Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
ASSIGNMENT 1Instrument Communication and Networking Report on
development of an Automated test and Measurement System Submitted
by SOUMYAJYOTI SENGUPTA Matriculation no: S0909491
1
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
2
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
OBJECTIVE:To set up an automated test and measurement system for
determination of the cut-off frequency of a low pass filter. And
this task has to be completed in two different stages, as described
below: Stage 1 Setup an appropriate hardware and design a program
in Labview for determination of the cut-off frequency of a low pass
filter. Stage 2 Based on the program developed above determine the
cut-off frequency of ten different filters (identified by the
filter number or a serial number). The results should be written in
an automatically generated spreadsheet file by the application of
the program in Labview. HARDWARE PROVIDED: PXI with E series data
acquisition card GPIB instruments Signal generator or Function
Generator Oscilloscope Low pass filters Digital Multimeter SOFTWARE
USED: NI Labview
3
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
BASIC CONCEPTS:LOW PASS FILTER: Low pass filter passes a certain
low level frequencies and attenuates (reduces the amplitude) the
frequencies higher than the cut off frequency. Now cut off
frequency is the frequency boundary between pass band and stop band
of any filter. The cut-off frequency can be determined from the
frequency response curve which is as shown below. Now this it
should be noted that the frequency scale is logarithmic rather than
linear. The gain (in db) is given by, Gain (db) = 20* log10 Vout
Vin
For a low pass filter Vout is always lesser than the Vin. So the
logarithmic fraction of Vout and Vin will be negative. The ideal
cut-off frequency for a low pass filter is -3db. Therefore the
ratio of Vout/Vin is 0.707.Cut off frequency0 - 10 - 20 - 30 Gain
(db) - 40 - 50 - 3 db
Pass Band- 60 10 100
Stop band1000
Frequency (Hz)
4
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
TECHNIQUES USED FOR DETERMINATION OF CUT-OFF FREQUENCY: The
techniques used to determine the cut-off frequency of a low pass
filter are as follows: Successive Approximation, Curve fitting,
Interpolation etc. Apart from these methods cut-off frequency can
be determined manually by adjusting the amplitude and the frequency
of function generator and also by tuning the oscilloscope until the
ratio Vout/Vin becomes equal to or less than 0.707. SUCCESSIVE
APPROXIMATION: In this technique at the first loop, the initial
frequency and peak to peak input voltage are fed into the system
set up. Then the input and output voltages are measured and then
ratio and gain are calculated. When the ratio reaches the condition
, the first loop gets stopped and the signal is jumped to the next
successive loop. In this way by the gradual increment of frequency
input user can get the desired value of cut-off frequency.
INTERPOLATION: In this technique the user uses the known data
values to determine the unknown data values. There are different
types of interpolation methods: Linear interpolation Polynomial
interpolation So this method is a kind of mathematical process used
for getting the proper values of cut-off frequency. CURVE FITTING:
Here in this technique the curve is constructed to get the best fit
with the series of data points and through that way user can get
the desired cut-off frequency.
5
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
PART 1Here out of the three basic techniques of finding the cut
off frequency the first technique, - successive approximation is
going to be used.
HARDWARE SET UP:The pictorial view of the hardware set up for
the purpose of determination of cut off frequency of any low pass
filter is as shown below. Here in this setting of the hardware it
is clearly shown that signal generated by the signal generator
passing through the low pass filter and gets entered in the
oscilloscope at two different points, input and output named as X
and Y. Before starting of the experiment the signal generated
should be set at initial frequency and at the peak to peak input
voltages. And the signal generator should also be set for high
impedance. And PXI has been used for communicating with the signal
generator and the PC. From oscilloscope the waveform of both
voltages and the point where they get tripped or the cut off
frequency point can be achievable. Description of different parts
of this hardware set up and the techniques are in the following
section.
6
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
Descriptions of two basic instruments used in this technique:y y
Signal generator: Agilent 33220A 20MHz Function/Arbitrary Waveform
Generator Oscilloscope: Agilent DS03102A Digital Storage
Oscilloscope
Brief review of all the components used:PXI: PXI stands for PCI
Extensions for Instrumentation. It is a rugged PC-based which is
used for measurement and automation applications that require high
performances. It provides synchronization between the buses and the
software features used in the system. GPIB: The GPIB stands for
General Purpose Interface Bus. Its role is to communicate between
the computers and the instruments for transferring the data between
them. This is also called as IEEE-488. GPIB consists of 16 signal
lines and 8 ground lines. Out of those 16 signal lines, there are 8
data lines 3 handshake lines and 5 interface management lines. And
out of 8 ground lines there are 5 lines twisted as pairs with the 5
interface management lines. The role of all the lines has been in
the following section. The data transfer rate of this bus is up to
1Mbps. TYPES OF GPIB: GPIB can be talkers, listeners and/or
controllers. Talker transmits the data onto the interface bus and
there can be a single talker at a particular time. Listener
receives the information from the interface bus and there can be
more than one listener at a time.
7
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
Controller manages the flow of information on the interface bus
by sending commands to all devices and there can be only one
controller at a time. All GPIB devices communicate between
themselves through the way of sending messages. And there are two
types of GPIB messages: Device-dependent messages (Data messages):
This kind of messages contains device specific information: program
instruction, measurement results and data files. Interface
messages: This kind of messages manages the bus and usually
performs the functions such as, initializing the bus, addressing
the devices and setting device modes for the remote or local
programming. DESCRIPTION OF ROLE OF DIFFERENT LINES: First of all
the schematic diagram of GPIB device is as shown below.
Schematic diagram of GPIB device DATA LINES: The eight data
lines carry both data and command messages. The state of the
attention line (ATN) determines whether the information is data or
commands. All the commands or most of the data use the 7 bit ASCII
code set and the eighth bit either be unused or used for parity.
HANDSHAKE LINES: Three lines asynchronously control the transfer of
message bytes between devices. This process is called 3-wire
interlocked handshake. It assures the status of the message bytes
that they are sent and received without transmission error. The
three lines are:
8
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
y y y
NRFD: It stands for not ready for data because it indicates when
a device is ready or not ready to receive a message byte. NDAC: It
stands for not data accepted because it indicates when a device has
or has not accepted a message byte. DAV: It stands for data valid
because it indicates when the signals on the data lines are stable
(valid) and can be accepted safely by devices.
INTERFACE MANAGEMENT LINES: Five interface management lines
manage the flow of information and those are as follows: y ATN
(attention): The controller drives ATN as true when it uses data
lines to send commands and drives ATN as false when a talker can
send data messages. IFC (interface clear): The system controller
drives the IFC line to initialize the bus and become CIC. REN
(remote enable): The controller drives the REN line which is used
for setting devices at remote or local mode. SRQ (service request):
Any device can drive the SRQ line to asynchronously request service
from controller. EOI (end or identify): This line has two main
purposes: talker uses this line to mark the end of message string
and controller uses this line to tell devices to identify their
response in a parallel poll.
y y y y
Here in this task the oscilloscope, function generator and the
computer are connected through the GPIB cable. The controller is a
board installed in the PC. The oscilloscope and the function
generator are usually talkers and listeners. All these GPIB
instruments are addressed with unique GPIB address. The address
string for the used oscilloscope (Agilent 3000/5000 series) is 7and
that of function generator (Agilent 33220A) is 7. The low pass
filter is connected to the function generator and oscilloscope so
that the output and the input voltage can be measured. After the
hardware setup the next step is to design the software for
performing the task through the writing the program in Labview.
9
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
SOFTWARE DESIGN:Here the program is written in Labview
programming for the purpose of determination of cut off frequency
of a low pass filter. For writing the program following function
palettes are used: GPIB Write:
(timeout ms) specifies the time that the function waits before
timing out. (address string) contains the address of the GPIB
device with which the function communicates. (data) is the data
which is written to the device by the function. (mode) indicates
how to terminate the GPIB write. (error in) describes error
conditions that occur before the function or VI runs. The default
condition is no error. When any error occurred before the VI runs
the error in value gets passed to the error out. And if any error
occurred during the running of VI it sets its own error status in
error out. (status) is true if any error occurred before the VI
runs and it is false when no error occurred before the VI runs.
Here the default is false. (code) is the error or warning code. The
default is zero (0) for this case. If the status is true, code is
an error code. (source) is the origin of the error or warning. The
default is an empty string. The data given to the function
generator are frequency and voltage. The commands are FREQ N HZ
which sets the frequency to the value N in Hz and VOLT N VPP which
sets the amplitude to the value N peak to peak.
10
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
For the oscilloscope the commands are (:MEAS:VPP? CHAN1) which
returns the peak to peak voltage on channel 1 and (:MEAS:VPP?
CHAN2) which is used for channel 2. GPIB Read:
(timeout ms) specifies the time (in milliseconds), that the
function waits before the timing out. For disabling timeout
(timeout ms) should be set to 0. (address string) contains the
address of the GPIB device with which the function communicates.
(byte count) specifies the number of bytes the function or VI reads
from the GPIB device. (mode) specifies the conditions, other than
reaching byte count for the termination of read. (error in)
describes error conditions that occur before the function or VI
runs. The default condition is no error. When any error occurred
before the VI runs the error in value gets passed to the error out.
And if any error occurred during the running of VI it sets its own
error status in error out. (status) is true if any error occurred
before the VI runs and it is false when no error occurred before
the VI runs. Here the default is false. (code) is the error or
warning code. The default is zero (0) for this case. If the status
is true, code is an error code. (source) is the origin of the error
or warning. The default is an empty string. (data) is the data the
function or VI reads. Here the number of byte count given is
1042.11
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
Brief description of the labview program: First of all the
function generator is initialized to 1 VPP and set to high
impedance load using the command (OUTP:LOAD INF). The reason of
setting at high impedance is to block the higher frequency signals.
The oscilloscope gets initialized to auto scale using the command
(:AUTO). The pictorial view of initializing of system is as shown
below.
The program consists of two loops. The first loop starts with
frequency of 50 Hz and gets incremented with the frequency of 50 Hz
and the second loop has the increment of frequency of 1 Hz. The
frequency data is given to the function generator using the
concatenated string. And before using the concatenated string the
number has been converted to decimal string. Between the generator
and the oscilloscope time delay function has been used so that the
oscilloscope can get time to get adjusted to the commands each time
it runs in the loop. The generator and the oscilloscope are also
has been addressed properly. The data from the GPIB read has been
converted to number to get the output voltage and the input voltage
of the filter. For determination of cut off frequency, the Vout/Vin
should be less than or equal to 0.707. This condition has been
calculated using the formula 20*log(Vout/Vin) = - 3 db. During the
running of the loop whenever this condition reaches the first loop
gets stopped and the signal jumps to the next loop where the
increment of frequency is 1 Hz. Thus in the second loop user can
get the proper value of cut off frequency using the same condition
(Vout/Vin 0.707).12
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
The pictorial views of loop 1 and 2 are as shown below. LOOP
1
LOOP 2
13
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
The pictorial view of the front panel is as given below.
PART 2In this part the same labview program has been used. The
differences are the serial number of the filters should be recorded
and the program should prompt for the serial number of the filter.
And moreover the data for each of the filter should be written in a
spreadsheet file. Here in this report the block (prompt user for
input) has been used for prompting for the serial number and for
the spreadsheet file where the data for each filter should be
written which is as shown in the next page.
14
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
Apart from this block an array has been used for writing a
series of data in a spreadsheet file. And few number strings have
been used for getting different values from where it is required.
Here in this part of the task the cut off frequencies of ten
different filters have been recorded in a spreadsheet file. The
whole block diagram is as shown in the next page.
RESULTS: The values of the cut off frequencies, different values
of ratio of Vout/Vin, values of Vout and Vin of ten different
filters with their serial numbers are given below in tabular form
those which have been written in the file soumyo.xls.
FILTER NO. Vout/Vin Vin Vout 15 0.73 1.06 0.688679 8 0.75 1.08
0.694444 6 0.75 1.08 0.694444 11 0.67 0.97 0.690722 3 0.73 1.05
0.695238 24 0.73 1.06 0.688679 16 0.75 1.08 0.694444 2 0.73 1.05
0.695238 18 0.75 1.08 0.694444 13 0.71 1.01 0.712941
CUT OFF FREQUENCY 1553 232 361 8556 1352 958 97 1472 79 1701
15
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
BLOCK DIAGRAM OF THE PART 2:
16
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
EVALUATION OF PERFORMANCE OF THE SYSTEM: RESOLUTION: The system
has a resolution of 1Hz. The program in Labview has been made in
such a way that the finest value of the cut off frequency can be
measured because as described before in the Labview program after
getting the probable range of cut off frequency from the first loop
the increment of frequency in the second loop is only one hertz. So
the resolution of the system is 1Hz. Here cut off frequency of one
filter (filter number 16) has been recorded for seven consecutive
times for evaluating the performance of the system properly. The
readings are tabulated as shown below. FILTER NUMBER 16 16 16 16 16
16 16 CUT OFF FREQUENCY 97 96 97 101 99 98 95
UNCERTAINTY: Uncertainty is the range of values of the
measurement within which the true value of the measurement should
lie. Type A: The best estimate ( ) of the cut off frequency is
97.57 Hz. Then the standard deviation of population (n-1)
is 1.998 Hz.
The standard deviation of the measurement is 0.755 Hz.
17
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
TYPE B: Applying the rectangular probability distribution, the
standard uncertainty of the cut off frequency can be calculated as
SN = (1/2)/ = 0.2887 Hz. So the combined uncertainty will be as
follows, SN = (0.7552 + 0.28872) = 0.8083Hz. This value of combined
uncertainty (= 0.8083Hz) indicates that the uncertainty is
moderate. This uncertainty has been calculated based on the values
of cut off frequency of a particular filter. It may be of higher
value for other filters. This uncertainty is due to the different
factors such as, systemic error, environmental disturbances, noises
present in the system or any other external factors.
REPEATABILITY: The cut off frequency of filter number 16 has
been recorded for 7 consecutive times. The values are tabulated
below: FILTER NUMBER 16 16 16 16 16 16 16 CUT OFF FREQUENCY 97 96
97 101 99 98 95
From the table shown above it is clear that the values of cut
off frequency of the filter number 16 are likely different every
time. The values are plotted in the following graph from where the
repeatability of the value of cut off frequency can be observed
clearly. From the graph shown in the next page it is clear that the
value of the cut off frequency is not at all constant for any
filter. Although it is not varying abruptly but the difference in
measured values is quite clear.
18
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
102 f r e q u e n c y 101 100 99 98 97 96 95 94 0 1 2 3 4 5 6 7
8
serial no of measurement
RESPONSE TIME: The response time of the system is moderately
fast as the frequency in the first loop gets incremented by 50Hz
and then in the second loop it increases by 1Hz in the range of
frequency of 50Hz. But the readings of the system are quite
efficient because of its resolution of 1Hz.
19
Soumyajyoti Sengupta (S0909491)
MSc in Applied Instrumentation & Control (Instrument
Communication & Networking)
2010-2011
CONCLUSION:Here an automated measurement system has been
developed based on the Labview programming where the cut off
frequencies of 10 different filters can be measured very
efficiently. Development of the system can be suggested depending
on the following factors: Here after investigating the uncertainty
for the particular filter (filter no. 16) it can be suggested that
by reducing the systemic error, by making the hardware set up
properly and by providing the environmental surrounding as good as
possible the uncertainty can be reduced. And at the same time by
making these factors (described before) proper repeatability also
can be improved. Response time can be improved by employing more
loops where values of frequency input can be started from any
higher value of frequency such as, 1000Hz with 1000Hz increment and
then increment of 100Hz in the following loop and so on. But for
getting efficient reading I think the system and program presented
in this report is the appropriate one.
REFERENCES:y y y Caledonian University, 2009. Measurement Theory
and Devices [Course Module Handbook] Caledonian University, 2010.
Instrumentation Communication and Networking[Course Module
Handbook] http://www.hit.bme.hu/~papay/edu/GPIB/tutor.htm
20