Fluorescent lights hum in a repair bay as an installer taps a spec sheet. You want clear answers on kWh and kW before signing an order. This guide keeps choices simple and grounded in examples.

At a Glance

  • A common residential baseline pairs a 5 kW inverter with roughly 12 kWh of capacity for short backups.
  • Use the runtime rule: hours = capacity ÷ continuous power. Example calculation: 12 kWh ÷ 5 kW = 2.4 hours (example calculation).
  • Round-trip efficiency (RTE) drives usable energy. RTE is the percentage of energy returned after charging and discharging.
  • For cost context, roughly $0.18 per kWh applies as a retail example (for example).
  • Homes usually favor higher inverter power for short outages. Commercial sites often prefer larger kWh stores for demand shaving.

These bullets set practical anchors. The next sections explain how to apply them in design.

Sizing across home, commercial and utility scales

Match instantaneous power, in kilowatts, to your peak load. Match stored energy, in kilowatt-hours, to the hours you need. A simple rule links them. Capacity equals power multiplied by hours.

Start with your outage or dispatch target. Then scale numbers to your site. If a small shop needs four hours at a steady 50 kW, plan for about 200 kWh. That example scales both down and up. Choose hours that reflect your real usage or outage pattern.

Homes often see short spikes when HVAC or cooking starts. That pattern favors more inverter kW relative to stored kWh. Many installers start with a mid-sized inverter for essential loads. Then they add energy capacity as budget allows.

Commercial sites usually need longer duration to reduce monthly demand charges. Here, the right size often centers on the billing interval. Cutting a 15-minute peak can deliver outsized monthly savings. Test the window length that sets your bill before buying more energy capacity.

Chemistry matters by site. Lithium-ion fits tight spaces and daily cycling. Flow batteries fit long-duration needs and simple capacity scaling. Second-life modules can work where budgets are tight and space exists.

A quick metering exercise helps. Gather at least one week of one-minute load data. Peak points define needed kW. The area under your typical peak event helps define needed kWh.

In practice, teams that gather high-resolution load data decide faster. The data show whether to increase kW, kWh, or both. I have seen projects trim weeks off design by charting one-minute peaks.

Evaluating inverters and PCS specs: what matters and why

The power conversion system (PCS) is the inverter and its controls. First, match continuous AC power to sustained loads. Then add headroom for motor starts and compressor surges.

Specify short-term overload capability in seconds. Short overload capability prevents nuisance trips during appliance starts. For instance, a 7 kW inverter with a 150% ten-second overload handled an 10.5 kW start spike in testing.

Check waveform quality measured as Total Harmonic Distortion (THD). THD is the harmonic content percentage in the AC waveform. Lower THD reduces motor heating and false trips. Keep THD under 5% for sensitive electronics.

Look at inverter efficiency at partial loads, not only at peak. Losses at 25% or 50% load often dominate real performance. Verify islanding and blackstart functions before purchase. Islanding keeps power on when the grid fails. Blackstart means starting from batteries without grid power.

Use round-trip efficiency to set usable energy expectations. For planning examples, use an 85% RTE. Apply that assumption consistently across calculations.

Design the PCS so internal losses stay small versus usable energy. If the inverter is inefficient at your normal dispatch level, you lose effective kWh and value. In field commissioning, overload and THD specs often decide success more than peak ratings.

Costs, value streams and economic sizing

Decide objectives before running numbers. Backup, self-consumption, demand reduction, and grid services each change the preferred kW:kWh mix. Set one primary target for the first build. Add flexibility where future needs are likely.

Value streams convert delivered kWh into dollars. Delivered_kWh equals rated_capacity_kWh multiplied by the RTE assumption. Use the same RTE assumption consistently for fair comparisons. In this guide, that assumption is 85%.

Worked example: pick a 15 kWh battery. Delivered energy is roughly 12.75 kWh under the 85% RTE assumption. At approximately $0.18 per kWh (for example), that delivery offsets about $2.30 in energy costs per full cycle.

If the target is to dispatch 10 kWh daily, compute needed rated capacity. Required_charge_energy ≈ 10 ÷ 0.85 equals about 11.76 kWh (example calculation). With a planned Depth of Discharge (DoD) of approximately 90% (for example), rated capacity should be near 13.1 kWh. That sizing supports the daily delivery without over-stressing the pack.

Energy arbitrage focuses on buying low and using high. Success depends on daily price spreads and cycling limits. If spreads are thin, prioritize efficiency and low standby losses. Small efficiency gains compound over many cycles.

Demand-charge savings work differently from energy arbitrage. A right-sized inverter can shave a short peak and cut billed demand. Focus on the exact window that sets demand charges. If demand charges exceed roughly $10 per kW-month (for example), prioritize inverter power. That spending can beat adding more energy capacity.

Resilience value depends on outage load. Lower average load extends runtime far beyond nameplate hours. For example, a steady 1 kW load will stretch a modest pack for many hours. Owners report much longer runtimes when they reduce nonessential loads.

Model cycles per year for each value stream. Then model peak shaving outcomes. That side-by-side view shows where dollars really land. I have seen teams choose smaller, higher-power units for backup after modeling. They chose larger kWh packs for steady commercial savings.

Do not forget operating limits. Limit daily cycling if warranty terms cap full cycles. Add temperature constraints to your model. Cold or hot rooms change capacity and lifetime.

Include balance-of-system costs in your totals. Panels, wiring, disconnects, and commissioning all add up. Small projects feel these extras more. On larger projects, soft costs spread over more capacity.

Checklist for economic sizing:

  1. Define the primary objective: backup, arbitrage, or demand shaving.
  2. Choose a delivered kWh target and apply the 85% RTE and a planned 90% DoD to size rated capacity.
  3. Test short-window shaving impacts to compare added kW versus extra kWh.

Bottom Line

Choose kW and kWh based on your primary objective. Backup and solar self-consumption need more kW per kWh. Demand-charge reduction and duration shifting favor bigger kWh stores.

Specify inverter continuous power and credible short-term overload capability. Ensure THD and partial-load efficiency match your load expectations. Small efficiency gains cannot replace a PCS that runs loads without tripping.

Document the use case before quoting gear. Run simple formulas and size both sides of the system together. That approach avoids expensive rework at commissioning.

Practical next steps: measure high-resolution loads, pick an hours target, and multiply by required power. Use the delivered energy formula with your chosen RTE and DoD assumptions. Share results with vendors and demand clear overload and waveform specs. In my experience, teams that follow these steps see smoother commissioning and faster acceptance.