Thoughts about heat, sinks, and the stuff between

John Wilson

Posted Jul 28, 2009

A couple of days ago I was in the hallway conversing with a colleague about the power, pardon the pun, we have in our mobile phones.   Being thermal engineering types the conversation shifted towards how well the device could perform all of these amazing things while still remaining cool.   He commented on how well the thermal engineers did in designing the cooling solution for this particular product.  While I applaud all those involved in getting this device in my hands I do think that there is only so much credit the thermal engineer can take…

A quick aside… I discuss thermal issues with quite a lot of people, all of which are experts at something but that something isn’t always thermodynamics, heat transfer,  or fluid dynamics  (I don’t claim to be an expert but I have been able to make a living at it for the last 10 years).  Many times the conversation begins by them discussing their need “to get the heat out” of their product.  I know what they mean but just so we are on the same page, no matter what I do as a thermal engineer-the heat will get out.  My job as a thermal engineer is to manage the heat flow, fluid flow, and related temperatures in order to satisfy the requirements of the design.

To discuss my line of thinking on the  fancy hand held device, let’s make some assumptions:

1. The heat will get out.
2. The surrounding air is an infinite sink (and for 99.999% of all thermal applications is the final destination for the heat)
3. Let’s ignore radiation…conservative simplifying assumption

And refer to the artistic representation below:

In my simplified thermal world the surface temperature is governed by Q (heat), h (heat transfer coefficient), A (convective surface area), and  T (ambient temperature).  Let’s not get bogged down with the temperature we feel when we are holding the device (that was discussed here).  Of the parameters mentioned, there are a couple of that won’t change significantly, T and h, and  Q is totally out of my hands.  With regards to the surface area, I doubt I will be granted the flexibility to add many fins on the outside of  the hand held device.  I can, however, manage the heat flow on the inside of the device to improve the effective convective surface area.  Ideally, the convective surface is one uniform temperature, so as a thermal designer I could work to expand the heat flow path from the internals to the external surface.

While I owe thanks to the thermal engineers that are involved in putting all of this power in my hand I do believe there is a limited amount of strategies they have available to them.  In my world, the only thing they can do is drive the surface temperatures to not deviate from the bulk average governed by the simplified equation outlined above.  Anyone have any thoughts they would like to share about this, I’d love the comments.

More Blog Posts

Thanks for sharing your thoughts. I totally agree with you. In this kind of cases, heat spreading is an important point for the thermal management. If radiation is considered, another way to reduce the temperature is increasing the external surface emissivity, especially at the big screen area. I guess the normal LCD screen has low emissivity. Do you have any trick to improve that?

Single Chen
6:19 AM Jul 29, 2009

Thanks for sharing your thoughts. I totally agree with you. In this kind of cases, heat spreading is an important point for the thermal management. If radiation is considered, another way to reduce the temperature is increasing the external surface emissivity, especially at the big screen area. I guess the normal screen has low emissivity. Do you have any trick to improve that?

Single Chen
6:25 AM Jul 29, 2009

'as a thermal designer I could work to expand the heat flow path from the internals to the external surface' Moving the heat from inside to outside is mostly by conductive heat transfer path right ! I wonder what is the typical % of radiation in the total heat transfer from inside to outside ? On the same topic how much can we increase the emissivity of the hot component or board in the device. So is a black (not so shiny) Phone cooler than a shiny fancy cell phone both with same total power?