build the circuit with the shortest possible leads, with large conductors, use a ground plane.

Good advice.
Ideally you would want all connections, both signal and return, to be a PCB copper plane on adjacent layers.
That will greatly reduce the loop inductance.
Any wires that can't be on the plane should be twisted pair of signal and return or a coax line.

 

You said the pulse is only 50μs wide, so the total energy dissipated is low.
It's the energy dissipated, not the peak power that's important.
The resistor would only dissipate 700W*50μs = 35mJ of energy (equivalent to 35mW for 1 second).

Thank you very much! I managed to get 1us rise time using 22V supply. But the LEDs now have a bit of continuous glowing when I increase the voltage over a certain value. How should I deal with that?

 

If you are going for really short, high current pulses- pay attention to the physical layout.

build the circuit with the shortest possible leads, with large conductors, use a ground plane.
Place your gate driver as close as possible to the FET.

Build it like and RF amplifier- because it is.

Thank you! I'll remember that.

 

Thank you very much! I managed to get 1us rise time using 22V supply. But the LEDs now have a bit of continuous glowing when I increase the voltage over a certain value. How should I deal with that?

You mean it glows when the MOSFET is supposedly off?
If so you have a circuit problem with a sneak current path.
Post your latest schematic.

 

You mean it glows when the MOSFET is supposedly off?
If so you have a circuit problem with a sneak current path.
Post your latest schematic.

You mean it glows when the MOSFET is supposedly off?
If so you have a circuit problem with a sneak current path.
Post your latest schematic.

Yes.
The drop across the LED is about 6.5V, and a supply voltage of 22V gives me 50A. When the voltage is below 18.5V, I don't see any continuous glowing whereas when it's above, I see the continuous glowing on top of the pulses.To see what's going on, I turned off the signal generator, connected the gate to the source and increased the supply voltage from 0V. The LEDs start glowing at about 8V. I guess the MOSFET is damaged somehow so it can't be fully switched off?

Thank you.

 

Yes.
The drop across the LED is about 6.5V, and a supply voltage of 22V gives me 50A. When the voltage is below 18.5V, I don't see any continuous glowing whereas when it's above, I see the continuous glowing on top of the pulses.To see what's going on, I turned off the signal generator, connected the gate to the source and increased the supply voltage from 0V. The LEDs start glowing at about 8V. I guess the MOSFET is damaged somehow so it can't be fully switched off?

Thank you.

Your MOSFET driver IC is only rated to 18V supply. I would guess that running it on more than that is your problem.

 

Your MOSFET driver IC is only rated to 18V supply. I would guess that running it on more than that is your problem.

I use two power supplies, one is constant at 10V for the driver IC and another one is adjusted to get the desired current. I was talking about the later one. So it's not a problem, but still thank you.

 

I use two power supplies, one is constant at 10V for the driver IC and another one is adjusted to get the desired current. I was talking about the later one. So it's not a problem, but still thank you.

I have almost no experience with driver ICs like this one, but doesn't operating the IC on 10V mean that the highest control voltage it can apply to the gate is 10V, and that if the supply voltage to the LEDs gets high enough, you're biasing the MOSFET on when the gate is at 10V?

So, output low puts the gate at ground potential with the MOSFET fully on, and output high puts the gate at 10V. With 18.5V supply voltage minus 6V across LEDs, that leaves 12.5V, which is still greater than the 10V gate voltage, so the MOSFET will be partially activated?

 

I have almost no experience with driver ICs like this one, but doesn't operating the IC on 10V mean that the highest control voltage it can apply to the gate is 10V, and that if the supply voltage to the LEDs gets high enough, you're biasing the MOSFET on when the gate is at 10V?

So, output low puts the gate at ground potential with the MOSFET fully on, and output high puts the gate at 10V. With 18.5V supply voltage minus 6V across LEDs, that leaves 12.5V, which is still greater than the 10V gate voltage, so the MOSFET will be partially activated?

Sorry I am a bit confused about what you are saying. Are you saying that the MOSFET is on when the gate voltage is at ground potential? For my MOSFET and circuit, 0V gate voltage switches the MOSFET off but 10V gate voltage switches it on.

 

Sorry I am a bit confused about what you are saying. Are you saying that the MOSFET is on when the gate voltage is at ground potential? For my MOSFET and circuit, 0V gate voltage switches the MOSFET off but 10V gate voltage switches it on.

I'm so sorry. I see my mistake now. I was getting P-channel/N-channel high-side/low-side stuff turned around in my head. For some reason, although it feels like second nature thinking about BJT circuits, I sometimes transpose things when thinking about MOSFETs. I think it's partially because those pesky schematic symbol arrows go "the wrong way" to my eyes. Anyway, I feel dumb. Ignore my last comments.

 

The LEDs start glowing at about 8V. I guess the MOSFET is damaged somehow so it can't be fully switched off?

If the gate is a 0V, then yes, it would seem there's something wrong with the MOSFET.

 

If the gate is a 0V, then yes, it would seem there's something wrong with the MOSFET.

I think I have been operating it rather carefully, what could be the reason it is damaged? The reverse voltage created by the inductance or maybe there was too much heat when I soldered it?

 

I'm so sorry. I see my mistake now. I was getting P-channel/N-channel high-side/low-side stuff turned around in my head. For some reason, although it feels like second nature thinking about BJT circuits, I sometimes transpose things when thinking about MOSFETs. I think it's partially because those pesky schematic symbol arrows go "the wrong way" to my eyes. Anyway, I feel dumb. Ignore my last comments.

Don't worry

 

I think I have been operating it rather carefully, what could be the reason it is damaged? The reverse voltage created by the inductance or maybe there was too much heat when I soldered it?

Hard to say.

You should add a transient suppression diode between the MOSFET drain and the power supply (anode to drain).
A 1N400x type should work.

 

If the gate is a 0V, then yes, it would seem there's something wrong with the MOSFET.

I tried with a new MOSFET but the LEDs still start glowing when the voltage reaches about 18V, is it possibly related to the leakage current of the MOSFET?

 

I tried with a new MOSFET but the LEDs still start glowing when the voltage reaches about 18V, is it possibly related to the leakage current of the MOSFET?

The data sheet says the leakage current is in the order of uA when Vds is 80V, so it should not be the problem

 

The data sheet says the leakage current is in the order of uA when Vds is 80V, so it should not be the problem

Can you put a multimeter in series with the LEDs to measure the current?

 

Can you put a multimeter in series with the LEDs to measure the current?

Actually I just found out the problem, it is because the diode I used to protect the transistor was conducting. I forgot to include that in my schematic.

I have been trying to increase the voltage I can use by increasing the resistor value, but they are all turned to open circuit gradually. The resistors I used were two 1ohm in parallel, so the instantaneous power would be 25*25*1W each and average power would be 31mW, which is less than the 0.25W rating. Is it because of the high instantaneous power so they get damaged gradually?

 

t is because the diode I used to protect the transistor was conducting.

How is that diode connected?

Is it because of the high instantaneous power so they get damaged gradually?

Yes, if the peak power is high enough it can damage a resistor, even if it average power is below its rating.
How wide is the pulse width you are using?

 

How is that diode connected?
Yes, if the peak power is high enough it can damage a resistor, even if it average power is below its rating.
How wide is the pulse width you are using?

I used two diodes, one to the ground to protect from high negative voltage and one to the power supply to protect from high positive voltage. I somehow connected the later one to the IC's power supply so when the LED's power supply gets greater than that of the IC, the diode starts to conduct.

I was using 50us pulse width. I tried 1us afterwards and they seemed to be fine.