FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 

Analogue ICs - why do they die?
MUFF WIGGLER Forum Index -> General Gear  
Author Analogue ICs - why do they die?
I guess it's kind of a dumb question, but why do we commonly associate a lifespan to ICs, partcularly the rare crucially important unobtanium ones? They don't have moving parts to wear out (obviously), but I suppose being heated to operating temperature (or above) and then cooling down a few thousand times over must take its toll - or does it??

Do ICs die purely from physical abuse, or due to related parts failures leading to over-voltages or static shocks, or do they sometimes just power down one day and decide to head for that big silicon wafer in the sky all of their own accord?

And are their any mods, techniques or habits of usage that we can employ to prolong their lives indefinitely?

I ask as someone who would rather have a great old piece of gear knowing it may end in a puff of smoke at any time (there are others here who feel the same - you know who you are), than someone who would prefer to live with a plasticky new toy just because it will outlive themselves...
i have got bad experiences with an Prophet 5 Rev 2,its the only one but did leave a very bad impression,since then i sweared myself never ever buying a vintage polyphonic synth anymore.
Fair enough - it's precisely these situations that I'm talking about. Did you ever find out what the malfunction was?

My own experience was strangely not with a vintage piece, but with a very bland digital box - an E-Mu Orbit 9090 MIDI module, that I bought when they came out around '96. It cost a lot of money for me at the time, and it died a matter of weeks after the warranty period expired. Of course, consisting internally of just one SMT board and a power supply, I was quoted pretty much the retail replacement cost just to repair it meh d'oh!
Because you touch yourself at night.

...or because I did something stupid and killed them - or at least that was the case when I was still spending a lot of time with DIY stuff.

Amazingly enough, though, even with my horrible luck, I've never had anything that couldn't be fixed with easily obtainable parts go wrong with any of my vintage gear seriously, i just don't get it
So far we have as possible causes of death:

1. Stabbing with a luke-warm soldering iron and, er,
2. Excess humidity eek!
Just found this while searching out of curiosity...
The microscopic traces can change shape over time, I was told...not quite in the manner in which glass is a liquid, but similarly..
chips are basically teeny tiny components in a particular sort of format, aren't they? like, inside the chip, it's resistors and transistors and whatnot..

..i'd assume that whatever kills them is whatever kills other electronics.. maybe more sensitive because it's microscopic traces and pieces, so the same size 'zap' can take out a bunch of pieces?

note the question marks! i don't know, i'm just guessing.
science wrote:
Just found this while searching out of curiosity...

So this electromigration appears to be the ticking time-bomb at the heart of all silicon chips, even under optimum conditions. Basically the atomic structure of the silicon degrades from being continually exposed to electromagnetic energy, until one day... Poof. Simply using your equipment is bad for it!

According to Wikipedia this problem increases with the degree of ‘miniaturisation’ on the chip (which we all know has increased over the years), but this is counteracted by the fact that chip designers have only really started addressing the problem with workarounds since the late 80s. So the chances are the CEMs and SSMs we know and love are living on borrowed time...

I guess one day soon there will be a market for modern-day substitutes for these impossible-to-find ICs. Not that long ago I saw a Prophet VS go on eBay that had a custom-made circuit board with a modern (PIC?) processor, taking the place of the CEM 5530 S+H and thus ensuring years of trouble-free operation (until any of its other incredibly rare and critical parts fails...)
Ranxerox wrote:
Fair enough - it's precisely these situations that I'm talking about. Did you ever find out what the malfunction was?

The reason i don't know,but the sequence was

first bad memory rams,then electronic switches ic's,then some ssm's ,finaly parts that takes cares for the prontpanel scanning.

a mate fixed the frontpannel scanning part, then i sold it as-is

i did now the rev2 is a very fragile item,but not at this level.
at the same time i did restore my memorymoog and when i discovered the 40+ 3360 vca's into i decided to sell all my polyphonics
There's a nice document here that speaks a bit to the metal migration phenomenon, as it relates to synth repairs:

I've found that advice (and 2012's above) to be sound, at least in my own experiences. I've shotgunned fixes on my OB-8 and X-911 by taking inventory and replacing old CMOS. I had a Moog synth with dead osc due to a bad cap. Space echo died recently with op-amp failure. My tube preamp and Pro-one are both onb the fritz - failures yet to be determined. And meanwhile, I'm trying to proactively replace power supply caps in my elderly gear to try and head off any problems.

Never a dull moment! Why can't electronics just get better with age like guitars?! waah
chromium wrote:

Never a dull moment! Why can't electronics just get better with age like guitars?! waah

The problem was that back in those days i had a lot of vintage synths and drum machines and i was very productive,in fact there was not a week that pased away without anything that went wrong on one of the machines.
Even the first 'new' analogs went wrong like the SE-1 midi that locking always.
MUFF WIGGLER Forum Index -> General Gear  
Page 1 of 1
Powered by phpBB © phpBB Group