thanks bob and wayne.  I am more intend to looking for next gen tech,  
such as phase transition material build in core of the PWB.  with  
high density and super hot chip down to the road with compound semi,  
something need to be addressed on the thermal side.  Yes, I am fully  
aware of the FEM power, however, the limitation still in  the  
thermal  conduction - BN and diamond coating are still pricy and not  
compatible to many of the HDI cct.  Just try to see if there is any  
new trick out there.  Thanks again.
Best regards,
           joyce
On Jul 17, 2014, at 9:40 AM, Bob Landman wrote:

> Hi Joyce,
>
> Well, even PC's these days esp. with very powerful processors) use  
> heat pipes.  Of course they also use huge fans.  (warning! Pun  
> ahead!!)
>
> What's really cool about heat pipes is they spread the heat and  
> conduct it to the exterior of the enclosure where there can be a  
> radiator.  My Apple MacBook Air gets so hot I can't rest it on my  
> lap!  The aluminum case spreads the heat.  There is a tiny fan  
> inside but the fan's vents are under the laptop so if you do place  
> it on your lap, or when it bed reading, on a pillow, the vents are  
> blocked. Haven't had it crash on me "yet".  Why they didn't put the  
> vents on the sides? Esthetics, I'll bet, knowing Steve Jobs.
>
> E,g, http://www.thermacore.com/industries/communications.aspx    
> http://www.aavid.com/product-group/microprocessors/notebook
>
> They offer design assistance.
>
> Bob
>
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Wayne Thayer
> Sent: Thursday, July 17, 2014 9:20 AM
> To: [log in to unmask]
> Subject: Re: [TN] LIst of thermal conduction effective design
>
> Hi Joyce-
>
> Basically, you have to model the bejeesus out of the arrangement.  
> Nowadays that modeling is CFD (Computational Fluid Dynamics). Big  
> finite element model. And you really need a physical model in  
> addition to make sure all of the fudge factors are properly  
> calculated. Then, after you begin production, you will find your  
> customer changed something in the environment which rendered most  
> of the modeling moot! Getting good representations for actual  
> airflow in the application can be difficult to do also.
>
> If you are longer on time and short on money, a bunch of physical  
> models can also get you there, but if you are going to be playing  
> with the number of copper planes in the pcb that can get expensive  
> too. Used to be that someone, like a university, would run a bunch  
> of those kinds of modeling tests and publish results so you could  
> interpolate for your problem, but not any more: the math models  
> have taken over.
>
> Wayne
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Yuan-chia Joyce  
> Koo
> Sent: Thursday, July 17, 2014 6:34 AM
> To: [log in to unmask]
> Subject: [TN] LIst of thermal conduction effective design
>
> Gurus,
> The most challenge in my short career of 24 years is application of  
> thermal
> components:  the requirements just go up non stop.  How do you  
> handle the thermal load in terms of static,  dynamic or transient  
> thermal load on high density PWB/PWA, besides known build in heat  
> sink.  I am seeking for the low provide conduction type, with  
> limited or no convection cooling.  Many thanks.
> (IBM's metal backing PWB is nice, but for multi-layer, >8 is a bit  
> difficult. Besides, double side or high density might have some  
> limitation - any new development in the field? ) Thanks in advance.  
> (net a bit quiet, I guess many on summer vacation mode with BBQ  ;-).
>              joyce
>
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]  
> ______________________________________________________________________
>
>
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud  
> service.
> For more information please contact helpdesk at x2960 or  
> [log in to unmask]
> ______________________________________________________________________


______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________