issue: July 2004 APPLIANCE Magazine
Engineering Motor Control
Brushless D.C. Motor Control for Refrigerator Applications
Email this Article
by Padmaraja Yedamale, senior applications engineer,Microchip Technology, Inc.
Evolving market needs and changing energy policies push designers to consider energy efficiency, audible noise, product lifetime, maintenance, and the cost of initial products and operation.
Refrigeration performance largely depends upon the control capabilities
of the compressor and cooling fan. The compressor is instrumental in cooling
the refrigerant, while the fan blows cold air throughout the appliance.
Traditionally, Permanent Split Capacitor (PSC) single-phase induction
motors are used for the compressor. Shaded pole motors, which are also
a single-phase induction motors, are used for fan control.
The PSC motors have a non-linear speed versus torque characteristic. For
this reason, using a variable-speed drive does not give good torque control
over the speed range, thus, the motors have low-energy efficiency and
tend to be noisy compared to brushless d.c. (BLDC) motors. PSC motors
are not ideal for energy-efficient or quiet refrigerators.
Most of today’s refrigerators switch on at full speed when the temperature
increases above a trip temperature. The compressor is turned off when
the refrigerator reaches the set temperature. This operation is noisy
because the motor runs at full speed with a “bang” ON and
“bang” OFF. Also, the compressor motor draws a large starting
current. A typical curve, as shown in Figure 1, demonstrates the starting
characteristics of the compressor motor. This may introduce a voltage
dip in the main-power supply and cause a momentary brownout situation.
1. Motor Current Versus Time When an Induction Motor
Brushless D.C. Motors
Brushless d.c. (BLDC) motors are gaining popularity due to their performance
advantages over PSC and brushed d.c. (BDC) motors, including the following:
• BLDC motors have a relatively flat speed-torque characteristic
(see Figure 2). This enables the motor to operate at lower speeds without
compromising torque when the motor is loaded.
• The ratio of output power to frame size is higher in BLDC motors.
This reduces the size and weight of the product. This also saves the
cost of motor mounting and shipping expenses.
• The BLDC motors operate at higher-power efficiency compared to
induction motors and BDC motors because they have permanent magnets on
the rotor and there are no brushes for commutation.
• Brush inspection is eliminated, making them suitable for limited-access
areas like compressors and fans. This also increases the life of the motor
and reduces the service requirements.
• They operate much quieter compared to BDC motors since brushes
make audible noise.
• BLDC motors have less electromagnetic interference (EMI) generation.
With BDC motors, the brushes tend to break and make contacts while the
motor is rotating, resulting in the emission of electromagnetic noise
into the surroundings.
However, there are two concerns with BLDC motors. First, BLDC motors can
be more expensive; however, the performance advantages override this concern.
In addition, the cost delta is reduced as BLDC motors become more common.
Second, BLDC motors need electronic commutation. The stator windings are
commutated based on the rotor position. This requirement can be turned
into an advantage. The same electronics used to control the commutation
can also provide speed control. This paper discusses the control methodology
for BLDC motors.
2. Typical - Speed Torque Characteristics of a BLDC Motor
Versus PSC Induction Motor
Figure 2 shows the typical-speed torque characteristics of a BLDC
motor. During continuous operations, the motor can be loaded up
to the rated torque. The torque remains constant for a speed range
up to the rated speed. The torque generated by a BLDC motor is governed
by following equation:
T µ NBlrI
Where N is the number of stator turns per phase,
B is the rotor flux density,
l is the length of the rotor,
r is the radius of the rotor,
I is the current flowing in the stator winding.
Once the motor is designed, all parameters (except current) remain
constant. Therefore, torque becomes proportional to the current
drawn by the motor. At lower speeds, the average voltage supplied
to the motor is less than the rated voltage. The power drawn by
the motor is a product of the current drawn and the voltage. When
the motor runs at lower speeds, the average power consumed by the
motor also reduces proportionately to the speed. This adds up with
the inherent energy efficiency, saving a significant amount of power
compared to the ON/OFF induction motors in traditional refrigerators.
3. Control Block Diagram
BLDC motors have alternate north (N) and south (S) permanent magnets
on the rotor. The stator has stacked-steel laminations with windings placed
in the slots that are axially cut along inner periphery. To rotate the
BLDC motor, the stator windings should be energized in a sequence. Knowing
the rotor position is important in order to follow the commutation sequence.
Rotor position is sensed using Hall-Effect sensors embedded into the stator.
Most motors have three Hall sensors embedded into the stator on the non-driving
end of the motor. Whenever the rotor-magnetic poles pass near the Hall
sensors, they give a high or low signal, indicating that the N or S pole
is passing near the sensors. Based on the combination of the three Hall
sensor signals, the exact sequence of commutation can be determined. Table
1 provides a typical commutation sequence with respect to Hall sensor
With every 60 electrical degrees of rotation, one of the Hall sensor outputs
changes its state from high to low, or from low to high. Given this, it
takes six steps to complete one electrical cycle. In a synchronous pattern
corresponding to these 60 electrical degrees, the sequence of phase-voltage
switching should be updated. However, one electrical cycle may not correspond
to a complete mechanical revolution of the rotor. The number of electrical
cycles that must be repeated in order to complete a mechanical rotation
is determined by the rotor-pole pair. For each rotor-pole pair, one electrical
cycle is completed. So, the number of electrical cycles per rotation equals
the rotor-pole pair.
As Table 1 indicates, every sequence has two of the three phases connected
to the power supply, with the third phase left open. Figure 3 shows a
simplified BLDC motor control block diagram. The 8-bit Flash microcontroller
(MCU) in the figure is used to control the power switches. Matching drivers
are used to give the appropriate gate drive to the power switches. The
MCU family has six Pulse-Width Modulated or Modulation (PWM) channels
with a programmable PWM frequency and duty cycle. Phases A, B, and C are
connected to the center point of each half H-bridge PWM0 to PWM5 control
the power switches, and Q0 to Q5 are connected in a three-phase inverter
bridge configuration. The MCU has three input-capture pins indicated as
IC1, IC2, and IC3. The input-capture module has a mode in which Timer5
captured every transition on any of the input-capture pins. This mode
is suitable for interfacing Hall sensors to the MCU. With every Hall sensor
transition, an interrupt is generated, and the Timer 5 value is captured.
This captured value corresponds to the motor speed.
The motor will run at a rated speed when the signals marked by PWM0 to
PWM5 are switched ON or OFF according to the sequence (see Figure 3).
This is assuming that the d.c.-bus voltage is equal to the motor-rated
voltage, plus any losses across the switches. To vary the speed, these
signals should be PWM at a much higher frequency than the motor frequency.
As a rule of thumb, the PWM frequency should be at least 10 times that
of the maximum frequency of the motor. When the duty cycle of the PWM
frequency is varied within the sequences, the average voltage supplied
to the stator decreases, thus reducing the speed.
Another advantage of having PWM is that when the d.c.-bus voltage is much
higher than the motor-rated voltage, the voltage supplied to the motor
can be limited to the motor-rated voltage by limiting the percentage of
the PWM duty cycles corresponding to that of motor-rated voltage. This
adds the ability to use the refrigerator control circuit in multiple countries
with different a.c. inputs. The a.c.-voltage inputs are converted to d.c.
using a diode-bridge rectifier, and the average voltage output by the
controller is matched to the motor-rated voltage by controlling the PWM
There are different approaches for control. If the PWM signals are limited
in the MCU, the upper switches can be turned ON for the entire time during
the corresponding sequence, and the corresponding lower switch can be
controlled by the required duty cycle on the PWM.
In Figure 3, the temperature is set by the user. The internal refrigerator
temperature is measured using temperature sensors. Depending upon the
size of the refrigerator, there may be more than one temperature sensor
placed at different locations inside the refrigerator. The set temperature
and actual temperatures are read using an on-chip Analog-to-Digital Converter
(ADC). When the difference in temperature is more than a predetermined
hysterisis (typically less than 2°F), the motor starts rotating at
a lower speed. If the temperature difference is larger (i.e., the refrigerator
door is left open), then the motor should run at a higher speed. A relationship
between the temperature difference and speed can be derived based on the
size of refrigerator. As the difference between set and actual temperature
changes, the compressor and fan speed can be changed accordingly.
4. Flow Chart for a BLDC Motor Control with Hall Sensor Commutation
Control of a BLDC Motor
Sensorless control of a BLDC motor calls for commutation based on the
back EMF produced in the stator windings. Hall sensors are not required
for this method. Sensorless control has two distinct advantages: increased
reliability and lower costs. Inherently, systems with fewer components
are more reliable. The compressor generates heat and an elevated temperature
accelerates Hall sensor failures. The cost savings are achieved, not only
with the piece part pricing of the Hall sensor, but also since the Hall
sensor wires are not necessary. At least five wires can be eliminated.
1. Typical Commutation Sequence with Respect to the
Until now, we have seen the commutation based on the rotor position from
the Hall sensor. Manipulating the back EMF signals, instead of the Hall
sensors, can commutate the BLDC motors.
The amplitude of back EMF depends upon three factors: the angular velocity
of the rotor, the number of turns in the stator winding, and the rotor-magnetic
field. Once the motor is designed, the rotor-magnetic field and the number
of turns in the stator windings remain constant. The factors that govern
the back EMF are the angular velocity or speed of the rotor. The back
EMF increases proportionally to the rotor speed. However, back EMF can
be estimated for a given speed using the back EMF constant provided in
the motor data sheet.
The relationship between the Hall sensors and the back EMF, with respect
to the phase-voltage, is shown in Figure 5. As we have seen in the earlier
sections, every commutation sequence has one of the windings positively
energized, a second negatively energized, and a third left open. As shown
in Figure 5, the Hall sensor signal changes state when the voltage polarity
of the back EMF crosses from a positive to negative, or from a negative
to positive, with a phase difference of 30 degrees. In ideal cases, this
happens on zero crossing of the back EMF. However, there will be a delay
due to the winding characteristics. This delay should be compensated by
5. Relationship Between Hall Sensor Signals, Phases - Voltage,
Current, and Back EMF
Another aspect to consider is very low-speed operation. Because the back
EMF is proportional to the rotation speed, at very-low speeds, the back
EMF would have very low amplitude with which to detect zero crossing.
From standstill, motors need to be started in open loop. When sufficient
back EMF is built in to detect the zero cross point, the control should
be shifted to the back EMF sensing. The minimum speed at which the back
EMF can be sensed is calculated from the back EMF constant of the motor.
Some motors may have the Hall-sensor magnets on the rotor, in addition
to the main rotor magnets. These are used to simplify the process of mounting
the Hall sensors onto the stator, a scaled down replica of the rotor.
Therefore, whenever the rotor rotates, the Hall-sensor magnets give the
same effect as the main magnets. The Hall sensors are normally mounted
on a PC board and fixed to the enclosure cap on the non-driving end of
With a sensorless method of commutation, the Hall sensors, Hall sensor
magnets, Hall sensor wires, and the PC board can be eliminated. This
simplifies the motor construction, reduces costs, and increases reliability.
Figure 6. Back EMF Zero Cross Detection
EMF Zero Cross Detection
Detecting the back EMF zero-crossover point is crucial to the sensorless-control
system. Different techniques are used in order to determine the zero
crossover point. Every commutation sequence has one non-energized
winding. The back EMF zero crossover point is sensed from every winding
when it passes through the non-energized state in the commutation
sequence. Figure 6 shows a scheme used to detect the back EMF zero
crossover point. In this example, Phase A is connected to the positive
side of the power supply, Phase C is the connected to the negative
side (or return path) of the power supply, and Phase B is open. When
the back EMF is observed on Phase B, it builds up toward positive
d.c. and then falls to negative d.c. supply. A virtual-zero cross
point can be derived when it is compared with half of the d.c. bus.
Using an operational amplifier comparator, the zero crossover point
can be determined. Figure 6 shows the zero cross-detection circuitry
for Phase B. Similar circuits should be used with Phase A and Phase
C to detect the back EMF zero crossover, when its respective windings
are not energized.
Another plan for detecting back EMF zero crossover is using the
ADC. Using a potential divider, the back EMF signal is brought down
to a MCU measurable level. This signal is sampled using an on-chip
ADC. These samples are continuously compared with a digital value
corresponding to the zero point. When these two values match, the
commutation sequence is updated. The advantage with this method
is that it is more flexible in terms of measurement. When the speed
varies, the winding characteristics may fluctuate, resulting in
variation of back EMF. In such situations, the MCU has complete
control over the determination of the zero crossover point. Also,
digital filters can be implemented to filter out the high-frequency
switching noise components from the back EMF signal.
The inherent advantages of BLDC motors can be used to make the refrigerator
control energy efficient and less noisy, while providing continuous
speed variation. However, BLDC motors require electronic commutation
that is controlled by the drive circuit. Using a drive not only allows
variable-speed operation, but also decreases the noise level. Additionally,
the system consumes less energy compared to the traditional way of
intermittent turn ON and turn OFF operation of the compressor, based
on the cooling load.
Yedamale is a senior applications engineer with Microchip
Technology, Inc. He has a B.S. in Electrical and Electronics
Engineering from the University of Mysore, India. Mr.
Yedamale has more than 8 years of experience in embedded
control, designing embedded systems hardware, and software.