David Griffith wrote: can you supply a picture? I've never seen this occur after time has passed. Other problems can cause this during wave solder... but that's a different issue. -brad

William Donzelli wrote: The differences are enough for many thermal cycles over a long period of time. It seems that the greater the temperature delta and the more frequently the temp is cycled the sooner it occurs. I've seen boards cycled between 20 and 300 degrees C five times a day that have wrinkles in about a year and trace lift by five years. I've seen similar wrinkling in boards cycled between 7 and 98 degrees C hourly that have been in service for as long as 20 years. Interestingly, the boards that are at a (more or less) constant 200 to 300 degrees C show no wrinkling or trace lift. The expansion coefficients may be have a relatively small delta but it is enough given enough time, enough temp cycling and enough of a temp range. [snip] Well, when someone involved in getting mil-spec and space grade acceptance for PCB's mentions it, one tends to believe him. == jd Real programmers don't draw flowcharts. Flowcharts are, after all, the illiterate's form of documentation. Cavemen drew flowcharts; look how much good it did them.

William Donzelli wrote: How does the copper stick to the substrate? Are you taking that into account? == jd [Prime Minister Joseph] Chamberlain loves the working man -- he loves to see him work. -- Winston Churchill

William Donzelli wrote: The adhesive loses it's youthful suppleness over time. It eventually becomes rock hard and brittle. Not a very good moment arm, then. I've got 2oz PCB's built in 1962 with traces just plain falling off the board because the adhesive hardened and quit holding. These were not stored in extreme temperatures, rather, the temp range has been very narrow. I had a PCB from an early 1980's microwave that had less than 1oz copper and the traces started visibly wrinkling some years ago. I attempted to contrast an extreme case to the more common smaller temp deltas seen in "normal use" to show how the factors of temp, temp cycling and time can affect wrinkling. Sorry that was missed. The point is that wrinkling occurs as a function (in part) of temp ranges and time. And, by the way, the weight of the copper also affects it. Less than 1oz copper goes faster than over 1oz copper. Et cetera, et cetera, ad nauseum, et al. In an extreme case wrinkling and lifting/peeling/whatever happen _much sooner_. Obviously it takes a lot more time under "normal conditions". And wrinkling is more likely to occur with thin copper than with thick copper, which is more likely to lift/peel/ whatever, obviously. Likely, that is, not an absolute certainty. A lot of consumer electronics were made with less than 1oz copper PCB's--and cheap ones to boot. It's been long enough that it is possible that the wrinkling was caused by these factors. I know wrinkling does happen. I've seen it plenty of times. Even the smallest degrees of contraction and expansion can eventually work far enough along to create wrinkles and lifting/peeling. If the OP is talking about the "wrinkles" caused by flow during production soldering then that's a whole different can of worms but it did not seem that is what was being asked about. == jd "I may appear to be just sitting here like a bucket of tapioca, but don't let appearances fool you. I'm approaching old age ... at the speed of light." -- Prof. Cosmo Fishhawk

Peter C. Wallace wrote: [snip] feh. You're right. That's the ambient temp just for that cabinet. With the clamp sinking the heat etc., the temp was actually closer to 200, maybe 250 on a good day. Only got above 300 when the exhaust fans stopped working, then the boards would brown real nice and the traces would fall off. == jd Major Premise: Sixty men can do a piece of work sixty times as quickly as one man. Minor Premise: One man can dig a posthole in sixty seconds. Conclusion: Sixty men can dig a posthole in one second. -- Ambrose Bierce, "The Devil's Dictionary"