At the Speed of Heat
July 31st, 2009, by Many analogies exist relating heat flow to fluid flow. Oops, there’s one. A good electronics thermal design is one that gets the heat out from the heat sources quickly and the cold in to quench them quickly. Before I scoot off to sunny (though heavy rain is forecast) Devon for a family holiday I thought I’d leave you with this…
Thermal power dissipation is measured in Watts (W)
A Watt is a Joule/second (J/s)
A Joule is a Newton meter (Nm)
So Power = m/s x N
A speed multiplied by a force.
Does this mean that there is a concept of a (steady state) ‘speed of heat’? If so what is it? These aren’t rhetorical questions by the way 
31st July Hampton Court
More Blog Posts
Add Your Comment
About Robin Bornoff's blog
Views and insights into the concepts behind electronics cooling with a specific focus on the application of FloTHERM to the thermal simulation of electronic systems. Investigations into the application of FloVENT to HVAC simulation. Plus the odd foray into CFD, non-linear dynamic systems and cider making.
Latest Posts
- Why Not Just Shove a Heatsink on Top of it? Part 2: Heat Flow Budgets
- Why Not Just Shove a Heatsink on Top of it? Part 1
- Experiment vs. Simulation, Part 5: Detailed IC Package Model Calibration Methodology
- CFD – Colourful Friday Distractions
- Experiment vs. Simulation, Part 4: Compact Thermal Models
- Experiment vs. Simulation, Part 3: JESD51-14
Comments (↓ Add Your Own)
One comment on this post
Commented on August 17, 2009 at 8:55 am
By Johannes
Dear Robin,
a really nice idea, but I don’t think it will come to a result because the equations of motion (Newton’s m*a=F) and heat transfer (Fourier + heat exchange term) are quite different in nature. Even if we consider as a mechanical counterpart a 3D air flow, we are comparing a hyperbolic with a parabolic partial differential equation. In simpler words, heat transfer is always tending to smooth gradients instead of “transporting a thermal signature” at some speed.
What we can do is to speak about time scales, i.e. the characteristic time where some temperature changes happen. AND we would have to include words about boundary conditions.
To make a simple example, let us consider a bar of some length L (to get meters into the game), feed in some heat signal on one side and watch the time when the heat is arriving at the other (cold) end. All circumferal planes shall be insulated to prevent other heat losses. In any case, the time(scale) at which the temperature signal arrives is always proportional to the square of the length of the bar: time=(density*specific heat)/(conductivity*length**2). This means that speed=length/time does not behave as it does in mechanics: double time = double distance at constant speed, but depends on geometry. Should that be another sort of relativity? (Joke)