1
\$\begingroup\$

I have some problem with a P-channel mosfet used as high-side load switch. Everything works fine beneath the power up moment.

You can find the schematic in the picture below.

enter image description here

During Powering the circuit with 28V the following happens (blue channel: Gate; green channel: Drain):

enter image description here

I suppose that this flicker occurs because of the Gate Capacitance of the MOSFET, but I am not sure. Does somebody of you have an idea on how to solve this problem? During this period the output signal should stay at GND level.

Furthermore, I want to tell that on the input there will be a 5 kHz PWM coming from a Microcontroller. As I said before, after the power-up, there are no problems. Edit: The PWM is started after some ms.

Regards Armin

\$\endgroup\$
  • \$\begingroup\$ Those R values seem rather high for gate control... especially if you plan on using it for PWM. \$\endgroup\$ – Trevor_G Aug 31 '17 at 14:27
  • \$\begingroup\$ You also seem to be driving that gate to significantly lower than the 10V VGS the part requires. \$\endgroup\$ – Trevor_G Aug 31 '17 at 14:41
  • \$\begingroup\$ Do you think that the gate current is too low with these resistors? I am driving it down to 8V which should be ok. I read +/- 20V Vgs in the datasheet. \$\endgroup\$ – Armin Aug 31 '17 at 17:09
  • \$\begingroup\$ -20V is the absolute max though... that leaves you no room for variation in the 28V and any other noise spikes. If you look at the other figures it's really defined at -10V, which makes that the manufacturers "normal" operating point. \$\endgroup\$ – Trevor_G Aug 31 '17 at 17:14
  • 1
    \$\begingroup\$ And yes I'd drop those resistor values by at least a factor of 10. R1 is also unnecessary. \$\endgroup\$ – Trevor_G Aug 31 '17 at 17:16
2
\$\begingroup\$

Those R values seem rather high for gate control... especially if you plan on using it for PWM. I'd drop those resistor values by at least a factor of 10. Lower resistances will also mean the MOSFET will switch faster and run cooler when you PWM.

Incidentally, R1 is also unnecessary.

You also seem to be driving that gate to significantly lower than the -10V VGS the part is specified to. You are driving it down to 8V, which is -20V VGS.

From the spec sheet, -20V is the absolute max value... that leaves you no room for variation in the 28V, part tolerances, and any other noise spikes. If you look at the other figures it's really defined at -10V, which makes that the manufacturers "normal" operating point.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.