EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: JAGUAT on October 12, 2018, 09:43:50 pm

Title: Never ending Mosfet temperature
Post by: JAGUAT on October 12, 2018, 09:43:50 pm
Hi guys im an electronics engineering student from Guatemala (not sure if you have ever heard about this country) but there is a something that is driving me crazy, i have made some calculations about the power a Mosfet is disspating if im not wrong it is 7.38 watss (mostly because its duty cicle is 50%) the Mosfet im using is the IPP034N08N https://www.infineon.com/dgdl/Infineon-IPP034N08N5-DS-v02_00-EN.pdf?fileId=5546d4624ad04ef9014ade84c8b17b40 (https://www.infineon.com/dgdl/Infineon-IPP034N08N5-DS-v02_00-EN.pdf?fileId=5546d4624ad04ef9014ade84c8b17b40)

in my design im talking about a 4.4 m \$\Omega\$ on resistance and accordring to the datahseet that will be true if i keep the junction temperature at    80 °C.
this place https://www.electronicdesign.com/power/determine-mosfet-junction-temperature-and-switching-losses-various-package-types (https://www.electronicdesign.com/power/determine-mosfet-junction-temperature-and-switching-losses-various-package-types)
says that the temperature of the draing is Td=tj-(Pj*Rjd) im not really sure if this refers to the case.
the datasheet says that Rjc is 0.9 K/W, so if i do the math that means that the Mosfet should be below 73 °C to keep the junction at that temperature. Im wondering if this is correct or is this not as straightforward as im doing it. 
Title: Re: Never ending Mosfet temperature
Post by: german77 on October 12, 2018, 11:04:01 pm
With only knowing the mosfet is imposible to help you. What current is going trough, gate voltage, configuration, schematic?

If a mosfet heats mostly comes to 3 factors. The gate voltage is not enough therefore the on resistance is much higher than 4.4m ohms, The gate rise and fall times are too slow and transients cause a lot of heat. And your frequency is too high for the mosfet to handle.

For the first factor just leave it on and check the temperature, if it's hotter that it should be, you need more voltage.
For the second keep the same duty cycle and check if slower frequencies generate less heat. If yes. Your rice and fall times are too slow. You need a way to push more current into the gate.
The third one is unlikely this mosfet can reach 12Mhz.
Title: Re: Never ending Mosfet temperature
Post by: JAGUAT on October 12, 2018, 11:39:21 pm
the Mosfets are in a full bridge configuration for a high frequency transformer (50 Khz), the load is 25 amps i calculated the switching and conduction losses. I think it wouldnt be precise butwhat im trying to do is to predict the temperature the mosfet will reach since i have to wait about 2 weeks to get the components. Im using the UCC27714 half bridge driver.

Or may be i will just have to build the circuit and what happens
Title: Re: Never ending Mosfet temperature
Post by: coppercone2 on October 14, 2018, 04:16:05 pm
https://assets.nexperia.com/documents/application-note/AN10874.pdf

This does not answer your questions directly but if you read through this document you can get an intuitive idea of how power electronics behave thermally with common configurations.
Title: Re: Never ending Mosfet temperature
Post by: tooki on October 15, 2018, 12:32:15 am
I don’t have anything to add to your question, but I’m half chapin myself! :p My dad is from Guate and I was born there. (My sister is actually there now for a visit!)
Title: Re: Never ending Mosfet temperature
Post by: Jwillis on October 15, 2018, 12:52:13 am
If you look at the package the Tab is also the drain which is typical of Mosfet design.So Td=tj-(Pj*Rjd) Refers to the Drain which also includes the Tab which will be attached to a heat sink.So temperature measurements are taken at the tab.Hope this helps somewhat.And yes I have heard of  Guatemala.Best wishes on your education.