# MOSFET vs IGBT: definition of switching parameters

#### gabyft45ol

Joined Mar 23, 2020
4
Hello,

I would like to compute the minimum dead time for two power inverters, the first one is based on IGBT and the second one on MOSFET. I noticed that the switching parameters (Tdon, Tr, Tdoff and Tf) are defined as:

- Tdon: 10% of gate voltage to 10% of collector current
- Tr: 10% to 90% of collector current
- Tdoff: 90% of gate voltage to 90% of collector current
- Tf: 90% to 10% of collector current.

for IGBTs, and for MOSFETs as:

- Tdon: 10% of gate voltage to 90% of collector voltage
- Tr: 90% to 10% of collector voltage
- Tdoff: 90% of gate voltage to 10% of collector voltage
- Tf: 10% to 90% of collector voltage.

Why do we use the collector current for one technology, and the collector voltage for the other one? Can I estimate the switching parameters of an IGBT using the collector voltage?

#### crutschow

Joined Mar 14, 2008
27,213
Where did you get those definitions?
They are not correct since MOSFETs do not have a collector, they have a source and drain.

Last edited:

#### gabyft45ol

Joined Mar 23, 2020
4
Where did you get those definitions?
They are not correction since MOSFETs do not have a collector, they have a source and drain.
Sorry, I mean the drain source voltage (Vds) for the MOSFET. I got those definitions from application notes and datasheets of Infineon, Microsemi, Toshiba,...

#### crutschow

Joined Mar 14, 2008
27,213
There typically should be no significance difference between using the MOSFET drain voltage and the IGBT collector current for the switching speeds if the load is mainly resistive, as should be for a power inverter.

Last edited:

#### gabyft45ol

Joined Mar 23, 2020
4
All the MOSFETs are characterized with a resistive load, and all the IGBTs with an inductive load (double-pulse test). For the MOSFETs, the switching times are the same if we use the drain current or the drain voltage, due to the resistive nature of the load. However, for the IGBTs, the switching times are not the same if they are calculated with the collector current or the collector voltage.

Why the manufacturers always use an inductive load for the characterization of IGBTs?

#### shortbus

Joined Sep 30, 2009
8,585
Why the manufacturers always use an inductive load for the characterization of IGBTs?
From my prospective on this is is because they are used more in induction motor control, where switching speed is not as high as mosfets. There is an app note for the difference and how to choose which to use.
https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d462533600a40153574048b73edc

And a good discussion on it. https://www.watelectronics.com/difference-between-insulated-gate-bipolar-transistor-igbt-and-mosfet/

#### gabyft45ol

Joined Mar 23, 2020
4
I need to determine the minimum dead time for two power converter, one made of SiC MOSFETs and the other one of IGBTs, regardless of the application (motor, PV,...). For this purpose, I need to measure the switching times of the semiconductors.

A test with a resistor is feasible, but what is the impact of the load (inductive or resistive) on the switching times? Can it lead to an underestimated minimum dead time?

#### shortbus

Joined Sep 30, 2009
8,585
For this purpose, I need to measure the switching times of the semiconductors.
Isn't that in the data sheets? And it will vary for part to part even in the same batch.