Hi all.

I'm working on a project to build a H-bridge with MOSFETs, up to 4.5kV @ 0.5mA max. It will be turn on for around 5 seconds max with a period of 10 seconds (rise and fall time about 0.5 second).
I have found a MOSFET rated at this range (IXTX1R4N450HV), but the forward-bias safe operating area suggest me to not operated this over 3000kV when it's on 'DC' (I assume that means a switching time more than 100ms, correct me if I'm wrong).

Another alternative here will then be stacking up multiple low voltage MOSFETs (4 x 1.5kV MOSFETs) in series for higher voltage rating. I have been researched for a while and the thing came out most frequently is the RC snubber circuit. Now, as far as I know, the RC snubber circuit is to eliminate the dynamic voltage difference in fast switching application. I will be curious if my application was to turn on/off the MOSFET in such a slow period, will it still matters to use a snubber circuit?

Thanks in advance to anyone!

 

Try to describe the load.
Is it resistive, inductive or capacitive?
If it is mixed (probably), then about how?

Btw, using 1.4A mosfet for 0.5mA load usually brings some problems.

 

It is a variable capacitor (I assume it would be capacitive load with some unneglectable resistance).

May I know what will be the problems of using 1.4A MOSFET for 0.5mA? My understanding was the 1.4A is a 'path' provided by MOSFET, and 0.5mA should just pass through it (It's a very 'physica'l understanding, while it may not be appropriate for electrical stuff..)

 

With capacitive load the problem are current spikes, not a voltage spikes like in motors, transformers, inverters…

Since the cap shouldn’t be switched to voltage source directly because ideally an infinitive current will flow thru it, you should switch the load to DC thru some small series inductor or resistor at least.

The mosfet has some residual (leakage) current parameter and its value shouldn’t be comparable with load current. Also, when 1.5A mosfet switches off just 0.5mA the mosfet is getting out of saturation very hard.

 

Good point -- does that means that if I switch the load with a series resistor, it will limit the current spikes based on the Ohm's law?

And also, is operating MOSFET in saturation region a bad thing? I can only see the disadvantages of high power dissipation and thermal runaway -- which might be able to add some heat sink to cool down the device.

Last question -- if I connect it to a capacitive load, do I still need the snubber circuit to limit the voltage imbalance across small MOSFETs?

Appreciate your help!

 

Ya, the resistor or inductor in series with capacitive load will limit the current.

You want to hold mosfet in saturation region. In linear the mosfet will generate the heat and cause waist. Also, the switching mosfets cannot handle the linear state much. See SOA you added how long the mosfet can be in linear region to not be destroyed.