How Do You Design Supercapacitor Backup Power for a Product?

Detailed Explanation

A supercapacitor (also called an electric double-layer capacitor, or EDLC) stores energy electrostatically rather than through the chemical reactions a battery relies on — ions from the electrolyte adsorb onto a very high surface area electrode, giving capacitance values from single farads up to thousands of farads in a package not much larger than a small battery cell, at a rated voltage typically around 2.5–2.7 V per cell. That combination — high capacitance, low per-cell voltage, and genuinely capacitor-like (not battery-like) discharge behaviour — is what makes supercapacitor backup power design different from simply picking a small backup battery, and is the reason this page treats it as a distinct design problem from the lithium-ion battery content covered elsewhere on this site.

Typical applications are short-to-medium hold-up needs: keeping an RTC or SRAM alive through a brief power interruption, giving a system enough time to complete a graceful shutdown and save state after mains power is lost, or riding through a brownout that would otherwise reset the system. Supercapacitors are a poor fit for hold-up measured in months, which is where a small primary or rechargeable battery still wins.

Sizing: Usable Energy, Not Total Stored Energy

A capacitor's stored energy follows E = ½CV², identical in form to any other capacitor, just at a much larger C. The complication for backup-power sizing is that — unlike a battery, which holds a comparatively flat terminal voltage for most of its discharge — a supercapacitor's voltage drops roughly in proportion to the charge removed, so the voltage keeps falling for the entire discharge, not just at the end.

This matters because most loads (a regulator, an MCU, an RTC) have a minimum operating voltage, below which the circuit stops working even though the supercapacitor still holds some charge. The usable fraction of the total stored energy, discharging from a full voltage V_full down to a minimum usable voltage V_min, is:

Usable energy fraction = 1 − (V_min / V_full)²

For example, a system that can operate down to half its nominal supercapacitor voltage (V_min = 0.5 × V_full) recovers 1 − 0.25 = 75% of the total stored energy — but a system that needs the voltage to stay above 80% of nominal recovers only 1 − 0.64 = 36%. This is the single most common sizing mistake: calculating a required capacitance from total stored energy (½CV²) without accounting for how much of that energy is actually usable above the load's dropout voltage, then finding the real hold-up time is far shorter than the spreadsheet predicted. Where the load voltage window is narrow, a boost or buck-boost regulator between the supercapacitor and the load can recover most of the usable-energy fraction that a simple LDO would otherwise strand, at the cost of the regulator's own quiescent current and complexity.

ESR: Peak Current and Heating

Every supercapacitor has an equivalent series resistance (ESR) that behaves exactly as it does in any other capacitor — an instantaneous voltage drop of I × ESR appears at the terminals the moment a load step is applied, before the capacitor's bulk charge has time to respond, and continuous current through that resistance dissipates I² × ESR as heat inside the cell. Supercapacitor ESR is generally low relative to the very high currents these parts can supply for their size, which is part of the appeal for a load with a large instantaneous current spike (a radio transmit burst, a motor start, a flash write), but ESR varies substantially between products — from single-digit milliohms on parts optimised for high-power pulses to several ohms on small, low-cost cells optimised for energy density instead. Confirm ESR against the specific part's datasheet for the actual peak current your application needs, rather than assuming "supercapacitor" implies uniformly low ESR across every product in the category.

Leakage Current vs Battery Self-Discharge

A supercapacitor draws a continuous leakage current while charged, roughly proportional to its capacitance and voltage, that behaves more like a constant current draw than a battery's much slower percentage-of-capacity self-discharge. For a hold-up application measured in seconds to days, this leakage is usually a minor factor next to the load current itself. For anything approaching weeks to months of unpowered standby, leakage current becomes the dominant factor determining how long the backup actually lasts — and a small coin-cell battery, with its much lower self-discharge rate over long timescales, typically outperforms a supercapacitor of comparable physical size for that specific use case. Confirm the specific part's rated leakage current (usually specified after a stated stabilisation time, since leakage current is initially higher right after charging and settles over the following minutes to hours) against your product's actual unpowered-standby requirement before committing to a supercapacitor for anything beyond short-to-medium hold-up.

Series-Stack Balancing

Because a single supercapacitor cell is rated for only about 2.5–2.7 V, any system voltage above that requires stacking cells in series — and, exactly as with a series-connected lithium-ion pack, manufacturing tolerance in each cell's capacitance and leakage current means the cells will not naturally share the stack voltage equally. Left unaddressed, this drift can push an individual cell above its rated voltage even while the stack as a whole reads a safe total voltage, degrading or damaging that cell.

Two balancing approaches are used, chosen mainly by stack size and how much standing current loss is acceptable:

• Passive balancing — a resistor across each cell continuously bleeds a small current, keeping cells close to equal voltage over time. Simple and low-cost, appropriate for smaller stacks where the resistor's standing current draw is an acceptable tradeoff.
• Active balancing — a dedicated balancing IC actively redistributes charge between cells rather than simply bleeding it away, appropriate for larger stacks or applications where the passive approach's standing current loss isn't acceptable. This is the same balancing-topology choice made for multi-cell lithium packs — see what is a BMS? for the passive vs active balancing tradeoff in more depth.

Supercapacitor vs a Small Backup Battery

The decision between a supercapacitor and a small primary or rechargeable battery for backup power comes down to a small number of genuinely deciding factors, not a single "better" answer:

A supercapacitor is the better choice when the hold-up requirement is genuinely short, the circuit power-cycles frequently enough that a battery's finite cycle life becomes a real constraint, the deployment temperature range is extreme, or the safety profile of avoiding a lithium cell entirely matters more than energy density. A small battery remains the better choice whenever the required hold-up time is weeks or longer, or the load's energy requirement during backup exceeds what a reasonably sized supercapacitor can practically deliver.

Design Considerations

• Size from usable energy, not total stored energy — apply the 1 − (V_min/V_full)² fraction to any hold-up-time calculation, or the real backup duration will fall well short of a naive ½CV² estimate.
• Always current-limit the charge path. An uncharged supercapacitor's low ESR means a direct connection to a power rail draws a very large inrush current — use a series resistor, a constant-current charge circuit, or a dedicated charging IC.
• Check ESR against actual peak current needs, not a general assumption about the category — ESR varies by an order of magnitude or more between parts optimised for power density versus energy density.
• Don't oversize for "safety margin" without checking leakage current. A larger supercapacitor stores more energy but also typically leaks more — oversizing without checking the leakage-current tradeoff can make a marginal hold-up-time design worse, not better, for a low duty-cycle application.
• Confirm regulatory and mounting requirements for larger cells. Larger supercapacitors are physically bigger than an equivalent small battery and some require reflow-compatible or through-hole mounting rather than a coin-cell holder — plan enclosure and PCB footprint space early. Zeus Design designs backup power and power-path circuits, including supercapacitor and battery hold-up systems, as part of complete product electronics design.

Common Mistakes

• Sizing from total stored energy (½CV²) instead of usable energy above the load's dropout voltage — the most common supercapacitor backup-power sizing error, and the one most likely to produce a real hold-up time far shorter than calculated.
• Connecting an uncharged supercapacitor directly to a power rail with no inrush current limiting, risking an overcurrent trip or brownout at every power-up.
• Assuming low ESR is a given for "a supercapacitor" without checking the specific part's datasheet, then finding the voltage sags more than expected under a real load step.
• Choosing a supercapacitor for a hold-up requirement measured in months, where its continuous leakage current makes a small coin-cell or rechargeable battery a substantially better fit.
• Series-stacking supercapacitor cells with no balancing at all, risking one cell in the stack drifting above its rated voltage even while the stack's total voltage looks correct.