What is the fastest power diode you know?

Quote

View Details

In other words, speed is limited by small-signal bandwidth, which is wholly dependent on layout geometry and switching impedance.

The Schottky diode is placed on the bottom side of the PCB, exactly below the EPC with 5 vias each connecting it. I am aware that the inductance is critical but there is no better way. I simulated the whole circuit and adjusted the diode inductance in the simulation until the result matched the measured curve, when the low side FET turns off and the high side FET turns on. You can see, that the voltage dips into the negative and returns to about -1V which is the Vf of the Schottky at load current. In the simulation I get the same dip when I insert an inductance of 100pH which is already quite low.
The strange thing is that when the high side FET turns off, the diode does not seem to do its job because the voltage drops way too low. The only difference to the previous case is that the diode has a reverse bias now before turning on. I don't know why it does that. It seems like it doesn't even turn on at all. Do you know why that could be?

Quote

Have you considered:
- Using a GaN-specific controller?

Yes, I would really like to and if you know one pleas tell me but to my knowledge there is no GaN specific buck controller going up to 75V. The MICA is listed on EPC's website as a suitable controller which is ironic because it clearly doesn't work and I already had a long video call with the chip developers from Microchip discussing this issue. They also have no better suggestion.

Quote

- External gate driver?
- Those GaN-driver-and-inverter-in-a-chip parts?

maybe possible but buck controllers of this kind have a current sense feature for which the switchnode voltage has to be connected to the controller independent of the gate driver.

Quote

A possibility is putting the series gate resistor in the negative (PGND?) path between controller and bridge, so that the negative swing can be clamped by a (smaller) diode, without upsetting the controller.  Whether this is feasible given other limitations (e.g. permissible AGND/PGND voltage, whether the controller itself needs to be referenced to a global GND e.g. because of a programmable voltage setpoint or something, etc.), you'll have to figure out.

Something simular was suggested by the Microchip developers but after thinking about that for a long time and simulating it I am pretty sure this would destroy the IC because it messes with the voltage reference for the gate path. If you are really interested I can show you a detailed analysis of this...

Quote

I'm going to guess that if these don't help, you're shit out of luck

made my day  looks interesting, I will look into it tomorrow

Thanks for all the helpful responses 

The Schottky diode is placed on the bottom side of the PCB, exactly below the EPC with 5 vias each connecting it. I am aware that the inductance is critical but there is no better way. I simulated the whole circuit and adjusted the diode inductance in the simulation until the result matched the measured curve, when the low side FET turns off and the high side FET turns on. You can see, that the voltage dips into the negative and returns to about -1V which is the Vf of the Schottky at load current. In the simulation I get the same dip when I insert an inductance of 100pH which is already quite low.The strange thing is that when the high side FET turns off, the diode does not seem to do its job because the voltage drops way too low. The only difference to the previous case is that the diode has a reverse bias now before turning on. I don't know why it does that. It seems like it doesn't even turn on at all. Do you know why that could be?Yes, I would really like to and if you know one pleas tell me but to my knowledge there is no GaN specific buck controller going up to 75V. The MICA is listed on EPC's website as a suitable controller which is ironic because it clearly doesn't work and I already had a long video call with the chip developers from Microchip discussing this issue. They also have no better suggestion.maybe possible but buck controllers of this kind have a current sense feature for which the switchnode voltage has to be connected to the controller independent of the gate driver.Something simular was suggested by the Microchip developers but after thinking about that for a long time and simulating it I am pretty sure this would destroy the IC because it messes with the voltage reference for the gate path. If you are really interested I can show you a detailed analysis of this...made my daylooks interesting, I will look into it tomorrowThanks for all the helpful responses

WEE Technology Company Limited
Manufacturer of Surface Mount (SMD) and Through Hole (DIP) Diodes & Rectifiers

Seven Main steps required for testing of a diode rectifier are as follows. Suggested by WEET engineer group.

1, Choose a superior quality digital multi-meter, which is very important. Good sensitivity and accuracy counts a lot in testing period.

2, Fix the testing range in the diode position. You need to check both terminal are set in proper position and connected well.

For more information, please visit Hornby Electronic.

3, On the display you are supposed to easily get a 3 or an infinite voltage reading depending upon the multi-meter used

4, Correctly connect the red probe to the cathode and the black probe to the anode of the rectifier. 

5, The digital multi-meter display will show the low forward voltage drop (rectifier diode) of around 0.6 volts at once

6, Now reverse the connections, the display will return to its original reading, which indicating that the diode is a good condition

7, If the meter displays any other reading, the diode may be leaky or faulty while a reading means a short circuit.

Need any technical support? Contact with WEET sales team right now, we will find the best solution for you.

https://www.weetcl.com/pdf/WEET_DO27_SF31G_THRU_SF38G.pdf

Want more information on diode ultra fast recovery? Feel free to contact us.