Probably 20+ years for a lithium coin cell... basically the shelf life of the cell.  I have a card of 24 year old CR-2032's that are still above 3V, and no sign of leakage.

BTW, never handle a coin cell (particularly in watch applications) with your fingers... your grubby fingerprints are rather conductive and can discharge a cell surprisingly quickly.

Also note the clock chip has 512 bytes of RAM in it.

rate on a reasonable battery?

Hi

Since we don’t often need the smallest cell made and we’re probably
talking lifetime of the cell….. does 22 na vs 33 na matter?

…. hmmm ….

CR2032 ( which is the smallest I would use) is rated at 0.22 AH.

A nano amp for a year is about 8 uA hours a year.

So 30 na for 20 years is 0.005 AH

Indeed, the self discharge of the cell (or life or whatever you want to call it)
will probably get you before the 30 na drain will.

Based on the previous data on the chip, I think I would just run the crystal
all the time.

Bob

On Wed, 7 Mar 2018 08:27:00 -0500
Bob kb8tq kb8tq@n1k.org wrote:

Not really. It starts to matter when you are space limited and don't
have space for a CR2032.

At this level, though, every tiny bit of leakage matters. Finger prints,
dust, humidity, FR4... Going below 1µA in current consumption is like
going below 10^-12 in frequency stability, suddenly 1M is a low resistance.

The the watches (quartz with analog dials) I have run >5 years
on their batteries, an I am pretty sure they don't have a CR2032.
And at least one of them must have a TCXO.

		Attila Kinali

--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson

Hi

I’ve designed watch guts (long ago). It was at a time that you used
an analog (motor) movement if you wanted really low power. The CMOS
/ LCD’s of that era were power hogs by comparison.

What you can put in a small / thin  watch isn’t what you would use on a
RTC board. My suspicion is that the leakage from a number of sources will
indeed dominate the actual power consumption in a random build
sort of application. (At least compared to 10’s of nano amps of power
into the chip).

Bob

It gets even more amazing ...

"A 1.5 nW, 32.768 kHz XTAL Oscillator Operational From a 0.3 V Supply"
https://web.northeastern.edu/ecl/wp-content/uploads/2016/12/xtal.pdf

/tvb

Hi

Ok, so energy harvesting from Lazy Bob in his arm chair makes a button cell
look like a giant power source …..

You likely aren’t going to win any ADEV competitions with that oscillator. They
did go to a lot of effort to squeeze out that last nano watt.

Bob

On Wed, 7 Mar 2018 12:06:43 -0500
Bob kb8tq kb8tq@n1k.org wrote:

Giant is an understatement. For comparison, a single brain cell uses
approximately 0.5nW of power. A human body dissipates something around
100 to 200W of heat, constantly.... but this is going a bit too OT I guess :-)

			Attila Kinali

--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson