Electromagnetic Emission of BLDC Motor

Controllers
István Bodnár Rafael Ruben Boros Dániel Erdősy
Department of Electrical and Department of Electrical and Department of Electrical and
Electronic Engineering, Institute of Electronic Engineering, Institute of Electronic Engineering, Institute of
Physics and Electrical Engineering Physics and Electrical Engineering Physics and Electrical Engineering
University of Miskolc University of Miskolc University of Miskolc
H-3515 Miskolc, Hungary H-3515 Miskolc, Hungary H-3515 Miskolc, Hungary
2021 22nd International Carpathian Control Conference (ICCC) | 978-1-7281-8609-2/20/$31.00 ©2021 IEEE | DOI: 10.1109/ICCC51557.2021.9454655

vegybod@uni-miskolc.hu elkruben@uni-miskolc.hu elkedani@uni-miskolc.hu

Dávid Matusz-Kalász
Department of Electrical and
Electronic Engineering, Institute of
Physics and Electrical Engineering
University of Miskolc
H-3515 Miskolc, Hungary
elkmkd@uni-miskolc.hu

Abstract—The electrically commutated brushless direct the permanent magnet rotor and the rotating magnetic field
current (BLDC) motors are constantly displacing conventional induced in the wound stator poles. Fig. 1. shows a simplified
brushed DC motors in industrial usage, especially in the illustration of a BLDC motor construction.
automotive industry. During the research, we investigated the
electromagnetic noise emission of a BLDC motor operating a car
engine cooling fan. During the determination of electromagnetic
noise, we performed near-field measurements and examined the
directed fields. The emission of the control electronics of BLDC
motors with different phase and pole numbers was measured in
three directions, with two antenna polarizations per direction.
These data can be used to outline a three-dimensional radiation
characteristic. The measurement results were compared, and
the result was obtained to what extent and to which controlled
electromagnetic noise emissions each component of the BLDC
motor is responsible.

Keywords—BLDC motor, EMC, driver circuit, 2-phase motor,

3-phase motor

I. INTRODUCTION
As a part of a bigger project, we were analysing an
automotive cooling fan module. For us, Electromagnetic
Compatibility (EMC) of the driving motor was in spotlight, Fig. 1. Simplified BLDC motor construction.
but we should have work and consult with colleges with other
specialization in order to achieve the best results. In this article Most BLDC motors have a three-phase winding topology
a brief summary is about BLDC motors in general to learn with star connection. A motor with this topology is driven by
about its working principle. Another short section is about energizing two phases at a time. The three-phase winding is
EMC and finally our measurement method, we performed on fed by a three-phase inverter and the electronic commutation
the mentioned driving motor and results are included. is implemented also by the same inverter. Fig. 2. (next page)
shows the three-phase windings in star connection and the
II. BLDC MOTORS terminal names which are corresponding to Fig. 1. Three-
BLDC motors are a type of synchronous motors. This phase motors are better than two-phase design, because the
means the magnetic field generated by the stator and the rotating magnetic field is much more symmetrical. The three-
magnetic field generated by the rotor rotate at the same phase windings make less EMC noise.
frequency. BLDC motors do not produce “slip” that is The rotor is made with permanent magnets and can vary
normally seen in induction motors. BLDC motors come in from two to eight pole pairs with alternate North (N) and
single-phase, 2-phase and 3-phase configurations. South (S) magnetic poles. Based on the required magnetic
Corresponding to its type, the stator has the same number of field density in the rotor, the proper magnetic material is
windings. Out of these, 3-phase motors are the most popular chosen to make the rotor. Ferrite magnets are traditionally
and widely used. used to make permanent magnets. As technology advances,
A brushless motor is constructed with a permanent magnet rare earth alloy magnets are gaining popularity. The ferrite
rotor and wound stator poles. Electrical energy is converted to magnets are less expensive, but they have the disadvantage of
mechanical energy by the magnetic attractive forces between low flux density for a given volume. In contrast, the alloy

978-1-7281-8609-2/21/$31.00 ©2021 IEEE

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.
 material has high magnetic density per volume and enables the
rotor to be compressed further for the same torque. Also, these
alloy magnets improve the size-to-weight ratio and give
higher torque for the same size motor using ferrite magnets.
Neodymium (Nd), Samarium Cobalt (SmCo) and the alloy of
Neodymium, Ferrite and Boron (NdFeB) are some examples
of rare earth alloy magnets. Continuous research is going on
to improve the flux density to compress the rotor further.

Fig. 3. The driver BLDC motor of the cooling fan module.

It needs 12 V DC power and a simple 100 Hz PWM signal

with 15-90% duty cycle to set the required speed. The rotor is
connected to the shaft with a simple ball bearing. Before we
have made modification plans, we had had to carefully analyse
the source of the radiated electromagnetic waves.
III. BLDC MOTOR CONTROLLERS
The BLDC motor controller contains power electronics,
Fig. 2. Three-phase windings with terminals.
microcontrollers, passive components (Fig. 4. shows). The
Unlike a brushed DC motor, the commutation of a BLDC input is connected to 12 V DC. An electrolytic buffer capacitor
motor is controlled electronically. To rotate the BLDC motor, reduces the voltage ripples. The red circle shows the
the stator windings should be energized in a sequence. It is conducted EMC filter. The block diagram (Fig. 5.) shows the
important to know the rotor position to understand which further component in the module. The controller PWM signal
winding will be energized according to the energizing is connected to the microcontroller. The modules do not
sequence. Rotor position is sensed using Hall effect sensors contain Hall-sensors, so the back EMF is needed to be
embedded into the stator. measured. The “sensorless rotor position detection” on the
block diagram means this. To implement the field-oriented
Most BLDC motors have three Hall sensors embedded control, current sensing is required.
into the stator on the non-driving end of the motor. Whenever
magnetic poles of the rotor pass near the Hall sensors, they The driver circuit of the power MOSFET is connected to
give a high or low signal, indicating the N or S pole is passing the microcontroller’s output. In our measurements two-phase
near the sensors. Based on the combination of these three Hall and three-phase motors are used, thus two-phase and three-
sensor signals, the exact sequence of commutation can be phase inverter is required. The two-phase inverter is not a
determined. conventional H-bridge, it contains a double H-bridge circuit
(Fig. 6.), the stator windings are connected to it. The three-
The Hall sensors increase the price of the motor. The phase modules are using a conventional three-phase inverter
sensors can be left if we measure the motors back EMF. When configuration which is showed by Fig. 7. The winding is
a BLDC motor rotates, each winding generates a voltage connected in delta configuration.
known as back Electromotive Force or back EMF, which
opposes the main voltage supplied to the windings according
to Lenz’s Law. The polarity of this back EMF is in the
opposite direction of the energizing voltage.
The motors of our project (Fig. 3. shows) compliance the
required emission rates, but in strict conditions, for example
in electric vehicles electromagnetic (EM) emission of them
should be reduced.
It has two phase winding, which is not the best choice if
EMC is important. Next to the winding, the rotor has
permanent magnets, 5 of them which can cause not some
mechanical but also magnetic issues, because they are not so
close to each other and the magnetic flux lines are not evenly
distributed inside the rotor. The control electronics are on the
back side, on a baseplate, which is also used for mounting the
motor onto the whole fan assembly. The controlling here
works with back EMF, so our motor does not contain Hall
sensors. Fig. 4. The controller of BLDC motor.

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.
 IV. MEASURING TECHNIQUES
All kind of electronic devices radiate electromagnetic
waves. Most of them might malfunction from a higher
radiation levels coming from other devices. These two
properties are called electromagnetic compatibility or EMC.
Emission rates, EM immunity and the different measurement
methods are restricted by local and international standards. In
automotive industry CISPR 25, CISPR 12 and ISO 11452-2
are the most important restrictions. Engineers must make their
plans, factories must produce devices, EMC laboratories must
measure in the way written in these restrictions.
All kind of electric motors have different EMC problems.
It depends on the size, the required voltage type, having
brushes or not, how speed controlling work, how the stator and
rotor produce the magnetic field, what materials are used in
the construction and so on. We had a BLDC motor, which type
has major advantages. The main reason it is used more
frequently nowadays, the lack of brushes and wide variety of
size and power, but it needs an electrical commutation, which
can cause some EMC problems.
We used the handheld spectrum analyser FSH-8 from
Fig. 5. The controller of BLDC motor. Rohde & Schwarz for the measurements and a computer to
collect the data with. The antenna was a HE300HF directional
antenna. Its frequency range is from 9 kHz to 20 MHz.
Considering the amount of resulting data we made decisions
how to reduce them. We made the comparison measurements
with the following settings and methods.
Evaluation method:
• Radial and axial measurements (in front of the motor
and into the motor)
• Horizontal and vertical antenna polarity
• The distance between the EUT (Engine Under Test)
and the antenna is exactly 30 cm
• Duty cycle of the controlling PWM signal: 0%, 25%,
50%, 70%, 90%
• Measurement frequency range: 100-1500 kHz
• Evaluated frequency range: 900-1100 kHz
• 4 distinct frequency have been chosen to evaluate the
amplitude: 900 kHz, 966.667 kHz, 1033.33 kHz, 1100
kHz
Fig. 6. The controller of a 2-phase BLDC motor. • In the examined frequencies RMS (Root Mean Square)
value of the amplitude is calculated (1)
• We calculated the ratio between the RMS values of the
original EUT and the modified EUT. It’s marked with
“H”. If H<1, then the modification made an
improvement, if H is about 1, there were no
improvement and if H>1, that means the modified
version performed worse.

RMS = sqrt [(1/N) ∙∑ (i=1; N) Xi2] (1)

where RMS is Root Mean Square, N is the number of

samples, i is the index number, Xi is the i-th value of X.
We made 36 different measurements on each EUT (2
different antenna polarity, 3 antenna line arrangement and in
one arrangement 6 different RPM). We created a specific
excel sheet to do the mathematical calculations. We used only
the data required for our evaluation method, because the
measurement makes almost 15,000 data points per EUT.
Fig. 7. The controller of a 3-phase BLDC motor.

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.
 EMC measurements require precise measurement because vertical radiation was much more intense at all
methods and equipment, showed on Fig. 8. modules. (We made all the measurements, but at the final
evaluation we used only the limited data). According to the
usage of these modules, important PWM duty cycles are 25%,
70% and 90%, so we could ignore more data. All these
considerations result in a massive decrease of evaluated data
points. From the 15,000 collected data, we only had to analyse
270. The settings of the spectrum analyser are in TABLE 1.

TABLE 1. Applied settings on FSH8

Settings Value
Middle Frequency 800 000 Hz
Frequency offset 0 Hz
RF dampening manual
RF dampening 0 dB
Preamp ON
RF input 50 Ohm
RBW 10 000 Hz
Fig. 8. Proper automotive EMC test method according to CISPR 25. VBW 100 000 Hz
Our measurements were made in one of our laboratories. Sweep time 200 ms
This lab is owned by the University of Miskolc, Institute of Type of measurement Average
Physics and Electrical Engineering, Department of Electrical Detection method RMS
and Electronic Engineering. It is in building A3, on the Primary transducer HE300A-HF
basement floor. This lab is not an accredited, proper EMC
laboratory, but it has an electromagnetic shielding (it shields Number of averages 10
till 1 GHz), and we made our best to build the required test
conditions, but we could not achieve the standards. The later V. MEASUREMENTS RESULTS
detailed measuring data is only for reference, it cannot be used The following tables (Tables 2-7.) give a sample of the results.
for qualification purposes. During the bigger project Antenna polarity was vertical, but results are from three
mentioned in the introduction, a proper EMC laboratory has different antenna positions (in front of, next to and behind the
been built in the University, so we will continue the
motor) and three different speeds (speeds belong to 25, 70 and
measurements there. The results of this paper were all made
90 % duty cycle PWM signal). We made evaluations with
in the shielded laboratory, so we will describe only those
measurements. horizontal antenna polarity and two other speed (50 and 78
%), but these results were not so significant.
Other test equipment was an oscilloscope (to set the
control PWM signal precisely), a simple analogue PWM TABLE 2. Normalized amplitude of electromagnetic emission of 2-phase
signal generator, a car battery (for supply DC power for EUT). motor (in front of the motor)
The EUT was mounted with wooden blocks on a metal plate
covered wooden table. The covering plate was grounded. We [dBμV/m] PWM
made an extra grounding pole into our lab to increase the Frequency [Hz] 25% 70% 90%
grounding properties to achieve higher shielding factors. One 900 000.00 7.500 15.559 1.901
of the measurement arrangements is showed on Fig. 9. 6.577 13.434 2.246
966 666.67
1 033 333.33 6.057 10.973 2.513
1 100 000.00 5.984 8.815 1.813
RMS 6.771 10.657 1.606
TABLE 3. Normalized amplitude of electromagnetic emission of 2-phase
motor (next to the motor)

[dBμV/m] PWM
Frequency [Hz] 25% 70% 90%
900 000.00 7.500 15.559 1.901
966 666.67 8.550 13.748 2.000
Fig. 9. Cooling fan module (EUT) under measurement. 7.989 14.274 3.282
1 033 333.33
The results of our measurements lead to the fact, that in 1 100 000.00 9.035 11.944 2.303
the radial direction modules produce less electromagnetic RMS 7.900 13.042 2.382
radiation. Due to this fact, we used the axial directional data
when comparing different modifications and modules. We
made another simplification, not to use the horizontal data,

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.
 TABLE 4. Normalized amplitude of electromagnetic emission of 2-phase Figure 11. shows the sum of all the measured amplitudes
motor (next to) (also summing the axial and radial measurements). It clearly
shows that the 2-phase motor emits almost twice as much of
[dBμV/m] PWM
electromagnetic noise than the 3-phase motor does.
Frequency [Hz] 25% 70% 90%
10.19 15.796 2.892 60
900 000.00
966 666.67 9.965 17.116 3.888 50

Magnitude [dBuV/m]
1 033 333.33 10.128 16.621 2.478 40
1 100 000.00 9.877 15.028 2.545
30
RMS 8.921 13.221 2.502
20
TABLE 5. Normalized amplitude of electromagnetic emission of 3-phase
motor (in front of axial) 10
0
[dBμV/m] PWM 3 phase 2 phase
Frequency [Hz] 25% 70% 90%
900 000.00 0.287 0.087 0.037 Fig. 11. Summarized EM emission rates of the two motors

966 666.67 0.180 0.269 0.289

To examine the results from different approaches, we
1 033 333.33 0.0168 0.396 0.224 made Figure 12. It shows the relative distribution of EM noise
1 100 000.00 0.147 0.099 0.117 emission rates in the same antenna positions and at the same
RMS 0.433 0.210 0.400 speeds. We can see that behind and next to the motor the
difference in distribution is smaller, than in front of the motor.
TABLE 6. Normalized amplitude of electromagnetic emission of 3-phase This can be traced back to the similar driver circuits of the
motor (into of axial)
two motors.
[dBμV/m] PWM
Frequency [Hz] 25% 70% 90%
900 000.00 3.372 2.081 2.409
966 666.67 3.351 1.784 1.933
1 033 333.33 3.860 1.865 1.929
1 100 000.00 3.723 1.872 2.030
RMS 3.339 1.468 2.098
TABLE 7. Normalized amplitude of electromagnetic emission of 3-phase
motor (next to)

[dBμV/m] PWM
Frequency [Hz] 25% 70% 90%
900 000.00 2.270 4.088 0.185 Fig. 12. Relative distribution of EM noise rates
966 666.67 4.283 3.763 0.890
4.403 4.334 0.579 The final evaluation table is Table 8. It clearly shows EM
1 033 333.33 noise emission rates of the 3-phase motor are at least 95 %
1 100 000.00 4.563 4.144 0.455 smaller than emission rates of the 2-phase motor. This result
RMS 3.520 4.082 0.977 can be described with the electronically symmetrical winding
Figure 10 shows the sum of measured amplitudes on the same of the stator. The driver circuit creates a smaller difference
speed. The curves show a similar nature but on 70 % there is between the two motors, because the column „behind” shows
a significant difference. only 37 % decreasing. The last column „next to” explains the
two other columns. Next to the motor both EM sources
40
(winding and the driver circuit) produce the noise but not as
Magnitude [dBuV/m]

30 much as in their “native direction”.

TABLE 8. Results of the evaluation method
20
Ratio of the 3 and 2-phase
in front of behind next to
10 motor

Approximated error percentages -2.720% 0.0087% -0.371%

0
0% 20% 40% 60% 80% 100% Linear value of changes 97.17% 37.22% 63.72%

RMS value of changes 95.48% 37.95% 66.94%

2 phase 3 phase

Fig. 10. Summarized EM emission rates at same speeds

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.
 VI. CONCLUSIONS [3] M. Perotti, F. Fiori, “Software based control of the EMI generated in
BLDC motor drives.” In: 2016 International Symposium on
In this paper we compared two different kind of BLDC Electromagnetic Compatibility - EMC EUROPE, 2016. pp. 417-421.
motors. One of them was a 2-phase motor, the other was a 3- [4] S. Lee, J. Hur, “Detection Technique for Stator Inter-Turn Faults in
phase motor. To understand the differences, we briefly BLDC Motors Based on Third-Harmonic Components of Line
Currents.” IEEE Transactions on Industry Applications. 2017. Vol. 53.
introduced the BLDC motors, their driver circuit, and their No. 1. pp. 143-150.
possible electromagnetic problems. We also made EM [5] H. Kim, J. Hur, “Dynamic Characteristic Analysis of Irreversible
measurements on the two motors, so we include a description Demagnetization in SPM- and IPM-Type BLDC Motors.” IEEE
of our measuring environment, test equipment and evaluation Transactions on Industry Applications. 2017. Vol. 53. No. 2., pp. 982-
method. We also wrote about some of the results and showed 990.
different graphs with which our theory is proven. According [6] G. Vidmar, B. Štibelj, B. Rihtarsic, M. Zagirnyak, D. Miljavec,
“Evaluation of different mitigation techniques for electromagnetic
to literature 3-phase windings are much more symmetrical, so interference caused by common mode voltage in BLDC outer-rotor
they produce less EM emission rates than 2-phase ones if motor.” 2012 15th International Power Electronics and Motion Control
excited by the same currents. We had to rule out the EM noise Conference (EPE/PEMC). 2012. DS1c.3-1-DS1c.3-5.
coming from the driver circuit, but the different antenna [7] J. Park, T.R. Wellawatta, Z. Ullah, J. Hur, “New Equivalent Circuit of
positions could show that the circuit does not play such a big the IPM-Type BLDC Motor for Calculation of Shaft Voltage by
Considering Electric and Magnetic Fields.” IEEE Transactions on
role such as the stator windings do. Industry Applications. 2016. Vol. 52. No. 5. pp. 3763-3771.
[8] Automotive EMC Testing: CISPR 25, ISO 11452-2 and Equivalent
ACKNOWLEDGMENT Standards.
“This research was supported by the European Union and [9] ISO-11452: Road vehicles – Component test methods for electrical
the Hungarian State, co-financed by the European Regional disturbances from narrowband radiated electromagnetic energy – Part
Development Fund in the framework of the GINOP-2.2.1-15- 2: Absorber-lined shielded enclosure Second Edition 2011-11-01.
2017-00090 project, titled „E-mobility from Miskolc: [10] CISPR 16-1-4: Specification for radio disturbance and immunity
measurement apparatus and methods Part 1-4 radio disturbance and
Improvement of Coolant Pump and Engine Cooling Fan immunity measuring apparatus – Antennas and test sites for radiated
Taking into Account the Higher Quality Requirements in disturbance measurements. 3rd Ed. IEC, Geneva, Switzerland (2010).
Electric Vehicles”.” [11] S. Baek, “Optimum shape design of a BLDC motor for electric
continuous variable valve timing system considering efficiency and
REFERENCES torque characteristics.” Microsyst Technologies. 2018. Vol. 24. pp.
[1] F. Krall, H. Gruebler, A. Muetze, “Angle Modulated Switching 4441–4452.
Strategy for Fractional Horsepower BLDC Motors for Improved [12] M. Gliga, S. Andreica, C. Constantinescu, C. Munteanu, “Study of
Electromagnetic Compatibility.” 2019 21st European Conference on Conducted Electromagnetic Emissions in Automotive Applications
Power Electronics and Applications (EPE '19 ECCE Europe). 2019. pp. using an Artificial Network.” 2018 International Conference and
1-9. Exposition on Electrical And Power Engineering (EPE). 2018. pp.
[2] J.E. Makaran, J. LoVetri, “Prediction of conducted RFI emissions in 0493-0497.
BLDC motors for automotive applications.” In: 2001 IEEE EMC [13] T. Rahimi, S. Yousefi Khangah, B. Yousefi, “Reduction EMI due to
International Symposium. Symposium Record. International di/dt and dv/dt DC and AC sides of BLDC motor drive.” The 5th
Symposium on Electromagnetic Compatibility (Cat. No.01CH37161). Annual International Power Electronics, Drive Systems and
2002. pp. 21-25. Technologies Conference (PEDSTC 2014). 2014. pp. 428-433.

Authorized licensed use limited to: KOREA ELECTROTECHNOLOGY RESEARCH INSTITUTE. Downloaded on November 28,2023 at 05:21:00 UTC from IEEE Xplore. Restrictions apply.