See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321017452

Air-Gap Heat Transfer in Rotating Electrical Machines: A Parametric Study

Article  in  Energy Procedia · August 2017

DOI: 10.1016/j.egypro.2017.12.343

CITATION READS

1 205

2 authors:

Md Lokman Hosain Rebei Bel Fdhila

Mälardalen University Mälardalen University
9 PUBLICATIONS   40 CITATIONS    29 PUBLICATIONS   416 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Air Flow in Large machines View project

All content following this page was uploaded by Md Lokman Hosain on 07 February 2018.

The user has requested enhancement of the downloaded file.

 Available online at www.sciencedirect.com

AvailableScienceDirect
Availableonline
onlineatatwww.sciencedirect.com
www.sciencedirect.com

ScienceDirect
ScienceDirect
Energy Procedia 00 (2017) 000–000
www.elsevier.com/locate/procedia
Energy
EnergyProcedia 142
Procedia 00(2017)
(2017)4176–4181
000–000
www.elsevier.com/locate/procedia

9th International Conference on Applied Energy, ICAE2017, 21-24 August 2017, Cardiff, UK

Air-Gap Heat Transfer in Rotating Electrical Machines: A

The 15th International Symposium on District Heating and Cooling
Parametric Study
Assessing the Mdfeasibility
Lokman Hosain of b,a,
using
*, Rebei theBelheat Fdhilademand-outdoor
a,b

temperature function for a long-term district heat demand forecast

ABB AB, Corporate Research, SE - 721 78, Västerås, Sweden
a
b
Mälardalen University, School of Business, Society & Engineering, P.O. Box 883, SE-721 23, Västerås, Sweden
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc
Abstract
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
c
More than halfDépartement Systèmes
of all electrical Énergétiques
energy et Environnement
is consumed by motors -and IMTgenerators
Atlantique, in
4 rue
an Alfred Kastler, 44300
industrialized Nantes,
country. France
About 5-25% of this
energy is lost and converted to heat. This heat produced by the losses has adverse effect on the lifetime and performance of a
machine. A machine has to be operated at a given temperature to achieve maximum efficiency, therefore heat transfer study of
machines is of special interest to rotating machines manufacturers. In this paper we investigate the heat transfer in the air-gap
Abstract
between the rotor and the stator of a simplified induction motor using Computational Fluid Dynamics. We consider three different
air-gap widths and rotation speeds to explore the change in air-gap heat transfer when changing the air-gap width and the rotation
District
speed. Theheating networks
simulated are heat
average commonly
transferaddressed
coefficientsin the
for literature as oneare
all the models of inthegood
mostagreement
effective solutions for decreasing
with the correlations fromthe
greenhouse
published gas emissions
literature. from the building
The Taylor-Couette sector.
vortical flowThese systems
pattern requireinhigh
is observed investments
the air-gap in ourwhich are returned
simulation results through the heat
for the models
sales.
with Due
large to the The
air-gaps. changed climate
numerical conditions
results show thatand building of
the presence renovation policies,
Taylor-Couette heat in
vortices demand in the
the air-gap futurethe
enhance could decrease,
heat transfer.
prolonging the investment return period.
The heat transfer coefficient increases with the increase in the rotation speed and decreases with the decrease in the air-gap width.
The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand
©forecast.
2017 TheThe districtPublished
Authors. of Alvalade, locatedLtd.
by Elsevier in Lisbon (Portugal), was used as a case study. The district is consisted of 665
buildings that vary in both construction period
Peer-review under responsibility of the scientific committee and typology.
of theThree weather scenarios
9th International (low,onmedium,
Conference Appliedhigh) and three district
Energy.
renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were
comparedAir-gap;
Keywords: with results
Rotatingfrom a dynamic
electrical heatCFD
machine; demand model,
simulation; previously
Thermal developed
analysis; and validated
Taylor-Couette flow by the authors.
The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
(the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation
1.scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered).
Introduction
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the
decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and
Motors and
renovation generators
scenarios are oneOn
considered). of the
the other
largest energy
hand, consumers
function interceptinincreased
the worldfor [1]. Several studies
7.8-12.7% highlighted
per decade (depending that
onthe
the
energy
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered,[3].
consumptions by motors and generators amount to between 60-80% within the European Union (EU) [2], and
The use of
improve themotors
accuracyand generators
of heat demandare increasing day by day, thus the demand of energy is also increasing from this
estimations.
sector. To meet the increasing demand, the energy providers often need to use the non-renewable energy resources
which
© 2017increase the greenhouse
The Authors. Published bygas emission
Elsevier Ltd. [4]. Therefore, energy savings from motors and generators in industrial
and commercial
Peer-review sectors
under becameofa the
responsibility keyScientific
focus in Committee
many EU countries [5].
of The 15th An increase
International in the efficiency
Symposium of Heating
on District standardand
motors
Cooling.
and generators by 2% will reduce the energy consumption by 25% [6].
Keywords: Heat demand; Forecast; Climate change

* Corresponding author. Tel.: +46 760 98 30 82

E-mail address: md.lokman.hosain@mdh.se
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
1876-6102 © 2017
Peer-review underThe Authors. Published
responsibility by Elsevier
of the Scientific Ltd. of The 15th International Symposium on District Heating and Cooling.
Committee
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy .
10.1016/j.egypro.2017.12.343
2 Md Lokman Hosain and Rebei
Md Lokman Bel et
Hosain Fdhila / Energy
al. / Energy Procedia
Procedia 00(2017)
142 (2017)4176–4181
000–000 4177

Nomenclature

𝐷𝐷ℎ =2g, Hydraulic diameter (m) Tacr Critical Taylor number

g =r2-r1, Gap size (m) Nu Nusselt number
k Thermal conductivity of air (W/m-K) Nuavg Average Nusselt number
𝑟𝑟1 Outer radius of the rotor (0.075m) 𝜈𝜈 Kinematic viscosity of air (m2/s)
𝑟𝑟2 Inner radius of the stator (m) Ω Angular velocity of the rotor (rad/s)
𝑟𝑟1 +𝑟𝑟2
𝑟𝑟𝑚𝑚 = , Mean radius (m) Ω𝑐𝑐𝑐𝑐 Critical angular velocity (rad/s)
2
Ta Taylor number Pr Prandtl number

About 5-25% of total electrical energy consumed is lost during the machine operation [7]. This lost energy is then
converted into heat which increases the temperature inside the rotating electrical machine. The rise of temperature has
a great impact on the performance and lifetime of a machine. A machine must be operated under a given temperature
to achieve a specified efficiency and maintain its lifetime. Therefore, the heat transfer inside the machine is of special
interest to the motors and generators manufacturers. To operate rotating machine, variable speed drivers are most often
used which control the speed of a machine based on load demand to achieve higher efficiency. The major heat transfer
occurs in the air-gap between the rotor and the stator which is mainly driven by the air flow pattern. Therefore, it is
crucial to understand the relationship of air-gap width and rotor’s rotation speed to the heat transfer to optimize
machine design, to reduce the losses, to achieve proper cooling and to operate the machine at an optimal condition.
Air-gap heat transfer has been subject to many other numerical and experimental studies both for the enclosed
machines and the open machines [8]. The airflow inside an enclosed and an open machine is known as Taylor-Couette
flow and Taylor-Couette-Poiseuille flow respectively. In this article we consider an enclosed machine geometry thus
we focus only on Taylor-Couette type flow. Use of Computational Fluid Dynamics (CFD) to study the air-gap heat
transfer numerically are very few compared to the experimental studies [8]. Romanazzi and Howey [9] and Anderson
et al. [10] presented CFD simulation results to analyze the heat transfer and also proposed correlations to calculate
heat transfer coefficients. Previous experimental investigations provide correlations to calculate heat transfer
coefficients for a broad range of air-gap width and operating conditions [8], [11].
In a preceding investigation [12], we studied the transient heat transfer inside the annulus air-gap of a rotating
electrical machine. The main focus in [12] was to investigate the transient behavior of the airflow and its influence on
the heat transfer while taking into account the turbulence effect from the rotor wafters. In this paper, we present a
parametric study based on air-gap width and rotor’s rotation speed for the same model geometry as in [12] resembling
a low voltage induction motor. Such numerical parametric studies are rare in the available published literature. In this
study, we present the numerical results and discuss the changes in heat transfer behavior while changing the chosen
parameters to understand the relationship between each parameter and their overall impact on the heat transfer.

2. Numerical domain and operating conditions

The simplified numerical model presented in Fig. 1a represents a low voltage induction motor which mainly
consists of a smooth cylindrical rotor together with wafters and the internal surfaces of the stator and the housing. The
air-gap width in general is different for different machines depending on the size and power ratings of the machine.
Fig. 1b presents a 2D cross section of the domain with all the dimensions. The model geometry is parametrized to
study three different air-gap widths of 1mm, 3mm and 5mm where rest of the dimensions are kept exactly same. The
rotation speed of the rotor is also varied to 1500rpm, 5000rpm and 10000rpm for all the three air-gap widths to
investigate the impact of air-gap width and rotation speed on the heat transfer. The meshes (Table 1) for all the three
models consist only hexahedral structured cells where the cells close to the solid walls are refined to fully resolve the
boundary layer (Fig. 1c). In total nine different simulations (Table 2) are performed and the results are compared and
validated using available correlations from experimental data. To model the rotating machine efficiently the sliding
4178 Md Lokman Hosain et al. / Energy Procedia 142 (2017) 4176–4181
Md Lokman Hosain and Rebei Bel Fdhila / Energy Procedia 00 (2017) 000–000 3

mesh technique is used where a moving air zone rotates at a given rpm together with the rotor surfaces and a stationary
air zone stays stationary together with the stator and the housing surfaces. The stator’s internal surface and the housing
surfaces are modelled as stationary surface where a virtual wall thickness is considered to allow conjugate heat
transfer. The boundary conditions for all the surfaces in Fig. 1a are explicitly presented in Table 3.

(a) (b) (c)

Fig. 1. Numerical domain and mesh (a) Numerical domain in 3D; (b) YZ cross section plane at X=0 of the numerical domain with dimensions
(Line1 is a line in the air-gap along Y axis); (c) Mesh for the models with 5mm air-gap (other meshes for 1mm and 3mm models are similar)

Table 1. Mesh for three different models with three different gap sizes between rotor and stator
Mesh Mesh 1 (5mm gap) Mesh 2 (3mm gap) Mesh 3 (1mm gap)
Number of cells (Million) 1.3 2.8 5.5

Table 2. Simulated cases for the parametric study

Gap\RPM 1500 rpm 5000 rpm 10000 rpm
5mm gap Case1 Case2 Case3
3mm gap Case4 Case5 Case6
1mm gap Case7 Case8 Case9

Table 3. Boundary conditions for the wall surfaces of the domain (Fig.1a) [12]
Surfaces Wall motion Shear condition Thermal
Rotor, wafters and rotor shaft Moving wall No-slip 150°C constant temperature
Stator’s internal surface Stationary wall No-slip Convective heat transfer coefficient 10W/m2K
Wall thickness: 0.05m
Free stream temperature: 40°C
Surfaces of housing Stationary wall No-slip Convective heat transfer coefficient 10W/m2K
Wall thickness: 0.002m
Free stream temperature: 40°C

The air inside the machine is modelled as incompressible by solving the Reynolds averaged Navier-Stokes (RANS)
model using Finite Volume Method (FVM). To model the air flow and the heat transfer, the mass, the momentum and
the energy transport equations are solved under transient condition in ANSYS Fluent 18.0. To model the turbulence
and the boundary layer the k-ɛ two equation turbulence model and the enhanced near wall treatment are used
respectively.

3. Results and discussion

The non-dimensional Taylor number (Ta), which measures the importance of the centrifugal forces relative to the
viscous forces, mainly determine the flow characteristics inside the cylindrical annulus air-gap. The Taylor number is
calculated from the geometric dimensions, the operating conditions and the fluid properties. The rotation speed of the
 Md Lokman Hosain et al. / Energy Procedia 142 (2017) 4176–4181 4179
4 Md Lokman Hosain and Rebei Bel Fdhila / Energy Procedia 00 (2017) 000–000

rotor (Ω) and the air-gap width (𝑔𝑔)are the two main parameters influencing the flow characteristics. The Taylor
number can be expressed as [13],
𝑇𝑇𝑇𝑇 = Ω2 𝑟𝑟𝑚𝑚 (𝑟𝑟2 − 𝑟𝑟1 )3 /𝜈𝜈 2 . (1)
The air-gap flow are categorised into different flow regimes based on a critical value of the Taylor number
occurring at a critical speed of the rotor. The laminar flow breaks down into Taylor instabilities as a set of many
counter rotating vortices along the axial direction when a critical speed reached. For detail of these flow regimes we
refer to [8]. The Taylor solution for small gap approximation and the experimental study [14] indicate that the Taylor
vortices are most probably present in the flow inside the annulus air-gap when the Taylor number is larger than a
critical value. The critical Taylor number can be defined as [13]:
𝑇𝑇𝑎𝑎𝑐𝑐𝑐𝑐 = 1697𝐹𝐹𝑔𝑔 , (2)
where 𝐹𝐹𝑔𝑔 is defined as
𝜋𝜋4 𝑟𝑟2 −𝑟𝑟1 −2
𝐹𝐹𝑔𝑔 = (1 − ) 𝑆𝑆 −1 , (3)
1697 2𝑟𝑟𝑚𝑚
𝑟𝑟2 −𝑟𝑟1 𝑟𝑟2 −𝑟𝑟1 −1
and S is defined as, 𝑆𝑆 = 0.0571 [1 − 0.652 𝑟𝑟1
] + 0.00056 [1 − 0.652
𝑟𝑟1
] . (4)
The critical angular velocity of the rotor is calculated as,
Ω2𝑐𝑐𝑐𝑐 = 1697 𝐹𝐹𝑔𝑔 𝜈𝜈 2 /𝑟𝑟1 (𝑟𝑟2 − 𝑟𝑟1 )3 . (5)
For 0 < 𝑇𝑇𝑇𝑇 < the average Nusselt number can be calculated from the following correlation [8],
108 ,
𝑁𝑁𝑁𝑁 = 0.42(𝑇𝑇𝑇𝑇. 𝑃𝑃𝑃𝑃)0.25 (6)

(a) (b) (c)

Fig. 2. Non-dimensional velocity profiles on Line1 in the air-gap for different rotation speeds (a) Profiles for models with 5mm air-gap; (b)
Profiles for models with 3mm air-gap; (c) Profiles for models with 1mm air-gap

(a) (b) (c)

Fig. 3. Velocity contours on the YZ cross section plane (Fig. 1b) (a) Velocity contour for model “Case 3”; (b) Velocity contour for model “Case
6”; (c) Velocity contour for model “Case 9”
4180 Md Lokman Hosain et al. / Energy Procedia 142 (2017) 4176–4181
Md Lokman Hosain and Rebei Bel Fdhila / Energy Procedia 00 (2017) 000–000 5

Fig. 2 presents the velocity profiles for all the simulated cases (Table 2). 𝑋𝑋 = −1 and 𝑋𝑋 = 1 in Fig. 2 represent the
stator and the rotor respectively. The velocity profiles are presented in non-dimensional numbers with the same scaling
to be able to compare. The velocity profiles are quite similar for 3mm and 5mm air-gap width however, for 1mm air-
gap width the velocity profiles are almost laminar. For higher rotor speed the profile gets steeper close to the wall and
more flat in the middle of the air-gap. Fig. 3 and Fig 4 presents the velocity contours on the YZ cross section plane
and the temperature contours on the stator’s internal surface for Case 3, Case 6 and Case 9 respectively. The Taylor
vortices in pairs are observed only in 3mm and 5mm air-gap models (Fig. 3a, 3b). Similar effect from the Taylor
vortices are also observed in the temperature contours (Fig. 4a, 4b). However, for 1mm air-gap models there are no
Taylor vortices observed (Fig. 3c) for the considered rotation speeds and thus the temperature on the stator surface is
almost constant (Fig. 4c).

(a) (b) (c)

Fig. 4. Temperature contours on Stator’s internal surface (a) Temperature profile for model “Case 3”; (b) Temperature profile for model “Case
6”; (c) Temperature profile for model “Case 9”

Fig. 5. Nusselt number curve on a line along the length on the rotor surface for 9 different cases (a) Nusselt number for models with 5mm air-gap;
(b) Nusselt number for models with 3mm air-gap; (c) Nusselt number for models with 1mm air-gap
 Md Lokman Hosain et al. / Energy Procedia 142 (2017) 4176–4181 4181
6 Md Lokman Hosain and Rebei Bel Fdhila / Energy Procedia 00 (2017) 000–000

Table 4. Comparison of the average Nusselt number for the rotor surface from simulations with the correlation (6)
Nuavg Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7 Case 8 Case 9
Correlation 11.59 21.15 29.91 7.87 14.37 20.33 3.44 6.28 8.88
Simulation 11.25 21.47 30.12 8.62 14.68 19.94 3.20 5.95 9.21

The average Nusselt number can be calculated from correlation (6). Fig. 5 presents the Nusselt number on a line
along the length of the rotor surface. The heat transfer coefficient increases with the increase in the rotation speed and
the increase in the air-gap width. The Nusselt number curves show a periodic oscillating pattern due to the Taylor
vortices in the flow for 3mm and 5mm air-gap width models (Fig. 5a, 5b). Moreover, the higher the rotation speed the
higher the difference in the maxima and the minima of the Nusselt number curve (Fig. 5a, 5b). For 1mm model the
Nusselt number curve is almost flat with much lower values compared to that of the 3mm and 5mm models (Table 4).
Fig. 5 and Table 4 also indicate that the Taylor vortices enhance the heat transfer inside the air-gap. The average
Nusselt number of the rotor surface for all the models are compared with correlation (6) with an overall good
agreement. The results presented in this paper provide information on the overall heat transfer behavior for different
air-gap widths and rotation speeds which can be used to improve machine design and to achieve better control and
cooling of rotating electrical machines in general.

4. Conclusion

The results presented in this paper illustrate the complex flow inside the air-gap and highlight the importance of
different parameters related to the heat transfer. The models predict the heat transfer coefficient with an overall good
accuracy when compared to correlation derived from experimental data. The simulation results indicate strong
influence of the air-gap width and the rotation speed on the heat transfer. The heat transfer coefficient increases with
the increase in rotation speed and the air-gap width. The results also show that the presence of the Taylor vortices in
the air-gap flow enhance the overall heat transfer.

Acknowledgements
The authors acknowledge the financial support for the project from ABB.

References

[1] R. Hanitsch, “Energy efficient electric motors,” RIO-02 World Clim. Energy Event, January 6-11, 2002.
[2] P. Waide and C. U. Brunner, “Energy-Efficiency Policy Opportunities for Electric Motor-Driven Systems,” OECD Publishing, 2011.
[3] B. C. Mecrow and A. G. Jack, “Efficiency trends in electric machines and drives,” Energy Policy, vol. 36, no. 12, pp. 4336–4341, 2008.
[4] E. A. Abdelaziz, R. Saidur, and S. Mekhilef, “A review on energy saving strategies in industrial sector,” Renew. Sustain. Energy Rev.,
vol. 15, no. 1, pp. 150–168, 2011.
[5] “Commission Regulation (EC) No 640/2009, implementing Directive 2005/32/EC of the European Parliament and of the Council with
regard to ecodesign requirements for electric motors,” Off. J. Eur. Union, 2009.
[6] R. Saidur, “A review on electrical motors energy use and energy savings,” Renew. Sustain. Energy Rev., vol. 14, no. 3, pp. 877–898,
2010.
[7] J. C. Andreas, Energy-efficient electric motors : selection and application. M. Dekker, 1992.
[8] M. Fénot, Y. Bertin, E. Dorignac, and G. Lalizel, “A review of heat transfer between concentric rotating cylinders with or without axial
flow,” Int. J. Therm. Sci., vol. 50, no. 7, pp. 1138–1155, 2011.
[9] P. Romanazzi and D. A. Howey, “Air-gap convection in a switched reluctance machine,” in 2015 Tenth International Conference on
Ecological Vehicles and Renewable Energies (EVER), pp. 1–7, 2015.
[10] K. R. Anderson, J. Lin, C. McNamara, and V. Magri, “CFD Study of Forced Air Cooling and Windage Losses in a High Speed Electric
Motor,” J. Electron. Cool. Therm. Control, vol. 5, no. 2, pp. 27–44, 2015.
[11] D. A. Howey, P. R. N. Childs, and A. S. Holmes, “Air-Gap Convection in Rotating Electrical Machines,” IEEE Trans. Ind. Electron.,
vol. 59, no. 3, pp. 1367–1375, Mar. 2012.
[12] M. L. Hosain, R. Bel Fdhila, and K. Rönnberg, “Taylor-Couette flow and transient heat transfer inside the annulus air-gap of rotating
electrical machines,” Appl. Energy, 2017.
[13] K. Becker and J. Kaye, “The influence of a radial temperature gradient on the instability of fluid flow in an annulus with an inner rotating
cylinder,” J. Heat Transfer, vol. 84, no. 2, pp. 106–110, 1962.
[14] G. Taylor, “Stability of a viscous liquid contained between two rotating cylinders,” Philos. Trans. R. Soc., 1923.

View publication stats