Effective Thermal Conductivity of Multi-layered PCB

Rt_design · May 3, 2024, 05:53

Hi everyone,
I work as a thermal design engineer for aircraft and EW applications. and we usually work with multilayered PCB like fr4, rogers etc.

My question
1. is there any hand calculations that thermal design engineers use to get the rough estimate of the effective thermal conductivity of such PCB.
i usually use parallel and series formulae for calculation of effective thermal conductivity. but the results are way off to the practical values i get when being calculated in the software ( i use flotherm XT)

2. How do you model vias under the PCB for doing thermal simulation as the vias will take a lot of time for both to model and simulate.

native · June 5, 2024, 09:39

Hi Karnataka,

1. Yes, series and parallel connection of the layers should work for calculating rough values. Of course, you need to use the right area percentage per copper layer. For more detailed simulations you need to spatially resolve the conductivities on the PCB.

2. Also for vias you could calculate effective thermal conductivities for example from the thermal resistance. Then replace the via array in your model by a solid body and assign the effective values to it.

What are the values you calculated and what do you expect from Flotherm?

Regards, Florian

May 3, 2024, 05:53	Effective Thermal Conductivity of Multi-layered PCB	#1
Rt_design New Member RaviTeja Join Date: Feb 2022 Posts: 1 Rep Power: 0	Hi everyone, I work as a thermal design engineer for aircraft and EW applications. and we usually work with multilayered PCB like fr4, rogers etc. My question 1. is there any hand calculations that thermal design engineers use to get the rough estimate of the effective thermal conductivity of such PCB. i usually use parallel and series formulae for calculation of effective thermal conductivity. but the results are way off to the practical values i get when being calculated in the software ( i use flotherm XT) 2. How do you model vias under the PCB for doing thermal simulation as the vias will take a lot of time for both to model and simulate.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
effective thermal conductivity and mass weighted thermal conductivity	Weiqiang Liu	FLUENT	8	June 19, 2020 01:08
Question about anisotropic thermal conductivity	Szweyk	STAR-CCM+	2	June 30, 2016 09:26
Thermal conductivity issue	Jared1986	CFX	3	March 20, 2014 11:38
IAPWS water properties + orthotropic thermal conductivity	Chander	CFX	1	February 29, 2012 18:26
Porous Media Effective Thermal Conductivity	Greg Perkins	FLUENT	3	August 5, 2001 22:53

June 5, 2024, 09:39		#2
native New Member Florian Join Date: Feb 2015 Location: Germany Posts: 18 Rep Power: 11	Hi Karnataka, 1. Yes, series and parallel connection of the layers should work for calculating rough values. Of course, you need to use the right area percentage per copper layer. For more detailed simulations you need to spatially resolve the conductivities on the PCB. 2. Also for vias you could calculate effective thermal conductivities for example from the thermal resistance. Then replace the via array in your model by a solid body and assign the effective values to it. What are the values you calculated and what do you expect from Flotherm? Regards, Florian