PCB reliability 1

[Ste134]

Hi,
I am struggling with life prevision of bare Printed Circuit Board (PCB).
The approach is to submit a bunch of PCBs to climate chamber accelerated life 885°C/85%) for a certain amount of time.
The problem then is the correlation between such accelerated life time to teh standard/specs operating life time.
How much does 1hr in the chamber (accelerated life) correspond to operating (T°C) life time ?
Can anyone provide simple method, material (white paper, presentation, ..) to correlate the two ?
Thank you in advance for any help....
Stefano

 

[VEBill]

They won't last long at 885°C.

 

[VEBill]

Google Scholar with the terms 'accelerated life temperature' brings up plenty of formal papers on the subject.

Normal Google has plenty of explanations on various Wikis.

Good luck.

 

[IRstuff]

A websearch reveals:

https://en.wikipedia.org/wiki/Highly_accelerated_stress_test

http://www.tdipower.com/PDF/white_paper/hass.pdf

http://arxiv.org/pdf/0708.0369.pdf

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[hendersdc]

How much will you learn from subjecting bare PCBs to life testing? You will need to repeat the testing on populated boards again won't you?

 

[DHambley]

IRStuff, Thanks for those links. It was weird reading the reasoning in Weglinski's paper; He decides to only consider thermal effects even knowing that power supplies do fail often at turn on, and noting that the turn-on surges degrade components considerably. He wrote that the decision to not do that testing was because few people had written a paper on it.

Darrell Hambley P.E.
SENTEK Engineering, LLC

 

[MacGyverS2000]

In other words, "I can't find any shoulders to stand upon, so I won't offer my own shoulders..."

Dan - Owner

http://www.Hi-TecDesigns.com

 

[IRstuff]

Things are not quite as they were many eons ago w.r.t. turn-on. Since the surge currents are relatively understood, the circuits and power supplies can be designed to accommodate that. One thing that is rarely talked about is the "dormancy factor," which is the percent of the active power failure rate assigned to when the equipment is turned off. Generally, that's in few percent range, but high temperature accelerate that. Certain failures can still be accelerated solely be temperature. A example might be metal spikes penetrating integrated circuit junctions due to high currents. Once the spike is formed, temperature alone can get the metal to continue to migrate into the junction and short it out.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[Ste134]

Hello and thank you All for feedbacks.
Actually, I was not looking for theoretical, school or Wikipedia's material but rather something more operating and real experience-based.
I got plenty of materials from Internet but I was asking real field experience and help.
We experienced failure in prototype flex pin-to-pin PCB due to (we assume) bath infiltration under the cover lay and we need to perform in-house accelerated test to understand risk on parts in the field.
Do anyone know a quick-to-apply relation or procedure to evaluate something like "1hr test @ certain T/RH conditions = X hrs at operating conditions ?"
I understand it is something like a magic recipe but it is what I need .....
Thank you to All for help.

 

[IRstuff]

So, the equations are simple, as shown in the Wikipedia article:

https://en.wikipedia.org/wiki/Highly_accelerated_stress_test.

The difficulty is in getting meaningful activation energies or equivalent.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[Ste134]

It seems that Ea= 0.6eV is a commonly accepted reference value.

I have also found a relation showing the acceleration due to RH and it could be another acceleration factor in my tests.
A=t_test/t_opearting=(RH_test/RH_opeating)^3* <Arrhenius equation>
Where:
RH_test= RH during test
RH_perating= max RH in operating conditiosn
t_test= test time
t_operating=life under operating conditions