at the level 1.5m, after passing through

the inverting amplifier has a value less than -10V (negative of ref).The value of the maximum

allowed gain inverting amplifier is given below

max output required

10
max Inverting Amplifier gain 
1.6

max Inverting Amplifier gain  6.25

Page 8 of 13
 Using the standard available resistance we have designed the inverting amplifier of gain.

6.2 . The schematic of the designed inverting amplifier is given below

So the overall block diagram of the our system interfaced with ADC is shown below

Page 9 of 13
Sensor Output Gain for Relay Operation Relay

Inverting Amplifier Dual Slope ADC Optional Microcontroller

Page 10 of 13
 References

1. Sedra, Adel S., Dean Emeritus Adel S. Sedra, Kenneth Carless Smith, and Kenneth C.

Smith. Microelectronic circuits. New York: Oxford University Press, 1998.

Page 11 of 13