Just to throw a wrinkle in the temperature rise rates.
If you have high thermal mass parts, cycling past a certain rate will
result in massive temperature imbalances on the panels and within the
chamber. These imbalances can result in water condensing on high
thermal mass parts such as transformers. If power is applied, the dirt
washed off the transformers and relocated into standing water will
cause electrolytic growth. This does not occur when ramping is slowed
so that no condensation occurs, power is applied after all moisture
has evaporated, the board (or test fixture) is designed to minimize
this failure mode, or moisture is maintained below the dew point of
the lowest temperature of the cycle until all parts have achieved
ambient temperatures.
Also, in some older chambers, if you ramp fast enough, with humidity,
it can rain within the chamber.
Aric Parr
______________________________ Reply Separator _________________________________
Subject: Test temperature range
Author: [log in to unmask] at internet
Date: 1/29/97 1:43 PM
Hi Werner, Hi Dave
I think, if one performs accelerated tests on solder joints it is more
important not to activate a deformation mechanism which does not occur in
reality than stick to any temperature range. Tin lead solder deforms with
two different deformation mechanisms: Grainboundary sliding (GBs)and
dislocation climb (DC). Which deformation mechanism is activated depends on
the deformation rate ( temperature exchange rate ) and the temperature.
Grain boundary sliding is diffusion controlled and occurs at lower
deformation rates and at higher temperatures than dislocation climb. At
-40 C the strain rate activating primarily GBS is approx. 10E-6, at 125 C
approx. 10E-2. For a ceramic SMT capacitor 1210 on FR4 with a soldergap of
30um this results in temperature exchange rates of 0.1 C/min, and 500 C /min
respectively.
In performing accelerated testing with a Coffin- Mansion plot to perform a
lifetime prediction one must run temperature cycle tests with 2 or 3
different temperature ranges. In these test you shouldn't change the ratio
of GBS and DC to keep the Coffin Mansion exponent constant. Here lays the
rabbit in the pepper ( just a joke for those who know German ). It is not
easy to determine this ratio. We are working on it to prepare some diagrams
that bring all that theoretical stuff in a form easy to use. However,
measurements in automotive under hood electronics showed temperature
exchange rates of approx. 2 C/min. This induces mainly GBS at temperatures
around - 0 C. In office equippment we measured approx. 0.5 C/min. I think
for accelerated testing of the majority of electronic applications mainly
GBS can be assumed. In this case it might be wise to extend the temperature
range to higher temperatures (avoiding the glass transition temperature of
the board). If one extends the temperature range to lower temperatures use
1 C / min. from -20 C to -10 C, 2 C / min from -10 C to 0 C, 4 C / min.
from 0 C to 20 C and above 20 C as fast as your equipment runs.
Another point is the dwell time. At high temperatures ( 100 C ) 5 minutes
dwell time are enough to relieve all the stress induced. The lower the
temperatures the longer one must wait. At -20 C at least 30 min dwell time
are necessary, otherwise a lot of strain is stored elastically in the PCB
or in the leads of the IC's. It is questionable, whether one safes time if
the temperature range is extended to lower temperatures.
Best regards
Guenter Grossmann
***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to: *
* [log in to unmask] with <subject: unsubscribe> and no text. *
***************************************************************************
* If you are having a problem with the IPC TechNet forum please contact *
* Dmitriy Sklyar at 847-509-9700 ext. 311 or email at [log in to unmask] *
***************************************************************************
***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to: *
* [log in to unmask] with <subject: unsubscribe> and no text. *
***************************************************************************
* If you are having a problem with the IPC TechNet forum please contact *
* Dmitriy Sklyar at 847-509-9700 ext. 311 or email at [log in to unmask] *
***************************************************************************
|