On Tuesday, March 18, 2014 3:22:47 AM UTC-7, iavine wrote: > > A lot of the time the fault was with the end customer using the device > out of spec. Prototype 1, 2 and 3 work. Production run of 1000 gave a high > failure rate. > > One of the common sayings, of one of my old bosses was, "there is nothing more treacherous than a working prototype".
I remember one that he made, and tested, then had a limited run of 25 made. They failed. The prototype had a bad diode, that was shorted. So, in this case the fix was simple. I'm not too keen on that Microchip App Note, either. But they published it. Personally, I use a buffer stage (NPN xstr) with a simple RC filter. If it goes into a uC input, then I do further filtering in software. Usually, a variation of switch debounce routine. I always like using buffer stages. If nothing more than just something "cheap" to blowout first, before something more expensive, with more leads to desolder, pops. -- You received this message because you are subscribed to the Google Groups "neonixie-l" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/a2b0b93e-d03a-430b-8057-28706584bef8%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
