Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Monitoring power FET/IGBT temperatures?

Status
Not open for further replies.

cupoftea

Advanced Member level 6
Advanced Member level 6
Joined
Jun 13, 2021
Messages
3,058
Helped
62
Reputation
124
Reaction score
139
Trophy points
63
Activity points
15,957
Hi,
Above 1KW , its best to have NTC’s glued to your power mosfets/IGBTs, and then gradually reduce the power throughput of the converter if the fets get too warm.
https://www.farnell.com/datasheets/3155179.pdf

NTC is best for this, because you can put say 4mA through it, and get a low noise reading out of this, as long as the wiring to the NTC is twisted.
Would you agree?...or is there a better way of monitoring FET temperatures?
 

"Glued to" sounds like DIY design. Transistor temperature monitoring is useful in some situations. IGBT and MOSFET modules are mostly equiped with NTC for this purpose. For discrete transistors, heat sink mounted temperature sensors are a popular solution.
 
heat sink mounted temperature sensors are a popular solution.
Thanks, yes, now, i can think of those NTC's built onto a metal ring terminal, so it can be tightly screwed to the heatsink....right next to the fet/igbt in question.

like this....

...but they all seem so very expensive, i wonder why. I think there must be some way of DIY'ing these with a ring terminal and some glue for the NTC?
 
Last edited:

Sure you can make the sensor mount yourself. But cheaper? I doubt.

Major disadvantage of the usual ring terminal mounted NTC is limited insulation rating, by the way.
 
Usual is type K thermocouple, fine wire, twisted tightly and then glued or clamped to the device.

We used to use Al2O3 with Cu track on both side, 3 x 3mm you can solder the type K to one side and solder the other side to the source or drain ( the static one ) on a TO-247 - this gave the best performance - and isolated as a bonus.
 
We used to use Al2O3 with Cu track on both side, 3 x 3mm you can solder the type K to one side and solder the other side to the source or drain ( the static one ) on a TO-247 - this gave the best performance - and isolated as a bonus.
Thanks, but i wasnt sure how you were using Aluminium Oxide? (you mean as in making your own thermocouple?)
Also, i wasnt sure where the isolation came in?

Usual is type K thermocouple, fine wire, twisted tightly and then glued or clamped to the device.
Thanks for this, i wondered with thermocouples though, the noise might be an issue?

Major disadvantage of the usual ring terminal mounted NTC is limited insulation rating, by the way.
..Yes i agree, an isolated small PSU would be needed to power the little micro that would measure the NTC current (via a voltage AYK)....then the result passed digitally by digital isolator, say.
 
Last edited:

The previous posts make me wonder if you are asking about temperature monitoring tools for development or series design. I was talking about series design so far.
--- Updated ---

@Easy peasy Where do you buy copper cladded Al2O3 substrate cut in 3x3 mm pieces?
 
Last edited:
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top