See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309384158

Design and Fabrication of a turbine Flow meter

Conference Paper · December 2015

CITATIONS READS
5 9,964

4 authors, including:

Mohammed Liaket Ali Mohammed Badsha Alamgir

German Aerospace Center (DLR) North South University
13 PUBLICATIONS 40 CITATIONS 2 PUBLICATIONS 7 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Mohammed Liaket Ali on 07 January 2022.

The user has requested enhancement of the downloaded file.

 Design and Fabrication of a Turbine Flow Meter
1
Mohammed Liaket Ali, 2 Riaduzzaman Ridoy, 3Utsha Barua, and
4
Mohammed Badsha Alamgir
In the modern age of industrialization measuring the flow of fluid in an industry is
a major issue for controlling the production process. Along with the flow of fluid,
temperature is also a matter of concern for quality maintenance of the product.
Keeping those things in mind this project is so designed and fabricated to record
the flow rate and temperature of either a liquid (such as oil or water) or a gas
(such as natural gas) simultaneously and precisely. And the output is shown in a
liquid Crystal Display (LCD). Here a fluid (liquid) is passed through a turbine and
rotated it at a rate that is proportional to the rate of flow of the fluid. An
opto-sensor will then pick up the turbine rotation and send a pulse signal to the
Arduino. A temperature sensor is also used to measure the temperature of the
flowing fluid. We calibrated the Sensors with great care that’s why a highly
precise results of not more than 3% error were recorded and our data was
measured several times. Every time this instrumental setup gave us same stable
data with negligible fluctuations.

Field of Interest: Fluid mechanics, Renewable Energy, Industrial Measurement.

1. Introduction

Measuring the flow of fluid is a basic need in many industrial plants. In some operations, the
ability to conduct accurate flow measurements is so important that it can make a difference
between making a profit and taking a loss. In other cases, inaccurate flow measurements or
failure to take measurements can cause severe (or even disastrous) results. In our
undergraduate study, we did an experiment by measuring the flow and temperature manually,
which gave us various data. As a result, desired output varied on a large scale. After that
experiment, we planned to design a system that would lead us to a fixed output every time and
we also planned an automated system with a simple design. The turbine flow meter has many
benefits. One example would be the oil industry. If anyone gets a clear idea of exactly how
much oil are processing per hour he can estimate the productivity of the plant. That is not the
only benefit of this meter. It is also an effective way to keep industrial piping system safe. If the
meters are kept calibrated and checked frequently, the authority will be notified before
something goes wrong with system, possibly blowing a pipe. In industry, it is common to have a
piping system for more than one fluids, for example, natural gas and crude oil. This flow meter
could easily be inserted into the system and can be monitored very efficiently. Turbine flow
________________________
1
Mohammed Liaket Ali, Department of Mechanical Engineering, Chittagong University of Engineering and
Technology, Chittagong, Bangladesh, Email: [email protected]
2
Riaduzzaman Ridoy, Department of Mechanical Engineering, Chittagong University of Engineering and
Technology, Chittagong, Bangladesh, Email: [email protected]
3
Utsha Barua, Department of Mechanical Engineering, Chittagong University of Engineering and Technology,
Chittagong, Bangladesh, Email: [email protected]
4
Mohammed Badsha Alamgir, Department of Computer Science and Engineering, Chittagong University of
Engineering and Technology, Chittagong, Bangladesh,
Email: [email protected]
meters have been used to measure fluid flow for a wide variety of applications. One type of
conventional flow meter utilizes an axial flow turbine that is disposed within a cylindrical bore.
As fluid passes through the bore, it impinges upon the vanes of the turbine and causes rotation

1
which is proportional to the rates fluid flow. But in this project a Z axis turbine will be used,
which also exhibit similar kind of operation during water impinges.

When a liquid moving through the flow meter at moderate rate, the rate of rotation is a fairly
linear function of the fluid flow rate. Some conventional flow meters can be operated over a
limited or narrow range exhibit one percent or better accuracy. Many electrical hydraulic system
applications, product packaging applications involving the dispensing of predetermined
amounts of liquid such as the filling of beverage bottles, petroleum distribution such as retail
sales of gasoline and scientific instrument applications would greatly benefit from turbine flow
meter which exhibit a high degree of accuracy over a broad fluid range of two or more orders of
magnitude. Normally the problem of inaccuracy is most acute in the extremely low flow portions
of fluid portions of fluid flow range of the flow meter. Fluid flow measurement can be divided
into several types; each type requires specific considerations of such factors as accuracy
requirements, cost considerations, and use of the flow information to obtain the required end
results. Normally the flow meter is measure flow indirectly by measuring a related property
such as a differential pressure across a flow restriction or a fluid velocity in a pipe.

In this research, we tried to fabricate a turbine flow meter so as to get digital readings along
with a temperature readings. Some previous studies were conducted in which the flow and
temperature were measured manually. So, in this respect, our design is not relevant to others.

The paper is organized as follows. Section 1 presents the introductory thought and motivation
for this experiment. Section 2 describes the literature review on this topic of flow measuring
devices. Section 3 explains the methodology for this experiment and basic outlines. Section 4
presents the step by step outlooks of the experiment about how it works. Section 5 describes
physically how the mechanical and Electrical setup has been built and how they have been
interfaced. Section 6 comprises of series of calibration and the measuring demonstration.
Section 7 shows a series of experimental results that demonstrates the effectiveness of this
experimental device and the simulated results. Section 8 describes the final findings, limitation
and the future aspects of this experiment.

2. Literature Review

In the beginning of 5000 BC measurement of flow was introduced to control water distribution
through the ancient aqueducts of the early Sumerian civilization from the Tigris and Euphrates
rivers. Such systems were based on volume per time operated by diverting flow in a single
direction from dawn to noon and diverted it in another way from noon to dusk. Bernoulli (1738),
he outlined the basic principles of the conservation of energy for flow, In Hydrodynamics. He
focused on the reciprocal relation of kinetic energy and static energy which is the basis of
differential pressure flow measurement. Woltman (1790), a German engineer developed the
first vane type turbine meter to measure flow velocity in river and canals. But those meters
were not available in industrial market until World War 2. Emergence of this war created a
desire need for equipment of higher accuracy, greater versatility and quicker response under
critical conditions. Faraday (1831) who tried to use his electromagnetic induction laws to
measure flow. To measure the water flow of river Thames, Faraday used two metal electrodes
connected to a galvanometer from Waterloo Bridge. Later, Thurlemann (1941) worked on the
method of electrical velocity measurement for liquid and after that Shercliff (1962) established
firm principles of magnetic flow meters in his book “The theory of electromagnetic flow
measurement”. Just after three years of Faraday’s original experiment, Coriolis (1835) made
the discovery of coriolis effect which helped to construct the highly accurate direct

2
measurement mass flow coriolis meter later. In 1963 ultrasonic flow meters were first used in
commercial application. Before that positive displacement flow meters were common in form.
But due to misuse they gained bad reputation until further improvements which help them to
become world’s best. Now the market of ultrasonic flow meter is the quickest growing among
all flow meter. Yokogawa (1970) developed vortex flow meter which was first introduced by
Eastech (1969). Those flowmeters depend on the creation of vortices generated by a bulk
object placed inside flow stream. Although vortex meter are especially for steam measurement
but for gas and liquid measurement they can also be used. These have higher accuracy than
positive displacement based flow meter. Sierra (1997) Instruments introduced new
multivariable vortex flow meter which was capable to measure more than one process variable
and these meters were designed specially including pressure transmitter and a temperature
sensor. Zierke et al (1993) introduced a high capacity, low noise slip ring in a rotating frame to
measure flow where an optical shaft encoder provided rotor angular position and speed.
Kolhare and Thorat (2013) used magnet on one arm of turbine and a Hall Effect device on the
outer side of the pipe to measure flow of water in solar water heater. Garmabdari et al (2014)
used Hall Effect sensor based rotary encoder to measure algorithm and for monitoring water
flow rate. They computed the flowrate by dividing the number of counted pulses (C. P.) over the
actual time (Ta) converted to minute which give an output in round per minute (RPM).

3. Methodology

For this project, a standard ARDUINO is used for programming. From various options
ARDUINO is chosen for it’s simple design and lower power requirement to function properly. A
turbine with a stainless steel shaft is mounted in a rectangular box. A wheel encoder will be
attached with the shaft to count the rotation. The sensor pulses are then sent to the ARDUINO
to calculate and store data. The temperature sensor will also take a stream of data that will be
sent to the ARDUINO and displayed. The ARDUINO will then use the measurements to
calculate the rate of flow & temperature of the material. A digital display will be used as the
user interface, showing the data along with simple buttons.

4. Flow Diagram

Fluid or water will flow from the inlet at a high velocity. The high velocity of water will impact on
the turbine. According to the jet principle it will rotate for its shape. As the shaft rotates, the
turbine encoder will also rotate as it is attached with the shaft. Then the optical sensor will pick
up its output & send signal to the Arduino. Temperature sensor will also send data to the
Arduino. The microcontroller will process those data and show the flow rate and the
temperature of the flowing fluid.

3
 Figure 1: Flow Diagram

5. Structural Design

First of all a rotor is made just like a turbine that involves cutting sheet metal with a proper
design for bending and a compatible shape for better use. A mild steel rod serves the purpose
of the shaft of the turbine. The bearing setup will be such that, the shaft can rotates freely along
its axis.
There are two basic parts in the overall structure.
A. Mechanical setup
B. Electrical setup

A. Mechanical set up
Mechanical set up for this flow meter is based on some of the separate parts
1. Rotors
2. Shafts
3. Bearing
4. Reducer & socket

4
 Figure 2: Blade & Shaft Assembly with Optical Encoder

The rotor shaft assembly or the turbine will be then placed in a box. The box will be made with
sufficient dimension to make no trouble for the turbine to rotate and to make the fluid flow
imping on the turbine blades effectively. The making of the turbine involves cutting of sheet
metal with proper design for bending and give it a twist. Four twisted blades are used. Each
blade is having a height of 1.5 inches. Two circular disks having a diameter of 3 inches are
used to attach the blades in equal spacing. A mild steel rod is taken as the shaft of the turbine.
As the bearing is with inner diameter of 1cm, the shaft is designed to meet the requirement by
turning operation. The assembly of rotor and shaft is given in fig 03. The rotor shaft assembly
or the turbine is now placed in a box. The box is made with sufficient dimension to make no
trouble for the turbine to rotate and to make sure the water flow can impinging on the turbine
blades effectively. The dimension of the rectangular box is 3.7 inch by 3 inch.

Figure 3: Shaft and Turbine

Whereas the diameter and height of the turbine are 3 inches and 2 inch respectively. The
length of the rectangular box is 12 inches. A reducer is used at the incoming side of the meter
just to ensure the convenient flow from the supply line. The meter is facilitated with the option
of being attached in a line of piping system via a incoming reducer and outgoing socket. Two
bearing is provided for confirming the free movement or rotation of the shaft. Inner diameter of
the bearing is 1cm. To hold the bearing, two bearing cap is provided so that bearing can be
attached with rectangular. The bearing caps are attached with the rectangular box by drilling
and providing bolt and nut system.

Figure 4: Body Structure of the Meter

5
B. Electrical Set Up
The main parts of the electrical setup are-Optical encoder, ARDUINO, Data and display Cable.
Among them worth mentioning is the optical Encoder.

Optical Encoder

Encoders can electronically monitor the position of a rotating shaft. Our Absolute Encoders are
electro-mechanical devices that are useful feedback elements in closed-loop control systems
[in fig 3.7]. They provide position control in packaging, robotics, pick and place, lead/ball screw,
rotary table positioning and component insertion applications. Our Incremental Optical
Encoders provide low cost, smaller physical size, high frequency and high resolution. Our
accessories help you easily install and efficiently use our encoders. The two main parts are the
encoder wheel and the opto-sensor. The encoder wheel has evenly spaced holes around the
diameter and in this case is made from etched brass sheet. The opto-sensor is a dual channel
TCUT1200. As the encoder wheel passes over the opto-sensor.

Figure 5: Optical Encoder

the transitions from a gap to a non-gap are detected. The signals from the sensor can be
decoded to produce a count of the number of transitions. The count increments for one
direction and decrements for the other.
Overall test setup is shown in the Figure 5.

Figure 6: Test Setup

6. Calibration and Measurement

6
For getting accurate data it is very important to calibrate the sensors. The success of the whole
project depends on calibration. There are several calibration techniques.

a. Positive displacement
b. Time weight
c. Field power
d. Old stand by
e. Bucket and stop watch
Among the prescribed methods we preferred bucket and stop watch technique for our
convenience.

Bucket and Stopwatch Calibration

The simplest way to measure the volumetric flow is to measure how long it takes to fill up a
bucket of known weight. Converting the mass into volume using the density of the water
volumetric flow can be measured. A bucket is taken for the purpose. The stopwatch is started
when the flow goes to the output with a constant rpm of the shaft and stops after the bucket is
filled up. The bucket weight is then measured by the weight measuring device which can be
converted into cubic inches. With the help of the dimension, one can know the volumetric flow
rate by dividing the volume and the time taken to reach that volume. Here the calibration and
reading of rpm from the encoder is taken simultaneously. For a given rpm, the device is
attached with the hose and a bucket is positioned underneath the device. So we can have the
measured flow. LCD or laptop display gives us the flow rate for a constant rpm of the shaft and
stop watch gives us time reading. Rpm can be calculated from the encoder disc dimension and
structure.

7. Tests and Results

Our flow meter exhibits a satisfactory degree of accuracy over a broad fluid range of
magnitude. Normally, the problem of inaccuracy is most acute in the extreme high and low flow
portion of the flow range. The recorded data are presented below.

Table 1: Flow Rate & Rpm Reading

Observation Time of Mass flow Measured Rpm of the Actual flow rate
observation rate(kg/sec) flow corresponding from the
(min) rate(in3/sec) flow rate calibration
meter(in3/sec)
1. 5.02 1.46 120.51 81 123.33
2. 5.32 1.50 123.81 83 127.11
3. 5.14 1.52 125.46 87 129.2
4. 5.19 1.55 127.94 90 128.53
5. 5.24 1.56 128.76 96 130.1
6. 5.09 1.57 129.59 98 130.5

Table 2: Temperature Reading and Error Estimation

Observation Measured Actual Temperature(°C) Error (%)

Temperature(°C)
7
 01 38 38.9 2.31
02 36 36.5 1.37
03 41 41.8 1.91
04 40 40.7 1.72

Table 3: Error Estimation of Flow rate

Measured flow rate(cubic Actual flow rate from % Error

inches per second) calibration meter(cubic
inches per second)

120.51 123.33 2.29

123.81 127.11 2.60
125.46 129.2 2.89
127.94 128.53 0.46
128.76 130.1 1.03
129.59 130.5 0.68

7.1 Linearity Test

As the flow rate is proportional to the rpm of the shaft that’s why their graph is linear. In the
graph below we also plotted the measured flow rate versus rpm and compared it with the
actual linear one. Here we see that the experimental graph is almost linear and very close to
the actual graph.

Figure 7: Linearity Graph

7.2. Software Simulation

The circuitry portion is simulated with Proteus 8 professional. The codes that we developed in
the Arduino IDM works fine. It is compiled without any warning. We also debugged the code.
Every parts of the circuit acts as we expected. Here is the simulated result of the whole system.

8
 Figure 8: Simulation Result

During the simulation, the optical encoder continuously generated a pulse signal as the
encoder wheel rotates. We used a digital oscilloscope to detect the pulse signal. Here is the
simulated pulse signal result.

Figure 9: Pulse Generated by the Optical Encoder

8. Conclusion

From the result, we can see that the relationship between the measured rpm and measured
flow is almost linear. There is some error in the measured data comparing with the actual data.
It’s due to the water leakage from the bearing joint. There is some energy loss to overcome the
friction. Another major problem is that fluid could not impact the turbine with the same accurate
position. As we calculated flow rate and temperature from this simple setup and no other
design have not invented yet, this setup can be used universally. There are some limitations in

9
this project. The incoming fluid should have enough speed to rotate the turbine. So this flow
meter can only be used in a high-speed fluid supply line. Moreover, the sensor that we used is
rated for -55 ͦ C to 150 ͦ C range. That’s why this meter is not compatible with extreme condition.

References

9. References

Baker, R. C., 2000, ‘Flow Measurement Handbook: Industrial Designs, Operating Principles,
Performance, and Applications’, 2nd Edition, Cambridge University Press.

Castorina, J. S., Naval Ship Engineering Center, Philadelphia Division, Project A-771, Task
3950, 30 june,1968

Castaiier, L., Jimenez, V., Dom'nguez, M., Masana, F., and Rodriguez, A., 1997, ‘Design and
fabrication of a low cost water flow meter’, IEEE International Conference on
Solid-State Sensors and Actuators, Vol. 5, pp. 159-162. Digital Object Identifier:
10.1109/SENSOR.1997.613607
Crabtree, M. A., 2006 ‘Industrial Flow Measurement’,The University of Huddersfield,
England

Dong, G. Q. F., 2006, ‘Study on wide range turbine flow meter’, Proceedings of the Fifth
International Conference on Machine Learning and Cybernetics IEEE,
pp. 775- 778

Enggcyclopedia, ‘Turbine Flow meters’, 22.08.2015, 12:45 PM. Available

<http://www.enggcyclopedia.com/2012/01/turbineflow-meters/>

Gatabi, J. R., Forouzbakhsh, F., Darkhaneh, H., Rezanejad, Z., Janipour, M., Gatabi, I. R.,2010
Auxillary Fluid Flow Meter, European Journal of Scientific Research, Vol. 42, pp.
84-92

Garmabdari, R., Shafie, S., Garmabdari, A.,Jaafar, H.,Aram, A. K., 2014, Labview Based Flow
Rate Monitoring and Measurement Algorithm for Rotary Encoder, Special Issue
on SAES 2013, Journal of Engineering Science and Technology, Vol. 9, pp. 66-
75

Han, T. H., Myaing, O. M., 2011, ‘Design and Construction of Microcontroller-Based Water
Flow Control System’, International Conference on Circuits, System and
Simulation, Volume 7, pp. 304-309

Kolhare, N. R.,Thorat, P. R., 2013, An Approach of Flow Measurement In Solar Water Heater
Using Turbine Flow Meter, International Journal of Engineering Research &

10
View publication stats

Technology (IJERT), Vol. 2, pp. 1-4

Miller, R. W., 1989 ‘Flow Measurement Engineering Handbook’, Second Edition, McGraw-Hill
Publishing Company

Spink, L. K., 1978, Principles and Practice of Flow Meter Engineering, Ninth Edition, The
Foxboro Company, Foxboro, Massachusetts. USA

Shercliff, J., A., 1962, The Theory of Electromagnetic Flow-Measurement, Cambridge

University Press. Vol. 12,Pp. 10-22

Upp, E. L., LaNasa, P. J. 2002, Fluid Flow Measurement: A Practical Guide to Accurate
Flow Measurement, 2nd Edition, Gulf Professional Publishing

Zierke, W. C., Straka W. A., and Taylor, P. D., 1995, An Experimental Investigation of the Flow
Through an Axial-flow Pump, Journal of Fluids Engineering, ASME, Volume 117,
Issue 3, pp. 485-490, Digital Object Identifier: 10.1115/1.2817288

11