Howard,
What you describe is not at all unusual. We have many customers who actually cycle 10,000 times at –65DegC to +130DegC. However, you do not mention humidity, but I guess that you will be up pretty high and this will do as much, if not far more, damage than the temperature extremes alone.
A Conformal Coating will certainly help you qualify, but thicker is not better and the CTE mismatch with a silicone will kill the assembly if you are employing any COB packages, BGA’s etc.. Get a copy of the IPC-CC-830 spec and the IPC-TM-650 documents.
You will need to look carefully at the issue of flux residues that you mention. These may lead to electro-chemical reactions that may also “kill” the assembly, apart from the mechanical issues you have encountered.
I strongly recommend the use of a conformal coating, and you would be wise to use one that is qualified to the old MIL-I-46058 specification which will give you a high assurance to its performance. An acrylic would probably give you the best results coupled with easy reworkability. If you have a high salt environment, then choose a polyurethane.
"Howard P. Stevens" wrote:
A customer of Metallic Resources (manufacturers of bar solder) assembles circuit boards. One of the many tests the boards go through is a thermal cycling test to see if thermal shock can cause failure in the circuit board.
The customer's test specification is Spec Number TSC7514g, section 5.4, reference specification TSC 7010g. (We do not know what this means.) The customer cycles the circuit board from -30 degrees celcius to +80 degrees celcius. The board temperature is dropped to -30 and held for thirty minutes. The board temperature is then raised to +80 and held for thirty minutes, until the temperature is lowered again and so forth for a total of 3,500 hours. The customer is now noticing solder joint failures on this double sided board on the top, which is both through-hole and SMT (failures of solder paste joints on the top), and on the bottom, which is just through-hole (failures of bar solder joints). Solder joints, both top and bottom, show much cracking and some flux residue.
The first question is: is this customer using their own test or is there an IPC test for thermal cycling of a completely assembled circuit board?
The second question is: is the customer's temperature range too extreme, which is causing severe thermal shock, which is then causing joint failures due to different co-efficients of expansion and contraction (board, solder, lead)?
The third question is: is the use of 3,500 hours of thermal cycling realistic and/or too long?
Any comments would be appreciated.
Sincerely,
METALLIC RESOURCES
Howard P. Stevens
---------------------------------------------------
--
Regards, Graham Naisbitt
[log in to unmask]
Mobile: +44 (0)79 6858 2121
Concoat Limited - Engineering Electronics Reliability
NEW – TechShot 05 now available please email us for your copy – NEW
NEW – SoldaPro Wizard please email us for your copy – NEW
Web: www.concoat.co.uk AND www.concoatsystems.com
Sales:
Millfield House, Fleet Road
Fleet, Hampshire GU51 3QF UK
Phone: +44 (0)1252 813706
Fax: +44 (0)1252 813709
Manufacturing and Head Office:
Alasan House, Albany Park
Camberley, Surrey GU16 7PH UK
Phone: +44 (0)1276 691100
Fax: +44 (0)1276 691227