IEEE TRANSACTIONS ON MAGNETICS, VOL. 42, NO. 10, OCTOBER 2006 3389 A New Concept: Asymmetrical Pick-Ups for Inductively Coupled Power Transfer Monorail Systems Grant A. J. Elliott, Grant A. Covic, Dariusz Kacprzak, and John T. Boys Department of Electrical and Computer Engineering, School of Engineering, The University of Auckland, Private bag 92019, Auckland, New Zealand Inductively coupled power transfer systems have recently proven to be popular in moving vehicle monorail systems situated in difficult environments such as clean rooms. In such applications the magnetic design is critical if low weight high power pick-ups are to be realized. Early designs were largely experimental and used magnetic shapes that easily fit the existing structure. Modern finite-element-modeling packages are now being used to achieve significant alterations to the pick-up shape under the assumption that the monorail structures can be changed. This paper introduces several important magnetic design metrics that must be considered in such a design process, and applies these to various newly proposed and unconventional asymmetrical pick-up shapes. These new pick-ups are shown to be capable of achieving comparable power output with lower ferrite volume/length of the pick-up structure, or significant increases in output power for identical volume/length, to a conventional E-shaped pick-up. Index Terms—Electromagnetic coupling, electromagnetic induction, energy conversion, magnetic circuits. I. INTRODUCTION NDUCTIVELY coupled power transfer (ICPT) systems are widely used in monorail transportation areas and are especially important for clean room applications due to their noncontact operation and the ability to seal all electronics ensuring no residues. A typical ICPT system is composed of a primary track and a secondary pick-up [1]. The pick-up is usually attached to a moving unit (bogie) and is designed to travel along the primary track. The majority of monorail ICPT systems are based on E-pick-ups [2] since these pick-up shapes were readily available and easy to fit into existing structures. Three-dimensional finite-element-modeling (FEM) packages (such as JMAG Studio) can now be accurately applied to improve the magnetic design of the pick-up structure, enabling new and improved magnetic structures to be rapidly explored. Recently, an S-pick-up shape was proposed that has significantly higher power transfer capability than a conventional E shape [3], but this design requires significant modification to a monorail support system in which most applications are fitted and these track supports would in many applications interfere with the S pick-up during operation. Fig. 1(a) shows both the conventional E and this S pick-up on a single track. The track support structures are indicated in Fig. 1(b). This paper introduces a number of new asymmetrical magnetic topologies (in addition to the S-shaped pick-up) that can be used on monorail systems and determines magnetic parameters that enable the usefulness of each magnetic design to be compared against the conventional E pick-up. I II. MAGNETIC DESIGN METRICS The analysis of these systems can be undertaken in terms of coupled and uncoupled flux components. The coupled compo- Fig. 1. (a) E and S power pick-ups as positioned along a track section. (b) Cross sections of pick-up and track with added support structures. nent results from mutual flux that links the track and pick-up in the proper manner and is described in coil inductance terms of a coupling factor given by (1) [4] (1) is the mutual flux linking and the track is the where total flux in . The uncoupled flux has no useful purpose, but links both wires of the track together and reduces the power transfer efficiency. This results in reduced flux in the pick-up core and is described here in terms of an inter conductor coupling factor (ICCF). Fig. 2 shows the cross section of an E pick-up and the flux components generated in relation to the current carrying conductors. With one track excited as shown in Fig. 2(a), the can be detotal flux generated by current in conductor A scribed by the following equation: (2) Digital Object Identifier 10.1109/TMAG.2006.879619 0018-9464/$20.00 © 2006 IEEE 3390 IEEE TRANSACTIONS ON MAGNETICS, VOL. 42, NO. 10, OCTOBER 2006 Fig. 3. Analyzed pick-ups and their power. the secondary leakage reluctance paths. As such, the total flux linking the secondary coil can be approximated as Fig. 2. (a) E-core pick-up with only one track conductor excited. (b) E-core pick-up with both track conductors excited. (5) Thus, (1) can be rewritten as where mutual flux linking ; common mode flux linking conductor B; leakage flux that combines all remaining fluxes. A similar expression exists when only conductor B is excited. With both track conductors excited such that conductor B is the return path for current in conductor A, these common mode flux components linking each track and are equal and opposite and effectively cancel [Fig. 2(b)]. In the majority of ICPT systems, the pick-up core sits above an elongated track so that only a small proportion of the total flux along the track intersects the core in some manner. As such, when comparing the relative performance of various pick-up magnetic structures, only the proportion of each of the flux comand that actually intersects the pick-up core, deponents fined here as and , respectively, is of interest. Thus, for example (2) can be rewritten as (6) The reluctance of the total flux the pick-up magnetic structure is given by (7) is the magnitude of the track current and is the Here, number of turns of the track. is The nominal power output of an untuned pick-up given by the product of the secondary coil’s open circuit voltage and short circuit current [2], [4] as (8) (3) [the coupled component of in the core shown in Fig. 2(a)] is a measure of the field that should be available to be coupled to the coil but is lost due to the presence of a common mode reluctance path between conductors A and , this common mode reluctance B. In order to maximize should be made as high as possible and ideally infinite so that is zero. Such a design will maximize the mutual flux when both track conductors are excited. The ICCF for a magnetic pick-up structure can therefore be expressed (using only the coupled flux components) as the proportion of the total coupled flux linking the pick-up core that is canceled with conductors A and B excited out of phase (4) ICCF and are dimensionless quantities (that scale from 0–1) and depend only on the two-dimensional cross-sectional geometry of the magnetic system (as shown in Fig. 2). Neither are substantially affected by the pick-up length (parallel to the track). The reluctance paths of the primary leakage components in ) are essentially the same as the core (giving that couples Here frequency of the track current; mutual coupling between and the track; number of turns of the pick-up coil. and combining (5)–(7) results By definition, in (9) For constant and , then determines the power transfer capability of the pick-up core. III. PICK-UP ANALYSIS A number of new asymmetrical pick-up structures (as shown in Fig. 3) were analyzed along with the conventional E-shaped pick-up using FEM software JMAG Studio to determine how usefully they might be applied to ICPT applications. The track spacing, track current magnitude, and frequency were all identical. All pick-ups were assumed to be constructed from identical sized elements of linear magnetic material having a , a depth of 120 mm, a height [and width from Fig. 1(a)] of 60 mm, and a 10 mm thickness. The Z and U pick-ups use three of these elements while the other pick-ups require five elements ELLIOTT et al.: ASYMMETRICAL PICK-UPS FOR ICPT MONORAIL SYSTEMS 3391 Fig. 5. Fig. 4. , ICCF, and reluctance. in their construction. The secondary coil of each model was assumed to be positioned in the center vertical-standing element directly between the track conductors. The key parameters ( , ICCF, and ) resulting from the analyses are shown in Fig. 4. The calculated power [as determined from (8)] is shown in Fig. 3 alongside each pick-up. These power levels were independently confirmed by measurement in the laboratory using similar sized ferrite cores. There is surprisingly little variation in between the various designs, indicating that the proposed changes to the magnetic structure have a similar effect on both the mutual reluctance and leakage paths. However, significant power improvements result in the S- and -shaped pick-ups predominantly because their respective ICCFs are lower than in the E pick-up. The S pick-up, in particular, has significantly more flux in the outer legs and the ferrite shape ensures that this captured flux naturally couples the secondary coil positioned on the central leg. As shown in Fig. 1(a), this comes at a cost of adding significant complexity to the track support system. The -shaped pick-up is slightly easier to fit in an industrial application but results in a larger ICCF and reduced that is only partially offset by an improved . The Z pick-up is a simplification of both the S and -shaped pick-ups and in comparison with the S has significantly worse , resulting in a loss of half the available output ICCF and power. Despite this, the Z compares favorably with the E-shaped pick-up, as the relative increase in is largely offset by a reduction in ICCF. Similar comparisons can be made between the and U, and the E and U, shaped pick-ups. Ratio of power to volume of ferrite for each pick-up. In practice, it is desirable to reduce the weight of any pick-up on a moving vehicle/bogie; therefore, the power/volume of each pick-up is often a useful comparative measure and these are shown in Fig. 5. Considering these results, the S pick-up is only marginally better than the , Z and U, and consequently the choice of pick-up in these applications may well be determined by how easy it is to construct the system to fit and support the pick-up rather than the power improvement available. IV. CONCLUSION The magnetic design of power pick-ups in ICPT applications has until recently been defined by the application. Modern FEM techniques have made it possible to explore new designs— three of which are introduced in this paper. In order to fairly compare each proposed pick-up’s performance, a set of new design metrics were introduced in this paper. These enable useful insight into the underpinning magnetic relationships and a method of optimizing pick-up design for new and existing ICPT applications. REFERENCES [1] E. Abel and S. Third, “Contactless power transfer—An exercise in topology,” IEEE Trans. Magn., vol. MAG-20, pp. 1813–1815, Sep.–Nov. 1984. [2] J. T. Boys, G. A. Covic, and A. W. Green, “Stability and control of inductively coupled power transfer systems,” IEE Proc. Elect. Power Appl., vol. 147, no. 1, pp. 37–43, Jan. 2000. [3] D. Kacprzak, G. A. Covic, and J. T. Boys, “An improved magnetic design for inductively coupled power transfer system pickups,” in Proc. 7th Int. Power Engineering Conf. (IPEC), 2005. [4] J. T. Boys, G. A. J Elliott, and G. A. Covic, “An appropriate magnetic coupling co-efficient for the design and comparison of ICPT pick-ups,” IEEE Power Electron. Lett., submitted for publication. Manuscript received March 12, 2006 (e-mail: d.kacprzak@auckland.ac.nz).