Detail is welcome....actually detail is what we focus on. Typically we have
been able to achieve a < 1 degree delta at any part of the up/down ramp
between devices under test to ensure that all DUT's are seeing identical
profiles.

Best regards,

John

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Grossmann, Günter
Sent: Monday, June 27, 2016 12:55 AM
To: [log in to unmask]
Subject: [TN] AW: [TN] Thermal Cycle Testing

John

If you want to test the degradation of solder joints with thermal cycles you
should consider the stress occurring in the field and the mechanical
behaviour of the solder. Solder degrades due to time dependent plastic
deformation. This means if you change the temperature you apply mechanical
stress due to the varying CTEs of the materials used. Over time this stress
relaxes due to creep of the solder and this deformation degrades the solder
joint. The colder the solder is the slower it creeps. If a solder joint has
a temperature of 20°C and is heated up to 80°C and back again in working day
cycle the entire difference of the thermal cycle will be induced into the
solder joint. If you test the specimen in this range with a gradient of 2°C
per minute you don't really need a dwell time since the entire deformation
takes place in the ramp which is a good way to accelerate a test (well in
such a case we have a dwell of 15 minutes to be sure that the entire
specimen gets the final temperature). If you cool a specimen from 80°C to
-40°C with a steep ramp, let's say 10°C/min, you'll have to wait approx. 5h
until most of the deformation is induced. Thus, extending the temperature
range and running steeper ramps does not necessarily mean accelerating a
test.
On the other hand, extending the temperature range is a way for
accelerating. To extrapolate the test results to reality you may stick to
Coffin- Manson and, given the large spread in the results obtained, use a
quadratic law. Meaning doubling the temperature swing gives a quarter of the
lifetime.
But be aware that with passive cycling you make an idealised test. In
reality each component has its own temperature swing and actually that's
what needs to be tested. But maybe I'm getting too much into detail now.

Best regards


Günter 

> -----Ursprüngliche Nachricht-----
> Von: TechNet [mailto:[log in to unmask]] Im Auftrag von John Burke
> Gesendet: Donnerstag, 23. Juni 2016 18:13
> An: [log in to unmask]
> Betreff: [TN] Thermal Cycle Testing
> 
> Good morning !!
> 
> 
> 
> A question came up on reliability testing.
> 
> 
> 
> Typically PCBA's are thermally cycled as a part of reliability testing 
> usually for
> 1000 cycles.
> 
> 
> 
> What is the group's experience for non-military thermal cycling ranges?.
> 
> 
> 
> I am often asked to use 0 - 100c sometimes -40 to 85c and sometimes 
> various other thermal cycles picked from the JEDEC standard.
> 
> 
> 
> What is the groups experience for this?. I am trying to establish "normal"
> practice.
> 
> 
> 
> Best regards,
> 
> 
> 
> John
> 
> 
> 
> __________________________________________________________
> ____________
> 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] 
______________________________________________________________________