IGBT Modules - Technologies, Driver and Application (Second Edition) - page 72

Fig. 1.53
Simulated switchingbehaviour of a JFET/MOSFET cascode
If there is a need to have higher blocking voltages than can be achieved with a SiC
JFET, the cascode can, in principle, be expanded in a serieswithmore JFETs, at which
point it is called a supercascode. To balance the voltage of the series connection,
additional standard silicon diodes, adjusted to the voltage, are required. If necessary,
these limit the voltage drop of each JFET by going into avalanche breakdown. Silicon
diodes are not required for the freewheeling mechanism, as the body diodes
) can take over this function if the JFET is designed to facilitate this. The
disadvantage of supercascodes is their high voltagedrop in conductionmode.
Whether the cascode has one or several JFETs, the topology creates high output
capacitances. These increase the turn-on losses of the cascode in comparison to an
individual JFET. For commercial applications, a modified variant known as "cascode
light", inwhich the JFET and p-channel MOSFET are controlled separately, is preferred.
In normal operation, the current flow is controlled exclusively by the JFET, while the
MOSFET is permanently turned on. The voltage drop across the MOSFET at this
operating point is negligible, as MOSFET are available with an R
of 1m
, for
example. In the event of a fault, when the supply voltage of the driver stage fails, the
MOSFET is automatically turned off and the topology converted into the "normal"
cascode variant. This means that there is a "normally-off" component that can safely
block the current flow. The disadvantage of a "cascode light" circuit is that two driver
channels are required, one for the JFET and one for the MOSFET. The driver stages
alsohave to be combined.
1...,62,63,64,65,66,67,68,69,70,71 73,74,75,76,77,78,79,80,81,82,...548
Powered by FlippingBook