# High-side gate driver not working under load high current

I have been trying to implement a current limiting circuit using a high power NMOS and high side gate driver.

My idea is to adjust the PWM duty-cycle of the gate driver to give an limited (average) current to the load. So far, I had good luck with a small load (8 Ohm at 24V). However, when I try a bigger load (2 Ohm at 24V), the gate driver stops working.

My original guess is that the NMOS gate is taking too much current form the gate driver output pin, but the simulations tell me that the TG pin is only outputting 2.4A max (LTC4440 can do 4A max for 1us). Another guess is that the inductance in the wire make the peak current higher than what the simulations show. Does anyone have a recommendation ?

The schematic below shows my setup. The plot shows the output voltage in green, and the current at at the gate of the NMOS in blue.

• Where is the current limiting? Or how is this supposed to work? I admit, I'm having trouble finding that out. You want current limiting with a PWM driver that has no filter at the output, yet it worked? Could you explain how? May 7, 2018 at 19:37
• Please explain how you are bootstrapping it. May 7, 2018 at 19:39
• Apologies for the confusion: a micro-controller adjusts the duty-cycle of the PWM according to the load current (a Hall current sensor monitors the load current at all time). For the bootstrap circuitry, I tried to match what was recommended in the LTC4440 datasheet. May 7, 2018 at 19:51
• To me, the circuit looks completely wrong, pretty much everything. May 7, 2018 at 19:57
• Where is your energy storage inductor? May 7, 2018 at 20:08

The gate driver and the MOSFET are working just fine.

However, with a source impedance (R2) of 0.6Ω, you're simply not going to get full voltage with a 2Ω load (R1).

These two resistors form a voltage divider that gives you the 18.46 volts that you see at the output.

What exactly is the purpose of R2?

• R2 is just to simulate the output impedance of the 24V (lithium-ion battery). If I remove it, I get the same waveform, but with an amplitude of 24V (instead 18V). As you said, the driver and the MOSFET work just fine on the simulation. That's why I'm puzzled that it does not work on the testbench. May 7, 2018 at 20:48
• If you're trying to debug a prototype, why does your question talk exclusively about a simulation? What exactly is going wrong with the prototype? Edit the information into your question. May 7, 2018 at 21:06
• OP may need a more robust MOSFET with a much lower ON resistance. 600 miliohms is actually a bit high-not meant to be a high current switch.
– user105652
May 7, 2018 at 21:29
• Dave, you are 100% right. I edited my question and added more info on my experiment. May 7, 2018 at 21:43