NPN transistor overheating even when used with appropriate parameters

In summary: It is important to clarify this when seeking advice. Additionally, it is important to consider all design specifications, such as inputs, control, output, power management, cooling, switching speed, PWM, slew rate, diode recovery time, efficiency, and maximum temperature rise. These factors will impact the overall performance and reliability of the circuit.
  • #1
Erik_clifton102
13
1
TL;DR Summary
Using TIP31C NPN transistor to amplify a current from 0A to 10A with 10V, but the transistor is over heating, max loads are 100V and up to 25A if I have read the data sheet correctly.
I have been trying to build a simple Current control circuit using a NPN TIP35C transistor but have run into the problem of it constantly over heating and being destroyed. The transistor is the only component that heats up whilst it is on. The supply voltage is 10V, well below the Max and at the moment I am only running a max of 7A through the transistor. If I have also done my calculations a am well below the power dissipation for the transistor.
what could I possibly be doing wrong or what have a missed? or do I just need a heat sink to dissipate the heat?
I have attached the Datasheet for the transistor below.
 

Attachments

  • ZT2293-dataSheetMain.pdf
    167.1 KB · Views: 86
Engineering news on Phys.org
  • #2
Erik_clifton102 said:
If I have also done my calculations a am well below the power dissipation for the transistor.
Did you do the calculations for the heatsink (you have a proper heatsink, right?) too?
 
  • Like
Likes DaveE
  • #3
Please describe the load resistance. Or even better show us your schematic. We can't answer without that information.

Have you done the thermal calculation with the worst case power dissipation and using the thermal resistance parameters in the data sheet. Using TJ ≤ 130C, TA ≤ 50C (a fairly conservative design), this device can only dissipate PD ≤ 2.24W without a heat sink. At IC ≤ 7A, you are limited to VCE ≤ 0.32V. So, yes, you undoubtedly need a heatsink, probably a big one.

Read these if you aren't familiar with this sort of thermal calculation, it can be a bit complicated.

https://www.ti.com/lit/an/slva462
https://components101.com/articles/selecting-the-right-heatsink
https://cheever.domains.swarthmore.edu/Class/e12Code/HEAT-NOTE.pdf
 
  • Like
Likes jrmichler and Rive
  • #4
Erik[QUOTE=Erik_clifton102 said:
TL;DR Summary: Using TIP31C NPN transistor to amplify a current from 0A to 10A with 10V, but the transistor is over heating, max loads are 100V and up to 25A if I have read the data sheet correctly.
...
what could I possibly be doing wrong or what have a missed? or do I just need a heat sink to dissipate the heat?
I have attached the Datasheet for the transistor below.

With Ic=25A, Ib=5A Vce(sat)=4V thus 100W and now you need a forced air cooled 100W heatsink.
This is a poor choice. Consider a Nch FET switch instead. rated > 100V with flyback diode added and possibly rated at 50A so you can operate at the 50% of max rated temp for better reliability.

regards, Tony EE since'75
 
  • #5
Rive said:
Did you do the calculations for the heatsink (you have a proper heatsink, right?) too?
Originally I didn't have a heat sink, looking into one now but from other advice I'm looking at using a FET instead.
 
  • #6
DaveE said:
Please describe the load resistance. Or even better show us your schematic. We can't answer without that information.

Have you done the thermal calculation with the worst case power dissipation and using the thermal resistance parameters in the data sheet. Using TJ ≤ 130C, TA ≤ 50C (a fairly conservative design), this device can only dissipate PD ≤ 2.24W without a heat sink. At IC ≤ 7A, you are limited to VCE ≤ 0.32V. So, yes, you undoubtedly need a heatsink, probably a big one.

Read these if you aren't familiar with this sort of thermal calculation, it can be a bit complicated.

https://www.ti.com/lit/an/slva462
https://components101.com/articles/selecting-the-right-heatsink
https://cheever.domains.swarthmore.edu/Class/e12Code/HEAT-NOTE.pdf
I thought I could originally use it without a heatsink as it was going to only be used for about half of its capabilities. but definitely looking into one now.
thank you as well for the links
 
  • #7
TonyStewart said:
With Ic=25A, Ib=5A Vce(sat)=4V thus 100W and now you need a forced air cooled 100W heatsink.
This is a poor choice. Consider a Nch FET switch instead. rated > 100V with flyback diode added and possibly rated at 50A so you can operate at the 50% of max rated temp for better reliability.

regards, Tony EE since'75Thanks for that advice, i did a bit more research and think that what you suggested would be better suited for my little project.
 
  • #8
- start with all design specs, then design it.
- must haves, nice to haves, inputs control, output , power cooling management switching speed PWM? slew rate, diode recovery time, efficiency, max temp rise, etc
 
Last edited:
  • Like
Likes DaveE
  • #9
I thought it was a linear regulator design, but now I don't know why. Maybe the use of the term "amplify". Anyway, some of the comments apply to switching topologies, some to linear topologies. Keep that in mind; know which sort of circuit you are asking about (extra credit: tell us too).
 
  • Like
Likes berkeman
  • #10
DaveE said:
I thought it was a linear regulator design, but now I don't know why. Maybe the use of the term "amplify". Anyway, some of the comments apply to switching topologies, some to linear topologies. Keep that in mind; know which sort of circuit you are asking about (extra credit: tell us too).
DaveE said:
I thought it was a linear regulator design, but now I don't know why. Maybe the use of the term "amplify". Anyway, some of the comments apply to switching topologies, some to linear topologies. Keep that in mind; know which sort of circuit you are asking about (extra credit: tell us too).
I'm trying to use the transistor to amplify and regulate the current for a load, I've been using a simple voltage divider paired with a pot to regulate the current. Anything over 5A at 10V starts to heat the transistor dramatically so I was wondering weather I need a heat sink or I need to change my way of regulating the current allowed through the transistor, or maybe even change the transistor to a MOSFET.
 
  • #11
1st you need to answer my questions in #8.

what is the load and voltage drop * current on every part in schematic. It may need to be redesigned.
 
  • #12
Belated, but please allow for 'Stinkin' Hot Days' with your heat-sinking.
Else, sure as Murphy, you'll get a day that's pushing past 40ºC in the shade...
 

Similar threads

Replies
68
Views
4K
  • Electrical Engineering
Replies
7
Views
6K
Replies
21
Views
1K
  • Electrical Engineering
Replies
28
Views
4K
  • Electrical Engineering
Replies
2
Views
1K
  • Electrical Engineering
Replies
10
Views
5K
Replies
2
Views
2K
  • Electrical Engineering
Replies
3
Views
6K
Replies
15
Views
2K
  • Electrical Engineering
Replies
4
Views
3K
Back
Top