Environmental Engineering Reference
In-Depth Information
f
p
f
p
f
p
f
p
Figure 5.24 Flux squeeze interior permanent magnet machine
The rotor centre of the flux squeeze design must be made of non-magnetic
material or some design of iron bridges and cavities so that the ends of the rotor
magnets are not shorted out. In the flux squeeze IPM, the magnets are mounted in the
tangential direction and the flux path is completed as shown by the soft iron wedges set
between the magnets. By designing the rotor so that the magnet face area is large
compared with the airgap surface, the flux in the gap may be very high. For example,
with ceramic magnets having remanence
B
r
ΒΌ
0.25 T, the flux density in the gap may
be 0.82 T. This provides rare earth magnet performance for approximately a tenth of
the magnet cost, but at the expense of a much larger diameter rotor. Because a portion
of the magnet lies at the machine airgap, the magnets themselves must have suffi-
ciently high coercivity so as not to demagnetize beneath the strong demagnetizing
fields of the stator. High coercivity magnets such as barium ferrite and NdFeB rare
earth magnets do well in this geometry. It is also noteworthy that for the flux squeeze
design, the saliency ratio
x <
1since
L
ds
>
L
qs
.The
d
-axis in fact lies completely in
rotor iron, and the
q
-axis interestingly lies completely inmagnet material. The saliency
ratio can be very low in this regard, or viewed from another perspective,
x
1
... 1.
Another difference of the flux squeeze compared to the buried magnet IPM is that now
the mmf across eachmagnet is twice the mmf across the airgap. This is true because the
airgap flux over the soft iron pole face is the composite of flux from two magnets.
This type of machine has many proponents for various hybrid propulsion sys-
tems. Honda, for example, uses a variation of the flux squeeze IPM in its hybrid
designs. The reason for this is that the volumetric and gravimetric power output of
electric machines for hybrid propulsion, as well as aerospace, must be as high as
possible and the flux squeeze IPMdoes deliver high specific output, but for relatively
small ratings. This latter fact does not appear to have been made sufficiently clear in
the hybrid propulsion design camp. In Reference 11 the investigators proposed an
optimization procedure in which contours of constant volume are presented for IPM
machines in the 0.25-10 kW range and for speeds in the 10-100 krpm range. It is
