Hi all, to elaborate a little further, 30 out of 30 1st-article pcbs=20 with the invensense MEMS were destroyed by ultrasonic cleaning of the=20 pcb. 100% failure rate of a relatively small sample I suppose=20 statistically but way larger than one wants as an engineer trying to=20 finish a project. :) The Invensense datasheet warns about this as I=20 recall. I have heard the ultrasonic welding of cases causing grief with MEMs and=20 xtals, as people have discussed. For this project, there are 4 very=20 small self-tapping screws to close the case, no welding. That said, pretty clearly other wearable bluetooth devices (requiring at=20 minimum an xtal in the 16-32Mhz range, and often a 32.768Khz as well if=20 it is bluetooth low-energy) are welded shut. What devices like the=20 Fitbit and the like (MEMs and xtals) do would be interesting to know.=20 Presumably they have reliably solved the ultrasonic damage issue. J Robert Dvoracek wrote: > I believe this would fall under the category of standing waves, which > are affected by all kinds of things, up to and including the way the > operator holds its mouth during the "sound off". A single well > placed self-tapping screw is nearly as fast and allows later > disassembly. Unless of course you really need a hermetic seal for > waterproofing, etc. In that case, a weighted object near the > sensitive component(s) to dampen the vibrations may help. As others > have mentioned, post assembly testing would be key here. > > Cheers, Robert > > Sent from my iPad > > On Dec 18, 2014, at 5:31 AM, RussellMc wrote: >>>>> ... MEMES devices are usually destroyed by ultrasonic >>>>> cleaning >> >>>> Does the same apply to assemblies ultrasonically welded into a >>>> box I wonder? If so, this post could have saved us a bit of >>>> future strife - >> --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .