IEEE Consumer Electronics Magazine - March/April 2020 - 73

power loss in the converter is a much more
important quantity for a designer compared to
its efficiency. In other words, efficiency is a relative parameter of a converter and power-loss is
an absolute parameter. A typical efficiency curve
of a converter is shown in the Figure 5. It shows
that a converter efficiency varies as a function of
output power. The peak efficiency does not
always occur at the rated power output of the
converter and is design dependent.

Figure 5. Typical efficiency curve of a power
converter.

which leads to smaller magnetic size. Their
applications are still emerging in some areas
as summarized in Figure 4. Specifically, they
are suitable for high power converters such
as EV chargers, which require fast charging.3
5) Gallium Nitride Mosfets: This is a relatively
new wide-bandgap device and its applications are still emerging. As they are still not
in heavy production, their cost is higher than
Silicon Mosfets, and therefore, not that
widely used in commercial applications.
They promise significantly lower switching
loss and better thermal performance compared to Silicon Mosfets. They are available
below 650 V voltage level. These devices are
perceived to be better than Mosfets when
voltage rating is higher than 200 V.

Efficiency (Relative Quantity) Vs Power Loss
(Absolute Quantity)
Efficiency of a converter is defined as the net
input power required to produce a given output.
It can be expressed as
h¼

Pout ðPin À Ploss Þ
¼
) Ploss ¼ Pin Á ð1 À hÞ:
Pin
Pin

As an illustration, for an efficiency (h) of 0.95,
if the converter input power (Pin) is 10 W, its loss
is 0.5 W. For the same efficiency, if the input
power (Pin) is 500 kW, its loss is 25 kW. The
power loss in a converter directly affects the
thermal design and heat sink size. Therefore,

March/April 2020

Switching Frequency
Inductors are used as storage elements or filter components in a converter. Making switching
frequency higher reduces the size of the filter.
Therefore, a very high switching frequency is
required to keep the form factor of the converter
small and power density high. As the switching
frequency goes up, the eddy current loss in the
core of the inductor also becomes higher. Eddy
current loss is directly dependent on the square
of the switching frequency and inversely dependent on the resistivity of the core.
At higher switching frequency, it is desirable to
have a core with higher resistivity to reduce the
eddy current loss. As the inductor requirement is
smaller at higher switching frequency, a lower permeability core material is an acceptable tradeoff.
Table 1 provides a summary of important
cores used in power converters and their typical
operating frequency range for the magnetics.
CRGO core is used at utility frequency level and
NiZn core is used at multi-MHz operating frequency. A very high resistivity of NiZn ferrites
results in acceptable eddy current loss at MultiMHz frequency of operation.

OUTPUT POWER, SWITCHING
FREQUENCY, AND MAXIMUM
EFFICIENCY
Conventional Designs
It is a general belief that a converter should
have very high efficiency irrespective of its power
level. Even if it is a logical assertion, it is not practical. At lower power levels, economics generally
rules the choice of converter topology and inturn its efficiency. Adding more power components to push efficiency higher by some percentage may not be economically viable. This is

IEEE Consumer Electronics Magazine - March/April 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - March/April 2020