Long-lasting battery

do or what your goals are.

So the requirement is 4 uW for 2 weeks or 1.344 mWHr. So round up to 2 mWH r.

ears is too short.

l be exceedingly hard to come by.

ng a turbine?

I'm pretty sure that's not accurate. There are clocks that have been runni ng for longer than 40 years... mechanical clocks. You don't actually say a nything about being unattended, but I assume that is the case. The 10,000 year clock requires winding by a human.

But the Atmos clock does not.

echargeable?

You also don't say how often it will need to supply the 2mWHr of electricit y. That's why I can't agree that a mechanical approach is out of the realm of possibility. If it will be used once a month for two weeks, that might be a wear problem, but even then probably not. The turbines in the Hoover Dam worked for 40 years before being upgraded. I'm pretty sure a 4 uW gen erator of some sort can run for 40 years.

A TEC would not need much temperature difference to generate 4 uW. I guess the trick is preserving the heat in a reservoir for two weeks. Perhaps a chemical fuel that can be released gradually? A reliable slow release coul d depend on diffusion I expect. But if you need to turn it on and off as n eeded that might not be so good.

I'm thinking a clock works would be practical. But I don't have the experi ence to know how best turn that into electrical energy. I just know 2 mWHr isn't very much. A literal clock could do it perhaps. Not that this woul d be what you are looking for, but it certainly is possible.

--

  Rick C. 

  + Get 1,000 miles of free Supercharging 
  + Tesla referral code - https://ts.la/richard11209

Dunno about how one could go about figuring "battery life" or self discharge rate.

But have a 30 year old (appx) Sanyo Cadnica Extra KR-5000DEL

1.2V,5000mAh cell that measures 0.66V open circuit.

No date designation.

I have been keeping it as a curiosity.

Thanx.

I have a Tadarin 3v Li battery and a 1M resistor and a green LED on the shelf near my bed. It helps navigate in the dark. It's been on for maybe 8 years now.

In an emergency, a strategically placed tongue can shunt the resistor and make a usable hands-free flashlight.

--

John Larkin         Highland Technology, Inc   trk 

The cork popped merrily, and Lord Peter rose to his feet.   
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"

Supercaps dry out, or spoil by internal chemical reaction in high humidty; batteries dry out or leak; and so do electrolytic caps.

On the bright side, I calculate a 1uF polypropylene cap charged to

4kV would cover you energy needs perfectly! Now all you need is a 4kV to 3.3V @ 2uA converter...

Grins, James Arthur

No need to grin, this is becoming increasingly clear to me that only the HV scheme would match my expectations. I was thinking more about a 50uF energised to 600V than 1uF at 4kV, but the principle remains the same.

Evans claims unlimited shelf life and only 25 years when charged, so three options come to mind:

1. An electromechanical converter based on this idea:

Way too crazy, but it works.

1. Keeping the energy in the PP cap when there is a surplus of it and then blasting it in one shot to a supercap when there is a power loss.

2b: Secondary side-driven hysteretic converter and multiple charging bursts. With the recently discovered 150nA comparator, it might even work.

1. Multiple supercaps, of which only one is charged. Then cycling through them in a round-robin fashion.

Best regards, Piotr

So you want to ring a bell at subaudible levels??? I hope you realize the power drain is probably much below even your needs.

I thought the problem with the supercap was it might not survive for 40 years?

So supercaps will last for 40 years if you don't use them? I haven't see any indication of the duty cycle. I thought it was rather low. That's why I think electromechanical is a good approach.

--

  Rick C. 

  -- Get 1,000 miles of free Supercharging 
  -- Tesla referral code - https://ts.la/richard11209

Don't know in general, but this is precisely what Evans claims:

But still it is only 1F max.

Best regards, Piotr

"25 years", that seems pretty clear to me.

--

  Rick C. 

  -+ Get 1,000 miles of free Supercharging 
  -+ Tesla referral code - https://ts.la/richard11209

"All Capattery series capacitors have an unlimited shelf life" is even clearer.

And there is this general rule of thumb saying that decreasing the supercapacitor capacitor voltage by 10% doubles its lifetime. Some (electro)chemical processes must be going on slowly when the part is energised and CO2 is one of the products:

"The Capattery improves upon previous double layer capacitors because of its patented Permselective valve, which allows the escape of CO2 generated by all double layer capacitors, while it maintains its sealed construction."

Best regards, Piotr

Valves don't sound very "unlimited life" to me. Ask them if they give your money back if it only lasts 39 years. All 100,000 units!

You still have not indicated the duty cycle you expect (maybe I missed it). Will this thing need to run just once in 40 years or be a monthly occurrence?

--

  Rick C. 

  +- Get 1,000 miles of free Supercharging 
  +- Tesla referral code - https://ts.la/richard11209

Sure, but it is a top-notch stuff. Some are not even allowed for export from the US. If these guys claim that 25 years is "limited", then their understanding of "unlimited" must imply quite a ridiculous value. The cheap/regular competitors specify several kilohours of useful life (several weeks!) and ~5 years of the shelf life.

"These hybrid capacitors, known as EVANSCAPS, provide significant savings of space and weight compared to other capacitor technologies, in a high reliability ruggedized package providing MTBF > 5 million hours."

Merely 570 years. Looks sufficiently "unlimited" to me.

It's just a backup power source for the critical circuitry. Assume month of charging followed by a month of discharging. It is the absolute upper limit, way beyond the requirements of the most realistic scenarios.

Best regards, Piotr

That doesn't sound consistent. MTBF is measured in operation. How can it be 570 years when they only claim 25 years?

There is also the 100,000 cycles...

I'm just asking what the design specs are. So 6 x 40 years is well below 100,000 cycles.

I thought that earlier you had rejected supercaps because they could not survive for 40 years under power, now you seem to be saying they will. Ok, good enough for me.

--

  Rick C. 

  ++ Get 1,000 miles of free Supercharging 
  ++ Tesla referral code - https://ts.la/richard11209

A burp-mode nanopower 1kV to 3.3V converter could be done with just a twist or two on the SMPS you recently finished.

If the high voltage is known, and the 3.3V capacitor value and voltage are known, then you'll know what charge time or current to charge your known inductor to in order to yield the desired output voltage increment at the 3.3V output capacitor.

A uC-based device could calculate it all for you, or a uC or analog control circuit determine the correct increment empirically.

Disconnect the HV cap with a relay when storing energy, switch the inductor through a SiC FET and burp once each time you need power.

/ K1 .----o o-------. | |_

--- 5uF/1kV _)|| _.-->|---+---> +3.3V

--- C.hv _)||(_ | | _)|| | --- C3.3V

--- | --- --- ||--' | ||

This is the solution I've been holding in my head as read the thread, If th e supercaps have "unlimited" shelf life when discharged, this could be a si mple as running on one for 20 years, then switching to a second one (maybe even leaving the first in place) for the next 20 years. Not particularly co mplicated, I would think.

Also, it would be useful to know the operating environment. Is this device on an orbiting satellite? On the moon? On mars? Down an oil well?