I'm working on a circuit with a power mosfet and a gate to drive it. I've quickly realized there is not any good way to activate the Vgs of my power mosfet with my gate driver that only outputs around 10V. I've looked online at some other gate drivers, and all of them only output low voltages around 20-30V MAX.

Every single datasheet for gate drivers intended to be used with power mosfets describes their output in terms of current. Sink or source, up to 4A/6A, etc.

These are 2 of the gate drivers that I have, one of them is also an isolated gate driver specifically used for power mosfets.

http://www.ti.com/lit/ds/symlink/ucc20520.pdf

https://www.mouser.com/datasheet/2/698/isl55110-11-1302115.pdf

I understand that the mosfet gates activate upon charging to a certain Qg, but if the gate voltage needs to be 10V > than Vs, and Vs is 200V+, how is it possible to turn the gate on when the drivers don't output anywhere near that?

 

The driver’s low side voltage would have to be referenced to that higher source voltage. The MOSFET source pin is normally tied to ground, as would be the gate driver.

 

how is it possible to turn the gate on when the drivers don't output anywhere near that?

Please post a schematic so we can understand more clearly what you're talking about.

 

Please post a schematic so we can understand more clearly what you're talking about.

 

You're trying to drive your load high side with an N channel MOSFET, and you have the MOSFET wired backwards. As wired, the body diode will always conduct.

We usually use a P channel MOSFET to drive high side and an N channel MOSFET to drive low side.

If you swap the source and drain terminals and drive the load low side, the 10V from the gate driver may be sufficient. I haven't read either of the PDF's referenced and no part number was provided.

 

The max Vgs to fully turn off a typical MOSFET is never more than 10V.
Remember, Vg to ground is not always the same as Vgs.

In your circuit you need to ground the transistor source and place the 200V in series with RL and the transistor drain.
That way the Vgs of 10V will fully turn on the MOSFET and apply 200V across RL.

If you need to have RL grounded, then you would need to either use a P-MOSFET with a high driver, or an N-MOSFET with a bootstrap driver.

 

Right, place the MOSFET so that it is controlling the return-to-ground on the low side of your load. The load is attached to +200V on its high side. This is called a low side switch. You can use a high-side switch instead by using a P-channel MOSFET but that would require a different driver.

 

Right, place the MOSFET so that it is controlling the return-to-ground on the low side of your load. The load is attached to +200V on its high side. This is called a low side switch. You can use a high-side switch instead by using a P-channel MOSFET but that would require a different driver.

What driver is needed for a high side p-mosfet?

The max Vgs to fully turn off a typical MOSFET is never more than 10V.
Remember, Vg to ground is not always the same as Vgs.

In your circuit you need to ground the transistor source and place the 200V in series with RL and the transistor drain.
That way the Vgs of 10V will fully turn on the MOSFET and apply 200V across RL.

If you need to have RL grounded, then you would need to either use a P-MOSFET with a high driver, or an N-MOSFET with a bootstrap driver.

Low-side isn't an option unfortunately. I'll be looking into some P-mosfets for high side.

 

This might be a stupid question, but the datasheet for the transistor I have doesn't explicitly state if it is N or P type, and the name also hase both N and P in it... SiHP6N40D.

Check the little diagram. It says right there that it's N-type. The little arrow in the diagram also indicates N vs. P. It points away from the gate in a P-type.

 

What driver is needed for a high side p-mosfet?

If high speed is not required, you can use a high-voltage NPN transistor.

What frequency switching do you need?

 

What driver is needed for a high side p-mosfet?

Here's a very simple example.
https://hackaday.com/2015/09/16/learn-and-build-a-high-side-switch/
A little more here: https://www.electronics-tutorials.ws/transistor/tran_7.html

And a more elaborate solution: https://www.infineon.com/dgdl/dt92-4.pdf?fileId=5546d46254e133b401554de58b225d15

 

If high speed is not required, you an use a high-voltage NPN transistor.

What frequency switching do you need?

High frequency is desired. I'm hoping to be able to open/close creating a 100ns pulse sent to RL.

 

High frequency is desired. I'm hoping to be able to open/close creating a 100ns pulse sent to RL.

I'm wondering if a better solution for this application would be some sort of "burst" circuit rather than just a switch. You would design it to store an amount of energy that won't damage the LED, and then when ready it would dump that energy to flash the LEDs. I've never built such a thing but I'm sure there are LED-as-flashbulb circuits out there.

 

High frequency is desired. I'm hoping to be able to open/close creating a 100ns pulse sent to RL.

What's the pulse frequency?

 

I'm wondering if a better solution for this application would be some sort of "burst" circuit rather than just a switch. You would design it to store an amount of energy that won't damage the LED, and then when ready it would dump that energy to flash the LEDs. I've never built such a thing but I'm sure there are LED-as-flashbulb circuits out there.

I'm using a capacitor that discharges the high voltage within 100ns. Kind of similar to your idea.

What's the pulse frequency?

My intended pulse frequency is 2kHz, or 500us cycle.

 

Here's an example driver circuit with its LTspice simulation, to generate a 200V, 100ns pulse.
Since the pulse has a low duty-cycle, it can be AC coupled to the high-voltage P-MOSFET.
(Note that the I only had a 200V P-MOSFET model, so you would need find a P-MOSFET with at least a 250-300V rating for the actual circuit.

 

I think that this might be a good option.

https://www.vishay.com/docs/91282/sihfr921.pdf

Will the same gate driver work for both n and p types?

 

High frequency is desired.

My intended pulse frequency is 2kHz, or 500us cycle.

Glad we got that cleared up. HF is generally in the MHz range.

Low frequency is in the 30kHz-300kHz range.

 

Glad we got that cleared up. HF is generally in the MHz range.

Low frequency is in the 30kHz-300kHz range.

If I can't blink my eyes faster than the frequency it is HF in my eyes

 

Here's an example driver circuit with its LTspice simulation, to generate a 200V, 100ns pulse.
Since the pulse has a low duty-cycle, it can be AC coupled to the high-voltage P-MOSFET.
(Note that the I only had a 200V P-MOSFET model, so you would need find a P-MOSFET with at least a 250-300V rating for the actual circuit.

View attachment 164384

I'm impressed you got that working so quickly. I've been getting some crappy graphs that take me about 20X longer than that. I'm hoping I'll be able to use only a single P mosfet as opposed to combining it with other BJTs, although I do have a bunch of old crusty BJTs laying in my circuit box.