Energy and Power Engineering, 2013, 5, 1-5
doi:10.4236/epe.2013.53B001 Published Online May 2013 (http://www.scirp.org/journal/epe)
Design and Application of a MCU-based Smart
Contactor System
Fuqing Huang, Weipeng Lin
Engineering & Technology Training Center , Civil Aviation University of China , Tianjin , China
Received 2013
ABSTRACT
This paper aims to design a MCU-based smart contactor system, which will realize functions of real time surveillance
of contactors’ work status and data exchange with host computer, thus instantly reflect different status of the system. In
case of abnormal status such as over-current or under-voltage, the contactor will be able to automatically cut off power
supply to protect electrical load and the circuit. Through ARINC485 bus, system computer will collect and record con-
tactor parameters, including contact voltage, contact current, supply frequency, contact temperature and contact status
to provide critical data to the examination and repair of contactors.
Keywords: Aerial Contactor; MCU; Testing; ARINC485
1. Introduction
Aerial contactors have many applications in aircrafts and
play an important role in power supply for on-board
equipments and instruments. A malfunctioned contactor
would cause the destruction of electric load and, even
worse, an on-board electrical fire. Hence, the performance
of contactors has direct impact on power supply quality
for on-board system devices and has a close relation with
flight safety. If it is possible to carry out real time sur-
veillance and recording of contactors during a flight, then
arising fault would be instantly detected and relative
electrical load be disconnected to ensure the normal
functioning of other electrical loads and power circuits.
By this way, accident caused by breakdowns such as
short circuit will be avoided and flight safety would be
greatly improved. Meanwhile, maintenance interval will
be greatly shortened and operation cost be saved as
maintenance personnel are directed to targets based on
system information, such as recorded contactor status.
2. System Design
First, we select Micro Controller Unit (MCU) as the sys-
tem’s main control chip to realize functions of surveil-
lance and communications. Second, make sure the device
is able to perform real time measuring of three phase
voltage and current as signal excursion may appear. To
enable an accurate measurement, we need to synchronize
the sampling frequency of MCU A/D with signal fre-
quency and phase position, which calls for three-phase
AC signal measuring. What’s more, enable the direct
control of contactor through the driving of contactor coil
via I/O signal and the receiving of external control level.
At last, make sure the Full-duplex communication by
way of two groups of half-duplex operation RS485. The
system structure framework is shown in Figure 1.
2.1. Hardware Design
The system controller adopts MSP430F149 [1] MCU, an
ultra-low power consumption unit produced by TI.The
MCU is especially suitable for battery applications or
handheld devices. Besides, the integration of a large
amount of peripheral modules into the controller unit
enables the MCU a perfect option for the design of SoC;
and abundant alternative components with various types
also enrich its flexibility for designers. As a 16 bit sim-
plified instruction architecture, the MCU has plenty of
working registers and data storage units, including
Figure 1. System structure framework.

F. Q. HUANG, W. P. LIN
2
calculation enabled RAM units. Except for the selection
of controller, it also satisfies the design of voltage and
current transducers.
2.1.1. Design of Voltage Transducer
Adopt an isolation transformer voltage drop method.
Take Model HPT205NB voltage transformer as the iso-
lation transformer. See its parameters as follows: Incom-
ing current: 0 - 2 mA, Output current: 0 - 2 mA, Preci-
sion: 0.1%, Linearity: 0.07%, Phase shifting: ≤11′，
Isolation withstand voltage: 2700 Vac, Internal resistance:
105Ω. Its main advantages are shown in the schematic
diagram of circuit (see Figure 2), including high signal
isolation, strong anti-jamming capability and conformity
to requirements for industrial anti-jamming system de-
sign.
2.1.2. Design of Current Transducer
Adopt A1302 linear Hall sensor [2] produced by Allegro.
A1302 is a successive type proportional linear Hall Ef-
fect sensor, which is able to provide precise proportion-
able voltage output values with the applied magnetic
field. It has a static output voltage which equals with
50% of supply voltage. Its output sensitivity is 1.3 mV/G.
See its application circuit diagram in Figure 3.
In Figure 3, power source of the transducer is pro-
vided by a DC/DC uncontrolled rectifier, which goes
through two levels II type filtering and a voltage regula-
tor tube. This enables the stability and precision of signal
measurement. After capacitor filtering and voltage partial
pressure, output signal from transducer will be sent to
MCU A/D for processing.
Figure 2. Step-down circuit of isolating transformer.
Figure 3. Circuit diagram of hall current signal conditioning.

F. Q. HUANG, W. P. LIN 3
2.2. Software Design
Considering the frequency jitter of tested signal is not
large and will not result in serious wave distortion, we
use input capture terminal to trigger the sampling, and
then continuously take samples at numerous points
(within the range of 1 - 1.5 circle) and then take down
results. Calculate the digital value of these signals’ am-
plitude, and then determine proportional relation of ef-
fective values through fitting method in experiments to
obtain effective value of the signals. Such a solution of
sampling and calculation not only saves hardware re-
sources, reduces the size of circuit board, but also saves
precious RAM resource in MSP430F149, thus enhancing
the utilization of stack and acquiring a qualified speed for
real time surveillance. The controller software flow path
is shown in Figure 4.
3. System Testing
3.1. System Testing Platform
A testing platform is required for a comprehensive test-
ing of system functions and system performance. The
platform consists of power source, electrical load, pro-
tection device, standard measuring equipments and etc.
See block diagram of the platform in Figure 5.
3.2. Analysis of Experimental Data
Use fitting method to obtain proportional relation of
measured signals’ virtual values and then determine the
final virtual values.
Figure 4. Controller software flow chart.

F. Q. HUANG, W. P. LIN
4
Figure 5. Block diagram of system testing platform.
3.2.1. Analysis of Current Data
Given a polynomial as equation (1), then a corresponding
normal equation is obtained as equation (2).
22 01 () .
x
aaxax    (1)
2
0
23
1
23 42
2
26 ii
ii ii
ii ii
xx
Xyi
x
xxXxyi
X
x
xx xyi

 
 




 






  



x
x
(2)
Calculate with the help of MATLAB to get fitting
polynomials of phase A, B and C as equation (3).
2
2
02
()11.1562(- 0.0018)(2.6088e -
()10.3004(- 0.0013)(2.2989e -
( )10.5804(-1.9585e-004)(9.191
A
B
C
xx
xx
006)
006)
6e -07)
x
xx



 
 

 

(3)
Conduct fitting process on each point’s electrical cur-
rent scaling factor to get a proportionality coefficient
curve. Together with the proportionality coefficient of
each point obtained through experimental data, it is shown
in Figure 6.
Based on proportional coefficient of each point ob-
tained through fitting, determine corresponding nominal
current value and display the contrast relation with actual
current values in Figure 7.
Final experimental results indicated that the deviation
between obtained three-phase experimental current value
and the actual current value is about 0.1%, which is able
to meet engineering design requirements.
3.2.2. Analysis of Voltage Data
Given a polynomial as equation (4), then a corresponding
normal equation is obtained as equation (5).
01
() .
x
aax

 (4)
0
21
12 i
i
ii
x
X
yi
X
xyi
xx
 
  


 

  
 (5)
Calculate with the help of MATLAB to get fitting
Figure 6. Proportional coefficient curve obtained through
fitting of currents.
Figure 7. Contrast of experimental current value and vir-
tual current value obtained through fitting.
polynomials of phase A, B and C as equation (6).
()- 0.0025x+ 24.397
()- 0.0025+ 24.4434
()- 0.0027+ 24.2588
A
B
C
x
xx
xx









(6)

F. Q. HUANG, W. P. LIN 5
Figure 8. Proportional coefficient curve obtained through
fitting of voltage.
Figure 9. Contrast of experimental voltage value and virtual
voltage value obtained through fitting.
Conduct fitting process on each point’s voltage scaling
factor to get a proportionality coefficient curve. Together
with the proportionality coefficient of each point ob-
tained through experimental data, it is shown in Figure
8.
Based on proportional coefficient of each point through
fitting, determine corresponding voltage value and dis-
play the contrast relation with actual voltage values in
Figure 9.
Final experimental results indicated that the deviation
between obtained three-phase experimental voltage value
and the actual voltage value is about 0.3%, which is able
to meet engineering design requirements.
4. Conclusions
As experiments and analyses showed, this innovative
smart contactor is featured by the realization of the on-
off control of three-phase 115 V/400 Hz AC power sup-
ply, the real time adjustment of contactor status as per
bus command and the real time surveillance of
three-phase load voltage within the safety range of 60 -
150 V and the AC within the safety range of 10 - 60 A as
per bus command, as well as the surveillance of contac-
tor’s ambient temperature fluctuating within ±1℃under
the scope of -40℃ - +150℃. The contactor is also able
to perform real time uploading of present working pa-
rameters as per requirement of data processing devices to
realize remote control.
REFERENCES
[1] X. L. WEI, “Example of MSP430 Series MCU Interface
Design and System Design,” Publishing House of Beijing
University of Aeronautics and Astronautics, Beijing,
2002.
[2] Allegro MicroSystems Inc, “A1302 Databook,” 2005.