Hi,

what might be the expected lifespan of these in a backup application?

https://media.digikey.com/pdf/Data%20Sheets/TDK%20PDFs/B73180A0101M062_DS.pdf

Assuming a very low number of charge-discharge cycles, say 10, and
constant charging, are there any significant degradation mechanisms to
be taken into account? Given the datasheet doesn't mention that, what
would be your guess?

Best regards, Piotr

On Wed, 6 Apr 2022 08:00:44 +0200, Piotr Wyderski
<bombald@protonmail.com> wrote:

>Hi,
>
>what might be the expected lifespan of these in a backup application?
>
>https://media.digikey.com/pdf/Data%20Sheets/TDK%20PDFs/B73180A0101M062_DS.pdf
>
>Assuming a very low number of charge-discharge cycles, say 10, and
>constant charging, are there any significant degradation mechanisms to
>be taken into account? Given the datasheet doesn't mention that, what
>would be your guess?
>
> Best regards, Piotr

100 uAH? Why bother? You can get amp-hours of primary lithium that
will last decades. Or use a supercap.

The 1812 size is what might make it interesting, but just barely. May
as well use flash or eeprom.

--

I yam what I yam - Popeye

jlarkin@highlandsniptechnology.com wrote:

> 100 uAH? Why bother?

I have a number of 1uA-ballpark almost always-powered circuit that need
to operate without interruption. N*100h is plenty, especially that I can
add N chips like that in parallel/series, as needed. This chip is
otherwise tailor-made for me, but they don't bother to specify its
longevity.

> You can get amp-hours of primary lithium that
> will last decades.

Not really, the self discharge rate is too high for "decades". The best
thionyl chloride primary cells last as many as 25 years, then even they
are dead.

> Or use a supercap.

No hermetic variants available, all drying out. Wet tantalums cannot
even remotely approach that level of capacitance and they have terrible
leakage/self-discharge currents. Not an option.

> May as well use flash or eeprom.

The circuits are actually doing something, it's not storage backup.
I'd use FRAM/MRAM for that.

Best regards, Piotr

On Wed, 6 Apr 2022 20:53:35 +0200, Piotr Wyderski
<bombald@protonmail.com> wrote:

>jlarkin@highlandsniptechnology.com wrote:
>
>> 100 uAH? Why bother?
>
>I have a number of 1uA-ballpark almost always-powered circuit that need
>to operate without interruption. N*100h is plenty, especially that I can
>add N chips like that in parallel/series, as needed. This chip is
>otherwise tailor-made for me, but they don't bother to specify its
>longevity.
>
>> You can get amp-hours of primary lithium that
>> will last decades.
>
>Not really, the self discharge rate is too high for "decades". The best
>thionyl chloride primary cells last as many as 25 years, then even they
>are dead.

The Tadiran parts claim proven 40 year life.

>
>> Or use a supercap.
>
>No hermetic variants available, all drying out. Wet tantalums cannot
>even remotely approach that level of capacitance and they have terrible
>leakage/self-discharge currents. Not an option.
>
>> May as well use flash or eeprom.
>
>The circuits are actually doing something, it's not storage backup.
>I'd use FRAM/MRAM for that.
>
> Best regards, Piotr
--

If a man will begin with certainties, he shall end with doubts,
but if he will be content to begin with doubts he shall end in certainties.
Francis Bacon

John Larkin wrote:

> The Tadiran parts claim proven 40 year life.

Indeed, in their marketing materials. The actual datasheets of the XOL
line don't go that far. On top of that, they are bulky and solve a
different problem than mine: they provide energy even when it's not
needed, i.e. in a typical backup scenario. You have plenty of energy an
suddenly you have nothing. So the 100uAh capacity can actually be a lot,
provided you can depend on this capacity.

Best regards, Piotr

On Wednesday, April 6, 2022 at 12:21:35 PM UTC-7, John Larkin wrote:
> On Wed, 6 Apr 2022 20:53:35 +0200, Piotr Wyderski
> <bom...@protonmail.com> wrote:
>
> >jla...@highlandsniptechnology.com wrote:

> >> You can get amp-hours of primary lithium that
> >> will last decades.
> >
> >Not really, the self discharge rate is too high for "decades". The best
> >thionyl chloride primary cells last as many as 25 years, then even they
> >are dead.

> The Tadiran parts claim proven 40 year life.

Last year, I replaced a lithium cell (Tadiran) that
had a 1969 datecode. It failed some time in the last five years,
as a clock backup.

jlarkin@highlandsniptechnology.com wrote in
news:ai9r4h9ndlnhfkd056f8372i2b6huc0pod@4ax.com:

> On Wed, 6 Apr 2022 08:00:44 +0200, Piotr Wyderski
> <bombald@protonmail.com> wrote:
>
>>Hi,
>>
>>what might be the expected lifespan of these in a backup
>>application?
>>
>>https://media.digikey.com/pdf/Data%20Sheets/TDK%20PDFs/B73180A0101M
>>062_DS.pdf
>>
>>Assuming a very low number of charge-discharge cycles, say 10, and
>>constant charging, are there any significant degradation
>>mechanisms to be taken into account? Given the datasheet doesn't
>>mention that, what would be your guess?
>>
>> Best regards, Piotr
>
> 100 uAH? Why bother? You can get amp-hours of primary lithium that
> will last decades. Or use a supercap.
>
> The 1812 size is what might make it interesting, but just barely.
> May as well use flash or eeprom.
>
Sodium Ion is supposed to be the next battery technology.