Magnetic Properties of Soft Magnetic Powder Composites at Higher Frequencies in Comparison with Electrical Steels

Magnetic Properties of Soft Magnetic Powder Composites at Higher Frequencies in Comparison with Electrical Steels Andreas Schoppa, Patrice Delarbre, E...
Author: Irene Park
16 downloads 2 Views 304KB Size
Magnetic Properties of Soft Magnetic Powder Composites at Higher Frequencies in Comparison with Electrical Steels Andreas Schoppa, Patrice Delarbre, Elmar Holzmann, Maximilian Sigl PMG Füssen GmbH Hiebelerstr. 4, 87629 Füssen, Germany e-mail: [email protected]

Abstract—To ensure a fair comparison of magnetic properties between electrical steels and SMC the measurements should be done on samples with a similar geometry. At a certain transition point of frequency the specific core losses of SMC become lower then those of electrical steels. Thus, the application of SMC instead of electrical steels in electrical machines operating at elevated frequencies can improve their efficiency. Keywords—SMC; soft magnetic powder composites; high frequency; 3D magnetic flux; core loss; permeability

I.

INTRODUCTION

Driven by requirements of e-mobility, electric motors bring new challenges to the traditional laminated motor construction to reach high power density. It is well known that traditional laminated motor construction is limited to 2D magnetic flux to minimize losses in the direction perpendicular to the steel lamination. Because soft magnetic powder composites (SMC) can be used for a 3D magnetic flux path, it is an ideal solution for applications such as transverse flux motors [1].

developed exclusively for quality reasons and ignores the negative influence of manufacturing process of electric components on their magnetic properties. To correlate these properties measured on Epstein samples with a real motor the designers apply various correction factors for different types of electric machines as a solution to this problem. In contrast, the magnetic properties of SMC are typically measured on compacted rings, considering all magnetizing directions and thus correlating with real motors. III.

MANUFACTURING PROCESS OF MAGNETIC COMPONENTS

It is a matter of common knowledge that the magnetic properties of electromagnetic components depend strongly on their manufacturing process [3]. 2.0 Epstein sample

1.8

Epstein sample air-gap corrected loose laminations

Powder metallurgical manufacture has the singular ability to produce near net shaped products (gear box parts, motor parts) for the automobile industry. SMC materials coupled with the P/M production process open new possibilities in the design and manufacture of parts for electrical applications. II.

METHODS OF MAGNETIC MEASUREMENTS

For magnetic measurements of electrical steels the Epstein method is widely used. This standardized method is applied on sheet samples which are cut longitudinal and perpendicular to the rolling direction of the steel strip. The Epstein method was

1.6

[T]

automatically stacked core

magnetic polarisation J

Also, in high frequency alternating magnetic fields, SMC materials act as an insulator to the eddy current and thus provide overall low iron losses. The optimal use of SMC can increase the power density of the electric machine [2]. For electrical steel based asynchronous and transversal flux machines the power density, expressing power which can be performed by a certain size of the motor, can reach values of 0.1 – 7.0 kW/dm³. According to the own expertise these values can be substantially increased by the use of SMC in transversal flux machines and by optimization of design.

1.4

rivet core aut. stacked in Al-frame

1.2

1.0

0.8

0.6

0.4

0.2

0.0 10

100

1000

magnetic field strength H

10000

[A/m]

Fig. 1. Influence of manufacturing steps on the magnetizing behaviour of 1.1 kW asynchronous motor [3].

In the case of electrical steels this process includes punching, various assembling methods of laminations (automatic stacking, riveting, welding…) and pressing into the motor frame. These processing steps cause an interior deformation of the material resulting in deterioration of magnetic properties. An example for this deterioration after various steps of manufacturing process is presented in Fig. 1 and Fig. 2.

M270-35A and NO20 were compared with SMC-grades Siron® S280b, Siron® S300b, Siron® S360, Siron® S400b, Siron® S720 manufactured by PMG Füssen GmbH. The results at J=1T are presented in Fig. 6. 2.0

1.8

10

1.6

magnetic polarisation J [T]

Epstein sample loose laminations 8

automatically stacked core

[W/kg]

rivet core aut. stacked in Al-frame

specific core loss P s

6

4

1.4

1.2

1.0

0.8

M330-35A, Epstein, 50 Hz

0.6

M330-35A, Epstein, 5000 Hz

0.4

M330-35A toroidal; 50 Hz

0.2

M330-35A toroidal; 5000 Hz

0.0 10

100

1000

10000

100000

magnetic field strength H [A/m] 2

Fig. 3. Influence of sample geometry (Epstein vs. toroidal) on the magnetizing behaviour of electrical steel grade M330-35A.

0

0.45

0

0.5

1

magnetic polarisation J

1.5

2

[T]

Fig. 2. Influence of manufacturing steps on the specific core loss of 1.1 kW asynchronous motor [3].

IV.

MAGNETIC LOSSES OF SMC IN COMPARISON WITH ELECTRICAL STEELS AT ELEVATED FREQUENCIES

The distribution of magnetic losses in SMC cores deviates from the behaviour in laminated cores because of a different structure of ferromagnetic material components. The hysteresis losses of SMC are higher and the eddy-current losses result from the internal losses within the particles. This general characteristic is illustrated in Fig. 5 [4]. The specific core loss of standard electrical steel grades used traditionally at elevated frequencies: M330-35A,

M330-35A, Epstein

0.35

M330-35A; toroidal

0.3

P s / f [J/kg]

To ensure a fair comparison between the magnetic properties of electrical steels and SMC, the measurements were done on samples with the same geometry (toroidal samples; OD = 55 mm, ID = 45 mm, thickness = 5 mm). The influence of geometry on magnetic properties of electrical steel grade M330-35A is presented in Fig. 3 and Fig. 4. Additional measurements on different grades with thicknesses of 0.200.35 mm confirm this tendency.

0.4

0.25

0.2

0.15

0.1

0.05

0 0

1000

2000

3000

4000

5000

f [Hz]

Fig. 4. Influence of sample geometry (Epstein vs. toroidal) on the specific core loss of electrical steel grade M330-35A.

P/f

The frequency value of the transition point depends on the nominal thickness of comparable electrical steel and can vary for the typical commercial grades between 500 Hz…1500 Hz. So, the application of SMC becomes interesting for machines operating at elevated frequency or for machines with a substantial amount of higher harmonics. Additionally, Fig. 7 gives an overview of core losses at various frequencies and at various values of polarization for two selected materials (SMC and electrical steel M330-35A). 10000

1000

f Fig. 5. Frequency behaviour of SMC and electrical steels; general view [4].

Ps [W/kg]

100

10

1

M330-35A, 5000 Hz

0.5

SIRON® S360, 5000 Hz

NO20

M330-35A, 500 Hz

M270-35A

0.1

SIRON® S360, 500 Hz

M330-35A 0.4

M330-35A, 50 Hz

SIRON® S280b

SIRON® S360, 50 Hz

SIRON® S300b

0.01 0

SIRON® S360

0.5

1

1.5

2

J [T]

0.3

P s / f [J/kg]

SIRON® S400b

Fig. 7. Comparison of frequency behaviour between electrical steel grade ® M330-35A and SMC-grade SIRON S360.

SIRON® S720

Below the transition point of frequency electrical steels have lower specific core losses in the whole range of magnetic polarization then SMC. Above this point (approx. 500 Hz - if compared SIRON® S360 with M330-35A) the specific core losses of SMC become lower. This can be observed in the whole range of magnetic polarization as well.

0.2

0.1

V.

MAGNETIZATION PROCESS OF SMC IN COMPARISON WITH ELECTRICAL STEELS

0 0

1000

2000

3000

4000

5000

f [Hz]

Fig. 6. Comparison of frequency behaviour between selected SMC grades and electrical steels at J=1T.

The magnetizing behavior (J vs. H) was determined as well. The results are presented in Fig. 8. The magnetization process of SMC is hindered by the typical structure of SMC including pores (Fig. 9) [5]. The electrical steels have a “smooth” microstructure with some minor impurities but without a typical porosity. The porous

microstructure of SMC and the resulting lower density of ferromagnetic element iron (Fe) in comparison with electrical steels is a reason for lower permeability of SMC.

µ



1.6

=

1 (1)

l 1 + L µr l Fe

NO20

where:

M270-35A

1.4

µ* - overall permeability of the magnetic circuit, µ r - relative permeability of soft magnetic material, lL - length of air gap, lFe - length of (soft) magnetic path (e.g. Fe).

M330-35A SIRON® S280b

magnetic polarisation J [T]

1.2

SIRON® S300b

According to this equation the difference between the permeability of SMC and of electrical steel becomes negligible with increasing length of the air gap. For typical transversal flux motors and machines with permanent magnet excitation the resulting air gap is substantially higher then for e.g. asynchronous machines, decreasing the significance of permeability of applied soft magnetic materials.

SIRON® S360

1

SIRON® S400b SIRON® S720

0.8

0.6

2000

0.4

M330-35A

1800

M270-35A 0.2

1600

0

NO20

1400 100

1000

10000

100000

magnetic field strength H [A/m]

Fig. 8. Comparison of magnetizing behaviour between grades and electrical steels at f = 5000 Hz.

selected SMC

relative permeability µ r

10

1200

1000

800

600 ®

SIRON Smaterials

400

200

0 10

Fig. 9. Microstructure of soft magnetic composites after compacting and curing [5].

The designers however have to decide whether this fact is relevant for the calculated magnetic circuit because of the resulting ratio between the iron path and the air gap. Fig. 9 shows the permeability of tested materials “as measured” and Fig. 10 under consideration of an air gap (1 mm) according to the simplified equation (1) [4]:

100

1000

10000

100000

magnetic field strength H [A/m] Fig. 10. Comparison of permeability of selected electrical steels and SMC ® materials (SIRON manufactured by PMG Füssen GmbH) at f=1000 Hz .

Additionally, the advantage of higher permeability of electrical steels becomes substantially reduced with increasing frequency. Thus, for the appropriate design of electric

machines a combination of all factors have to be considered: operating frequency, permeability and the air gap of the magnetic circuit, to achieve the best possible performance.

dimensional design solutions with minimal iron losses and optimized copper winding. The on-going development in the area of soft magnetic composites proceeds as follows: • • •

2000 M330-35A 1800

M270-35A



NO20

1600



SIRON®S720

relative Permeability µ r

1400

SIRON®S280b

These improvements are accomplished through optimisation of the compacting and curing process as well as the addition of special binders or lubricants.

1200

1000

VII. CONCLUSSIONS 800

600

400

200

0 10

100

1000

10000

100000

magnetic field strength H [A/m]

Fig. 11. Comparison of permeability of selected electrical steels and SMC ® materials (SIRON manufactured by PMG Füssen GmbH) at f=1000 Hz under consideration of 1 mm air-gap according to equation (1) .

Electrical machines with three-dimensional magnetic flux are needed for high efficiency motor applications. At the same time, new applications requiring high operating frequencies are becoming more relevant and available. Soft magnetic powder composites are the upcoming development in the powder metallurgy offering optimal magnetic properties at elevated frequencies and contributing to the increase of the power density and to miniaturization of electric machines. This makes SMC perfect for applications with limited space e.g. in the automotive industry, robotics or selected home appliances. In these fields of electrical applications SMC can even outperform the commercially available electrical steels. VIII. [1]

[2]

VI.

SMC AS ALTERNATIVE MATERIAL FOR ELECTRIC APPLICATIONS

In comparison with widely used electrical steels, SMC have advantages making them suitable for special constructions of electric machines. These advantages are: • • •

improvement of magnetising behaviour improvement of saturation polarisation shifting of transition point of the eddy-current loss (see Fig. 6) to lower the values of frequency optimal choice of application according to the relevance of permeability (see Fig. 11) improvement of mechanical strength.

High power density by 3D magnetic flux conduction Lower core losses at elevated frequencies compared with electrical steel Good formability; complex shapes can be directly compacted without destroying the material structure and resulting deterioration of magnetic properties.

Since the magnetic cores get their final shape after compacting and their final magnetic and mechanical properties after curing, they can be immediately wound with wires and assembled into the motor frame. This enables the magnetic core manufacturer to scale the design, and simplify both the core winding geometry, and the motor manufacturing process. The isotropic natures of the SMC material combined with the net shaping possibilities allow us to introduce new three-

[3]

[4]

[5]

REFERENCES

Ola Anderson, Paul Hofecker, “Advances in Soft Magnetic Composites – Materials and Applications”, Advances in Powder Metallurgy & Particulate Materials – 2009; Proceedings of the 2009 International Conference on Powder Metallurgy & Particulate Materials, Las Vegas. A. Schoppa, P. Delarbre, A. Schatz, “Optimal use of soft magnetic powder composites (SMC) in electric machines”, Proceedings of the 2013 International Conference on Powder Metallurgy & Particulate Materials, Chicago, to be published A. Schoppa, “Influence of manufacturing process on the magnetic properties of non-oriented electrical steel”, doctoral thesis, Aachen, Germany, 2001. R. Boll, „Soft Magnetic Materials” (German: „Weichmagnetische Werkstoffe“), Page 87 and 101, ISBN 3-8009-1546-4, Vacuumschmelze Hanau, Germany, 1990. P. Delarbre, E. Holzmann, A. Schoppa, PMS500 – a modern soft magnetic composite material for electric machines, 5th International Conference on Magnetism and Metallurgy WMM’12, Proceedings, Page 326 – 334, 2012, Ghent, Belgium.

Suggest Documents