I found Altium document describing the power dissipation in DC DC converters. The power loss is the difference between the input power (Pin) and output power (Pout). The efficiency is Pout/Pin. In most DC DC converts the efficiency is 70 to 80 %. This can be higher for very efficient DC DC converters.

https://resources.altium.com/p/know-power-dissipation

Can we calculate the Junction Temperature (Tj) of DC DC converters, provided we know the power loss, thermal resistance and ambient temperature ? I have done such calculations Junction Temperature (Tj) for LDOs where the power loss is (Vin - Vout) x Current. Is there any difference in calculating Junction Temperature (Tj) for DC DC converters ?

One more question regarding ambient temperature. I found another document from Texas Instruments.

https://www.ti.com/lit/an/slyt793a/slyt793a.pdf?ts=1725538993221&ref_url=https%3A%2F%2Fwww.google.com%2F

On the last page, there is a figure which says that ambient temperature decreases as we increase the output current. This is not clear to me. The figure is also attached. Kindly have a look and explain how the ambient temperature decreases with increase in output current.

 

simple cause and effect...

it is not that ambient gets cooler the more current you run. quite the contrary. ambient temperature represents a limit - in this case a thermal envelope. if the limit is exceeded things fail.

term derating suggests that such limit is at play. if you want to use max current, must make sure that ambient temperature is sufficiently low.
normally the axes of the graph would be swapped but since derating is taking place, it is the output current that is set on horizontal axis (independent variable) and resulting thermal envelope is on the vertical axis (dependent variable). i can see how this can cause confusion.

forgot to mention... the last page shows no mention of current or temperature. it is "important notice and disclaimer". page before that lists relevant links.... the page with the chart is 3rd last page.

 

The limitation of all semiconductors is their JUNCTION temperature. The lower the ambient temperature at high output currents the more room there is between the ambient temperature and the maximum junction temperature. Similarly, the lower the current output current the less room you need between the ambient temperature and the maximum junction temperature.