# Calculator for heat rise in materials from applied power

Discussion in 'Mathematics and Physics' started by riccardo, May 9, 2015.

1. ### riccardoMember

Joined:
Apr 15, 2010
Messages:
107
Likes:
2
Location:
United Kingdom
2. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
ONLINE
Hi,

One of the harder things to calculate is the heat loss, and material temperature gradient.

When the heat loss equals the applied heat we reach thermal equilibrium, but if we dont know the heat loss it's hard to say what the final temperature of the component will be.
There are exponential approximations, but then we also might be concerned about the gradient between different material layers so that the temperature at the very center of the component is hotter than the layers closer to the outside of the component where they contact the cooler ambient air.

3. ### Tony StewartWell-Known MemberMost Helpful Member

Joined:
Aug 31, 2012
Messages:
3,152
Likes:
282
Location:
Richmond Hill , ON Canada near Toronto
The biggest lesson I taught myself in thermal design was how to drop the hotspot temperature in a 180W open PSU inside an enclosed 1U (1+7/8" high) rack.

I started with dual fans and laminar flow and had a worst case hot spot of 60'C rise.
After smoking in the office with a flashlight one evening I got it down to 10'C rise with the same air flow.

The trick is to create a thin spoiler on air intake to start up eddy currents instead of laminar flow. THis increased the radial air velocity in contact with the part, which is more important than linear air velocity (over top of the trees) which is much more than important than volumetric air flow. (CFM) which takes path of least resistance.

Joined:
Jan 12, 1997
Messages:
-
Likes:
0

5. ### MrAlWell-Known MemberMost Helpful Member

Joined:
Sep 7, 2008
Messages:
11,049
Likes:
961
Location:
NJ
ONLINE

Hi,

Tony your scope and depth of knowledge truly amazes me, and i am a very hard person to amaze.

Your solution makes a lot of sense. If the air flows too systematically it may systematically never reach some areas. I found that if i hang something in front of a fan that can move with the air forward, backward, and side to side a little (like a ribbon or ribbons) the air gets diverted in somewhat random directions rather than straight out from the front. I believe it is called "turbulent flow".

6. ### Tony StewartWell-Known MemberMost Helpful Member

Joined:
Aug 31, 2012
Messages:
3,152
Likes:
282
Location:
Richmond Hill , ON Canada near Toronto
Correct. In my case i bought 100 ft spool of 1mm mylar, then cut shape to serve two purposes.
1) satisfy UL,coke spill test over HiV to DC open frame enclosed PSU
2) spoiler fold down on intake to create turbulent eddy currents to reach surface of all MOSFETs and ferrite cores.

Then I found 1.5" fans too noisy at full speed so I used LM317 , pot , thermistor epoxied the ferrite core, 1signal transistor bridge amp to drive ADJ and used FR4 waste tabs to use as an insulated nut for self tap nylon screw to chassis.
Then fan was 0 Rpm up to 80W load and gradually increased to 50% speed at full load 40'C ambient from 45 to 50C inside at hot spot.
Total budget I designed for was 2\$ including connector for precision fan speed control, 1k/mo

Who knows Al, we may have had connections from my days in Downingtown, Paoli , Rockaway (HP PSU factory) or elsewhere nearby. Its hard to be humble, but I have done a lot.