Maker Pro
Maker Pro

Transistors for higher voltages

pidja105

Oct 16, 2015
106
Joined
Oct 16, 2015
Messages
106
Hello,
How to wire two 40Volt transistors(mosfet) for 80Volt drain-source voltage?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Best to get a single 80v (or better 100V) device.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
It does not matter which way you wire it, you will get the same result either way: a blown transistor.

Bob
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
That's not strictly true. In theory a pair of mosfets can be placed in series to withstand a higher voltage. In some exotic applications I've seen people trying to use three or more.

When turned on there is no problem.

When turned off, there are a few problems, most of which can be worked around with resistors to balance the voltage across them (at the cost of some leakage).

The first hint of real complexity happens when you turn them off. Typically this is done by pulling the gates to their respective sources. No matter how this is done, unless there is active control to keep the voltage between them to a mid point, one mosfet will turn off faster than the other causing the other to have greater than half the voltage across it. This will cause the transistor to avalanche. Whilst this will bring the transistors closer to a balanced voltage, it requires a mosfet that has an adequate avalanche rating.

If you think that sounds like fun, turning them on again is even more of a challenge. Whilst off, the gates are essentially V/2 volts apart. When on, both are (say) 10V above their sources, but their sources are essentially at the same voltage. Again, without active control, your task is to raise the voltage at the gate wrt the source, while one of the source voltages plummets from V/2 too a value close to 0. Again, you need to do this without causing either of the transistors to go into avalanche. This is more difficult while turning on as it is far more likely that the imbalance will dump the full voltage across the slower transistor in a situation where the gates are being controlled in a more complex manner and where current is increasing.

Turning the compound device on and off without avalanche requires active control of one of the pair of mosfets aiming at maintaining an equal voltage across each mosfet. In order to do this, the automatically controlled mosfet needs to have a greater range of available gate voltages to compensate for the fact that delays in the feedback loop mean it needs to respond faster than the transistor it is tracking (mostly to overcome the RC delay caused by the gate driver impedance and the gate capacitance).

The practical upshot? If you're switching a low current, get a higher voltage device. If you're switching a high current (where the avalanche energy will be much larger) consider an IGBT.

The complexity of doing it right is significant.

Also consider the failure modes. So many more things can go wrong with a more complex circuit.
 
Last edited:

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Somehow I missed the "two" in the OP. It was the 80V source to drain and the 40V rating that I saw. If you wire two in series taking precautions to make sure neither has over 40V across it, that is a different story.

Bob
 
Top