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

[radiolistener]

Why would anyone go thru that many troubles instead of just using the cheapest components possible.

There are at least two reasons for that.

First, if you manufacture first product with cheap components, your product will become an outcast due to the fame of it problems. So, no one want to buy it.

Second, you can have too many expenses on RMA, so you will not be able to cover all expenses from earned income.

The more smart decision will be to use a better components at first release, it will minimize risk of expenses on RMA and make good reputation for the product. Then you can measure it's lifespan more precisely and apply patch with more cheap components for expenses optimizations.

[SteveThackery]

They were trying to mitigate the effects of failing batteries, a component we have already discussed.

I knew it, you have a reason for everything... 

Of course, you can come up with some reason and just close your eyes on obvious reason. But it just shows that you don't want to know the root of cause why it happens.

I have no idea what you are saying - sorry.

And like we've already mentioned, it is trivial to make a product with a microcontroller "fail" after one year, or any other interval.  To be honest, it's a far easier way than the ridiculous lengths you claim are routinely done.  Can you point us to devices where this is done?  And are you still asserting that "most" electronic devices are designed to fail after a year, maybe less?  You did say that.

[Shock]

There are methods and approaches scientifically (expensive) where you can extend a device past 100 years. 20 years is easy with today's technology, correct component selection and construction techniques.

But it's becoming more and more imbalanced with cheaper products on the market manufactured with low quality alloys, rubbers and plastics. So if you allow these to get into your the supply chain it can lead to all these unexpected failures in components due to thermal expansion/contraction, corrosion and off-gasing etc. This is even before poor design of the actual device or shortcuts in component selection/tolerances in order to cut costs are made.

1000 years would be quite challenging but if you had the time, money and did the research I think possible. It would be nothing like a miniaturized consumer type device or even military like in appearance. Expect it "more like" a quantum computing device in a multi-chambered and vacuumed enclosed housing. Most parts would need to be directly coupled for reliability so no solder or pcb, no moving parts or plastics/rubbers.

[SteveThackery]

Sorry, chaps, this is fun but it's almost 2am in my part of the world and I must get some sleep.  All the best - talk later.

[radiolistener]

To be honest, it's a far easier way than the ridiculous lengths you claim are routinely done.  Can you point us to devices where this is done? 

Almost any product which was manufactured according to ISO9001 uses this approach. It includes almost all known electronics manufacturers.

And are you still asserting that "most" electronic devices are designed to fail after a year, maybe less?  You did say that.

I already provided you with example of device with no microcontroller - electric stove which consists of electric heater, power regulator and mains cable... It is failed just after 5 months due to failed power regulator. It is even hard to imagine how to make so simple thing with so small lifespan. But they do that... 

[Phil1977]

I listed what could and would go wrong with this kind of "planned obsolescence" many posts ago.

As long as you don't use a sophisticated timer (like a microcontroller with RTC) it will be a lottery game that usually hurts the selling company the most. If you use a sophisticated timer and someone finds out about it, it hurts the manufacturer too.

Especially cheap induction hobs have always been a lottery for the consumer. Sometimes they last forever, sometimes they give magic smoke with the first power up. It seems that resonating power circuits and inexperienced quality engineers are just no good combination.

[SteveThackery]

Almost any product which was manufactured according to ISO9001 uses this approach. It includes almost all known electronics manufacturers.

I am intimately familiar with ISO9001, having once been a Divisional Quality Manager and having brought an organisation up to ISO9001 compliance no less than four times. I have authored entire Quality Management Systems (QMSs).

There is literally nothing in ISO9001 that describes, or requires, the ridiculous antics that you claim are commonplace.

I already provided you with example of device with no microcontroller - electric stove which consists of electric heater, power regulator and mains cable... It is failed just after 5 months due to failed power regulator. It is even hard to imagine how to make so simple thing with so small lifespan. But they do that... 

Are you serious? That is ONE example!  And I've provided one example of a motherboard that is still working fine after 12 years. Both are of no value because we need quantities: failure rates and lifespans are statistical phenomena.

I suspect in the case of your stove, one or more components were operating significantly outside their specifications, hence an early failure. What it does NOT demonstrate is that some engineer sat down and said "OK, how can I design this such that it fails after 5 months?".

You made two claims:

1/ Electronic devices can be made to fail after one year, with an accuracy of weeks or even days.

2/ Most devices are designed to fail after a year or less.

In support of item 1/ you described a frankly extraordinary process which would need to last several years, and which relies on finding electronic components with an almost inconceivably short lifespan (or short MTBF). I would remind you that the reliability of electronic components is specified in "FITs" - failures per billion hours.

In support of item 2/ you quote one example: a stove which failed after 5 months. But most people's experience of electronic products is that they last much, much longer. Almost indefinitely unless they contain a battery.

In my house there is only one electronic product less than a year old - a new laptop. There are many products in my possession that are between 1 and 13 years old, and a handful that are a few decades old: the tablet I'm typing on, two remote controls, a mobile phone, a TV, two TV set-top-boxes, several electronic watches (some going back to the 1970s), two novelty electronic toys, eight or nine quartz clocks, several smart-home light bulbs, three laptop PCs, a desktop PC, numerous pieces of test equipment in my workshop, including an oscilloscope, a signal generator, two bench PSUs (which were made in the 1980s), two electronically controlled soldering stations, an electronically controlled lathe, five multimeters (although one is not really electronic). Also around are another TV, a surround sound system that is 16 years old, I think, a DVD player, three Bluetooth headphones, numerous wall wart power supplies.....

I'm wracking my brains to remember the failures I've had.  The 13-year-old Sony hybrid works fine except that the bottom fifth of the screen no longer responds to touch, so there is that. I also recall a wall wart failing, but I've no idea how old it was - several years, I think.  Like most people I am literally surrounded by electronic products, in my case all but one being over a year old, and a good number are decades old. They are all still working apart from those two I mentioned. Oh, I've remembered another: an LED light bulb that failed after about a month.

This is the reason why I think your second assertion - that "most" electronic devices are designed to last "a year, often less" - is simply nonsense. It is blatantly, obviously not true, and I think that seriously undermines your first claim as well.

Anyway, this thread is surely exhausted by now.  Please feel free to have the last word - I'm sure you will want it. 

[radiolistener]

I am intimately familiar with ISO9001, having once been a Divisional Quality Manager and having brought an organisation up to ISO9001 compliance no less than four times. I have authored entire Quality Management Systems (QMSs).

I have no doubt about it, I immediately guessed why you are trying to defend planned obsolescence 

There is literally nothing in ISO9001 that describes, or requires, the ridiculous antics that you claim are commonplace.

The standard itself really don't have something like that. This is just a tool and it depends on you how you use it.
When you use it, it's not so hard to implement planned obsolescence. This is what I talked about.

Are you serious? That is ONE example!  And I've provided one example of a motherboard that is still working fine after 12 years. Both are of no value because we need quantities: failure rates and lifespans are statistical phenomena.

Are you kidding? Almost all modern products works the same. And all know that. With some minor exceptions for expensive brands, but they also tend to use the same practice. 
Over the past year, I have encountered at least a dozen such cases. Maybe even a hundred.

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

I suspect in the case of your stove, one or more components were operating significantly outside their specifications, hence an early failure.

I have no doubt that you will certainly blame the users for incorect usage, bad conditions, etc. But the fact remains - the service life is about one year.

You can name it "bad users". But I name it "planned obsolescence".

[Phil1977]

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

Bosch would need to have prophetic qualities to know how often you use the mixer for what. And beside that can nylon gears be much more reliable (less abrasion) than cheap metal gears.

[radiolistener]

I don't know what material they using, but it fails after a year of regular moderate using. It is reported by different buyers and the same reports from service. But it's fair to say that other brands suffers from the same issues. So if you want to find something better, you can't. The only alternative is to buy very expensive industrial grade equipment

[Phil1977]

Yes, that´s really annoying thing that even medium-priced appliances often have very low-cost internals.

But anyhow, I think it´s usual "optimization" of the companies. To really target a failure 1 day after end of warranty you can't rely on material weakness alone.

And, trust me, the load on a mixer is extremely different depending on what you prepare with it. Working on a heavy yeast dough is absolutely not comparable with preparing whipped cream. But the mixer should survive both for a long enough time.

[SteveThackery]

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

 

For goodness sake!! What is wrong with you? This thread is about ELECTRONICS! Unlike electronic components, mechanical components have many ageing mechanisms which are very well understood and no doubt exploited by manufacturers. Up the thread I explained (but you weren't listening, obviously) how car manufacturers specify the service life of components and the suppliers engineer their devices to meet, but not exceed (as far as possible) the specified life. I gave the example radiator fans, and how it is possible to predict with impressive accuracy the wear rate on the brushes, and thus the service life of the fan.  It was intended to be an illustration of how very different mechanical and electronic devices are.

This thread is about ELECTRONICS!

(Damn! I got sucked in again.  )

[coppice]

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

Bosch have been making fragile mixers for a long time. There is a weird dynamic there. If you just make cakes a Bosch is OK. If you knead dough it isn't, and it takes very little time to break them on these heavy loads. This would be OK if they just made low power mixers that were clearly not designed to knead dough. However, they have a 1600W model, where everything but the motor is built lightweight.

A more interesting case is Kitchenaid. They made reliable mixers. They moved production to China. The mixers sucked. Oh no, we have to move production back to the good old US. Actually, the problem had nothing to do with location. They had stripped the design bare when they moved production. They realised they'd overdone it, and rather than just admit it, take a hit on warranty claims, and make the US model in China, they had a whole charade around the issue. I assume they ended up with higher costs than they needed to. As I understand it, a new Kitchenaid today is pretty solid, although they make some flimsy attachments they sell alongside more robust versions.

[tooki]

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

Bosch have been making fragile mixers for a long time. There is a weird dynamic there. If you just make cakes a Bosch is OK. If you knead dough it isn't, and it takes very little time to break them on these heavy loads. This would be OK if they just made low power mixers that were clearly not designed to knead dough. However, they have a 1600W model, where everything but the motor is built lightweight.

A more interesting case is Kitchenaid. They made reliable mixers. They moved production to China. The mixers sucked. Oh no, we have to move production back to the good old US. Actually, the problem had nothing to do with location. They had stripped the design bare when they moved production. They realised they'd overdone it, and rather than just admit it, take a hit on warranty claims, and make the US model in China, they had a whole charade around the issue. I assume they ended up with higher costs than they needed to. As I understand it, a new Kitchenaid today is pretty solid, although they make some flimsy attachments they sell alongside more robust versions.

I do not believe that story is correct. As I understand it, KitchenAid stand mixer manufacturing has always remained in USA. I cannot find any claims of one made in China, never mind evidence thereof, nor about any such “charade” and drama around it. Literally not even mention of it.

Other KitchenAid products are made in various countries, but the stand mixers are all made in USA. They do, however, use imported parts, and some of those are from China.

[CatalinaWOW]

Lots of loose ends and thoughts in this thread.  While the question started with pulling a jet fighter out of storage after millennia, the specific question was life of "simple electronic components".

What is an electronic component?  Are pots and relays electronic components?  Switches?  Keyboards? 

Most people would consider a laptop computer an electronic device.  Same with a desktop computer.  To me that implies that the devices use to build them are "electronic components".    Now consider designing each for a 1000 year life.

For the desktop it is fairly easy to imagine things you could do to make it more likely to last 1000 years.  Eliminate the real time clock.  It isn't a core functionality.  Eliminate fans.  There are passive cooling mechanisms that are bulkier than forced air systems, but along with some compromise in speed it seems likely that they could be made to work.  Assume that thick film resistors and ceramic capacitors have no aging mechanisms that cannot be defused by proper storage techniques.  You can eliminate spinning rust with solid state memory, though I would assert that really long life, even non-operating for a device with actual countable numbers of atoms per memory cell is very much terra incognita.  But a keyboard and display are still necessary.  Maybe the keyboard can be a non-tactile capacitive sensor or a virtual keyboard based on LEDs and optical sensors.  The display is tougher.  I don't think anyone would say that LCDs or OLEDs or any other display technology has unlimited life.  There are many other details that could trip up a multi-thousand year lifetime.  For example the thermal paste that is normally used to improve heat paths is an issue.  Maybe an Indium puck?  Would have great thermal properties initially, but would it stay put for millennia?

The laptop has even more issues.  The mere concept of laptop makes the size compromises for thermal management impractical so things like a long life heat pipe or other more difficult solutions would be needed.  Even something simple like a flex harness to allow opening and closing said laptop (surely a harness is an electronic component) could outgas, depolymerize or whatever other process occurs which makes these things get stiffer with age, leading to breakage and opens circuits upon opening after such a long time.

[coppice]

If you want more example - modern mixer Bosch, it uses plastic gears which wear out in about a year (exactly the same reports from many users) and the mixer becomes unusable. Old mixers working for more than 10 years.

Bosch have been making fragile mixers for a long time. There is a weird dynamic there. If you just make cakes a Bosch is OK. If you knead dough it isn't, and it takes very little time to break them on these heavy loads. This would be OK if they just made low power mixers that were clearly not designed to knead dough. However, they have a 1600W model, where everything but the motor is built lightweight.

A more interesting case is Kitchenaid. They made reliable mixers. They moved production to China. The mixers sucked. Oh no, we have to move production back to the good old US. Actually, the problem had nothing to do with location. They had stripped the design bare when they moved production. They realised they'd overdone it, and rather than just admit it, take a hit on warranty claims, and make the US model in China, they had a whole charade around the issue. I assume they ended up with higher costs than they needed to. As I understand it, a new Kitchenaid today is pretty solid, although they make some flimsy attachments they sell alongside more robust versions.

I do not believe that story is correct. As I understand it, KitchenAid stand mixer manufacturing has always remained in USA. I cannot find any claims of one made in China, never mind evidence thereof, nor about any such “charade” and drama around it. Literally not even mention of it.

Other KitchenAid products are made in various countries, but the stand mixers are all made in USA. They do, however, use imported parts, and some of those are from China.

Interesting. I can find several references to "all Kicthenaid stand mixers have been made in the US since 1941". In the early 2010s we were in the market for a mixer, and there were lot of people complaining about the issue I described. The two are not necessarily incompatible. "Made" is usually little more than final assembly and packing. Perhaps they moved the production of the main mechanics to China around that time, and cheapened it.

[coppice]

The laptop has even more issues.

When looking at electronics lifetimes, I think you need to separate anything portable. The life of so much portable equipment is limited more by mechanical robustness issues than electronic ones. Things have a hard life in pockets and rucksacks.

[Phil1977]

The laptop has even more issues.

When looking at electronics lifetimes, I think you need to separate anything portable. The life of so much portable equipment is limited more by mechanical robustness issues than electronic ones. Things have a hard life in pockets and rucksacks.

And the more constraints you have regarding size and weight the more you may rely on materials like flexpcbs, plastic and glue. For me it seems well plausible that some military radio that just consists of ceramic ICs and wet tantalum capacitors with 7-segment-led displays in a rigid metallic housing may be equipped with a fresh battery in 200 years and just works. But an iphone with it´s oled display and plastic housed semi-sealed MEMS-parts??

New electronic devices are not generally unreliable, but if society wants to keep nowadays standards then we really need to make production sustainable. You won't be able to switch on a todays iphone in 50years, but (hopefully) we are able to recycle today's phones into new ones in future.

[radiolistener]

When looking at electronics lifetimes, I think you need to separate anything portable. The life of so much portable equipment is limited more by mechanical robustness issues than electronic ones. Things have a hard life in pockets and rucksacks.

I think on the contrary, mobile devices have better reliability, because they are designed to be used in very aggressive environment and conditions. But their thermal design is really worse due to size limitations and if it working at high temperature it leads to shortened lifetime.

For me it seems well plausible that some military radio that just consists of ceramic ICs and wet tantalum capacitors with 7-segment-led displays in a rigid metallic housing may be equipped with a fresh battery in 200 years and just works.

I'm very sceptic about it. Modern military radio is full of MCU with flash memory. I bet it will lose data earlier than 10-20 years. And even if it uses OTP memory, I'm very skeptic that it can run after 50 years with no issue. Most of all some components will fail and some going out of specification, but it is possible that some instances will continue to work.

[Shock]

I don't disagree that low cost devices fail way sooner but it's obvious that many survive a year. Due to compromises (not only in labor cost) low cost manufacturing has a huge ecological and consumer cost. Just as a brand can't stay in business selling faulty products (unless it specifically catches unsuspecting customers) an expensive brand can't exist if there are better quality alternatives at a cheaper price (unless it has a mindless fan base). A mix of these actual four conditions occur at the same time, so arguing over specific cases is pointless.

[tooki]

When looking at electronics lifetimes, I think you need to separate anything portable. The life of so much portable equipment is limited more by mechanical robustness issues than electronic ones. Things have a hard life in pockets and rucksacks.

I think on the contrary, mobile devices have better reliability, because they are designed to be used in very aggressive environment and conditions. But their thermal design is really worse due to size limitations and if it working at high temperature it leads to shortened lifetime.

Yeah, but the end result is nonetheless quite positive: in the early 2000s (that is, before iPhone, before smartphones as we understand them today), the average replacement cycle for mobile phones was 18 months. It’s now 3.6 years. So clearly the hardware has remained reliable enough that, as smartphones matured and got powerful enough that most people do not benefit from upgrading every year or two, most people can hold onto their phone for years.

I also consider your thermal design argument to be fundamentally flawed. No manufacturer wants their phone to run hot, and hardware durability certainly is a concern, the primary reason to avoid running hot is that any heat is wasted energy from the battery, so running hot means reducing the battery life without any benefit, if it can be done cool. And this is why we design mobile SoCs the way we do, with low-power cores to run everyday tasks, and high-performance cores used only when absolutely necessary, and dedicated hardware for all manner of tasks, so that things like audio/video/image compression is done with far less power than if a CPU core had to do it.

Since phones are not intended to be used for running long computing jobs, they are instead optimized for basically running little “spurts” of high performance computation to respond swiftly to user interaction, with plenty of pauses in between during which heat can be dissipated.

[SteveThackery]

The recent posts do seem to undermine @radiolisteners claim that "most" electronics are designed to fail after one year, (maybe less); that failures can be made to happen with an accuracy of weeks, even days; that "most" electronic products do, indeed, fail after about a year.

Like most people, I'm surrounded by dozens of electronic devices, in my case nearly all of which are between 1 and 13 years old. I also have a few at least 40 years old.  These are all working great.  I remember a wall wart power supply failing a few months ago. Beyond that, I'm really struggling to remember the last failure of an electronic device (excluding batteries reaching end of life, of course).

I'm willing to bet that most people's experience is very similar, especially when you remember just how ubiquitous electronic gadgets are.  There just seems to be no evidence to support @radiolistener's "about one year" claim.

[tooki]

Indeed.

I mean, the “within days” claim is absolutely preposterous; even if we did design in deliberate failure mechanisms (other than an RTC with suicide timer), individual usage patterns, storage conditions and duration before sale, and things like environmental conditions during use would all make accurate failure impossible. (I mean, if we say ±4 days, for example, that’s 1% accuracy!)

But obviously no manufacturer wants its products to fail the second the warranty expires, since that would ruin their reputation. This is the blatant fact that “planned obsolescence” loons always forget or ignore.

[coppice]

Yeah, but the end result is nonetheless quite positive: in the early 2000s (that is, before iPhone, before smartphones as we understand them today), the average replacement cycle for mobile phones was 18 months. It’s now 3.6 years. So clearly the hardware has remained reliable enough that, as smartphones matured and got powerful enough that most people do not benefit from upgrading every year or two, most people can hold onto their phone for years.

That's a complex issue. In the early 2000s phones were changing so fast, people really wanted the latest new thing. They were so eager to change they were dumping them within warranty, rather than exercise a free repair. The phones were not so physically robust, and that was OK for the time. Now they are not changing as fast. I think once most people have settled into their phone they actually want to keep it unless it starts to annoy them. Makers have learned how to get greater robustness in a slim package, and third party cases and screen protectors are getting pretty good at absorbing bumps and scratches. So, I think things have fortuitously aligned. People want to keep phones longer, as a new one is not a noteworthy improvement, and makers have learned to make them physically robust enough to last a lot longer. Even non-exchangeable batteries aren't proving to be the limiting factor one might have expected from earlier phones.

[rstofer]

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

I would expect them to be 'hobble' jets and 'jumping' would be a thing of the past.  Pretty much me trying to do jumping jacks.