# How long will a DS1307 RTC run on a CR2032 coin cell battery?

I have a homemade clock based on an atmega328 and a DS1307 Real Time Clock. When off, the DS1307 is powered by a 3V CR2032 coin cell, as usual.

If I leave the clock off most of the time, how long should I expect the coin cell to last?

I'm aware of the formulas for calculating the battery discharge time, but I'm probably failing at applying them right (I think). According to my calculations, it should last approximately 50 years based on 210mAh capacity of the battery and 500nA of current draw from the datasheet. But the current draw is so low that there must be other issues dominating the battery discharge rate.

So, does anyone have any better figures than I have? Any real experience? Did anyone's battery die before 10 years?

• possible duplicate of How to Calculate the time of Charging and Discharging of battery? Oct 18, 2013 at 17:33
• The datasheet says: "A lithium battery with 48mAh or greater will back up the DS1307 for more than 10 years in the absence of power at +25°C." I have a DS1307 with a CR2032 battery used infrequently, still "reasonably" accurate after three years. (The DS1307 isn't all that accurate to begin with!) Oct 18, 2013 at 17:35
• @Daniel Grillo: I was trying to apply the formulas, but was failing miserably at it (I think). According to my calculations, it should last 50 years based on 210mAh capacity of the battery and 500nA of current draw from the datasheet. But the current draw is so low that there must be other issues dominating the battery discharge. I think I should have added this to the question... Oct 18, 2013 at 17:42
• @Ricardo battery self discharge, battery chemistry life, and minimum voltage for operation of the ds1307. Oct 18, 2013 at 18:21
• @Ricardo that calculation is only for a perfect battery. A typical cr2032 has a ten year shelf life at best. You won't get 50 years out of it. But CR2032s are so common and mass produced you won't get a better priced lower sized battery either. Oct 18, 2013 at 19:32

There are few factors:

1) As it was mentioned - battery self-discharge. I've checked few datasheets and it's 1-2%/year at 20-23C. If your device will stay in hot place (70C under direct sunlight for example) - it will self-discharge much much faster. Also, if device will be frozen accidentally - battery might be damaged and loose capacity.

2) PCB leakage - can reach hundreds of nA in case of severe surface contamination in wet conditions. If you manufacture PCB in good place, clean & completely dry it, and then cover it all (including chips) with some protective layer (polyurhetane or something) you can virtually guarantee no leakage through device life.

So, if leakage won't be an issue, no extreme thermal conditions - device will likely work for about 20-30 years.

• 70°C is 158°F. What kind of sunlight are you under? Oct 18, 2013 at 21:02
• That's right, dark surface under direct summer sunlight could reach that temperature. Oct 18, 2013 at 21:13
• @Passerby Outdoor temperature in the shade hits 50 degrees Centigrade in many parts of the world, so an additional 20 degrees due to a dark casing is not entirely out of the realms of feasibility. I'd have bet on a more commonplace 60 rather than 70 degrees, though, at least here in India. Oct 19, 2013 at 6:13
• Finally, park a car in sunlight. It's easy to get 30K more inside than outside, and ambient temperatures of 30-40°C are not uncommon. Mar 26, 2015 at 14:04
• Jack Ganssle has a great write-up: ganssle.com/reports/ultra-low-power-design.html#cr2032behavior Jun 15, 2016 at 12:13