One of the quietest ways a battery sizing calculation goes wrong is by assuming a battery's amp-hour rating is a fixed pool of energy you can draw at any speed. It is not. The faster you pull current out of a lead-acid battery, the fewer usable amp-hours you actually get, and the effect can be large enough to break an autonomy estimate that looked perfectly reasonable on paper. The Peukert effect is the name for this rate-dependent capacity loss, and the Peukert exponent is the number that quantifies it. This page explains why the effect happens, how the C-rating on the datasheet encodes the rate the capacity was measured at, and why matching that rating to your real load rate is the difference between an honest and a fictional runtime.
Peukert Effect in one line: The Peukert effect is the fall in a lead-acid battery's usable capacity as the discharge current rises, so a battery delivers fewer amp-hours when drained quickly than when drained slowly. The Peukert exponent is a number, typically a little above one, that captures how steeply this happens for a given battery, with a higher exponent meaning a worse fall-off. It matters for sizing because amp-hour ratings are quoted at a specific slow rate, usually a twenty-hour or hundred-hour discharge, and if your actual load draws faster than that rating assumes, the true usable capacity is lower and your autonomy math is optimistic.
A lead-acid battery does not behave like a bucket of water that empties the same regardless of how fast you pour. When you draw current slowly, the chemical reaction has time to reach deep into the plates and use most of the active material, so you extract nearly all the theoretical capacity. When you draw current quickly, the reaction is confined to the plate surfaces, the electrolyte cannot diffuse fast enough to keep up, and the battery reaches its cutoff voltage while a substantial amount of usable material is still unreacted. The energy is not destroyed, but it is inaccessible at that rate, so the battery effectively delivers fewer amp-hours the harder you pull.
The Peukert effect is the empirical description of this behavior, and the Peukert exponent is how it is quantified. For an ideal battery the exponent would be one, meaning capacity is independent of rate, but real lead-acid batteries have an exponent somewhat above one, and the further above one it sits, the more sharply capacity collapses at high discharge rates. A battery with a low exponent holds its capacity well as the rate climbs, while one with a higher exponent loses a large fraction of its rated amp-hours under a heavy load. The exponent is a property of the specific battery, influenced by its construction and chemistry, and better batteries tend to have exponents closer to one.
Lithium chemistries are far less affected, which is worth knowing because the Peukert effect is often described as if it applies to all batteries. Lithium iron phosphate has an exponent very close to one, so its usable capacity stays nearly constant across a wide range of discharge rates, and the rate-derating that dominates lead-acid sizing is a minor correction for lithium. This is one of the practical advantages that makes lithium easier to size confidently: its nameplate capacity is close to what you actually get, whereas a lead-acid nameplate has to be interpreted through the lens of the rate at which you intend to use it.
Because usable capacity depends on discharge rate, a bare amp-hour number is meaningless without stating the rate it was measured at, and that is exactly what the C-rating on a datasheet does. A capacity quoted at C20 was measured by discharging the battery over twenty hours, while a C100 figure was measured over a slow hundred-hour discharge, and because slower is more favorable, the C100 number is always larger than the C20 number for the same battery. Some manufacturers prefer to advertise the flattering C100 rating, so two batteries that look different on paper may be closer than they appear once you compare them at the same rate.
The trap for field sizing is mixing the rating rate with the actual load rate. If a battery is rated at C100 but your site discharges it over more like ten or twenty hours, the usable capacity at your rate is meaningfully less than the C100 label, and an autonomy calculation that plugs in the C100 number will overstate how long the site runs. The mistake is subtle precisely because both numbers are honest: the datasheet is not lying, it is simply reporting capacity at a slow rate that your load does not match. The fix is to either use the rating measured at a rate close to your discharge, or to apply a Peukert correction to translate the quoted rating to your real rate.
This is one of the most commonly missed gotchas in remote power design because the discrepancy is invisible until the site underperforms. A design that pencils out with the nameplate capacity can go dark noticeably earlier than the paper autonomy predicted, and a tech troubleshooting it finds nothing broken because nothing is broken; the sizing simply used a rating measured at a slower rate than the load actually draws. Sizing against a C-rating that matches, or is corrected to, your genuine discharge rate is what closes the gap between the runtime you calculated and the runtime you get.
An honest autonomy estimate at a remote site starts by identifying the rate at which the load actually discharges the bank, then using a capacity figure valid at that rate rather than the most flattering number on the datasheet. For a typical low, steady SCADA load the discharge is slow, often stretching over many hours or days of autonomy, so the C-rate is gentle and the usable capacity sits close to the slower ratings. For a site with heavier or bursty loads the discharge is faster, the Peukert penalty is larger, and the derating from nameplate to usable capacity is something the design has to account for explicitly rather than hope away.
The reason this matters operationally is that the Peukert effect makes the whole autonomy chain sensitive to the load, not just the battery. If a site's load creeps up over time, perhaps because equipment was added, the discharge rate rises, the Peukert penalty grows, and the usable capacity shrinks faster than the extra load alone would suggest, so autonomy can degrade more than a simple energy balance predicts. This compounding is why a site that was comfortable can become marginal after a seemingly modest load addition, and why the real discharge rate deserves as much attention as the raw amp-hour totals when a site starts underperforming.
Remote monitoring is what makes the gap between rated and real capacity observable. A cloud SCADA platform such as Merobix trending battery voltage and current together lets an operator see the actual discharge rate a bank experiences and how steeply its voltage sags under that load, which is the visible signature of high-rate capacity loss. A bank that reaches its cutoff sooner than the paper autonomy promised, with a voltage that drops off faster than expected as the load runs, is telling you that its effective capacity at the real rate is below the nameplate. Seeing that in the trend lets a designer correct the sizing with a proper rate-aware figure rather than repeatedly chasing a site that keeps going dark earlier than its calculation said it should.
Only very slightly. Lithium iron phosphate has a Peukert exponent very close to one, so its usable capacity stays nearly constant across a wide range of discharge rates, and the rate-derating that dominates lead-acid sizing is a minor correction for lithium. The effect is significant mainly for lead-acid batteries, whose exponent sits enough above one that fast discharge noticeably reduces the amp-hours you actually get.
Both state the same battery's capacity, but measured at different discharge rates: C20 is measured over a twenty-hour discharge and C100 over a slower hundred-hour discharge. Because a slower discharge is more favorable, the C100 figure is always the larger of the two for the same battery. Mixing these up in a sizing calculation is a common mistake, since using a C100 number for a load that actually discharges the bank in far less time overstates the usable capacity.
It makes the usable capacity depend on how fast you discharge, so plugging a slow-rate amp-hour rating into an autonomy calculation for a faster-discharging load overstates the runtime. The battery reaches its cutoff voltage while unreacted material remains, delivering fewer amp-hours than the label suggests. The fix is to size against a capacity figure measured at, or corrected to, the rate your load actually draws, which keeps the calculated runtime close to what the site delivers.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.