I'm using a TIP120 (Darlington pair BJT) brought to saturation for a project of mine. I have \$V_{CE(sat)}=1V\$, \$I_C=2A\$, \$V_{BE}=2.5V\$ and \$I_B=0.005\$A, which give me a total power dissipation of:

\$P_D=V_{CE(sat)}*I_C+V_{BE}*I_B \approx 2W\$

When I look up the component datasheet to check the absolute maximum ratings, there are two values given for power dissipation: one at 65W (@ \$T_C=25°C\$) and one at 2W (@ \$T_A=25°C\$), as seen on the image below:

Absolute Maximum Rating section of TIP120 datasheet

So my question is: what is the difference between the two values? What is the difference between \$T_A\$ and \$T_C\$?

Sorry if this is a common question, I've searched everywhere to try and answer that question, but search engines are not very helpful when I want to know the purpose of parameters found in electronic datasheets (if there exists a glossary for the most common parameters found in datasheets somewhere, and someone has a link, I'd be very happy to use it!).

I suspect that I should use the first value for some reason, but given that my calculated \$P_D\$ value is pratically the same as the second one, I don't want to take any chance and destroy my future setup, making all that magic smoke escape...



The ON datasheet is rather confusing (or rather doesn't explain its notations). The 65W refers to the [max] power dissipation if you manage to keep the case at 25C. The 2W refers to an ambient temp of 25C, but no restriction on the case temp. This is a bit more clear from the Bourns datasheet of their similar product.

enter image description here

What this means in practice is that 65W is the max you can hope for with an ideal [possibly very large] heatsink.

Both of these data are actually a rather convoluted way of saying the same thing, namely that the max junction temperature allowed is 150C. This can be verified using the following data:

enter image description here

  • 1.92*65 + 25 = 124.8 + 25 = ~ 150C
  • 62.5*2 + 25 = 125 + 25 = 150C.

Which is actually given as such in the datasheet:

enter image description here

Now for practical purposes, I would suggest using a small heatsink rather than betting you won't fry it at exactly the dissipation limit for use without one.

If you want to calculate the temp rise with a heatsink, say one which gives 13C/W, then you add the heatsink's thermal resistance to that of the case (1.92C/W) and the interface material, let's say 1C/W, which would give you about 16C/W total resistance. For 2W that translates into 32C temp rise over ambient, so at 25C you'd have 57C. That's pretty decent for not frying yourself when accidentally touching it.

  • \$\begingroup\$ Your answer is exactly what I was looking for (and even more since you've explained how to actually choose a heatsink in practice, which was unknown to me). Thanks a lot ! \$\endgroup\$ – MatLag Dec 6 '15 at 9:26

Ta - Ambient temperature
Tc - Case temperature. Temperature of surface of the selected IC package
Tj - Junction temperature

Most datasheets specify the Tc in their specifications.

A simplified way to think of it:

  • The value given for Tc is the maximum you are able to squeeze out of the component. It refers to the value achievable, providing the case is kept at the specified temperature.
  • The value given for Ta is the upper limit of what you should expect out of the component when little to no heat sink is used.
  • Tj is often only used to indicate what temperature range the FET junction can operate at.
  • If you use a heatsink you will be able to operate somewhere between the Ta and Tc value (depending on the characteristics of the heatsink).
  • The component can usually handle a short duration pulse up to the maximum Tc value.

[I have tried to keep the above text generic as these notations are used for both power dissipation and current draw.]


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.