IEEE Power Electronics Magazine - June 2017 - 42

4,000
Switching Energy (µJ)

Switching-Off Energy (µJ)

200
180
160
140
120
100
80
60
40
20
0

15
20
25
Load Current (A)
(a)

Switching-On Energy (µJ)

800
700
600
500
400
300
200
100
5

15
20
25
Load Current (A)
(b)

15
20
25
Load Current (A)
(c)

Total Switching Energy (µJ)

1,200
1,000
800
600
400
200
0

ROHM
Wolfspeed #1
Wolfspeed #2

SEI
Monolith #1
Monolith #2

FIG 8 (a) The switching-on losses, (b) switching-off losses, and
(c) total switching losses for all six SiC MOSFETs under test
versus load current at the total gate resistances shown in
Table 2 and at 25 °C.

an Agilent 4294A impedance analyzer while the drain and
source terminals were isolated. The third column of the
table shows the external gate resistance selected for each
device. The external gate resistances are selected to feature
the fastest switching speed for each of the devices before
significant switching transients start to emerge. The ROHM
device has the largest internal gate resistance, and even

3,000
2,500
2,000
1,500
1,000
500
0

40
60
Load Current (A)

100

FIG 9 The switching-on and switching-off losses for the
Wolfspeed #2 and Monolith #2 SiC MOSFETs versus current at
the total gate resistances shown in Table 2 and at 25 °C.

900

Monolith #2 Switching-Off
Wolfspeed #2 Switching-Off
Monolith #2 Switching-On
Wolfspeed #2 Switching-On

3,500

IEEE PowEr ElEctronIcs MagazInE

z June 2017

when applying zero external gate resistance, its switching
transients are not as large as the other MOSFETs. Finally,
the last column of Table 2 shows the total gate resistance
for each MOSFET.
For each power MOSFET, the DPT is performed from
a 5-A load current up to the device's nominal current or
100 A, whichever happens first. For better resolution of the
dynamic test results at lower currents, Figure 8(a)-(c) demonstrates the switching-off, switching-on, and total switching losses, of all of the devices up to a 35-A load current.
Then in Figure 9, the turn-off and turn-on switching losses
are shown for the Wolfspeed #2 and Monolith #2 power
MOSFETs, up to 100 A.
According to Figure 8, the Wolfspeed #1 MOSFET features the smallest switching losses among all of the DUTs
at both turn-off and turn-on. This is mainly attributed to
the small input and Miller capacitances of the Wolfspeed
#1 device compared to the others, although it has a relatively large total input gate resistance. On the other side, the
Wolfspeed #2 and Monolith #2 power MOSFETs have larger
switching losses, which is due to their larger current rating
(larger die), resulting in larger parasitic capacitances. As
previously mentioned, larger parasitic capacitances slow
down the switching transients of the device and increase
the switching losses.
Although the ROHM semiconductor has an input capacitance as small as the Wolfspeed #1 MOSFET, its larger
input gate resistance causes larger switching losses compared to the Wolfspeed #1. The Monolith #1 power semiconductor features around the same switching losses as
the ROHM MOSFET, since it has a larger input capacitance
but a smaller gate resistance, and the turn-off and turn-on
losses of the SEI MOSFET are larger than the Wolfspeed #1
but smaller than the other devices. Figures 8 and 9 reveal
that the majority of the switching losses in SiC MOSFETs
is caused when switching on. Thus, by using soft-switching
methods and eliminating turn-on switching losses, the total
system efficiency can be significantly improved.

Table of Contents for the Digital Edition of IEEE Power Electronics Magazine - June 2017