G2 18v25 catastrophic failure

Hi there,
I purchased the g218v25 to drive a dc motor at max 14amps 16.3v max (from a lithium battery)
The driver is driven using only pwn and direction pin (and ground of course) by a microcontroller which has a bug in the firmware which cause somethimes to go in a reset state with unknown GPIO state.
First time this reset happened the motor had no load and it started turning for 4 seconds before i disconnected the battery. The mosfet where really hot.

Today the same happened under load conditions ( 10amps load normally) and the mosfet burned in few seconds. I would expect that there is no possibility to have a direct flow from high to low side mosfets but i really cant imagine why such a microcontroller failure could cause something like that.

All the electronics around the g218v25 is alive.

Help me please cause im lost!!!
Br,
Marco

One more question… what happen if i drive faster then 100khz (maybe this happened during the microcontroller failure)? Is it possible that the transistors work in ohmic mode burning and cross conducting from high to low side? or simply the driver filter frequencies higher then 100khz? Any schematic available for the g2 18v25? Thanks.
Marco

Hello.

I am sorry you damaged a couple of your motor drivers. Can you tell me more about how you were using them? What microcontroller are you using? Can you provide a link to the datasheet for your motor or at least its name and where you got it from? Can you also post pictures of the damaged boards? (If you can, try to focus on any obvious signs that show specifically where the damaged occurred.)

As for your other questions, sending a pulse faster than 100kHz should not inherently damage the G2 driver. A pulse on the PWM pin must be high for a minimum duration of approximately 0.5 µs before the outputs turn on. So, when the pulse frequency increases to a high enough value, the minimum duration forms a significant portion of the signal and you lose the ability to effectively command the motors. For example 100 kHz, the pulse period is 10 µs, and the minimum non-zero duty cycle achievable is 0.5/10, or 5%.

-Jon