a 100y lifetime project : incredibly difficult ?

[ker2x]

So, first post here, never designed a project from idea to pcb to mini-production. (never design anything actually, i buy "starter kit".
i got a gift idea for a birth : let's offer a clock that will count day from birth to infinity and beyond.

That's around a 100 year lifetime, and it need to need a stability of 1 day per century.
it need to be rock solid, it need power, it need some kind of super simple display, it need to be switched on.
after a few minute of thinking about it ... doh! is it even possible ?

I'm stuck at the very first problem : power !

Battery :
does a battery that last 100y exist ?
solar panel ? can a solar panel live for 100y ? and it still need some kind of battery anyway
replaceable battery ? who know what kind of battery will be available in 100y ? Perhaps lead-acid battery will be forbidden in 50y, who know ?

Rust/moisture/oxydation :
i tought about sealing the project in a glass cube. Well not glass because the melting temperature is too hot for the electronic to survive, but some kind of transparent plastic ?
Transparent so the solar panel can be sealed, and the display is still readable but not exposed to the elements.
But if it's sealed, how to switch it on ? i switch it on before sealing it ? (there is no need to switch it off)

human interface :
What kind of display ?
It can't be always on, to save power.
But if it's sealed : no push button. perhaps an IR receiver ?

Power again :
i know it's hated but : "energy harvesting" ?
i mean... this think will probably need a few picoAmps (random prefix, i like pico... will it be nano ? femto ? micro ? dunno...), isn't it ?

Any random tought ? some "very important" stuff i didn't mention and forgot about it because i'm a noob ?
Perhaps it shouldn't be on a PCB, i don't know...

Thank you 

Code: [Select]

EDIT (where are we now) :

Mechanic :
- pure mechanical solution (eg : good old clock) is out of topic
- Mechanical part may be introduced if it improve the lifespan and do not require regular servicing (eg : lubrication). But it may greatly increase the manufacturing complexity & cost

Battery :
- no solution found yet
- battery self discharge is a major problem
- super/ultra capacitor are given for < 20 years

Display :
- LED (green one apparently have a better lifespan than red one)
- Electromechanical counter may be a viable option ( [url]https://www.google.com/search?q=electromechanical+counter[/url] )

Oscillator :
- Apparently, aging is not a problem.
- stability may be a problem
- frequency ?

Eprom/flash :
- 100 year doesn't seems to be a problem
- but we need redondancy/error detection

Microcontroller :
- it can be done.

Enclosure :
- need protection from rust/oxidation/moist
- http://en.wikipedia.org/wiki/Poly(methyl_methacrylate)#Acrylate_resin_casting ?

TODO :
- A lot :)
- Grab/post/archive datasheet for future reference
- Power is the main topic.



[EEVblog]

Interesting...
I'm sure there are countless issues involved in keeping something running for 100 years.
Solvable of course, but so many traps, and how do you test it?
I think a mechanical clock is probably the best way to go here?

and there is 10,000 year clock project that cost a measly $42M:
http://www.10000yearclock.net/learnmore.html
http://mashable.com/2012/11/30/jeff-bezos-10000-year-clock/

[ker2x]

IR emitter (like a remote command) may not be easily available in 100y. (we don't use it anymore in 2014  )
It need something as simple as possible.
no code or anything : if any kind of IR is received : display days for 1 second.

I think about engraving the instruction manual in the transparent sealing :
- how to read (eg : if it's a binary display. but 36500 (100years) is 16bit, it may be difficult to read but good for 179.5 years)
- how to switch on the display

Perhaps something about battery too
- eg : need 1h of light per day
- Which is by the way, not practicable. what if it's traveling on a boat for 4 days, it need to survive in dark for some time.

[ker2x]

Interesting...
I'm sure there are countless issues involved in keeping something running for 100 years.
Solvable of course, but so many traps, and how do you test it?
I think a mechanical clock is probably the best way to go here?

and there is 10,000 year clock project:
http://www.10000yearclock.net/learnmore.html

This is the project that gave me this idea 
About testing ... i don't know... probably can't. But the project need to provide as much warranty as possible, which may lead to something super expensive.

Priority :
- Reliability. of course, it's the ultimate priority.
- Cost : not as important, but don't be silly. Let's forget about gold, saphire, diamond. Well... not sure about gold. Considering the power usage, and considering it will be sealed, ultra thin gold wire, gold plated button, ... may be a possible option
- design : let's make it beautifull enough, it should be a gift 

EDIT : $100 doesn't look overpriced, it's (literally) a lifetime gift.

[FrankBuss]

I would use an epaper display:
http://www.frank-buss.de/raspberrypi/epaper/index.html
Needs only power to change the image, but no power to show it. But no idea about the lifetime of it, 100 year is a lot. Same for a microcontroller: the flash might not hold the charges so long, maybe use some old school CPU with an EPROM, but then it needs more power and you would need a big solar cell area. Maybe a smart algorithm which refreshs the flash every some years might work.

And then there are the problems when someone moves to a Mars colony. On mars the solar day is 24 hours and 39 minutes. So you should add some buttons to adjust the length of the day, and to adjust the time if someone just moves to another timezone, because it could display the time as well, and I think it should increase the day count always on midnight.

Accuracy should be no problem, because 100 years are 36,500 days. A 10 ppm crystal (for less than 2 dollar at Digikey) would have an error of 1/3 day after 100 years. But maybe worse after 100 years? Probably a good idea to add an IR sensor to do some day light synchronisation.

Please report back in 2114 in this thread how it did go

[Pedram]

a solar panel and a supercap may give you 20 years.

a primary lithium battery also may give you 25-30 years.


100years.... hmmmmmm

Please report back in 2114 in this thread how it did go

Let's hope Sagan keep this forum alive until 2114!

[ker2x]

And then there are the problems when someone moves to a Mars colony. On mars the solar day is 24 hours and 39 minutes. So you should add some buttons to adjust the length of the day, and to adjust the time if someone just moves to another timezone, because it could display the time as well, and I think it should increase the day count always on midnight.

Let's plan this feature for the version 2, when the v1 will reach its EOL 

[coppice]

Most electronics engineers have the attitude that getting rid of moving parts is generally getting rid of wear and tear, and increases reliability and life expectancy. The reality is its easy to make a mechanical timer than will run for 100 years, and really really really hard to make an electronic one do the same thing. You would need to develop new components for almost every part of the device, as off the shelf ones have something that will chemically degrade over 100 years.

[G7PSK]

For a 100 year plus battery life you would more than likely need a radio isotope thermal battery, the sort of thing NASA uses on some space probes and rovers.

 http://www.technologyreview.com/view/428751/nuclear-generator-powers-curiosity-mars-mission/

Mechanical clocks would certainly last a 100 years plus, I have worked on some clocks in the past that were 150 years old, but they only kept going due to regular servicing and they needed winding once a week or so cant see hou you would provide spring or weight power in one wind for so long.

[ker2x]

Yes, mechanical solution is out of topic.

Perhaps changing a battery every 20 years may be an acceptable solution, but it won't be sealed anymore and may lead to aging problem.

It could be a wearable product with kinetic energy harvesting. But who would want to wear a product all his life ? 

[Marco]

You would need to develop new components for almost every part of the device, as off the shelf ones have something that will chemically degrade over 100 years.

If you seal it from air/moisture/light what exactly is supposed to happen to dry components in 100 years? (Lets ignore electrolytics, they aren't necessary any way.) There are plenty of electronic devices which already have passed the halfway mark.

[FrankBuss]

Maybe use only low-tech components, like this mechanical dot matrix display:



And no highly integrated components, only components with big structures where the chemical degration don't influence it (Commodore had some problems with electromigration in some of there chips, which causes them to fail sometimes after 30 years). Only a lot of transistors to build a counter, and a simple RC oscillator, synchronized to an IR sensor.

Solar cells should work 100 years. One of the first solar cell from 60 years ago is still working:

http://inhabitat.com/worlds-first-modern-solar-panel-still-works-after-60-years/

But there might be a loss of power of 0.5% per year, so only half the power after 100 years.

[zapta]

Why battery? Make it working on mains power, store the start time in non volatile memory, user atomic clock radio signals and you are done. If the clock break, have a way to clone it with the same start time. Piece of cake. ;-)

[max_torque]

Your "Life time" is just the current time minus the time you were born.

Realistically, someone is going to have to set your birth time into the unit (i think mum and dad are going to be a bit busy at your actual birth to worry about pressing a few buttons on a clock!!)  That means you will need some form of "user input" unless you let people specific the "birth time" on ordering, and you write that into the firmware before the unit is sealed.


Current time is more complex.  It could be set by the user, or returned from another souce (GPS, radiodata, internet over wifi etc) but you will have to ensure those sources remain "live" for the duration of the lifetime.  I guess a combination of all four might  be necessary.  That means an onboard realtime clock, that can be set by the user or by input from GPS/radio data etc

If you want it to display simply "days alive" that only obviously changes once a day, then a epaper display makes a lot of sense, otherwise, the device may have to only show the "lifetime" when prompted by the user.

Any of those methods would perhaps be easiest with a wireless access (no plugs/sockets, and the device can be fully sealed), and you could use a "local" webpage to set it up and a remote webpage to supply it with the current time/date etc

Using an inductive charging loop means you could use an internal capacitor, and then give options for the "power supply", like, mains powered, solar powered, even "wind up" etc


interesting project, but very difficult to test over it's (your) lifetime ;-)

[coppice]

Why battery? Make it working on mains power, store the start time in non volatile memory, user atomic clock radio signals and you are done. If the clock break, have a way to clone it with the same start time. Piece of cake. ;-)

Do you think a compatible atomic clock radio signal will be there in 100 years?

[zapta]

OP, you want to check the movie In Time on Netflix. Every person there has a count down clock projected on his arm. When your clock get to zero you die. Lifetime is the currency and you can transfer lifetime between people as you would do with bitcoins (e.g. 1 year for a HP SMU).

[gxti]

GPS, radiodata, internet over wifi etc

... webpage ...

None of those will be around in the same, compatible form in 100 years (except maaaaybe GPS, or at least one of the newer GNSS coming up now). 802.11b will be gone in 25 years tops, and the newer wifi standards are lasting a shorter and shorter time before they get superseded. And the WWVB signal that radio clocks in the USA use just changed formats a few years back, which while it did not affect the cheap receivers did break a bunch of precision instruments. You never know which you can rely on. And in any case you don't need radio signals to keep time to 27ppm (1 day in 100 years) in the first place, let alone a web interface.

[Jeroen3]

There are indeed a few challenges with a project like this.
One of them is indeed the power. Maybe a thermoelectric solution can turn on the display when you hold it.
And maybe use an energy harvester from whatever source to keep an rtc on. The main disadvantage of batteries will be their chemicals.

You cannot have it work on any external time source since timezones and daylight savings will be unknown.
Not to mention the protocols or mediums involved.

It's like those ancient devices found in several sci-fi movies. After 10000 years the power modules/generators still work!

[FrankBuss]

Why battery? Make it working on mains power, store the start time in non volatile memory, user atomic clock radio signals and you are done. If the clock break, have a way to clone it with the same start time. Piece of cake. ;-)

First I thought data retention time of flash memory would be too short. But this application note says on page 4 that 100 years at 25°C is no problem at all and extremely conservative, it might work as long as 1324 years. I don't know if this is true, there might be other problems for this time, like bit flipping because of cosmic rays etc., but 100 years should be possible for a microcontroller with flash memory. Maybe save the program and data multiple times and add some error correction algorithm, for bit flips or broken bits.

[ker2x]

Thank you for all the ideas.

I have no definitive answers because i'm not the expert here, but :

- date set at build time : so yes, you buy it after the birth, which is not the best idea to sell a product. i'm not planning to sell it, i would if i could, but it's unlikely  perhaps build one product per day and sell it on auction  ?
- But if you sell it after the birth, and you only one per day. you can also engrave the name 
- make it an openhardware project sound like a better idea 

- no external timekeeping dependency. I don't believe stuff like DCF77 will still be here in 100 year

- ultra-low-tech : no firmware, no flash, only 7400 TTL logic ?  the size will probably be a problem.
- what about rad-hard low-tech IC ? are they expected to have a huge lifetime ? or is it only about radiation ? i have no idea bout the price of this kind of stuff but it's probably insane.
- i think LED have a huge lifetime, it need to be confirmed. But if you put 16 leds and engrave (in decimal) the value of each bit ... everybody know how to make an addition and i'm pretty sure we'll still use decimal in 100y. And even if we don't... we still know how to read roman numbers so it should be ok and we're good for 65535 days

- if solar panel can survive 100 year in a sealed environment (just pour plexiglass all over the circuit and make it a nice cube or something) that would be awesome. But it still need some kind of battery at night

- external battery : if we can find a way to have an external battery (gold plated contact ? it will probably wear out after 100y of battery switch) then we could have 2 battery.
At every birthday, on odd year : change the left battery. on even year : change the right battery. Instruction engraved, of course. and it still need to be able to survive a lot of years without battery switch, just in case.
But we don't know what battery will still exist in 100 year 

I'm pretty sure it should be kept as lowtech as possible, no memory/MCU/... so no e-ink, even if it's an awesome idea. 
Perhaps i underestimate their lifetime, dunno.

[SeanB]

Long life flash is only possible with old Eprom OTP packages, as these use a large cell size, so the stored charge is large. Current memory is getting down to the point where the stored charge can be counted in numbers of electrons in the floating gate. That is statistically going to lose charge at a rate that might do a decade or two. An old 2716 will likely still retain data after a dozen centuries if it is programmed using the slow method of a 50ms pulse per location. Fast programming will reduce the stored charge somewhat as it uses 1ms pulses until it verifies then gives a small extra charge.

If using eprom verify at 4V and again at 6V to ensure the voltage is over the threshold for it to be reliably read.

[Pedram]

it's pretty hard to keep time with 74xxx logic chip. possible but require at least 15+ different gates/chips.


you can use a device like DS1307  and a button to power up an MCU to  show the date and then power off.... with a simple C/D size primary battery you have to change battery every 25 years.


the problem is flash based MCUs don't last that long.

[KedasProbe]

Sounds more like simple app.

[ker2x]

Long life flash is only possible with old Eprom OTP packages, as these use a large cell size, so the stored charge is large. Current memory is getting down to the point where the stored charge can be counted in numbers of electrons in the floating gate. That is statistically going to lose charge at a rate that might do a decade or two. An old 2716 will likely still retain data after a dozen centuries if it is programmed using the slow method of a 50ms pulse per location. Fast programming will reduce the stored charge somewhat as it uses 1ms pulses until it verifies then gives a small extra charge.

If using eprom verify at 4V and again at 6V to ensure the voltage is over the threshold for it to be reliably read.

Good to know.
16 bit for the day storage, or we could go crazy and store a few more bit. 19 bit = 1436 years

If we can find a MCU that can survive more than a century, the firmware would be ultrasupertiny. Which mean we can store the time at different place in the eprom for redundancy. (and have a display a bit more complex than N leds)

[ker2x]

Sounds more like simple app.

it never was the problem. the problem is the lifetime.
What's your oldest electronic device ?

Mine is probably my FRG-7700 HF receiver : http://en.wikipedia.org/wiki/Yaesu_FRG-7700
it was in production from 78 to 82, the analog part is still working but the digital memory died a long time ago.

[PA0PBZ]

An interesting problem.
I can't think of a display that will last 100 years and is low power, exept maybe this:



The interesting part is that it is both the display and the memory, but not very attractive.
To make it work you need 1 pulse per day, that should be possible to build with long life components.
To power it maybe a combination of solar and different kinds of energy harvesting.
But a power storage device that will last 100 years... no idea.

[ker2x]

An interesting problem.
I can't think of a display that will last 100 years and is low power, exept maybe this:



The interesting part is that it is both the display and the memory, but not very attractive.

Interesting.
It's not really ugly. It look low tech.
Even the shinyest awesome hitech whatever we could build today will look (and will be) low-tech in a century.
So... who care ? Our today awesome stuff will have the same effect as http://www.rewindmuseum.com/vintagetv.htm in a century 

PS : I'm reading this : http://www.sitime.com/support2/documents/AN10025-SiTime-Reliability-Calculations.pdf

[DanielS]

For setting the DoB and current date, one could use NFC with a mask-programmable (ROM) microcontroller. Since the NFC interface is powered by the master device, the wireless connectivity comes at no cost for the sealed device's battery life.

Will NFC still be around 100 years from now? Probably not. But neither will most of what we use today 40 years from now anyway: the last remaining PC connection from 30+ years ago is the serial port but most modern PCs only expose it as a motherboard header if at all. If the unit fails and requires replacement before then, you can upgrade to whatever is current at that point. In principle, the only time you will ever need to interface with it is soon after purchase or replacement; not a century later.

Since there is no guarantee any external timing network in existence today will still be around even 50 years from now, you will need an internal clock and with a typical 1MHz watch crystal, you will need an extra 37 bits to count microseconds per day... so that would be 52bits total for 100 years.

Achieving 1day/century error could be difficult since crystals could have that much of an error simply from their orientation relative to gravity and temperature. The only way around that would be to use an atomic clock like others suggested but those usually consume too much power for something battery-operated intended to last for years.

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

[ker2x]

For setting the DoB and current date, one could use NFC with a mask-programmable (ROM) microcontroller. Since the NFC interface is powered by the master device, the wireless connectivity comes at no cost for the sealed device's battery life.
Will NFC still be around 100 years from now? Probably not.

Not a problem, DoB need to be set only once, and in easily predictable future.

Achieving 1day/century error could be difficult since crystals could have that much of an error simply from their orientation relative to gravity and temperature. The only way around that would be to use an atomic clock like others suggested but those usually consume too much power for something battery-operated intended to last for years.

Is it difficult ? I'm really against any external timekeeping device. 

[SeanB]

EPROM is only to store the MCU program. You would have to store the time likely in SRAM and use a supercap or a silver oxide cell ( rechargeable) that will store enough charge to run the clock oscillator chip and keep the memory alive. You store the start time multiply in the RAM and read them to get the best result of start then use the RTC to count from that. That way you only have the RTC drift to contend with, and this will be within your error margin for most RTC chips.

You will have to have redundant clocks and ram though to be able to recover from single point errors and single failures of the RAM or the clock. You probably will also have to have redundant batteries and switching to enable them to feed the RTC units and the RAM. Batteries would probably be best in separate sealed subcompartments to keep corroding cells from damaging other parts.

[ker2x]

From atmel AVR 8bit FAQ :

Mean Time Between Failures is an indication of the number of hours to pass between failures.

Please find below the predicted MTBF (Mean Time Between Failure) numbers of
our micro's at different temperatures. The statistical calculations are based
on current reliability qualification data in the "microcontroller
reliability data package".

Atmel has been shipping NVM parts for 20 years and Flash Micro's for 13
years. We have never experienced a long term reliability problem.

Here are the MTBF numbers calculated from life test and data retention
results:

65ºC 1.69x10e7 hours. => 1929 years
85ºC 4.46x10e6 hours. => 509 years
105ºC 1.34x10e6 hours. => 153 years

It's not a product supposed to have a very high lifetime, isn't it ?

- But if atmel can do it on their AVR... well, i guess using a microcontroller in this project is a viable option 
- firmware storage look possible too (according to previous post here)
- oscillator lifetime doesn't look like a major problem too.
- we have some options for the "user interface"

Lot of option to explore for the power, which is still the first and main problem 

Energy harvesting (solar/RF/themic,kinetic/...) , if possible, would be best but it can't be the only source of power.
I can't find a supercapacitor with more than 20y lifetime.

[Jeroen3]

The battery with the longest lifespan I could find were NiFe batteries. http://en.wikipedia.org/wiki/Nickel%E2%80%93iron_battery

What would the lifespan of a CR2032 battery be? Vendors don't seem to specify this. I've never seen one leak, as opposed to Alkalines.

[ker2x]

The battery with the longest lifespan I could find were NiFe batteries. http://en.wikipedia.org/wiki/Nickel%E2%80%93iron_battery

What would the lifespan of a CR2032 battery be? Vendors don't seem to specify this. I've never seen one leak, as opposed to Alkalines.

The highest i could find is http://en.wikipedia.org/wiki/Zamboni_pile but it's not an viable option 
I'm just checked some datasheet from energizer. Not much information about lifetime but the few i could find was below 20y.

Nickel-iron battery won't work too.
Energy density is very, very, VERY bad. 
Also : Self-discharge rate   20% – 30%/month 

About the CR2032, the datasheet from enegizer (i checked this one already) say "Self Discharge: ~1% / year"

From TI about rechargeable battery :

Code: [Select]

SELF-DISCHARGE
CELL TYPE        NI-MH NI-CD LI-ION
@ 20°C (%/MONTH) 20-30 15-20 5-10


[Corporate666]

What about a combined electrical and mechanical solution?

You would have to do some calculations to figure out the power involved... but I wonder how much energy a wound-up spring contains and how that compares to a battery? 

You could have a clock-like mechanism but beefier that is wound up when the unit is made... then some sort of latching (or breakable) device that keeps the unit in an inactive state.

When the unit is activated (perhaps by hitting it on a surface, breaking a tiny glass latch that then allows the mechanism to start running), the device starts counting from a set point - like 10 days or 30 days.  This gives the proud parents an opportunity to buy it and have it engraved.  If you can store enough energy in the mechanical mechanism, you will never need a battery and will never have to replace anything.  You could put the whole thing in some sort of glass or plastic container which has a vacuum (or nitrogen) inside so it never corrodes or wears out.

[ker2x]

What about a combined electrical and mechanical solution?

You would have to do some calculations to figure out the power involved... but I wonder how much energy a wound-up spring contains and how that compares to a battery? 

You could have a clock-like mechanism but beefier that is wound up when the unit is made... then some sort of latching (or breakable) device that keeps the unit in an inactive state.

When the unit is activated (perhaps by hitting it on a surface, breaking a tiny glass latch that then allows the mechanism to start running), the device starts counting from a set point - like 10 days or 30 days.  This gives the proud parents an opportunity to buy it and have it engraved.  If you can store enough energy in the mechanical mechanism, you will never need a battery and will never have to replace anything.  You could put the whole thing in some sort of glass or plastic container which has a vacuum (or nitrogen) inside so it never corrodes or wears out.

I love the idea about the breakable mechanism.   

[tszaboo]

100 years contains very low frequencies, so you have very big pink noise. This can include lightnings, nuclear wars, collisions with other space objects... So I would put as much protection devices on board as possible.
I would use an antifuse FPGA for this. The crystal has to be in a super package.
Einstein said, that human stupidity is infinite, so I would power it from that.

[ker2x]

100 years contains very low frequencies, so you have very big pink noise.

I don't understand this part

This can include lightnings, nuclear wars, collisions with other space objects... So I would put as much protection devices on board as possible.
I would use an antifuse FPGA for this. The crystal has to be in a super package.
Einstein said, that human stupidity is infinite, so I would power it from that.

I'll write a memo "warranty void if nuclear winter seal is broken" 
The problem with stupidity : even if it potentially hold an infinite amount of energy, its power efficiency sux 

[max_torque]

if you want to simply count "days" as your smallest unit, could you use some other stimuli to correct the slow drift of a conventional realtime clock?  ie it generally gets dark once a day, so could you use a light sensor with a very very slow "averaging" algorithm of some type to keep the RTO from drifting ??


(obviously, if the device is kept in a drawer for any length of time, that approach will fail!)

[T3sl4co1l]

Well...

Let's make a list of things that absolutely cannot possibly be used:


Energy Storage, Power Generation

- Battery: All types

(Arguably, a Zamboni pile not only could last extremely long, but may even be the only proven multi-century technology!  Even so, note the power density is extraordinarily minuscule.)

- Supercapacitors, electrolytics, polymers (all age limited, susceptible to dry-out or moisture ingress).
- Solar cells (PV) -- dubious; good enough for aerospace, but most are known to degrade over time; also, dust accumulation in dry climates.
- RTGs -- that is, if made with conventional fuels and materials.  With suitable choices, they could last very long indeed, however, and the added weight and space of lower density fuel, extra shielding and poor efficiency converters would be tolerable in a terrestrial application (like the millennium clock thing).


Active Components

- Flash memory, EEPROM (and the vast array of programmable devices which integrate these): charge decays over time, and programming causes incremental damage.

- Fine pitch, low voltage ICs.  Besides being more sensitive to ESD / overvoltage, they're more sensitive, thermodynamically speaking.  (One of the reasons a PC CPU shouldn't go over say 80C, or a GPU over 100C, versus a mil spec transistor over 200C.)  I would be fine with, say, CD4000 series logic at 5V (well within ratings), despite the ESD sensitivity.  74/LS logic should be okay too, but will take a whole lot more power.  Of course, you can't go wrong with discrete transistors.

- ESD, EMC, life in general: everything must be several times below ratings, all circuits well shielded, filtered and protected, with current limiting (of the "big dumb resistor" kind) and multi-level transient protection everywhere.  Most protection devices (MOVs, GDTs, etc.) are wear components, but with current limiting, sufficiently large TVSs should be okay I think?

- What else?


Things that are okay include a surprising variety of mechanical things, but they all have to be magnificently engineered with top-notch materials.  Oxidation, moisture, and lubrication (migration / degradation) are the greatest challenges here.  Examples of energy storage / power generation might include springs (well below ratings -- nearly infinite fatigue limit), weights (gravity isn't going away any time soon, but mind if it moves), and motors (induction motors are essentially indefinite; permanent magnet motors may be susceptible to demagnetization either over time or due to very strong ambient fields).  Electrical energy reserves would be best implemented as massive stacks of film or ceramic capacitors -- very bulky, but won't dry out.

Tim

[FrankBuss]

Let's make a list of things that absolutely cannot possibly be used:
[...]
Active Components

- Flash memory, EEPROM (and the vast array of programmable devices which integrate these): charge decays over time, and programming causes incremental damage.

Both Atmel (for AVR) and TI (for MSP430) claims at least 100 years data retention for their flash, so I don't see a problem. Of course, single bit failures could still happen, like from cosmic rays or if one bit is not as reliable. Best would be to use many microcontrollers and an external majority circuit, like used in safety critical systems like nuclear reactors and spacecrafts, and self testing and error correcting firmware.

LEDs might be a good idea. I found this report with MTBF of 2,000 years for green LEDs. Interestingly red LEDs have a MTBF of only 175 years. But I guess this would be much longer if you don't use the maximum rated current. But it requires more power, or you need a button to show the number of days.

I couldn't find any MTBF rating for epapers.

Good finding for the SiTime spec. So one of those SiTime oscillators, an AVR microcontroller, some kind of sealed mechanical display, solar cells for the power supply and some big conventional film or ceramic capacitors for the backup should last 100 years. You could seal the whole circuit, maybe add some LEDs for status display and some capacitive buttons for setup.

[EEVblog]

I'll write a memo "warranty void if nuclear winter seal is broken" 

Include one of these in your BOM:
http://www.maxwell.com/products/microelectronics/docs/hsn1000_rev3.pdf
When your /NED pin goes low, activate protection circuitry.

[EEVblog]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

From the website:

Underwater sensors for seismic research or gas and oil exploration

Yep! I worked in that field for 10 years, and yes, our biggest problem was always precisely that (pun intended), a precision autonomous low power clock.
When I left in 2007, the problem still hadn't been solved. Now solutions like this make me want to cry 

[coppice]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

Does anyone know how a device like this can be atomic, and yet have such a poor aging characteristic? Atomic clocks are supposed to use very fundamental qualities of matter. How can that age?

[ElektroQuark]

Make a clepsidra.

[mikerj]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

Does anyone know how a device like this can be atomic, and yet have such a poor aging characteristic? Atomic clocks are supposed to use very fundamental qualities of matter. How can that age?

It's aging is specified at 3.0E-10/month, doesn't seem too bad to me!

[ker2x]

but 120mW is HUGE 
We already have a major problem with battery without even considering any power consumption (the aging itself is a problem).

[mikeselectricstuff]

Power is clearly going to be the biggest issue, so if you can decouple the time count from any processing & display, that would probably simplify things a lot.

So for example find the lowest power RTC chip you can find  to keep time, and an MCU/display that's only powered when there is enough power available - I think Solar is probably the only option if you don't want a radioactive battery.
Maybe multiple RTCs with multiple power sources for backup.

Any storage capacitance probably doesn't want to have anything liquid in it, so maybe (a lot of) tantalum, polymer  or ceramic.

[Jeroen3]

A mechanical system (spring) that hits a piezo every once in a while to recharge your energy harvesting solution.
They've made switches that send wireless signals when switched. All powered from a piezo energy of the click itself.

[ker2x]

edited first post

Power is clearly going to be the biggest issue, so if you can decouple the time count from any processing & display, that would probably simplify things a lot.

So for example find the lowest power RTC chip you can find  to keep time, and an MCU/display that's only powered when there is enough power available - I think Solar is probably the only option if you don't want a radioactive battery.
Maybe multiple RTCs with multiple power sources for backup.

Any storage capacitance probably doesn't want to have anything liquid in it, so maybe (a lot of) tantalum, polymer  or ceramic.

Yes, power is the bigest issue. no solution found yet.
I wrote to energizer with a simple description of the project and a simple a question : "is there any battery form factor guaranteed to still exist in 100 years ?"
Perhaps some kind of industrial form factor have a guaranteed minimum EOL.
I'm not expecting a reply but... who knows ? 


Also... why a RTC ? It seems to be much more complex than it need to be.
Why not a simple counter ? It's all about counting days. (also, i didn't find RTC that will still work after 2099, which is a problem)

A mechanical system (spring) that hits a piezo every once in a while to recharge your energy harvesting solution.
They've made switches that send wireless signals when switched. All powered from a piezo energy of the click itself.

We still need to store this energy, it may solve the self-discharge problem, but not the mechanical/chemical lifespan.

Also there will be a problem with isolation from exterior. It would be nice of all the electronic part were completely isolated from air.
Perhaps something with magnet and antenna ?

[mikeselectricstuff]

Also... why a RTC ? It seems to be much more complex than it need to be.
Why not a simple counter ? It's all about counting days. (also, i didn't find RTC that will still work after 2099, which is a problem)

Because an RTC is specifically engineered for the job - the counter isn't a huge issue but the oscillator is, and is probably the largest power consumer. Date range is not an issue - just keep any higher-order count seperately - many RTCs have some extra registers for user data, so have a bit for year rollover

[ker2x]

Also... why a RTC ? It seems to be much more complex than it need to be.
Why not a simple counter ? It's all about counting days. (also, i didn't find RTC that will still work after 2099, which is a problem)

Because an RTC is specifically engineered for the job - the counter isn't a huge issue but the oscillator is, and is probably the largest power consumer. Date range is not an issue - just keep any higher-order count seperately - many RTCs have some extra registers for user data, so have a bit for year rollover

The problem is leap year managment :
If the year can be evenly divided by 100, it is NOT a leap year,
unless; The year is also evenly divisible by 400. Then it is a leap year.
This means that 2000 and 2400 are leap years, while 1800, 1900, 2100, 2200, 2300 and 2500 are NOT leap years.

[ker2x]

We could also say it's not a real problem and just go with it.

Some random datasheet :
http://www.nxp.com/documents/data_sheet/PCF8563.pdf

Edit : reading this http://www.microchip.com/pagehandler/en-us/technology/xlp/
Edit 2 : http://www.cymbet.com/products/

[mikeselectricstuff]

Also... why a RTC ? It seems to be much more complex than it need to be.
Why not a simple counter ? It's all about counting days. (also, i didn't find RTC that will still work after 2099, which is a problem)

Because an RTC is specifically engineered for the job - the counter isn't a huge issue but the oscillator is, and is probably the largest power consumer. Date range is not an issue - just keep any higher-order count seperately - many RTCs have some extra registers for user data, so have a bit for year rollover

The problem is leap year managment :
If the year can be evenly divided by 100, it is NOT a leap year,
unless; The year is also evenly divisible by 400. Then it is a leap year.
This means that 2000 and 2400 are leap years, while 1800, 1900, 2100, 2200, 2300 and 2500 are NOT leap years.

This stuff can all be dealt with easily in software whenever the main controller is running - you don't have to use the RTCs logic - all you need is an elaped time count

[coppice]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

Does anyone know how a device like this can be atomic, and yet have such a poor aging characteristic? Atomic clocks are supposed to use very fundamental qualities of matter. How can that age?

It's aging is specified at 3.0E-10/month, doesn't seem too bad to me!

Its not bad if you are used to crystals, but its horrible for an atomic source. My point is, where does it come from? If the source of the timing is an atomic physics measurement, how does that age so much?

[PA0PBZ]

The problem is leap year managment :
If the year can be evenly divided by 100, it is NOT a leap year,
unless; The year is also evenly divisible by 400. Then it is a leap year.
This means that 2000 and 2400 are leap years, while 1800, 1900, 2100, 2200, 2300 and 2500 are NOT leap years.

Why would you worry about leap years if all you do is count days?

[tszaboo]

100 years contains very low frequencies, so you have very big pink noise.

I don't understand this part

This can include lightnings, nuclear wars, collisions with other space objects... So I would put as much protection devices on board as possible.
I would use an antifuse FPGA for this. The crystal has to be in a super package.
Einstein said, that human stupidity is infinite, so I would power it from that.

I'll write a memo "warranty void if nuclear winter seal is broken" 
The problem with stupidity : even if it potentially hold an infinite amount of energy, its power efficiency sux 

Well, this comes from an article about pink noise I've read a long time ago. Since our electronics products have a lifetime which measures in years, the 0.00...01 Hz noise is going to be very large. The circuit is switched on, switched off, environment changes, parts are ageing... All of this can be inspected in the frequency region. 0.00001Hz region is something happening every day, and so on. The wider the frequency range on the lower end, the harder to make something work. So making your stuff work is not only a problem in the MHZ region, but also in the microhertz region. And your DC voltage in the specification becomes "1mHz is good enough".

[ker2x]

The problem is leap year managment :
If the year can be evenly divided by 100, it is NOT a leap year,
unless; The year is also evenly divisible by 400. Then it is a leap year.
This means that 2000 and 2400 are leap years, while 1800, 1900, 2100, 2200, 2300 and 2500 are NOT leap years.

Why would you worry about leap years if all you do is count days?

If you use a counter : don't care
If you use a RTC with "days since birth = today date - birth date" then you have a problem with leap year. All rtc seems to handle the 4 years leap, but not the 100 & 400 year leap 

[mikeselectricstuff]

The problem is leap year managment :
If the year can be evenly divided by 100, it is NOT a leap year,
unless; The year is also evenly divisible by 400. Then it is a leap year.
This means that 2000 and 2400 are leap years, while 1800, 1900, 2100, 2200, 2300 and 2500 are NOT leap years.

Why would you worry about leap years if all you do is count days?

If you use a counter : don't care
If you use a RTC with "days since birth = today date - birth date" then you have a problem with leap year. All rtc seems to handle the 4 years leap, but not the 100 & 400 year leap 

As long as you know how the RTC behaves, you can deal with it. This is an insignificant issue compared to everything else.

[janekm]

Yes, power is the bigest issue. no solution found yet.
I wrote to energizer with a simple description of the project and a simple a question : "is there any battery form factor guaranteed to still exist in 100 years ?"
Perhaps some kind of industrial form factor have a guaranteed minimum EOL.
I'm not expecting a reply but... who knows ? 

It's Tadiran you want to write to for this: http://www.tadiranbat.com. They "only" aim for 40 years lifetime, but given the tiny energy requirements you have for the RTC I suspect they'd manage. They might enjoy the challenge

[ker2x]

It's Tadiran you want to write to for this: http://www.tadiranbat.com. They "only" aim for 40 years lifetime, but given the tiny energy requirements you have for the RTC I suspect they'd manage. They might enjoy the challenge

i mail them today.
Thank you 

[ker2x]

About battery form factor, coin cell look ideal.
I'm pretty sure we can design a cell holder that can accept a wide range of coin cell form factor.

[G7PSK]

If you could make it a wearable item you could use body heat for the power source.

[ker2x]

TS3005: A 1.55V TO 5.25V, 1.35µA, 1.7ms TO 33hrs, PIN-PROGRAMMABLE TIMER IC
http://touchstonesemi.com/products/ts3005

[ker2x]

Tadiran's french sales manager replied to my mail   
I call them monday 

[Artraze]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

Does anyone know how a device like this can be atomic, and yet have such a poor aging characteristic? Atomic clocks are supposed to use very fundamental qualities of matter. How can that age?

It's aging is specified at 3.0E-10/month, doesn't seem too bad to me!

Its not bad if you are used to crystals, but its horrible for an atomic source. My point is, where does it come from? If the source of the timing is an atomic physics measurement, how does that age so much?

Rubidium references work by using a PLL to discipline a normal ovenized crystal to a ~7GHz microwave source that causes a 0.1% droop in the output of a rubidium lamp when it locks.  There are a lot of potential sources of drift, from the electronics to the resonant cell.  Even high performance devices are usually only 4e-11/month, not even an order of magnitude better.  When it comes right down to it, rubidium gas cells simply aren't very good atomic clocks.

Anyways, they also have their yearly aging rate given as 1e-9/yr, which means only a couple minutes error accumulated over 100 years.  Not too bad.  Unfortunately they usually only have a design life of something like ~20 years so wouldn't be usable in this application.

[DanielS]

Just for fun, I went searching for a low-power atomic clocks and found this:
http://www.microsemi.com/products/timing-synchronization-systems/embedded-timing-solutions/components/sa-45s-chip-scale-atomic-clock
Sub-120mW chip-scale atomic clock, sounds interesting. Only rated for 100k hours MTBF though. Sounds like a great fit/upgrade for those Agilent frequency counters with lousy stock oscillator.

Does anyone know how a device like this can be atomic, and yet have such a poor aging characteristic? Atomic clocks are supposed to use very fundamental qualities of matter. How can that age?

The same way rubidium and cesium clocks age: the material used in the oscillator evaporates, embeds itself in the resonating cavity's walls, is no longer part of the resonant cavity and the loss of vapor pressure is enough to cause minor drift. You also have the atoms' decay rate, possible impurities, the VCSEL and sensor's aging, etc. National standard atomic clocks require periodic maintenance to maintain their stability so expecting perfect long-term stability out of a sealed, maintenance-free autonomous module is not very realistic.

Dave did a video about a museum HP/Agilent cesium clock source and the HP guy said those cesium sources have a useful life of about 20 years due to material evaporation... but that is for a lab-grade time reference; not a national-grade one.

Even those million-dollars national time standards cannot agree perfectly with each other and require cross-checking with other national-grade references to correct their drift.

[Pedram]

TS3005: A 1.55V TO 5.25V, 1.35µA, 1.7ms TO 33hrs, PIN-PROGRAMMABLE TIMER IC
http://touchstonesemi.com/products/ts3005

jellybean DS1307 consumes only 400nA

so 1.35uA is very high.



Also check this one:
http://www.abracon.com/Precisiontiming/AB18XX.pdf

[ker2x]

TS3005: A 1.55V TO 5.25V, 1.35µA, 1.7ms TO 33hrs, PIN-PROGRAMMABLE TIMER IC
http://touchstonesemi.com/products/ts3005

jellybean DS1307 consumes only 400nA

so 1.35uA is very high.



Also check this one:
http://www.abracon.com/Precisiontiming/AB18XX.pdf

 thx

[Corporate666]

Someone who is less lazy than me still should work out the math... what would the TOTAL power consumption of the device be over 100 years?

Because I don't think you can get around the power issue unless you use nuclear materials (might be hard to source now that Ghaddafi is gone) or go with the mechanically stored power.

Anything else - heat, light, batteries or whatever else will always require human interaction, and I think the fact that it will run for 100 years untouched is the appeal of the device.

If the total amount of power needed was known, the mass and height of a weight could be calculated needed to power the device.  Plus, it would be pretty cool that, year-after-year, the weight drops fractionally... like a few millimeters or less.  Almost like... "when the weight reaches the bottom... you're dead"

[AG6QR]

The Atmos clock solved the problem a while back, and the earliest ones seem to have made significant progress toward the 100 year lifetime.

http://en.wikipedia.org/wiki/Atmos_clock

They get their energy based on small variations in temperature and pressure.  As long as the weather doesn't stop, they'll have energy to keep running.

[joneikifi]

If the total amount of power needed was known, the mass and height of a weight could be calculated needed to power the device.  Plus, it would be pretty cool that, year-after-year, the weight drops fractionally... like a few millimeters or less.  Almost like... "when the weight reaches the bottom... you're dead"

If my Wolfram Alpha Fu is correct, a device using 3 volts at 1 uA for ONE year would use the potential energy of a 9.6 kg weight lifted by one meter.

Code: [Select]

(3V * 1 uA * 1 year) / (1 m * 9.81 m/s^2)

[johansen]

edit: and its gone

[ker2x]

Someone who is less lazy than me still should work out the math... what would the TOTAL power consumption of the device be over 100 years?

Because I don't think you can get around the power issue unless you use nuclear materials (might be hard to source now that Ghaddafi is gone) or go with the mechanically stored power.

Anything else - heat, light, batteries or whatever else will always require human interaction, and I think the fact that it will run for 100 years untouched is the appeal of the device.

If the total amount of power needed was known, the mass and height of a weight could be calculated needed to power the device.  Plus, it would be pretty cool that, year-after-year, the weight drops fractionally... like a few millimeters or less.  Almost like... "when the weight reaches the bottom... you're dead"

Nuclear will probably require servicing too.
I can't think of any solution that doesn't require some kind of battery change or regular servicing (even if "regular" = 40 years)

A device that require a new battery every 40 years could be an acceptable solution... perhaps... i don't tknow 
And perhaps in 40y we'll have battery that last 100years ?

But there is the form factor problem. What kind of battery form factor will be available at this time ?

[mikeselectricstuff]

Someone who is less lazy than me still should work out the math... what would the TOTAL power consumption of the device be over 100 years?

1uA over 100 years is 876mAh so in principle an AA cell or 2477 lithium, but shelf-life is going to be the dominant factor by far with any chemical battery.
I'd say for a solution based on readily available tech, solar plus large value low-leakage capacitor probably has the best chance, as long as it's kept somewhere that  receives some light each day.

[Jeroen3]

Dave found some lithium powered sram seemlingy working after 27 years. (ep 615)

[tszaboo]

Someone who is less lazy than me still should work out the math... what would the TOTAL power consumption of the device be over 100 years?

1uA over 100 years is 876mAh so in principle an AA cell or 2477 lithium, but shelf-life is going to be the dominant factor by far with any chemical battery.
I'd say for a solution based on readily available tech, solar plus large value low-leakage capacitor probably has the best chance, as long as it's kept somewhere that  receives some light each day.

Not to mention he wants display.

[charlespax]

There always a nuclear option. It looks like a NanoTritium will last about twenty years. Maybe a few of these?

You can even get one on TaoBao

[ker2x]

Not to mention he wants display.

Some kind of display indeed.
It need a way to show the days. but temporarly , not a permanent display.

[ker2x]

There always a nuclear option. It looks like a NanoTritium will last about twenty years. Maybe a few of these?

You can even get one on TaoBao

Perhaps 
But if they all die of old age after 20y, it won't help to have 1 or many of them 

[charlespax]

Perhaps 
But if they all die of old age after 20y, it won't help to have 1 or many of them 

The page states
The NanoTritium™ betavoltaic power source provides a source of continuous nanoWatt power for twenty years or more in microelectronic platforms. Sounds like it's only a matter of half-life. 

[ker2x]

The page states
The NanoTritium™ betavoltaic power source provides a source of continuous nanoWatt power for twenty years or more in microelectronic platforms. Sounds like it's only a matter of half-life. 

 

Available commercially, the device is expected to be valued in the “couple thousand dollar range” at first, Cabauy said, but in time as the company produces more the price may become less. The battery is currently available in “engineering” quantities, according to the company, up to 1,000 a year, and is assembled in the company’s lab.

[janekm]

Perhaps 
But if they all die of old age after 20y, it won't help to have 1 or many of them 

The page states
The NanoTritium™ betavoltaic power source provides a source of continuous nanoWatt power for twenty years or more in microelectronic platforms. Sounds like it's only a matter of half-life. 

Yeah, but the half-life seems to be around 12 years... The  "capacity-halflife" of the lithium thionyl chloride batteries at <1% self discharge should be >60 years theoretically, but perhaps  they degrade in somer other way.

[sunnyhighway]

Not to mention he wants display.

Some kind of display indeed.
It need a way to show the days. but temporarly , not a permanent display.

Mike already tackled that issue here, by having the display and its display logic powered from a solar power alone. Much like those purely solar powered calculators. No buttons required to activate it either, just supply enough (solar)light to read the display.

[Zero999]

It's certainly possible with today's technology.

There's the Oxford electric bell which is powered from batteries and has been ringing since 1840.
http://en.wikipedia.org/wiki/Oxford_Electric_Bell

[PA0PBZ]

It's certainly possible with today's technology.

There's the Oxford electric bell which is powered from batteries and has been ringing since 1840.
http://en.wikipedia.org/wiki/Oxford_Electric_Bell

Yeah, try to run a clock oscillator and a display from that...

[Phyvel]

Firstly, I haven't the time to read all the comments, I won't pretend to have done so:

Power - Solar could work just fine, given the right amount of cells, it would last an overnight discharge process. However, you'll run into problems during winter, and rainfall. It is entirely possible to keep a container sealed from the elements, with it being glass, even if you have a cable running into it; You simply need to ensure all holes are sealed. Epoxy resins can do this rather well.

Moisture, et cætera - You would be surprised how much heat electronics can survive, for intermittent periods, as it would take to seal a glass container. Regardless, I would try treating the whole thing like a vacuum tube, and flash some kind of "getter" into the sealed container, to remove an gasses which might develop over time; As gases are emitted from glass and other elements over time ... Especially in a vacuum. 

Interface: It could be powered on/off with a magnetic switch. No need for any fancy switching/remotes (which die, or get lost, and replacing them can be difficult, in say, 50 years). Magnets will be around for a looooooong time; some might even say "forever."

PCB: That, I would say, is personal preference. I like seeing things which are like a sculpture, not flat. Those things rarely get done, anymore.

Anyway, good luck. I assume you'll post something herein.

[lapm]

Maybe instruction of maintenance could be made in clay tablets? Some of those has already survived collapse of civilization that created them...

[ker2x]

Firstly, I haven't the time to read all the comments, I won't pretend to have done so:

Power - Solar could work just fine, given the right amount of cells, it would last an overnight discharge process. However, you'll run into problems during winter, and rainfall. It is entirely possible to keep a container sealed from the elements, with it being glass, even if you have a cable running into it; You simply need to ensure all holes are sealed. Epoxy resins can do this rather well.

Moisture, et cætera - You would be surprised how much heat electronics can survive, for intermittent periods, as it would take to seal a glass container. Regardless, I would try treating the whole thing like a vacuum tube, and flash some kind of "getter" into the sealed container, to remove an gasses which might develop over time; As gases are emitted from glass and other elements over time ... Especially in a vacuum. 

Interface: It could be powered on/off with a magnetic switch. No need for any fancy switching/remotes (which die, or get lost, and replacing them can be difficult, in say, 50 years). Magnets will be around for a looooooong time; some might even say "forever."

PCB: That, I would say, is personal preference. I like seeing things which are like a sculpture, not flat. Those things rarely get done, anymore.

Anyway, good luck. I assume you'll post something herein.

The problem with solar+cell is that the cell will die after ~20years
The magnetic switch is a good idea
About sculpture : agreed 

[Zero999]

It's certainly possible with today's technology.

There's the Oxford electric bell which is powered from batteries and has been ringing since 1840.
http://en.wikipedia.org/wiki/Oxford_Electric_Bell

Yeah, try to run a clock oscillator and a display from that...

Come on, you know what I meant. If it couldbe done back then, it can be done today but better.

[FrankBuss]

The problem with solar+cell is that the cell will die after ~20years

Do you have a reference for this? In this posting I wrote about solar cells. And for even higher lifetime, you don't need to exposure it to full sun light, like the solar cell calculators, which work in the shadow or dim artificial lighting, too.

[ker2x]

The problem with solar+cell is that the cell will die after ~20years

Do you have a reference for this? In this posting I wrote about solar cells. And for even higher lifetime, you don't need to exposure it to full sun light, like the solar cell calculators, which work in the shadow or dim artificial lighting, too.

Sorry, misunderstanding here.
i was thinking about battery to support the solar cell but that's not what you wrote. 

[ker2x]

That, would be incredibely awesome if it didn't suck so much power :


Edit : well, i'm assuming it use an insane lot of power compared to what could provide a small solar panel. But i may be wrong.

[FrankBuss]

I see, you meant solar cell + battery cell. Just "solar" means "relating to the Sun", only, for me, if you don't say "solar cell" or "solar panel", but this might be because English is not my native language

A vacuum fluorescent display needs a heated filament. Wikipedia says at least 0.2 W (but it doesn't say if this is per digit or per display). Might work, if you have a big solar panel and a lot of ceramic backup capacitors.

[ker2x]

I see, you meant solar cell + battery cell. Just "solar" means "relating to the Sun", only, for me, if you don't say "solar cell" or "solar panel", but this might be because English is not my native language

A vacuum fluorescent display needs a heated filament. Wikipedia says at least 0.2 W (but it doesn't say if this is per digit or per display). Might work, if you have a big solar panel and a lot of ceramic backup capacitors.

And since english isn't my main language isn't english either : we're in troubles 

[Jeroen3]

There is only one power source capable of supplying a project like this, standalone, for a long time.
A radioisotope thermoelectric generator, the Russians powered lighthouses from that.  I even think electric cars can be equipped with these if they weren't dangerous when opened.

[ker2x]

There is only one power source capable of supplying a project like this, standalone, for a long time.
A radioisotope thermoelectric generator, the Russians powered lighthouses from that.  I even think electric cars can be equipped with these if they weren't dangerous when opened.

Too expensive 

[miguelvp]

It serves Voyager well, maybe a smaller version that don't need that much power?

http://en.wikipedia.org/wiki/MHW-RTG

But back to the solar panels, if a solar calculator can last as long as they do, heck even battery power calculators I do have one that I never changed the batteries and still goes after 20 years with a couple of LR44s (or equivalent).

Actually I opened the case and they are energizers 357 I'm not sure if they are OEM anymore, I do think so.

I can envision an old solar power calculator being around after 100 years, then again not sure how often it's been used, like mine I don't use it but it's around and I power it every now and then, it was the only programmer's decent calculator back then, or at least I thought so. (1990s or so)

[tautech]

Or the 30 yr old Texas Instruments one I have,
BTW Made in China

[janekm]

Solar cell should be fine indeed, so long as it's sufficiently over-dimensioned to allow for the drop-off in power output: http://energyinformative.org/lifespan-solar-panels/

Looks like real-life figures are even better: http://www.engineering.com/ElectronicsDesign/ElectronicsDesignArticles/ArticleID/7475/What-Is-the-Lifespan-of-a-Solar-Panel.aspx

I was curious about the drop-off since a solar cell seems like a fairly stable device to me, looks like it's mostly due to mechanical stresses.

The display output is more tricky in my view... All the obvious choices either have known lifetime issues or not proven yet:

• the liquid crystal in all types of LCDs are known to degrade over time
• E-ink isn't proven and manufacturer claims of 5 years: http://www.eink.com/sell_sheets/pearl_spec_sheet.pdf
• LEDs have a theoretical life expectancy of 20000 - 50000 years (but real-life figures are variable)
• Mechanical devices have obvious wear issues

Any other ideas?

[Rerouter]

a very small flip dot display for the day fields onwards, and an lcd for the hours minutes seconds, that stays on for a few minutes after pushing a button? in my own experience an LCD lasts longer when not constantly on and not in even medium brightness sunlight for extended periods,

I Commonly deal with automotive displays from the early 90's which have suffered mechanical faults but have not faded by any great measure, By mechanical faults it is generally from a poorly designed plastic housing warping over time and stressing the die on glass controllers connections, or the flat flex traces debonding from the material, in the direct sunlight cases,

That on its own would likely be a decent place to look, as there are a large number of automotive LCD clock displays out there, and i have no doubts if you kept a few spares tucked under some cover it would not be cheating as long as it did not loose time while replacing,

[ker2x]

a very small flip dot display for the day fields onwards, and an lcd for the hours minutes seconds, that stays on for a few minutes after pushing a button? in my own experience an LCD lasts longer when not constantly on and not in even medium brightness sunlight for extended periods,

I Commonly deal with automotive displays from the early 90's which have suffered mechanical faults but have not faded by any great measure, By mechanical faults it is generally from a poorly designed plastic housing warping over time and stressing the die on glass controllers connections, or the flat flex traces debonding from the material, in the direct sunlight cases,

That on its own would likely be a decent place to look, as there are a large number of automotive LCD clock displays out there, and i have no doubts if you kept a few spares tucked under some cover it would not be cheating as long as it did not loose time while replacing,

This project only count days 

[coppice]

• the liquid crystal in all types of LCDs are known to degrade over time
• E-ink isn't proven and manufacturer claims of 5 years: http://www.eink.com/sell_sheets/pearl_spec_sheet.pdf
• LEDs have a theoretical life expectancy of 20000 - 50000 years (but real-life figures are variable)
• Mechanical devices have obvious wear issues

I don't know about LCD material degrading over time. I thought LCD displays only degraded with use, because of asymmetry in the waveforms. If you only run the LCD when it is needed I think the lifetime can be extremely long. I believe cholesteric displays have an enormous expected lifetime, because you don't need to keep refreshing them.

[ziggyfish]

There are a couple of ways you could do it. But I would recommend:

oscillator + microcontroller + solar + mechanical counter.

Day = oscillator hz * 86400

Program the microcontroller with a simple digital counter (store in one of the controllers registers), everytime the digital counter gets up to the value Day. Move the mechanical counter up by one, and set the digital counter back to zero.

Its that simple.

Thanks

Brendan

[mikeselectricstuff]

There are a couple of ways you could do it. But I would recommend:

oscillator + microcontroller + solar + mechanical counter.

Day = oscillator hz * 86400

Program the microcontroller with a simple digital counter (store in one of the controllers registers), everytime the digital counter gets up to the value Day. Move the mechanical counter up by one, and set the digital counter back to zero.

Its that simple.

Thanks

Brendan

Instead of a mech counter, how about taking a battery operated wall clock, replacing the scale so that the long hand counts 1-10 years, the short hand shows 10s of years (up to 120), and the sweep hand shows (approx) weeks.
 This would only need enough power to kick the motor every 6-ish days. The avarage draw would be in the sub-uA range
Put some solar cells on the face, and enough dry capacitors to hold for a few days.

If placed on an indoor wall it should easily get enough light to manage 1uA avarage.

[PA0PBZ]

Instead of a mech counter, how about taking a battery operated wall clock, replacing the scale so that the long hand counts 1-10 years, the short hand shows 10s of years (up to 120), and the sweep hand shows (approx) weeks.

How is that not a mech counter? 

[ziggyfish]

• Mechanical devices have obvious wear issues

Using a mechanical counter, the device will change at max 36,524 times over its lifetime. Considering your car's odometer changes 15,000 times on average per year. I don't think there it would be that much of an issue.

[janekm]

But does the odometer from a car that's been sitting in storage for 100 years still work? Or has the lube dried out? The gears rusted? The plastic degraded? (in fact, a watch that only moves once a day might seize up more easily!)
Mechanical watches are intended to be services every 5 years or so, at a minimum to get re-lubricated...

My gut feeling is that trying to build a mechanical indicator that is guaranteed to work without maintenance for 100 years would be trickier than finding a solid-state type of display that would.

[mikeselectricstuff]

Instead of a mech counter, how about taking a battery operated wall clock, replacing the scale so that the long hand counts 1-10 years, the short hand shows 10s of years (up to 120), and the sweep hand shows (approx) weeks.

How is that not a mech counter? 

it's a dial, not a counter

[Rerouter]

Janekm is dead on the money with the odometer example, i have worked on a 1890's military jeep's instrument cluster, and based on the owners knowledge it had to be serviced and the entire gear set replaced every 30 years or so,

When we had gotten to it, the lubrication had well past evaporated, the metal on metal shaft had rusted seized, which munched the brass gears, (it was interesting as we had no chance of there being spare parts....), automotive clocks between the late 60's to late 80's tended to have metal gears and metal bearings, over time the bearing would widen from the weight of the gears and eventually the gear stops making contact with the rest of the assembly.

Next along, metal shaft, plastic bearing, a large number of automotive dashes with number wheel displays face this problem, some form of oil or binder eventually sweats out of the plastic, which even on daily driven vehicles will add enough friction that it will eventually munch the plastic gears used in these models, drilling in and fitting a brass bearing gets almost another 15-20 years out of it, but it then suffers from the above when the metal shafts coating gets scraped by the brass and it rusts,

and finally plastic on plastic, I more commonly see this with modern automotive clocks, say 1995 onwards in vehicles not using a digital display, I believe a change in plastics has resolved most of the sweating issue, but in its place the plastic soaks up any solvent in the air so 10 years down the track it just crumbles to dust if you try and set the time,

For this project i would actually recommend metal on plastic, if you had to go mechanical, if the gears where also metal i believe these automotive clocks which should be going through far worse conditions could have worked up to 40 or 50 years if the plastic gears could not crumble,

[Prime73]

there are self lubricating ceramics which potentially could be used to create gears, ball bearings, etc for the mechanism. I'm not sure about ceramics gears, however ceramics in general survive hundreds of yeah and more.

[mikeselectricstuff]

Didn't the watchmaking industry solve this a while ago with jewelled bearings?

[denelec]

Janekm is dead on the money with the odometer example, i have worked on a 1890's military jeep's instrument cluster, and based on the owners knowledge it had to be serviced and the entire gear set replaced every 30 years or so,

A Jeep in 1890???  Steam powered?

[lewis]

I don't see this project as being massively difficult, as long as you don't expect it to be entirely self-powered, with a display, for 100 years:

# Low power RTCC chip for timekeeping, or ridiculously low power MCU on its own. Flash retention issue solved by self-reprogramming the MCU 'periodically' (every 10 years). http://ww1.microchip.com/downloads/en/DeviceDoc/20005010F.pdf achieves <1uA and has leap-year compensation built in to year 2399.

# Crystal accuracy can be improved by calibrating for precise frequency during manufacture, and temperature compensating during use. Use redundant oscillators and RTCCs if you're paranoid, at the expense of power consumption.

# Energy harvesting power source - the key is to power it from as many sources as possible. Mains power + Lithium battery + solar cells, thermocouples, microphone, piezo, coil antenna, wind-up mechanism, whatever etc, all charging a supercap for powering the RTCC. We must be able to get a few microwatts out of the air somewhere. Oversize the supercap and battery to allow for degradation. LT make some energy harvesting stuff: http://www.linear.com/parametric/energy_harvesting

# "Mains Power" means 3-30V AC/DC - who knows what sort of voltages will be around in 100 years, but it's a safe bet we'll be able to feed this thing with 12V from somewhere, somehow. Since this is the only direct electrical interface to the outside world, protect it heavily with TVSes and redundant polyswitches.

# Use whatever display you like, but only light it up when it's powered from the mains. Gives incentive for customer to mains power it thereby increasing the time the unit is powered from a reliable source. LED is probably the best lifetime - they reduce in brightness, but never 'die' unless you overdrive them. Mains power also charges internal lithium battery and supercap.

# Make a mould, and pot the whole thing under light vacuum in a cube of transparent ('optically clear') polyurethane resin. Sort of like this (but use good resin) http://www.ebay.com/itm/EPOXY-RESIN-CRYSTAL-CLEAR-CASTING-IMBEDDING-1-5-GAL-KIT-/220350141369. Use IP-rated vandal-resistant switches for setting the countdown time, or use capacitive or optical sensing only when mains power is present or there's significant charge in the battery. Potting will help prevent the supercaps drying out and make it "infinitely" rugged (no tin whiskers, moisture ingress, corrosion, or susceptibility to mechanical damage to worry about). Transparent shows off the internal construction, and you don't need to worry about holes to see the display through.

Not a trivial project, but I don't think it's beyond the realms of the hobbyist with today's technology. Nuclear power sources notwithstanding.

[coppice]

Janekm is dead on the money with the odometer example, i have worked on a 1890's military jeep's instrument cluster, and based on the owners knowledge it had to be serviced and the entire gear set replaced every 30 years or so,

A Jeep in 1890???  Steam powered?

I wonder if Stanley or Doble ever made a 4 wheel drive? 

[Rerouter]

Sorry, 1920, should not have gone off memory alone,

[janekm]

Didn't the watchmaking industry solve this a while ago with jewelled bearings?

Apparently they still need lubrication. Possibly to slow down the wear of the metal? The jewels can have Mohs hardness 9 after all...

[SeanB]

You can slow down the wear with jewelled bearings and ceramic shafting along with aluminia gears, but you will need to hermetically seal the whole movement in a dry atmosphere, preferably He, or Ar. That will last a century before wear gets too bad in a clock mechanism with a 20 second pendulum period. Driving the mechanical counters with a hour pulse and having a series of dials that move around without a Geneva to click over ( extra wear points) you likely could get close to a thousand years of operation out of it. That is doable with a solar panel to provide power and a large multilayer high voltage ceramic capacitor ( run at low voltage but using the high voltage rating to derate the stress on the insulation) to store power for night use.

Another would be to have the solar power only operate a counter mechanism to increment the day count, and simply use the electronics to implement a 12 hour delay between clock pulses.

[Marco]

Any other ideas?

I think some form of bistable electrowetting is the best bet.

http://www.adt-gmbh.com/en/technology/basics-patent-background.html

Or maybe float a dial in a liquid, centered with magnets and turned with electromagnets?

[mikeselectricstuff]

Didn't the watchmaking industry solve this a while ago with jewelled bearings?

Apparently they still need lubrication. Possibly to slow down the wear of the metal? The jewels can have Mohs hardness 9 after all...

If you do it as a clock where 24hrs = 120 years, wear won't be an issue

[ker2x]

They claim infinite shell life 
http://aquacellbattery.com/

No datasheet, of course 

[janekm]

They claim infinite shell life 
http://aquacellbattery.com/

No datasheet, of course 

Yeah, but only until activated so they won't deliver any power in the "infinite shelf life" state... After that they seem to be a regular zinc carbon battery, i.e. very crap (~10% per year self discharge...).

[FrankBuss]

And they don't even provide a capacity number (or I can't find it on their website). Probably because it is so bad compared to a standard battery.

[G7PSK]

Sorry, 1920, should not have gone off memory alone,

First Jeep was 1941.

[mechtheist]

Just an FYI some might find interesting: From wiki:  Longplayer is a piece of music that is designed to last for one thousand years. It started to play on January 1, 2000, and if all goes as planned, it will continue without repetition until December 31, 2999. It will restart on that date.

"Longplayer" is based on an existing piece of music, 20 minutes and 20 seconds in length, which is processed by computer using a simple algorithm. This gives a large number of variations, which, when played consecutively, gives a total expected runtime of 1000 years.

It plays at Bow Creek Lighthouse at Trinity Buoy Wharf, London.

using Tibetan singing bowls and gongs [not exactly a great way to go for 1000 years.]  Check out Tom Scott's take on it 

[janekm]

Just an FYI some might find interesting: From wiki:  Longplayer is a piece of music that is designed to last for one thousand years. It started to play on January 1, 2000, and if all goes as planned, it will continue without repetition until December 31, 2999. It will restart on that date.

"Longplayer" is based on an existing piece of music, 20 minutes and 20 seconds in length, which is processed by computer using a simple algorithm. This gives a large number of variations, which, when played consecutively, gives a total expected runtime of 1000 years.

It plays at Bow Creek Lighthouse at Trinity Buoy Wharf, London.

using Tibetan singing bowls and gongs [not exactly a great way to go for 1000 years.]  Check out Tom Scott's take on it 

Which appropriately is effectively a musical clock: http://longplayer.org/about/how-does-longplayer-work/

(though as far as I can tell the actual performing is done by humans?)

[miguelvp]

Just realized something, the energy density of methanol and other hydrocarbons and how a fuel cell could convert it to electricity little by little.

For example:
http://en.wikipedia.org/wiki/Direct_methanol_fuel_cell

But there are many fuel cell technologies and fuel combinations but that would be a good approach.
I wonder if using very low powered devices and eInk the butane in a disposable lighter might last 100 years for such a low powered device. Add some solar to assist and you are done.

[SeanB]

Problem then is the allowing of that horrible corrosive element Oxygen into the sealed enclosure. Fuel cells all degrade with time. Best would be around 3g of Americium and a few thermocouples ( peltier junctions are cheap these days) to convert the low grade heat to electric power. That at least will last the century or more required, and will fit in a small package and is relatively easy to shield with only a small amount of lead and Boron.

[miguelvp]

fuel cell related:

http://www.rsc.org/chemistryworld/News/2011/November/10111101.asp

not time analysis on how long it will stayed powered or decay.