I made a half-bridge following the recommended schematics from the IRS2184 datasheet. However when I powered it, it triggered over-current and got quite hot. Here is are schematic and routing. enter image description here enter image description here

I checked my pcb for shorts and didn't find any. At first, altium didn't place the pins at the correct place on the IC so I had to modify them according to this schematic enter image description here

What could make it work ? Thank you!

  • \$\begingroup\$ Power it with SD connected to ground and see what happens. Also - is the voltage regulator connected the proper way? - check it. \$\endgroup\$ – fifi_22 Feb 26 at 20:20
  • \$\begingroup\$ @fifi_22 that would (or should) shut it down... \$\endgroup\$ – user_1818839 Feb 26 at 20:32
  • \$\begingroup\$ @fifi_22 Thank you for your suggestions! I'll try that asap, as for the regulator, I do get 12V on the Ic's power pin when powered with 24 V \$\endgroup\$ – antoine serry Feb 26 at 20:55
  • \$\begingroup\$ THT in 2021? Really? Do you have a ground plane? Why isn’t your decoupling capacitors placed as close to your driver as possible? \$\endgroup\$ – winny Feb 26 at 21:44
  • \$\begingroup\$ @winny Hi, thank you for your feedback. I get why THT isn't the best but I'm working with what I have, this is nothing professional, just an after school project :) It shouldn't be the reason why it won't work though \$\endgroup\$ – antoine serry Feb 26 at 22:02

Make sure the 12V supply comes up first, it needs to happen before the 24V rail to ensure the gates are off. The big 380uF cap is probably preventing the 24V rail from coming up and depending on the supply the time constant could be in the millisecond range.

Put a scope on the gates of the fets (use GND of QL for the ground of the scope) when your powering up, they should be 0V while powering up. There may be a short amount of time where there is a transient from the board layout and both gates turn on at the same time.

  • \$\begingroup\$ Thanks for your answer, How could the 12V turn on before the 24 if it's created with a regulator? I'm not sure I understand what you mean. Ill be trying in a minute, is that why we often put pull down resistors on the gate in addition to the driver? If the gates pass the threshold it would short them and trigger the constant current, how would that damage the driver though? \$\endgroup\$ – antoine serry Feb 26 at 23:20
  • \$\begingroup\$ I remember somebody's text about those bootstrap drivers. He wrore: ALWAYS put pull down resistors (eg. 10k) on the mosfet gates) I'm not saying that this is the problem here, but I fried a lot of IR2110 before fixing that. \$\endgroup\$ – fifi_22 Feb 27 at 4:10
  • \$\begingroup\$ I forgot to mention that in the post, but if you do see voltage on the gates on startup then you'll definitely need a pull down resistor on the gates to keep the gate low while the power supplies are sequencing. \$\endgroup\$ – Voltage Spike Feb 27 at 5:21
  • 1
    \$\begingroup\$ Another problem is you might have an issue with fast transients and poor layout causing both gates to turn on but that would be unlikely it's more likely that it's the power supply start up \$\endgroup\$ – Voltage Spike Feb 27 at 5:21
  • \$\begingroup\$ Add some pull down between GS of the FET to ensure they are off when the 24V are activated as the 12V for the Gatedriver rises slightly later. \$\endgroup\$ – schwdk Feb 27 at 15:01

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.