Research on Modeling Method of Conducted Emissions Simulation for DC Brush Motors
Research on Modeling Method of Conducted Emissions Simulation for DC Brush Motors
Citation: Zhang, J. and Lv, X., “Research on Modeling Method of Conducted Emissions Simulation for DC Brush Motors,” SAE Technical
Paper 2020-01-1372, 2020, doi:10.4271/2020-01-1372.
Abstract
the impedance of a certain type of windowed DC brush motor,
T
he conducted emissions of DC motors have been a very and the impedance spectrum of the coil winding is obtained.
important content when testing electromagnetic noise. A second-order LC circuit model is established in PSPICE and
It has certain practical guiding significance to optimize fitted with the measured coil winding impedance. Secondly,
and improve the motor in the design stage through the simula- the commutation time is calculated by the motor parameters,
tion of conducted emissions circuit level. The existing litera- and an accurate model of the motor commutation process is
ture research shows that the simulation results can not reflect established in PSPICE. Finally, the time domain variation of
the test results of the conducted emissions of the motor well the voltage across the 50 ohm resistor in the LISN is obtained
because it is difficult to accurately model the armature winding by the voltage probe, and the voltage data is subjected to
and the commutation process of the motor in the electromag- Fourier operation to obtain the conducted emissions spectrum
netic simulation of the conducted emissions circuit level of of the motor. By comparing the simulation with the test
the DC brush motor. In this paper, an electromagnetic simula- results, it can be considered that the electromagnetic simula-
tion modeling method for conducted emissions of DC motor tion model established by the modeling method used in this
is proposed. The circuit model established in the PSPICE can paper can simulate the actual test results of the conducted
simulate the actual conducted emissions values of DC motor emissions of the motor, which has a good practical reference
by the method. Firstly, an impedance analyzer is used to test value for the motor design stage.
1. Introduction
the winding first, and then equivalent it to a lumped parameter
I
n the process of using the motor, the motor will produce circuit, so that the impedance of the equivalent circuit in the
some electromagnetic interference to the surrounding elec- whole frequency band is consistent with the impedance of the
trical equipment. These electromagnetic interference will measured winding.
cause certain hidden dangers to the control and normal opera- In reference [1-2], a two-stage circuit model is used to
tion of the equipment. It can save time and money for the later determine the parameters by curve fitting, and some experi-
EMC test of DC motor by establishing simulation model of ence values in reference are adjusted to make the impedance
conducted emissions to predict conducted emissions in motor characteristics of the model consistent with that of the
design stage. When conducting conducted emissions test for measured winding. The disadvantage of this method is that
DC brush motor, the test environment can be roughly divided it depends on experience. In reference [3], an impedance
into three parts: 12V DC power supply, two LISN and the model is proposed to simulate the complex impedance of a
motor. Among them, the DC power supply and LISN have motor in the range of 1G Hz. In addition, the skin effect of
corresponding standard models. Therefore, we need to focus the winding and the eddy current effect of the laminated core
on the modeling of DC motor in the simulation. are considered. In order to improve the model, a parameter
Under the excitation of motor commutation, the identification method is proposed in this paper. In reference
frequency spectrum characteristics of armature winding [4], considering some physical characteristics and phenomena
impedance will have a certain impact on the magnitude of related to motor armature, a high frequency modeling
frequency spectrum of conducted emission, so the modeling method for low-voltage DC motor armature is proposed,
of motor coil winding and commutation is the key content of which can better fit the armature test results. In reference [5],
the whole motor simulation. DC brush motor is generally a RLC parallel circuit module is proposed. Through the
multi winding structure, the impedance characteristics of a combination of multiple circuit modules, the number of
single winding will change irregularly with the increase of impedance peaks and troughs is simulated. Through the
test frequency, and there will be a certain number of peaks quantitative calculation of parameters, the impedance
and troughs on the spectrum. At present, a large number of spectrum simulation results of the circuit are well matched
methods used in the literature are to test the impedance of with the measured results.
Downloaded from SAE International by Jilin University, Friday, October 25, 2024
During the normal operation of the motor, the brush FIGURE 1 Armature winding of DC motor
inside the motor and the commutator will have alternating
contact, which will cause part of the windings in the armature
to be regularly short circuited, so in the whole commutator
process, regular current and voltage fluctuations will occur
inside the motor. These fluctuations will be transmitted to
other devices through the power line or space, and then
generate electromagnetic interference to other devices. It is
very important for the final result to simulate the commuta-
tion process of the motor accurately through the circuit model.
In reference [6], by analyzing the commutation voltage
equation of the motor during commutation, the voltage at the
motor terminal in a commutation cycle is calculated by
MATLAB, and the data is imported into PSPICE software for
simulation. In reference [7], a contact resistance model which
changes with time is constructed to simulate the change of
impedance when the motor is reversing, and then the voltage
fluctuation when the motor is reversing is simulated. In refer-
ence [8], the process of winding short-circuit is simulated by
building some switch models above the winding, but the
commutation burr caused by contact problem is not fully
considered. In reference [9], a commutation mathematical
model based on ANSOFT MAXWELL finite element simula-
tion is established. The model is applicable to all types of DC
brush motor. While calculating commutation current data, it
provides a lot of information related to commutation, which
is convenient and fast. In reference [10], the short circuit of
each winding in the commutation of DC motor is analyzed,
and the commutation circuit schematic diagram of the motor
is established by using Maxwell, and the current fluctuation
in the commutation of the motor is obtained.
© SAE International.
This paper is based on the research of previous literature,
through the accurate modeling of motor armature winding,
using the motor parameters to calculate the motor commuta-
tion time. In PSPICE, the switching element is used to simulate
the voltage fluctuation and high frequency noise when the
motor changes direction. The simulation results can fit the
actual test results well. The winding mode of the motor used in this test is shown
in Figure 2. It can be seen from the figure that there is a
complete winding between the two adjacent commutators.
Before the test, disconnect the winding connection on
2. Motor Winding the adjacent commutating piece, then select two joints of a
winding, and test the impedance of the winding by using the
Modeling impedance analyzer of stabco. The test layout and test results
are shown in Figure 3.
As shown in Figure 1, the motor armature used in this simula- Through the impedance spectrum of the winding, it can
tion has 8 windings, and the winding mode of 8 windings is be seen that the impedance of the winding presents a trend
single winding.
When modeling the armature, the focus of modeling is FIGURE 2 Connection mode of motor winding
on the modeling of single armature winding. According to
the connection mode of the winding, the armature model of
the DC brush motor is constructed by a single winding model.
FIGURE 3 Test experiment arrangement and test results FIGURE 4 LC series circuit model and impedance diagram
© SAE International.
1
f re = (1)
2π LC
be determined. Based on the parallel connection of L1 and FIGURE 7 Comparison between simulation results of circuit
C1, the value of R1 can be adjusted to ensure that the imped- model and test results
ance value of simulation calculation at the resonance point is
consistent with the actual test results. By calculating the
impedance of LC circuit in series at 24 MHz, the impedance
of resonant element structure at 24 MHz is determined, and
the value of R1 is determined to be about 8 Ω. The specific
calculation formula is as follows.
1 1
Z ( w ) = jwL + +R+ (2)
jwC 1 1
+ jwC1 +
R1 jwL1
© SAE International.
ance is basically consistent with the test results. In the whole
process of building the model, there is a theory for the value
of the components, and then it is compared with the actual
test results. Finally, the parameters of the electrical compo-
nents in the model are slightly adjusted. The advantage of this
method is that it can be used to model the windings of other
types of DC brush motors.
FIGURE 6 The coil winding model
Parameters Values
the gap between the motor and the shell can not be ignored.
R, R1, R2 0.2Ω, 8Ω, 1gΩ
It is necessary to test the parasitic capacitance between the
L, L1, L2 105nH, 2.7nH, 100H
motor winding and the shell to determine the value of the
C, C1, C2 230nF, 15nF, 1e-20nF parasitic capacitance.
Downloaded from SAE International by Jilin University, Friday, October 25, 2024
FIGURE 8 High frequency winding model FIGURE 9 Parasitic capacitance test and test results
© SAE International.
FIGURE 10 Comparison of test and simulation results FIGURE 11 Relative position of electric brush
and commutators
© SAE International.
commutation cycle.In the actual commutation of motor, the
commutator is moving. In order to facilitate understanding,
the commutation process of the motor is simplified as brush
© SAE International.
movement in the picture, and the brush and the commutation
plate are simplified as rectangles. The numbers in the rectan-
gular box represent the angle of arc relative to the center of
the rotor shaft.
The four figures in the figure correspond to four states
when reversing:
3.1. Determination of Motor
Commutation Time 1. At the beginning, the brush did not enter the
commutation (0 °)
During the commutation of motor, the winding of motor will 2. The brush is ready to enter the commutation (7 °)
be short circuited regularly with the increase of time in a
3. The brush enters the commutation (34 °)
commutation cycle.
During the commutation of motor, the winding of motor 4. End of commutation (4 °)
will be short circuited regularly with the increase of time in
The motor is of 8-winding structure, and the time corre-
a commutation cycle.The motor is of 8-winding structure,
sponding to 45 ° brush movement in the above process is a
with a set of symmetrical brushes and 8 commutators in a
commutation cycle. In a commutation cycle, when the brush
ring.When the brush of the motor is between the adjacent
moves 34 °, the two windings are in a short-circuit state, and
commutators, the electric potential of the two adjacent
when the commutation plate moves 11 °, the corresponding
commutators is the same, and two windings will be short
time is not in a short-circuit state.At this time, the rotation
circuited. When the brush is completely on the commutator,
speed of the motor is 3957r / min. After calculation, the
the armature will not be short circuited.
commutation period is about 1.8ms, in which the short-circuit
The summary of relevant parameters used to calculate
time is 1.4ms and the non short-circuit time is 0.4ms.
the short-circuit time of motor reversing is shown in Table 2.
In the actual motor, both the commutator and the brush
are arc-shaped, and there is an arc angle corresponding to the
center of the rotor shaft.Considering that the arc angle of the
3.2. Motor Commutation
brush relative to the shaft center is smaller than the arc angle Circuit Model
of the commutator, there is a certain period of time in the For the DC brush motor used in this simulation, the power
operation of the motor when eight windings are energized. on of each winding in the motor in a commutation cycle is
When the brush is between two commutators, the armature shown in Figure 12.
winding will be short circuited. In order to calculate the short The signal winding in Figure 12 can be replaced by the
circuit time of the armature in the commutation process, the winding model above.According to the above figure, it can
following position diagram can be drawn for the relative be determined that the 8-winding model is adopted in the
position of the brush and the commutation plate. simulation.When the motor winding is in the non short circuit
In Figure 11, the top black is the electric brush, and the state, two 4-windings in series are connected in parallel;When
bottom is the commutator. Four pictures represent a a short circuit occurs, two series 3 windings are connected in
parallel.In this process, the short-circuit time of motor
winding is a very important data, which has a great impact on
TABLE 2 Parameters related to motor commutation the final simulation results.According to the above circuit
Parameters Values
model and the armature short circuit time calculated above,
the circuit model is established in PSpice, as shown in Figure 13.
Rotor speed (rad/min) 3957
In PSpice, by adding switch elements TCLOSE and
© SAE International.
FIGURE 12 Four phases in DC motor commutation take a very short time and are easy to produce high-frequency
noise in the circuit.
There are many reasons for these burrs.On the one hand,
when the brush contacts the next commutator, the release of
electromagnetic energy in the circuit will lead to commutator
sparks, which will show irregular voltage fluctuations on the
commutator voltage, and the time interval is very small.On
the other hand, it may also be due to poor contact between
the brush and the reversing plate, mechanical friction and
other factors.From the reasons mentioned above, it is very
difficult to model the circuit of the burrs produced in the
course of commutation.
Refer to the armature short circuit model above, and add
some switches to simulate these burrs. To simulate the genera-
tion of voltage sudden change burrs in the whole commutation
process. In the simulation model, different short-circuit time
intervals can be set for the other three groups of windings.
© SAE International.
4. Circuit Model
Simulation Results
The conducted emissions results of the motor are obtained by
3.3. High Frequency CISPR25 voltage method. Figure 15 is the layout of conducted
Excitation in the emissions test for DC brush motor.
In Figure 15, the voltage source is a 12V battery, and the
Commutation positive and negative lines of the battery are connected with
In the process of DC motor commutation, there will be a lot two lisns respectively.During the operation of the motor, the
of high-frequency noise in the voltage waveform of its voltage fluctuation caused by commutation and other factors
harness, except the f luctuation of commutation voltage is transmitted to LISN through the harness.The receiver is
obtained by theoretical calculation.There are many tiny connected with LISN output terminal through coaxial
voltage burrs in the voltage time domain. These burrs usually shielded cable. Figure 16 shows the actual test result report.
FIGURE 15 Test diagram of conducted emissions FIGURE 18 Impedance simulation spectrum of LISN
equivalent circuit model
© SAE International.
© SAE International.
FIGURE 16 Motor conducted emissions test results
FIGURE 20 Simulation results of conducted emissions FIGURE 21 Comparison of simulation and test
© SAE International.
© SAE International.
© 2020 SAE International. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE International.
Positions and opinions advanced in this work are those of the author(s) and not necessarily those of SAE International. Responsibility for the content of the work lies
solely with the author(s).
ISSN 0148-7191