Delkrak

The STM32F405 chips do get a bit warm (either by themselves or courtesy of heat dissipated throughout the PCB from LDO power regulators, etc.). If it feels hot then something is wrong. If it just feels warmer than the air conditioning...that's pretty normal (at least when the chip is running at 168MHz constantly).

STM32F405 get significantly hotter than the voltage regulator, just to add a reference.

The Ethernet chip physically almost hurts to touch after a minute of run-time. (i was able to hook it up and receive packages through it, which indicate that it should be working)

On I2C: do you have a logic analyzer handy by any chance? If you can drive the pins high and low (using OutputPort) and read logic levels correctly with InputPort, then this is most likely not a hardware issue.

Think I found the problem, as SCA is constantly read as low. Now the next question is what can cause this to happen?

Still, it is wired that no change occur at SCD during I2C transfer, which should be independent of the read state of the clock line(SCA). Maybe there is something wrong within the I2C module of the chip as well.

I would love to get my hands on a sample code that for sure will change the I2C pins, just to make sure.