Design Analysis and Comparative Study of Hub Motor For An Electric Bike
Design Analysis and Comparative Study of Hub Motor For An Electric Bike
Design Analysis and Comparative Study of Hub Motor For An Electric Bike
net/publication/352808246
Design, analysis and comparative study of Hub motor for an electric bike
CITATIONS READS
10 1,991
4 authors, including:
Sumit Banerjee
Dr. B.C. Roy Engineering College
85 PUBLICATIONS 474 CITATIONS
SEE PROFILE
All content following this page was uploaded by Sandeep Kumar Chawrasia on 11 December 2022.
Design, analysis and comparative study of Hub motor for an electric bike
Sandeep Kumar Chawrasia 1*, Aakash Das 2, Chandan Kumar Chanda 3, Sumit Banerjee4
1
Electrical Engineering department, Indian Institute of Engineering Science and Technology, Shibpur, Howrah,
India.
2
Electrical Engineering department, Indian Institute of Engineering Science and Technology, Shibpur, Howrah,
India.
3
Electrical Engineering department, Indian Institute of Engineering Science and Technology, Shibpur, Howrah,
India.
4
Electrical Engineering Department, Dr. B C Roy Engineering College, Durgapur, India.
*[email protected]
Abstract: At present, the inner rotor or outer rotor BLDC motors are generally used in the electric motorbikes. However, the
outer rotor BLDC motors are fast finding applications in wheels of an electric bike for fewer core losses and higher
efficiency. But still, the speed of the bikes available in the market is quite low or if people go for some more top speed
electric bikes that are very costly. So, in this work, two high speed 36 slots, 24 poles exterior rotor motors are designed for
the high-speed, smooth operation of an electric bike. The design of the Hub motors is done based on the hand calculations
of the design parameters. Then the RMxprt models are designed using those calculated data, and various iterations are
performed for getting the required design. But here, just the final design sheets and plots are shown. Also, the RMxprt
models are converted to Maxwell 2D design model for further analysis and accurate and precise design. This paper consists
of a comparative study of both the designs based on these calculations, results and graphs. This research work gives a clear
concept to the reader for selecting a suitable model according to their applications.
1
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
Michael Faraday IET International Summit: MFIIS-2020, October 03 – 04, 2020, Kolkata, India
(Paper ID: 66)
3. Design parameter calculations
In this section, calculations of various parameters
required for designing both the models are discussed.
Designing of the motors require a lot of considerations for
the stator and rotor parameters, which must be calculated
before starting the designing of the motors. In this section,
the stator and rotor sizing, the air gap length calculations, and
the winding parameter calculations for motor designing are
discussed elaborately. The winding parameter calculations
are the most important of the lot because based on this, the
total number of conductors per slot can be determined, which
has a significant effect on the cost of our machines.
3.1. The stator and the rotor size
ߨ ଶ
ܶൌ ߪ݈ ܦሺͳሻ
ʹ
Where D is the rotor inner diameter. The shear stress
(ߪ) is given as a product of the magnetic loading (B) and the
electric loading (A), hence the formula for shear stress is:
ߪ ൌ ܣܤሺʹሻ
Usually, the value of B and A taken as 0.5 - 0.6T and
20 - 40 A/mm respectively. Considering B = 0.55 T and A =
30 A/mm, using (2) the calculated value of σ will be 16.5
KN/m2. Now taking l = 80mm and using (1) the calculated
value of the rotor inner diameter will be nearly 120mm, i.e.,
D = 120mm. Furthermore, to calculate the stator outer
diameter, the following formula can be used:
ܦൌ ܦ௦ ʹ݈ ሺ͵ሻ
Fig. 1. Flowchart of motor design stages [7]. Where Ds and ݈ are the stator outer diameter and the
air gap length, respectively.
Table 1 Bike specifications 3.2. Air gap length calculation
Specifications values
Gross weight (M) 150 kg With the help of power factor, overload capacity,
pulsation loss, unbalanced magnetic pull, cooling and noise,
Top speed (v) 100 km/hrs. the air gap length can easily be calculated using the formula:
Speeding (0-100 kmph.) In 25 seconds ݈ ൌ ͲǤʹ ʹξ݈ܦሺͶሻ
Tyre size 120 / 80 R 17 Considering D as 120 mm and l as 80 mm in (4), the
calculated value of air gap length is 0.5 mm. Using (3), the
Maximum. Grading (α) 100 value of the stator outer diameter is obtained as Ds = 119mm.
coefficient of rolling resistance (Crr ) 0.01 3.3. Calculation of winding parameter
At the first stage of the design algorithm, the results The main task of winding parameter calculation is
(Motor specifications) of various mechanical calculations finding the total conductors per slot for designing the motors.
and terminologies are tabulated below. However, some other parameters also come into play while
developing the motor models. They are:
Table 2 Motor specifications 3.3.1. Coil pitch: It is greatest integer value of the slots to
pole ratio, and the value must always be 1 or more. Here
Specifications values slots to pole ratio value is 1.5, so coil pitch is taken as 1.
Power 2.4 kW
3.3.2. Conductors per slot: The conductors per slot can
Torque 23 Nm be calculated from the formula, which depends upon the
ratio of total ampere conductors and circumference of air
Speed 1000 rpm gap. The formula is given by:
2
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
Michael Faraday IET International Summit: MFIIS-2020, October 03 – 04, 2020, Kolkata, India
(Paper ID: 66)
ܼܫ else various changes can be made in the initial data and
ܣൌ ሺͷሻ again, the results are verified until the desired output is
ߨܦ௦
Where, Z= total number of conductors, Ds = stator obtained.
outer diameter and I is the current rating. Here the design sheet and the plots for both models
are given below, and the explanations and discussions of
Winding parameter calculation for model-I: these results (plots) and design sheet-data are given in
Putting the value of A as 30 A/mm, rated current I as 20 section-6 and section-7.
Amp and the stator outer diameter in (5), the approximated
value of Z = 467. With 36 slots, the slots per conductor are Zs Table 3 Final design-sheet for Model-I
=12.972. Also for a double layer winding, the total
conductors per slot must be integer and multiple of 2, so it is Rated Data
better to take Zs = 14.
Output Power (kW) 2.4
Winding parameter calculation for model-II:
Similarly, Putting the value of A as 30 A/mm, rated current I Voltage (V): 120
as 33 amperes, and the stator outer diameter in (5), the
calculated value of Z=283. With 36 slots, the approximated Poles 24
value of slots per conductor Zs = 8.
Speed (rpm) 1000
3.4. Sizing of slots
Stator Data
Slot sizing can be determined from the parameters
such as slot fill factor, slot area, area of copper, and the Stator Slots 36
insulation area. The formula that combines the above
Stator Outer Diameter (mm) 129
parameters can be written as:
ሺܣ ܣ ሻ Stator Inner Diameter (mm) 25
ܣ௦ ൌ ሺሻ
ܨ௦ Stator Core Length(mm) 80
Where, As is the slot area, Ac is the copper area, Fs is
Steel Type steel_1010
slot fill factor, and Am is the insulation area.
3.4.1. Slot sizing of model-I: In model-I, considering Diameter of wire (mm) 0.302
strands per conductor as 15 and each strand having a wire
Slot Fill Factor (%) 36.8911
diameter of 0.302 mm, the copper area is calculated by using
the formula: Rotor Data
ܣ ൌ ܼ݊௦ ݀ ଶ ሺሻ
Where n is the number of strands, and d is the Length of Air Gap (mm) 0.5
diameter of the wire. Now putting the values of n, Zs, and d in Rotor Outer Diameter (mm) 175
(7), the copper area is calculated as 46.514 mm2. Ac = 46.514
mm2. Now, putting the slot fill factor as 39% and the Rotor Length (mm) 80
insulation area (Am) as 16 mm2 in (6), the net slot area is
calculated as: As = 161 mm2. Magnet Thickness (mm) 4
According to the slot area value, (slot depth and slot Magnet Type NdFe35
width) can be determined. Hence, on average we can take
slot depth 23 mm, and slot width 7 mm. Full-Load Data
3.4.2. Slot sizing of model-II: In model-II, considering Average Input Current (A) 21.6731
strands per conductor as eight and each strand having a wire
diameter of 0.511 mm and putting the values of n, Zs and d in Specific Electric Loading (A/mm) 32.2338
(7) the copper area is calculated as 37.6 mm2. Ac = 37.6 mm2.
Now, taking the slot fill factor as 43% and the insulation area Windage and Frictional Loss (W) 19.9998
(Am) as 7 mm2 in (7), the net slot area is As = 102 mm2.
Copper Loss (W) 175.532
Similar to model-I, using above calculations, on
average we can take slot depth as 14 mm, and slot width as Total Loss (W) 200.425
7mm [7].
Output Power (W) 2400.35
4. RMxprt design Input Power (W) 2600.77
RMxprt design is the primary step for designing any
motor models. It is a dedicated software package for the Efficiency (%) 92.2936
determination of simulation of electrical machines [8]. Data
Rated Speed (rpm) 999.994
required to create a RMxprt design can be obtained using
initial hand calculation. Then a basic design can be made, Rated Torque (Nm.) 22.9217
and the result of the designed model can be verified and
checked. If it is matching with the desired one, then it is ok, Locked-Rotor Torque (Nm.) 358.216
3
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
Michael Faraday IET International Summit: MFIIS-2020, October 03 – 04, 2020, Kolkata, India
(Paper ID: 66)
Table 4 Final design-sheet for Model-II
4.1. RMxprt curves and Plots for model-I: Rated Data
Output Power (kW) 2.4
Voltage (V): 72
Poles 24
Speed (rpm) 1000
Stator Data
Fig. 2. Plot of efficiency vs. speed.
Stator Slots 36
Stator Outer Diameter (mm) 129
Stator Inner Diameter (mm) 25
Stator Core Length(mm) 80
Steel Type steel_1010
Fig. 3. Plot of output power vs. speed. Diameter of wire (mm) 0.511
Slot Fill Factor (%) 45.2001
Rotor Data
Length of Air Gap (mm) 0.5
Rotor Outer Diameter (mm) 175
Full-Load Data
Average Input Current (A) 36.0075
Specific Electric Loading (A/mm) 27.8681
Windage and Frictional Loss (W) 19.9863
Fig.5. Plot of efficiency vs. speed. Copper Loss (W) 164.274
Total Loss (W) 192.223
Output Power (W) 2400.32
Input Power (W) 2592.54
Efficiency (%) 92.5855
Rated Speed (rpm) 999.656
Fig.6. Plot of output power vs. speed. Rated Torque (Nm.) 22.9292
Locked-Rotor Torque (Nm.) 379.313
5. Maxwell 2D design
After the desired result are obtained from RMxprt
models, the RMxprt designs are converted to their respective
Maxwell 2D design for further analysis like the saturation of
the core, the torque, etc., and to get a more precise result.
Fig. 7. Plot of output torque vs. speed. Here, for both the models, the different plots obtained from
Maxwell 2D design are given below:
4
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
Michael Faraday IET International Summit: MFIIS-2020, October 03 – 04, 2020, Kolkata, India
(Paper ID: 66)
Fig. 10. Plot of torque vs. time for model-1. Fig. 11. Plot of torque vs. time for model-II.
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
Michael Faraday IET International Summit: MFIIS-2020, October 03 – 04, 2020, Kolkata, India
(Paper ID: 66)
6. Comparative analysis 8. References
Based on the two models, the comparative analysis
[1] Design of Brushless Permanent Magnet Motors by J. R. Hendershot,
between the both models can easily be done based on some and T. J. E. Miller.
critical factors. The comparative analysis table is shown [2] Prasun Mishra, Suman Saha, and H. P. Ikkurti, "A Methodology for
below: Selection of Optimum Power Rating of Propulsion Motor of Three-
Wheeled Electric Vehicle on Indian Drive Cycle (IDC)", in
Table 5 Comparative analysis table. International Journal on Theoretical and Applied Research in
Mechanical Engineering [IJTARME], Volume 2, Issue 1, pp. 1, 2013
Parameters Model-I Model-II
[3] Immanuel Alphonse, Dr. S. Hosimin Thilagar, F. Bright Singh,
Rated Voltage(V) 120 72 "Design of Solar Powered BLDC Motor-Driven Electric Vehicle", in
International Journal of Renewable Energy Research, Volume 2, No.
Average Input Current(A) 21.6731 36.0075 3, 2012.
Diameter of wire (mm) 0.302 0.511 [4] Saeed Masoumi Kazraji, Saheb Khanabdal, Ramin Bavil Soflayi,
Mehran Sabahi, "A Micro-Scale Wind Turbine Fed BLDC Motor for
Stator Slot Fill Factor (%) 38.8149 43.4009 Electric Vehicle Drive Application", in International Journal of
Renewable Energy Research, Volume 4, No. 1, 2014.
Total Loss (W) 200.425 192.223
Efficiency (%) 92.3 92.6 [5] Shimon Y. Nof; Wilbert Wilhelm; H. Warnecke (1997). Industrial
Assembly. Springer Science & Business Media. p. 174. ISBN 978-0-
412-55770-5.
As Model-I is designed for higher voltage the input
current will have comparatively lower value. So, for Model- [6] Chunhua Liu, K. T. Chau, and J. Z. Jiang. “ A Permanent-magnet
Hybrid In-wheel Motor Drive for Electric Vehicles”, IEEE Vehicle
I comparatively thinner wire can be used in the winding Power and Propulsion Conference (VPPC), Harbin, China,
which decreases the slot fill factor. Also, Model-I will have September 3-5, 2008.
very less heating effect it increases the overload capacity if
the machine, Hence Model-II will have less overload [7] S. K. Chawrasia, C. Kumar Chanda and S. Banerjee, “Design and
capacity. Analysis of an In-Wheel Motor for an Electric Vehicle," 2020 IEEE
Calcutta Conference (CALCON), Kolkata, India, 2020, pp. 351-355,
DOI: 10.1109/ CALCON49167. 2020. 9106538.
From Fig. 8. and Fig. 9. It can be seen that Model-I is
having comparatively less saturation in the core, this is [8] ANSOFT RMXPRT Manual.
because it has less electric and magnetic loading than that of
Model-II. [9] H. V. Patel and H. Chandwani, "Notice of Violation of IEEE
Publication Principles: Review paper on Torque Ripple Mitigation
Based on the Maxwell 2D design of the two models, Techniques for PMBLDC Motor," 2017 International Conference on
Current Trends in Computer, Electrical, Electronics and
the ripple in toque is much lesser in Model-II than in Model-I, Communication (CTCEEC), Mysore, 2017, pp. 669-674, doi:
this is due to comparatively high numbers of winding. As 10.1109/CTCEEC.2017.8455143.
torque ripple can be reduced by increasing the number of
windings and the number of poles but here both the machines
are of 24 poles hence windings are increased in Model-II [9].
7. Conclusion
In this paper, two motor having high-efficiency and
high-torque are designed using RMxprt and Maxwell 2D.
After the completion of basic motor design using RMxprt,
the Maxwell 2D design is created to provide a more precise
and error-free result. The motor designing is done based on
parameter calculations and assumption which have been
adequately explained. Different graphs of both the models
are plotted and shown clearly in this paper. As shown in
fig.2 and fig.5, the efficiencies of both the models at rated
speed 1000 rpm are 93%. The efficiency is calculated in
RMxprt which neglect few losses and hence the practical
design will have efficiency of nearly 90%, still it is very
high efficiency for this application.
Similarly, based on fig.4 and fig.7, the output torque
characteristics have an inverse relationship with the speed.
Comparative analysis, as shown in Table 5 determines the
difference in the readings of the two designed models based on
some crucial parameters such as rated voltage, average input
current, stator slot fill factor, efficiency, saturation, etc. This is
an In-Wheel motor that can be used in high-speed bikes and
bikes to provide the necessary traction required in them.
6
Authorized licensed use limited to: Indian Instt of Engg Science & Tech- SHIBPUR. Downloaded on October 08,2022 at 15:19:28 UTC from IEEE Xplore. Restrictions apply.
View publication stats