The operating voltage occurring at the fuse terminals after it has blown must
not behigher than the fuse nominal voltage.
The voltage of the arc or the switching voltage must not be higher than the
voltage rating of thepower semiconductor.
The fuse turn-off integer I²t must be lower than that of the mains (rectifier
diodes or SCRs.
When using filter circuitry, check that the short circuit current is high enough to
blow the fuse.
The company SIBA, for example, offers design software for fuses in inverters as an aid
for the selectionof the right fuse.
13.12 Influenceof themodulation algorithm
The quality of the output current, as well as the switching losses of the inverter power
semiconductors, is determined to a large degree by the modulation algorithm used. A
standard modulation algorithm for a multitude of applications is pulse width modulation
PWM, of which there are various implementations. It is based on the voltage-time area
congruence concept. Chapter
ooked at the sine-trianglemodulation (SPWM) in
the context of the definition of themodulation factor. This concept, just like space vector
PWM (SVPWM) and third harmonic injection PWM (THIPWM), belongs to the
continuous PWM (CPWM) group. Thismeans that turning on and off occurs continually
within each period of the carrier wave. Another group of PWM algorithms is classed as
discontinuous PWM (DPWM). Here, some of the switches are not operated for a certain
time within the carrier period. Purely by numbers it can be derived from this that the
CPWM process will automatically lead to higher switching losses than DPWM. On the
other hand, DPWMwill render higher distortionof the output current thanwill CPWM.
The groupof theCPWMalgorithms include:
This concept samples a sinewave reference signal with a triangular signal of a
higher frequency. Each crossing of the triangular signal with the sine wave
reference signal leads to a switching command given to the power switches
This process is used in the majority of all applications. The expected output
voltage vector V
is calculated in phase and amplitude by mathematical and
time-based averaging of two neighbouring voltage vectors and the two zero
hows this with an example. The reference voltage vector
is emulated during the sampling time T
by the two active vectors V
aswell as the two zero vectorsV
. Vector V
is active for the time t
for the time t
. The two zero vectors are then used for the remaining
2 1 S 0
This technique, which is based on SPWM or SVPWM, tries to eliminate the
third harmonic in the output signal by overlapping the fundamental of the
reference signal with the thirdharmonic of it.