Automated Industrial Load Measurement System - AU Journal

AU J.T. 10(1): 23-28 (Jul. 2006)
Automated Industrial Load Measurement System
Seshanna Panthala, Nashtara Islam 1 , Sajed Ahmed Habib 2
Faculty of Engineering, Assumption University
Bangkok, Thailand
Abstract
Load measurement is an integral part of many process industries. Thus it is
essential to have a competent system for this measurement purpose. This paper deals
with the design and fabrication of a Personal Computer (PC) based, industrial load
measurement system. The main sensor used in this case is a commercially available
load cell and custom-made signal conditioning hardware and a data acquisition system.
The distinct advantage of this system is its cost effectiveness when compared with
conventional DAQ-based measurement systems frequently employed by industry.
Design, construction and testing of an innovative load measurement system are
discussed in this paper, along with some suggestions for further improvement.
Keywords: Strain gauge, load measurement, analog-to-digital converter,
assembly language programming, personal computer.
Introduction
Measuring load is an important and
essential part of many industrial and
commercial operations. It is crucial to have
accurate measurements of load, as small errors,
occurring repeatedly, can lead to substantial
loss of revenue.
One very common way to acquire load
measurements is to use the load cell, which is
quite effective and accurate; even though the
idea is relatively simple (Johnson 2003). The
load cell provides an output voltage depending
on the load placed on it. This cell is one of the
most important applications of a strain gauge
(SG) in an industrial environment.
The main theme of this research work is
to display how much load is placed on the cell.
Figure 1 gives the overall schematic idea of the
load measurement system.
The load cell output is a differential
voltage, which is dependent on the transfer
function of the cell itself. According to this
transfer function, the load cell provides a
certain voltage for a certain load level. The
output of the load cell is then amplified using a
differential instrumentation amplifier. Then the
instrumentation amplifier output signal is fed to
an Analog-to-digital converter (ADC) that will
provide digital output. Finally, this digital
information is sent to a Personal Computer
(PC) directly via the standard parallel port.
Sending information to a PC using a
parallel port is easily done, with the advantage
that there is no need for special circuitry and
algorithms to make the necessary changes to
achieve this operation. The standard parallel
port can be configured in several ways- to send
data, to receive data, and both i.e., bidirectional
operation (Hall 1992). In this case
data is being sent from an external device to the
PC. 1 2 The display area is the monitor of the PC,
which shows a screen with a prompt to place a
load. Then the load value is displayed in
kilograms. A program has been written to
manipulate the incoming digital information
from the ADC to show the correct load. The
software program to display load data can be
1
Mechatronics Research Center, Loughborough
University, Leicestershire, United Kingdom
2
Department of Mechanical and Manufacturing
Engineering, University of Calgary, Calgary, Canada

Fig. 1 Theoretical approach to the measurement idea
written in various languages (e.g., C, C++,
FORTRAN, Assembly etc.). In this case,
assembly language has been chosen for its
capability of having greater control over the
machine, i.e., the PC (Abel 2001) and for the
familiarity of the authors with this language.
Hardware
The main hardware components of the
system are discussed in considerable detail in
this section.
Strain Gauge (SG)
As discussed in the introductory part of
this paper, load cell is the main sensor used for
the measurement system. The load cell is
primarily made up of strain gauges.
V ref
R
R
R D
Fig. 2. Construction of a strain gauge (Johnson
2003)
Construction of strain gauge is shown in
Fig.2. It basically consists of resistive elements
in a Wheatstone bridge configuration.
R
The nominal values of the resistances are
equal under no-load conditions. Thus the
voltage output from the circuit is zero in this
setup. The resistance of R D changes in a linear
manner with the force acting on it.
The sensitivity of this bridge to strain can
be found by considering the equation for bridge
offset voltage. Suppose R 1 = R 2 = R D = R,
which is the nominal (unstrained) gauge
resistance. Then the active strain gauge
resistance will be given by (Johnson 2003):
R A = R (1+ΔR/R)
And the bridge off-null voltage will be
given by:
ΔV = V S [R D /(R D +R 1 ) – R A /(R A +R 2 )]
Finally, the expression of output voltage
is given in terms of strain as below:
ΔV = - (V S /4) GF (Δl/l)
This voltage is the output, which is fed to
an instrumentation amplifier for signal
amplification.
Instrumentation Operational Amplifier
There are many instances in
measurement and control systems in which the
difference between two voltages needs to be
24

conditioned. A good example is the
Wheatstone bridge where the offset voltage ΔV
= V a – V b is the quantity of interest.
Since the load cell is working with the
same principle as the Wheatstone bridge, the
ideal approach was to feed the load cell voltage
output to a differential instrumentation
amplifier.
An ideal differential amplifier will
provide output voltage with respect to ground
that is some gain times the difference between
two input voltages (Boylestad and Nashelsky
1999).
V out = G (V a –V b )
where, G is the differential gain and both V a
and V b are voltages with respect to ground.
Such an amplifier provides useful operation in
control and measurement.
There are a number of op-amp circuits
for a differential amplifier. The circuit uses two
pairs of matched resistors. But if the resistors
are not well matched, the Common Mode
Rejection (CMR) will be poor. Normally the
input impedance for such circuits is not very
high and it is not the same for two inputs.
For this reason, voltage followers are
often used on the input to provide high input
impedance. Such a device is called a
differential instrumentation amplifier. It is just
a differential amplifier with high input
impedance and low output impedance. Many
IC manufacturers package instrumentation
amplifiers in a single IC.
Instrumentation amplifier IC INA125P
from Burr Brown was used for this purpose, as
it offers all the aforementioned advantages.
V X = V ref (a 1 2 -1 + a 2 2 -2 + a 3 2 -3 + …….+ a n 2 -n )
where, V ref is the ADC reference voltage, V X is
the input analog voltage and n is the number of
bits.
In this case, the 8-bit ADC 0804 IC has
been used. This 8 bit digital information will be
sent to a PC for conversion to weight or a load
value using the appropriate software.
It is noteworthy that there is an inherent
uncertainly in the input voltage producing a
given ADC output, and this uncertainty is
given by:
ΔV = V ref 2 -n
Therefore, it is very important to take this
into account when using an ADC in industrial
instrumentation and control systems, and in
defense sector applications, where 100%
accuracy is required.
PC Parallel Port Interface
The parallel port is the most commonly
used port for interfacing applications involving
data transfer to and from the PC (Hall 1992).
This port allows the input of up to 9 bits or the
output of 12 bits at any given time, thus
requiring minimal external circuitry to
implement many simpler tasks.
Analog-to-digital Converter (ADC)
An analog to digital converter or ADC is
the next stage after the instrumentation
amplifier. An ADC takes analog voltage as an
input and provides a digital output. This digital
information is in binary format consisting of 0s
and 1s only. The ADC finds a fractional binary
number that gives the closest approximation to
the fraction formed by the input voltage and
reference. The following is the conversion
equation (Boylestad and Nashelsky 1999):
Fig. 3. LPT port connection diagram
The port is composed of 4 control lines, 5
status lines and 8 data lines. It is on the back of
a PC as a D-Type 25 Pin female connector.
25

The complete hardware interface diagram
is shown above in figure 5. A 74LS244 line
driver IC has been additionally used between
the LPT port of the PC and the ADC. This
setup has been implemented to electronically
isolate the two devices from any disturbances
in the connection (e.g., short circuit).
Although() not a necessity, it is always a good
practice to include such a buffer in the system.
Fig. 4. Complete hardware interface diagram
Software
The main tasks of the software were to –
1. Drive the parallel port for obtaining data
from the ADC.
2. Convert the digital data to an analog form
to provide the measurement value.
3. Provide an user interface for running the
program and manipulating the
measurement information.
To accomplish these tasks, an algorithm
was devised according to the flow chart shown
in figure 5.
Each process within the measurement
setup (e.g., convert ASCII to Decimal)
comprises a complex algorithm. Due to space
constraints, they are not discussed in detail in
this paper.
Fig. 5. Flow chart for load measurement system
26

Fig 6
Screen shot of the user interface
The user interface of the measurement
system looks like the one in figure 6. A few
options have been provided for the user of the
system.
The first option can be selected if the
user wants to perform only a single
measurement using the system.
Option 2 can be used to carry out a
succession of measurements with the help of
the system. The cumulative measurement value
will be displayed in the console when the user
decides to end the measurement session.
The third option allows setting a predetermined
limit for the measurements to be
carried out. If the user decides to measure up to
a certain weight (e.g., 1,000 kg), then the
program will warn the user, both with the help
of a text message, as well as an alarm sound
once that limit has been reached. This function
is extremely useful in packaging industries,
where each bag or sack has to be filled up to a
certain weight.
Discussion
A common method of acquiring load
measurement data from the load cell is using a
microcontroller. This setup ensures that the
system footprint is small and portable. The
acquired data is usually displayed in an LCD
display.
While this is a fairly straightforward way
of achieving the goal of load measurement, it is
not free of certain drawbacks. Firstly, the
measured data cannot be stored in a convenient
location (e.g., a hard drive) for future referral.
Secondly, the system can hardly be made an
interactive one, as demonstrated by the system
in this paper. And, finally, a microcontrollerbased
system cannot be easily reconfigured by
the user.
As mentioned earlier, various DAQ
products are being used in industries for
measurement purposes. The hardware can be
connected via PCI slots or the USB port of a
PC. The corresponding software is supplied
with the product or can be downloaded from
the internet. They are excellent in terms of ease
of use and robustness. However, cost analysis
(National Instruments 2006) shows that these
attributes come at a very high cost and can be
beyond the reach of small scale businesses.
This proposed measurement scheme can
be implemented in almost all industrial PCs
without the need for expensive hardware.
Although it doesn’t exactly offer the portability
of a microcontroller-based system, it can
certainly overcome all the aforesaid
constraints.
27

Further Work
A complete design of an automated load
measurement setup has been proposed,
describing both hardware and software.
However, it is possible to enhance the system
further. The following suggestions are made for
the improvement of the system–
1. A wireless method can be interfaced with
the measurement setup to realize remote
measurement schemes. Readily available
Bluetooth or ZigBee transceiver operating
in the ISM band can be used for this
purpose (Egan 2005).
2. Analog-to-digital converters with higher
resolution (e.g., 16 bit devices) can be used
to increase the accuracy of the
measurement system.
3. A Graphical User Interface (GUI) can be
developed to further enhance user
experience.
Conclusion
The research work discussed in this paper
has contributed to the development of a PCbased
load measurement system. A software
user interface along with the signal
conditioning and data acquisition hardware has
been fabricated. The system has been
successfully trialed to obtain various load
measurements. A few suggestions have also
been drawn up for the improvement of the
system.
As the system is easily customizable, its
seamless integration is always possible in a
highly automated industry. Thus, there is a
great potential for this system to be used in
numerous industries, where load measurement
forms a part of the process line.
References
Abel, P. 2001. IBM® PC Assembly Language
and Programming. 5 th ed. Prentice-Hall,
Englewood Cliff, NJ, USA.
Boylestad, R.L., Nashelsky, L.. 1999.
Electronic Devices and Circuit Theory. 7 th
ed. Prentice-Hall, Englewood Cliff, NJ,
USA.
Egan, D. 2005. The Emergence of ZigBee. IEE
Comp. Control Engin. Mag.. April - May
2005, pp: 14 - 19.
Hall, D.V. 1992. Microprocessors and
Interfacing - Programming and Hardware.
2 nd ed.. Macmillan/McGraw-Hill,
Singapore:.
Johnson, C.D. 2003. Process Control
Instrumentation Technology. 7 th edi.
Prentice-Hall, Englewood Cliff, NJ, USA.
National Instruments, 2006. UK Price List.
http://www.ni.com/pdf/branches/price_list_
uk.pdf
28