Sizing Battery Backup: From Load Profile to Usable Capacity

Legacy context

The old industrial drip lines ran on a simple logic: steady pressure, measured flow, and a reserve tank for the moments when the main supply faltered. That same principle of anticipating a drop in power now applies to the gear we rely on daily. Whether it’s a timing system at a track meet or a sensor array on a practice field, the equipment doesn’t care about the weather forecast—it cares about consistent voltage.

That’s where battery backup sizing enters the conversation. It’s not about buying the biggest unit on the shelf. It’s about matching stored energy to the specific draw of your devices, the duration of a typical outage, and the recharge window you actually have. A scoreboard clock might need only a few hundred watt-hours, while a portable cooling unit for sideline gear demands far more.

The old drip line crews knew that oversizing wasted water and undersizing starved the crop. The math here is similar, just measured in amp-hours instead of gallons per minute. Getting the reserve right means the system keeps running when the grid stumbles—no drama, just continuity. That’s the bridge from those early irrigation days to today’s portable power planning.

The Load Inventory: Defining What Must Run

Battery backup sizing begins with a complete inventory of the loads that must be served during an outage. This is not the same as the total facility load; it is the subset of loads deemed critical for the backup duration. For each load, the engineer must record two values: the steady-state power draw in watts and the energy consumption over the intended backup period in watt-hours. The energy demand is the product of power and operating time, and it must account for duty cycle—the fraction of time each load actually operates. A pump that draws 1 kW but runs only 30 minutes per hour contributes 500 Wh per hour, not 1,000 Wh. The sum of these hourly energy contributions across all critical loads yields the daily energy demand that the battery bank must supply. This inventory should be reviewed periodically, as loads change with facility operations. The evidence notes that effectively using storage requires a control algorithm that takes into account historical PV output and load to predict optimum dispatch set points [1]. That algorithm is only as good as the load inventory feeding it.

Usable Capacity: Depth of Discharge and Temperature Derating

The rated capacity of a battery—typically expressed in amp-hours or kilowatt-hours—is not the usable capacity. Two derating factors apply. First, depth of discharge (DoD) limits how much of the rated capacity can be withdrawn without damaging the battery or shortening its service life. The allowable DoD is a function of chemistry, discussed below. Second, temperature derating reduces available capacity when the battery operates outside its nominal temperature range. Cold temperatures increase internal resistance and reduce usable capacity; high temperatures accelerate degradation. The System Advisor Model (SAM) used for PV performance analysis explicitly includes temperature derating and balance-of-system efficiency as inputs [3]. The same principle applies to battery sizing: the required nameplate capacity is the daily energy demand divided by the product of the allowable DoD and the temperature derating factor. For example, if the daily critical load is 10 kWh, the allowable DoD is 80 percent, and the temperature derating is 90 percent, the nameplate capacity must be at least 10 / (0.80 × 0.90) = 13.9 kWh. This calculation is straightforward, but it is frequently skipped in favor of oversizing, which carries its own cost penalty.

Chemistry Differences in Allowable Discharge Depth

Battery chemistry determines how deeply a bank can be discharged on a routine basis. Lead-acid batteries, including flooded and sealed types, typically allow a shallower DoD—often in the range of 50 percent—to preserve cycle life. Lithium-based chemistries generally permit deeper discharge, commonly 80 to 90 percent, with less impact on cycle life. This difference has a direct effect on bank size: for the same energy demand, a lead-acid bank must be roughly 60 to 80 percent larger in nameplate capacity than a lithium bank. The trade-off is not simply capacity, however. Lead-acid batteries have a lower upfront cost per kilowatt-hour of nameplate capacity, but their shorter cycle life and shallower DoD mean they must be replaced more frequently. Lithium batteries cost more per nameplate kilowatt-hour but deliver more usable energy over their lifetime. The engineer must weigh these factors against the backup duration requirement and the expected number of discharge events per year. The evidence does not specify exact DoD limits for any chemistry, so the designer should consult manufacturer datasheets and warranty terms for the specific product under consideration.

Inverter Ratings: A Separate Constraint

Energy capacity is only one constraint. The inverter must also be sized to handle the instantaneous power demand of the critical loads. Two ratings matter: continuous power and surge power. Continuous rating is the power the inverter can supply indefinitely; surge rating is the power it can supply for a short period, typically a few seconds, to start motors, compressors, or other inductive loads. A battery bank sized for energy may be perfectly adequate for a 10-hour outage, but if the inverter cannot deliver the starting current of a large motor, the system will fail at the moment of the outage. The inverter rating is therefore a separate constraint from energy capacity, and both must be checked. The evidence notes that inverter capacity is a key input to PV system performance modeling [3], and that inverter topology and rated capacity are important system description parameters [5]. For backup applications, the inverter must be selected to match the peak surge demand of the critical load list, not merely the average demand. If the surge demand exceeds the inverter's capability, the designer must either select a larger inverter or remove the offending load from the critical list.

Recharge Rate: Closing the Loop

A battery bank that is discharged must be recharged, and in a PV system the recharge source is the array. The recharge rate determines how quickly the bank returns to full state of charge, which in turn determines whether the system can handle consecutive outage days or a single long outage followed by a short window of sun. The array must be sized not only for the daily energy demand of the facility but also for the additional energy required to recharge the battery. If the array is undersized, the battery may never fully recharge between events, and the backup duration will degrade over successive days. The control system must decide when to charge the battery and when to serve loads directly from PV, a problem that combines deterministic and stochastic effects [2]. For example, charging the storage in advance of high real-time energy prices would be desirable, but prices are not known and must be forecast [2]. Similarly, the availability of PV energy depends on insolation, which is variable. The recharge rate is therefore not a fixed number but a design parameter that must be evaluated against the expected duty cycle of the facility. The evidence does not provide a specific recharge rate standard, so the engineer must model the system using tools such as SAM, which combines system description with environmental parameters to calculate predicted performance [3].

Practical Sizing Procedure

A practical sizing procedure follows these steps. First, build the critical load inventory with duty cycles and sum to daily energy demand. Second, select a battery chemistry and determine the allowable DoD and temperature derating from manufacturer data. Third, compute nameplate capacity as daily energy divided by the product of DoD and temperature derating. Fourth, verify that the inverter continuous and surge ratings exceed the peak and starting demands of the critical loads. Fifth, size the array to meet both the facility load and the recharge requirement, using a performance model that accounts for temperature and balance-of-system losses. Finally, review the control algorithm to ensure it can manage the charge and discharge decisions under real-time conditions. The evidence emphasizes that the key to optimization is determining when and how to manage loads and when to charge or discharge storage [2]. This is not a one-time calculation but an ongoing operational consideration.