Vibration of PCB

[izax1]

Hi there
I've been searching and wondering how to correctly FE-model a populated PCB with components. I get excellent input from Steinberg book, but how to model thia with FE is still a mystery. The Componens will add both mass and stiffness to the PCB, and to model this corectly without using too much effort in detailed modelling of the components is for me a mystery.
In my case I need this for fatigue analysis of solder connection.
Any good ideas would be appreciated.

 

[drawoh]

izax1,

I would think that the FEA would be simple if you have an accurate model. Do you have the information you need to build the model?

--
JHG

 

[IRstuff]

Would your fatigue analysis be at all plausible if you cheat on the component models?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[izax1]

Yes, that's the thing. This is not an accurate model, because I don't have the stiffness of the components.

 

[drawoh]

izax1,

Can you do it by hand? You have a board of known stiffness with some sort of distributed mass. You can work out the bending radius under vibration. You can then evaluate the individual components strained to that radius.

I cannot see a chip manufacturer providing the information you need to create an FEA model. You can work out the bending and ask if their chip can handle that.

--
JHG

 

[izax1]

IRstuff, thanks for your feedback.

No, not at all. But I'm wondering: how can I get the correct stiffness of the components without coming down to the nitty gritty details of each component with soldering and cabling. Do you have any good suggestions on how to do this?
Or anyone else on the forum.
Thanks fo any feedback.

 

[izax1]

drawoh. Well you don't have a board of known stiffness. All the components will alter the PCB stiffness.

 

[drawoh]

izax1,

The components will make your board stiffer than you calculated, providing you with a conservative result. Do you really need to get this bang-on?

--
JHG