On Fri, Oct 7, 2016 at 6:45 PM, Mike Stein <firstname.lastname@example.org> wrote: > I guess technically 'reversible' means being able to put things back *exactly* the way they were, especially without any soldering, so cutting and restoring traces, adding and removing sockets etc. would not qualify as 'reversible'. > > I've often wondered why it's standard practice to install a socket when replacing a single bad chip; presumably the odds of the same chip failing again are slim, and it certainly is a glaring 'modification'. as I do quite a few repairs of old gears, I think I can give a number of reasons why I almost always install a socket: 1) most of the times, the failing chip is not alone, the replacement can be killed instantly by another fault: random example is a bus conflict where a failed input or output sinks/source current against another output, this in many cases makes the replacement fail soon again. 2) in the troubleshooting process, might often be good to isolate/replace a signal around a failed chip (because there're other related faults not yet identified), a socket makes this process easier. 3) it's cheap enough that even if you are sure there's no reason to fall in the 1) and 2) cases, you don't just take the chance to be wrong. (all IMHO) Frank Message was sent through the cbm-hackers mailing listReceived on 2016-10-07 18:00:02
Archive generated by hypermail 2.2.0.