Ada’s Calculation Audit: Chemistry Leak Versus Load

Ada treats self-discharge as a parallel current before trusting a cell’s capacity number

ADA · CALCULATION AUDIT

Ada's Calculation Audit: Chemistry Leak Versus Load

The chapter compares two chemistries for the same 15 uA sensor and a 10-year target. A 2600 mAh Li-SOCl2 cell leaks only about 3 uA, so it clears the target with margin; a 2000 mAh Li-ion cell leaks about 55 uA — nearly four times the load — and fails at 3.3 years. This audit treats self-discharge as a parallel current to show chemistry leak versus load, and why capacity alone does not decide.

Companion to the chapter Energy-Aware Case Studies — every number here comes from that chapter.

Treat self-discharge as a parallel current before trusting the capacity number. The same 15 uA sensor can pass or fail on chemistry alone.

Chemistry leak versus load

• Li-SOCl2 leak: 2600 mAh x 0.01 = 26 mAh/year; 26 mAh / 8766 h = 0.002966 mA = 2.97 uA.
• Li-SOCl2 operating draw: 15 uA + 2.97 uA = 17.97 uA, rounded in the prose to about 18 uA.
• Usable capacity with 20% reserve: 2600 mAh x 0.80 = 2080 mAh; lifetime 2080 mAh / 0.017966 mA = 115774 h; 115774 h / 8766 h/year = 13.21 years.
• Li-ion leak: 2000 mAh x 0.02 = 40 mAh/month; 40 mAh / 730 h = 0.0548 mA = 54.8 uA.
• Li-ion total draw: 15 uA + 54.8 uA = 69.8 uA; even with 2000 mAh usable, 2000 / 0.0698 = 28653 h = 3.27 years.

The physics check is the result: capacity is only useful after the chemistry leak is subtracted from the lifetime budget. The Li-SOCl2 case survives the 10-year target with reserve; the Li-ion case fails before temperature and ageing penalties are even added.

Every number above is taken from this chapter's own worked example and re-derived step by step.