# Maximum junction temperature

Discussion in 'General Electronics Chat' started by kanoun, May 21, 2015.

1. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
Hi guys
I would like to measure the maximum junction temperature that my devices could sustain. My devices are 5v mosfet. PN diode

2. ### Alec_t AAC Fanatic!

Sep 17, 2013
5,976
1,136
Really? The best you can hope for is to measure the device case temperature . The maximum junction temperature will be the temperature at which the junction vapourises .

3. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,066
You cant measure junction temperature. You can measure case temperature. You can only estimate junction temperature by knowing the thermal resistance from case to junction, and know the power being dissipated in the device.

4. ### ErnieM AAC Fanatic!

Apr 24, 2011
7,439
1,626
It's not completely impossible to measure a junction temperature as the voltage across a junction is a function of the junction temperature.

However, you need to pin down several other constants to get this to work, meaning you need a temperature controlled environment to conduct some tests, and a very good data reduction scheme. I'd be suprised if there are no documented ways to do this.

I have a dim memory of a commercial device that would pulse power into a transistor to measure the thermal impeadance. It was used where I worked to test the attach of transistors we had done. However, I was never involved in this so don't know any particulars.

However, when you are done all you have is the temperature. Then you have to do lifetime studdies to see how long your device will work at such a temperature.

A far better method is to read the manufacturer's data sheet for the safe opperating temperature and stick to their guidelines.

5. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
Many thanks for those replies ... Indeed, I know how to get a junction temperature but I want to size when only to know how could I get the max temperature the device could sustain befor thermal runaway ... I want to reproduce the max tj given in the datasheet

6. ### crutschow Expert

Mar 14, 2008
13,501
3,375
The max Tj given in the data sheet is generally not related to thermal runaway, it's the point above which the reliability of the device becomes seriously compromised.
If you note, even the maximum non-operating storage temperature of the device is unusually equal to or not much above the maximum operating temperature and that temperature obviously has nothing to do with its operating characteristics.

I don't see why you would want to try to determine that point yourself.

7. ### WBahn Moderator

Mar 31, 2012
18,087
4,917
How do you plan to get the junction temperature?

"The max temperature the device could sustain before thermal runaway" is pretty much a meaningless phrase. Thermal runaway is unrelated to max junction temperature. Do you understand what thermal runaway is?

8. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
i know what I'm talking about ... Do you know the difference between the thermal and electrical instability ... In fact when you increase the temperature above a certain temperature of a mosfet (ldmos) for instance, you decrease the vbe of the parasitic bipolar leading to the device destruction ... And this is because the current is very high and then the tenperature that the semiconductor can sustain is surpassed

9. ### WBahn Moderator

Mar 31, 2012
18,087
4,917
Okay, let's take a step back and try to get a firm handle on what it is you are trying to ask about.

Your OP said that you want to measure the maximum junction temperature your devices can sustain. That indicates that you are asking how to make a particular measurement, namely the junction temperature. But then you subsequently stated that you know how to get junction temperature, so what is it that you need help with? You then said that you want to reproduce the max junction temperature in the data sheet. What do you mean by that? That you want to know how to force a device to operate in such a way that it's junction temperature is actually at the max junction temperature in the data sheet? Or you want to determine what the actual max junction temperature is for a particular device in the same way that the max junction temperature was determined in the data sheets?

If it's the latter, then you need to understand what the metrics are that were used in establishing the max junction temperature in the first place. There are many and they are both acute and chronic -- meaning that some of them are related to things that go bad in a very short time frame when you exceed a certain temperature and others are related to things that reduce the life expectancy of the device if you exceed a certain temperature. Do you know what the relevant metrics are?

kanoun likes this.
10. ### crutschow Expert

Mar 14, 2008
13,501
3,375
The manufacturer states the maximum operating junction temperature of the device. The junction temperature can be determined from the device power dissipation, the ambient temperature, and the device thermal resistance from junction to air, so I don't see the purpose is of what you are trying to do.
You seem to be trying to reinvent the wheel.

11. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
Actually I want to know if the degradation of a device could impact its maximum temperature it could sustain

12. ### WBahn Moderator

Mar 31, 2012
18,087
4,917
What type of degradation? There are many types.

In general, I would expect most forms of degradation to impact the maximum junction temperature, but again it matters what you are talking about when you talk about the maximum temperature it could sustain. What are you metrics?

13. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
The degradation is a hci one and the metrics would be vth, ion, gm

14. ### DickCappels Moderator

Aug 21, 2008
2,774
671
If you stick to the manufacturer's recommendations you should be just fine.

15. ### crutschow Expert

Mar 14, 2008
13,501
3,375
Yes, it could.

16. ### kanoun Thread Starter Member

Jan 29, 2015
41
0
I need to check it and demonstrate it

17. ### crutschow Expert

Mar 14, 2008
13,501
3,375
Is this just for you own edification?
I see no other reason to do that since it's not needed to do a design.

Jan 29, 2015
41
0
Yes it is