What is the lifetime of common parts/circuits these days ?

[MathWizard]

I was watching a review of a bad/good movie 'Battlefield Earth', and in it they find some +1,000 year old fighter jets that still work perfect after sitting for that long.

Take for example, some BJT RC oscillator I soldered up, with cheap metal film resistors, cheap 50V  ceramic caps, and some TO-92 BJT's.

So how long might something like that last, if it was only powered up to test it, and then it just sits in climate controlled roomtemp box. Maybe it should be Lead-lined too.

Maybe some of the Lead/Tin solder would be the 1st thing to change, with tin whiskers. But what about common unpowered parts, like above ? What about common 8-pin DIP op-amps ?

IDK how long it takes for plastics or cermaics to change/breakdown. And molecules from the air can make their way into the materials too, and vise-versa. And what would happen to the metal film over the centuries ??

I would like to go 1,000yrs into the future and see what becomes of my little proto-board circuits.


I know there's old salt mines that have been converted to storage vaults for various things like documents, movie films, etc. But does anyone know if there's any amount of tech stored in these places ? I know there's tech/computer museum's, but yeah I wonder how much is in some secure vault, meant to last as long as possible.

[retiredfeline]

I don't think you can generalise, it depends very much on the component and how it was made. For example just read the repair threads where electolytic caps are routinely swapped out. On the other hand I have put half-century old ICs to use. Then also take into account the complexity of the circuit and how single failures can affect its operation.

[Andy Chee]

In a climate controlled environment, I'd imagine most electronics would last indefinitely.

So a related question might be, how long can one maintain a climate controlled environment with zero human intervention?

[coppice]

There isn't much electronics around today that will be functioning in 100 years, let along 1000. Most MCU makers target 100 years operation for flash memory integrity, but corrosion and other chemical degradation means few products using those MCUs are likely to be functioning in 100 years. Rust never sleeps. Things degrade whether used or not. They degrade faster in poor storage conditions. They just go down hill faster when used, especially if they run hot.

[coppercone2]

I think harriers could do it, their made in England after all.

[CatalinaWOW]

I agree it will be a case by case thing.  And depends a lot on what you mean by controlled environment.  A controlled oxygen and water free environment at absolutely constant temperature would mostly eliminate corrosion and fatigue failure from thermal cycling.  But over thousand year time scales diffusion gets to all sorts of things.  On your fight jets any rubber seals around the canopy, or on oxygen masks are likely to be very different than on day of manufacture.  That is a long time for the pot shots taken by cosmic rays to matter.  I doubt anything complex would last that long.

The real question is where would you find anyone that knew how to use them?  Or even could read the fantastic documentation that exists and was stored on some medium that still could be figured out and used.

[coppercone2]

https://www.eevblog.com/forum/mechanical-engineering/does-a-spring-under-constant-tension-age/

[ledtester]

https://www.popularmechanics.com/technology/gadgets/a22506/worlds-longest-running-battery-176-years/

[coppercone2]

https://en.wikipedia.org/wiki/Long-term_nuclear_waste_warning_messages

https://en.wikipedia.org/wiki/Timeline_of_the_far_future

scroll down to human constructs on that page

it gives some insight on materials

that memory crystal should be around for a while.


The first section gives some insight on atomic matter (starting at 2*10^36 AD)

[Solder_Junkie]

A more realistic goal is to build circuits that will last for your lifetime. Few, if any, worry about whether a circuit is still going to work in several centuries time.

I guess like many who have been in home building of electronic projects for many years, I still have projects that were built by myself around 50 years ago that work just as well as when first built. Clearly parts subject to excessive heat can be a problem and print boards benefit from a conformal coating (PCB lacquer). Old fashioned tin/lead solder is not a problem, neither have ordinary electrolytic capacitors failed in any of my projects.

Common problems that I come across include tantalum bead capacitors from the 1970 era that have a tendency to fail short circuit, and tin plated connectors that tarnish (especially EPROMs in test equipment). LCD displays appear to be an issue in a lot of equipment and some 7 segment LED displays have failed.

For those worried about the choice of parts and have deep pockets, you can buy wider temperature rated components that are industrial/military grade, use gold plated print boards and connectors, and directly solder ICs into boards... in many cases it's not worth the hassle and extra cost.

SJ

[Bicurico]

Side question: does it make sense to collect vintage computers like the C64 or the PET? Even if you have a working unit, how long will it continue to work, before failing? Like the CIA failed by itself due to oxidation (hello Intel)...

[janoc]

Side question: does it make sense to collect vintage computers like the C64 or the PET? Even if you have a working unit, how long will it continue to work, before failing? Like the CIA failed by itself due to oxidation (hello Intel)...

Depends?

Collecting is not necessarily about having the item in a working condition.

And if you do want the item in working condition then you shouldn't be caring about it being "pristine" as that may not be possible - if for nothing else because the plastics will be worn/brittle/broken, things will be yellowed, labels worn, etc. So you will need to repair.

For the unobtanium chips there are often modern replacements built using FPGAs, CPLDs and similar solutions, should the chip actually fail. Do your research for whatever device you want to get - unless it is something very obscure then it is likely someone has addressed a problem like that already. E.g. for the C64 CIA: https://1nt3r.net/j-cia/

Even then it is usually possible to obtain a salvaged IC from China somewhere because millions of these things were made and our electronic waste is ending up in landfills there. Buy a several of them and at least one will be good. I have bought the AY-8910 sound generators like that - 10 chip lot cheaply from AliExpress, clearly labeled as not new/salvaged (= a chance they are actually not fakes, those are being sold as new). I have tested them - and all of them actually do work, but two seem to have some damage/problem with one of their channels. Good enough for me for the 10 bucks or so I have paid for the lot. I have picked the best one and used that to repair my ZX Spectrum.

There are plenty of these machines around and working without issues, despite being decades old after one does the basic maintenance (cleaning, replacing old capacitors in power switching supplies, etc.). That one chip commonly fails doesn't mean that all of them will fail or that all of them fail before X years. So nobody can answer you a question like that.

[tszaboo]

If it's a power supply in an LG OLED TV apparently it's 3.5 years as I found out yesterday.  You know warranty time and just a little bit.

[radiolistener]

Take for example, some BJT RC oscillator I soldered up, with cheap metal film resistors, cheap 50V  ceramic caps, and some TO-92 BJT's.

So how long might something like that last, if it was only powered up to test it, and then it just sits in climate controlled roomtemp box. Maybe it should be Lead-lined too.

Most of modern electronics is designed to failure within 1 year. If it works longer, then just feel lucky.

Did you hear about new Intel technology? Their 13/14 Gen CPU live for just several months... 

[Zero999]

I was watching a review of a bad/good movie 'Battlefield Earth', and in it they find some +1,000 year old fighter jets that still work perfect after sitting for that long.

Take for example, some BJT RC oscillator I soldered up, with cheap metal film resistors, cheap 50V  ceramic caps, and some TO-92 BJT's.

So how long might something like that last, if it was only powered up to test it, and then it just sits in climate controlled roomtemp box. Maybe it should be Lead-lined too.

Maybe some of the Lead/Tin solder would be the 1st thing to change, with tin whiskers. But what about common unpowered parts, like above ? What about common 8-pin DIP op-amps ?

IDK how long it takes for plastics or cermaics to change/breakdown. And molecules from the air can make their way into the materials too, and vise-versa. And what would happen to the metal film over the centuries ??

I would like to go 1,000yrs into the future and see what becomes of my little proto-board circuits.


I know there's old salt mines that have been converted to storage vaults for various things like documents, movie films, etc. But does anyone know if there's any amount of tech stored in these places ? I know there's tech/computer museum's, but yeah I wonder how much is in some secure vault, meant to last as long as possible.

It depends on they type of ceramic capacitors? The higher energy density types tend to lose capacitance over time, even when not used. IF you've used  ones with a C0G dielectric then they should last forever.

[Phil1977]

Most of modern electronics is designed to failure within 1 year. If it works longer, then just feel lucky.

You forgot 1 to 2 zeros...

[coppice]

If it's a power supply in an LG OLED TV apparently it's 3.5 years as I found out yesterday.  You know warranty time and just a little bit.

LG frequently throw in a 5 year warranty during promotions, so I assume they expect more than 3.5 years of life in general. On the other hand, we have an LG TV that had a one year warranty, and they wanted a crazy monthly fee to extend the warranty on that.

[MrAl]

Hello,

I've wondered the same thing about those spring clamps.  These are the things you squeeze the handle and the jaws open, then you place the jaws on something you want to clamp and when you release the handles the jaws close and clamp the object or objects with the force of the built in spring.

I've had one clamped to a shelf one time and one day without warning the plastic on one of the jaws broke and the clamp went flying.  If it hit someone in the face it could have been a big problem.  The shelf wood thickness was not especially thick either maybe 3/4 of an inch thick and the clamp was not especially small either maybe the 5 inch type.

I figure the plastic of the jaws get fatigued over time and crack, then break right off.  That's when they are under tension from the spring, and the spring can be quite heavy duty and so very strong.  I noticed the plastic on some of these clamps seems to be of a better quality than others.  I have seen several clamp jaws break from the same cheaper set while none of the ones from a slightly more expensive set never broke.

[BILLPOD]

Good Morning Everyone, I read through this thread carefully, but I didn't see
MTBF mentioned.   Don't datasheets of components have this information any longer

[mikeselectricstuff]

Anything with flash memory/eeprom etc. won't last indefinitely whatever conditions you store it in. Pretty much everything has a MCU in it these days, and maskrom isn't nearly as common as it used to be

[coppice]

Anything with flash memory/eeprom etc. won't last indefinitely whatever conditions you store it in. Pretty much everything has a MCU in it these days, and maskrom isn't nearly as common as it used to be

Masked MCUs are become quite rare. Most new product ranges just don't bother with them. Its all flash, either with a pump so it can self program, without a pump to save die area, or serial flash.

[Phil1977]

But chances are not that bad that an industrial grade MCU that´s specified for 80'000h (ca. 10years) at 105°C will keep it´s memory even for 100years at normal operating/storage conditions.

Flash of this kind can not be compared to the extremely GB/$ optimized SSD or SD-card memory.

[madires]

ATmega's flash memory:
- 20 years at 85 °C
- 100 years at 25 °C

[SteveThackery]

Speaking as an ex-Reliability Engineer, there are remarkably few ageing mechanisms in modern electronics. In the past we had tin whiskering in some transistors, and wet electrolytics drying out. Oh, and ESD zaps to CMOS chips often induced a rapid ageing effect causing them to fail weeks or months later. Sometimes a couple of years.  Plus, obviously, vacuum tubes, CRTs, VFDs, tungsten filament bulbs, Nixie tubes - they all had well-understood ageing mechanisms.

Nowadays it will almost all be down to the environment (including the operating environment). Thermal cycling can disconnect lead-outs. Mechanical vibration can cause fatigue failures in leaded components. Any humidity over 0% has the potential to cause corrosion of the metallic parts. We should also recognise that most memory technologies have a (theoretically) limited retention time (although back when it was specified as 10 years, many such chips are still doing their job decades later). Exposure to UV, even tiny amounts, will eventually soften the storage elements in a windowed chip.  High energy particles from outer space can "soften" the data in EPROMs and flash memory.

Ensuring a device is never operated outside its safe operating area is also very important, but the damaging effects don't just instantly appear on the wrong side of the SOAR line and disappear just inside it, so it is wise to stay well within the safe operating area.

In general, assuming those factors (and others I have forgotten) are adequately addressed, there are little or no detectable ageing effects. This is actually amazing, and almost unique in the world of technology.

By the way, this does not mean they will last forever. Most devices have an intrinsic failure rate, with such failures manifesting as statistically random. Ageing manifests as an increase in the failure rate as time passes. Monitoring the failure rate over a sufficiently large population allows these measures to be meaningful and helpful in deciding when to retire the product being monitored. An aircraft engine, for example.

EDIT: I forgot to mention packaging. Modern plastic packages are not considered as hermetically sealed (at least not by designers of equipment that must be as reliable and durable as possible). This includes spacecraft, military equipment, undersea cable repeaters. Oh, and automotive electronics as well, due to there being some potential safety aspects. Ceramic packaging can meet the requirements for hermetically sealed, which is why such high reliability and durability electronics can be expensive.

Finally, we must not forget the soldered joint. A perfectly-made joint shows virtually no ageing, but even small imperfections can get worse as oxygen and water vapour go to work (plus mechanical disturbance, of course).

[SteveThackery]

Good Morning Everyone, I read through this thread carefully, but I didn't see
MTBF mentioned.   Don't datasheets of components have this information any longer

MTBF does not address ageing effects. It is a measure of the intrinsic steady-state failure rate.