Odd PWM Waveform on MOSFET Gate Pin

[jmd13391]

Hi TheDIYGuy999 !! Thank you for your excellent work on this PWM5 variant. I have taken your design and attempted to adopt it for use with a 24v,18Ah battery and a 37v, 70w solar panel. Although I am using a different MCU, all discrete components are as you call-out in the original schematic with the exception of the voltage divider resistors R5 & R6 (47K, 5.6K) and C5 (10uF). Here is my question / issue:

I am seeing an odd "inverted shark fin" PWM waveform at the MOSFET gate and was hoping that you might have insight as to how to correct it. See photo. CH1 (yellow) is the PWM waveform from the MCU. CH2 (blue) is the PWM waveform at the MOSFET gate.

MCU PWM Freq. = 300Hz. (MCU max is 1KHz.)
MCU Charge Pump Freq: 4.14KHz. (this is the MCU max)

Thank you in advance...

[TheDIYGuy999]

Hi,
Looks like the mosfet gate is not discharged fast enough. This is a potential issue of Julians gate driver circuitry. T3 provides a fast discharge path, but does not work at low voltages, because of the D5 voltage drop. That's why the voltage starts dropping fast and then slow.
The charge pump frequency may also be too low for a stable gate supply. Have you confirmed the stability of this rail with the scope? I assume, you have used a bigger C5 because of this potential issue.
Also make sure, that you don't exceed the gate and base voltage limits of your transistors, because of your high voltages.

[jmd13391]

Hi TheDIYGuy999, Thank you for your prompt reply.

"T3 provides a fast discharge path, but does not work at low voltages, because of the D5 voltage drop."

Would replacing D5 (1N4148) with a lower voltage drop Schottky diode improve the discharge rate?

"The charge pump frequency may also be too low for a stable gate supply."

I had a typo in my original post... the Charge Pump pulse frequency of each square wave is ~4.14Khz., CP1 pulse & CP2 pulse waves are 180 degrees out of phase with a 10uS state change delay:

"Have you confirmed the stability of this rail with the scope?"

Yes. Rock solid at 29.1v under the following conditions - Battery: 26.81v, Solar Panel: 33.52v

"I assume, you have used a bigger C5 because of this potential issue."

Yes, Sir. ;-)

Do you have any suggestions as to how the circuit can be modified so as to improve the trailing edge of the MOSFET gate pulse?

Thank you again for your assistance with this issue...

[jmd13391]

Closing due to inactivity.