Solar Batteries in California 2026: NEM 3.0 Sizing, Cost & Payback Guide
The honest version
Under NEM 3.0, batteries went from optional to essential. Solar-only systems still work but pay back in 12–15 years instead of 7–10 with a battery. Tesla Powerwall 3 ($13.5K–$15.5K installed) is the default; Enphase IQ Battery (modular 5kWh or 10kWh) is the alternative. Right size for a typical California household: 20–30 kWh of storage (two Powerwalls or three Enphase 10P). Skip the federal 25D credit — it terminated for 2026 installs — but check SGIP for state rebates if you’re in fire-prone or low-income tier.
Why batteries went from optional to essential under NEM 3.0
Pre-April 14, 2023, California solar customers operated under NEM 2.0: every kWh you exported to the grid earned you full retail credit ($0.30–$0.55/kWh in California). Solar-only systems were standard. You could overproduce in the day and bank credits to use at night. Storage was a nice-to-have, mostly for outage backup.
NEM 3.0 changed the economics overnight. Now every kWh you export earns you the Avoided Cost Calculator (ACC) rate — about $0.05–$0.10/kWh, roughly 25% of retail. But the rate you pay to import that same energy at night? Still $0.40–$0.55/kWh during peak hours. Exporting and re-buying became a structural loss.
The math forced a redesign. Instead of overproducing for export, NEM 3.0 systems aim to store and self-consume: capture daytime production in batteries, discharge during 4–9 PM peak hours when retail rates triple, avoid grid imports at the most expensive hours.
The dollar difference is real. A correctly sized battery saves a typical California household $2,500–$4,000 per year vs. solar-only on NEM 3.0. Over 25 years, that’s $60,000–$100,000 in lifetime savings. The battery cost ($25K-30K for a two-battery system) pays back in 7–12 years even without the (now-terminated) federal tax credit.
How a solar battery actually works
A residential solar battery is a lithium-ion storage unit with three modes:
- Daily cycling: Charges from solar during 10 AM–3 PM (peak production hours). Discharges during 4–9 PM (peak rate hours). Cycles once per day under normal use.
- Time-of-use optimization: Smart firmware shifts loads to align with rate plans. Batteries can also charge from grid during off-peak ($0.13/kWh) and discharge at peak ($0.55/kWh) on days solar is insufficient — though this is less common in California.
- Outage backup: When the grid goes down, the battery isolates the home (islanding) and runs essential or whole-home loads from stored energy plus continuing solar production. Most systems can run a typical home for 1–3 days off battery alone, indefinitely if there’s sun.
The control logic happens automatically — you don’t manually decide when to charge or discharge. The battery’s management system reads the rate plan, current solar production, current home loads, and grid status to optimize in real time.
The three brands you’ll encounter
For residential California installations in 2026, the market is dominated by three options:
Tesla Powerwall 3
Tesla’s current-generation residential battery, released in 2024. 13.5 kWh capacity per unit, 11.5 kW continuous AC output, integrated solar inverter (no separate string inverter needed), DC-coupled with solar for higher round-trip efficiency. Warrantied 10 years.
Best for: most installations where capacity and simplicity matter. Single-unit deployments cover small-to-medium homes; two-unit deployments cover larger homes including pool + EV.
Cost installed: $13,500–$15,500 per unit.
Enphase IQ Battery 5P / 10P
Modular battery sizing in 5 kWh (5P) or 10 kWh (10P) increments. Plays well with Enphase microinverters (already on the roof for many newer installations). Modular design lets you match capacity precisely — want 25 kWh? Five 5P units. 30 kWh? Three 10P units. Warrantied 15 years.
Best for: homes already on Enphase microinverters (seamless integration), homes wanting redundancy (one unit failing doesn’t take the whole system down), homes with very specific capacity targets.
Cost installed: $7,500–$9,000 per 5P unit; $11,000–$13,000 per 10P unit.
Sunrun-branded battery (typically Tesla Powerwall 3 underneath)
When you sign a Sunrun PPA, lease, or purchase, the battery hardware is usually a rebranded Powerwall 3 with Sunrun’s firmware and warranty wrapper. The economics are similar to a self-purchased Powerwall but with Sunrun’s 25-year warranty, maintenance, and monitoring included in the contract.
Best for: customers wanting a single-vendor contract that covers solar + battery + warranty + maintenance for 25 years.
Cost: bundled into total PPA or purchase price; not separately itemized.
Sizing the battery — how to pick the right kWh
The right battery size matches your peak-hour evening consumption plus a buffer for cloudy days and outage backup. Working math:
- Estimate your daily kWh use: Annual usage / 365. Typical California household: 12–18 kWh/day. Pool + EV households: 25–40 kWh/day.
- Estimate peak-hour use (4–9 PM): Typically 35–45% of daily total for a typical home. So 12 kWh/day household has ~5 kWh of peak-hour use; 30 kWh/day household has ~12 kWh peak-hour use.
- Add reserve for cloudy days + backup: Aim for batteries to cover at least 1.5x peak-hour use, so cloudy-day production gaps and outage backup are covered.
For a typical California household:
- 12–15 kWh/day usage: One Powerwall 3 (13.5 kWh) is borderline; two units (27 kWh) is comfortable. Or one 10P + one 5P Enphase (15 kWh).
- 15–20 kWh/day usage: Two Powerwall 3 (27 kWh) is the right answer. Or three 10P Enphase (30 kWh).
- 20+ kWh/day with pool, EV, or high cooling: Two-three Powerwall 3 (27–40 kWh). Larger Enphase configurations also work.
Undersizing is worse than oversizing. An undersized battery cycles to empty mid-evening and forces grid imports at the most expensive hours — defeating the purpose. Oversizing wastes capital but doesn’t hurt performance.
Common installer pattern: Some installers will size for one battery to keep the price down at the proposal stage, then upsell to two batteries during the home eval. The honest answer is: if your daily use exceeds 15 kWh or you have an EV/pool, two batteries is right from the start. Single-battery configurations on NEM 3.0 with that profile typically lose value to grid imports at peak hours within months of install.
Cost reality — equipment vs installed
The number you see online ($10,500 for a Powerwall 3) is the equipment cost, not installed cost. Installed cost includes labor, permits, electrical panel upgrades if needed, mounting hardware, and integration with the solar system.
| Battery | Equipment Cost | Installed Cost (CA) |
|---|---|---|
| Tesla Powerwall 3 (13.5 kWh) | ~$10,500 | $13,500–$15,500 |
| Enphase IQ Battery 5P (5 kWh) | ~$5,500 | $7,500–$9,000 |
| Enphase IQ Battery 10P (10 kWh) | ~$8,500 | $11,000–$13,000 |
| Two-Powerwall config (27 kWh) | ~$21,000 | $25,000–$30,000 |
| Three Enphase 10P config (30 kWh) | ~$25,500 | $33,000–$39,000 |
Installed cost premium over equipment cost: ~25–35%, depending on home wiring, panel age, distance from solar inverter, etc.
Battery payback math under NEM 3.0
For a typical California household with 15 kWh/day usage:
- Solar-only system (no battery): Production exports at $0.07/kWh ACC; imports at $0.50/kWh peak retail. Annual savings vs no-solar: $3,500–$5,500.
- Solar + 1 battery (13.5 kWh): 70–80% of production self-consumed. Annual savings vs no-solar: $5,500–$7,500. Battery delta vs no-battery: ~$2,000/year.
- Solar + 2 batteries (27 kWh): 90–95% of production self-consumed. Annual savings vs no-solar: $6,500–$8,500. Battery delta vs single-battery: ~$1,000–$2,000/year.
Battery cost payback: with $25,000–$30,000 in installed cost on a two-battery configuration adding $2,500–$4,000/year vs solar-only, payback runs 7–10 years. Combined with the 25-year solar life, the battery investment generates 15–18 years of pure savings after payback.
The federal tax credit reality for 2026
The federal Residential Clean Energy Credit (Section 25D), which previously gave 30% off batteries 3 kWh or larger, terminated for systems placed in service after December 31, 2025, under OBBBA (signed July 4, 2025). Battery purchases in 2026 do not qualify for the residential 25D credit.
Sunrun-owned PPA paths use a separate commercial credit Sunrun claims internally; their pricing reflects whatever credit they receive. But it’s not a credit the homeowner files on their personal taxes.
California’s Self-Generation Incentive Program (SGIP) still offers state rebates for batteries, especially for households in:
- Equity Resiliency tier: Up to $1.00/Wh for low-income customers in fire-prone areas
- Equity tier: Up to $0.85/Wh for low-income customers
- Standard tier: Smaller rebates that have been declining as funds deplete
Worth checking SGIP eligibility through PG&E or your installer before signing.
Backup power vs daily cycling — different value propositions
Customers often want batteries primarily for outage backup (PSPS events, wildfire-related grid shutdowns). The economic justification under NEM 3.0 is daily cycling, not backup.
Battery sized for backup only: 1 Powerwall (13.5 kWh) covers basic loads (refrigerator, lights, internet, basic outlets, one HVAC zone) for 1–2 days during an outage. ~$13.5K installed.
Battery sized for daily cycling on NEM 3.0: 2 Powerwalls (27 kWh) optimizes time-of-use savings AND provides 2–3 days of whole-home backup. ~$25–$30K installed.
The marginal cost from backup-only to daily-cycling is one extra battery (~$13K). The marginal benefit is $2,000–$4,000/year in savings AND much better outage coverage. Almost always worth the upgrade.
Common mistakes
From talking to homeowners after they’ve installed:
- Sizing for current use, not projected use. Customers add an EV or pool 18 months after install and discover their single battery is undersized. Plan for 5-year horizon, not current snapshot.
- Not pairing battery with smart panel. A SPAN, Schneider Square D Energy Center, or similar smart panel makes the battery vastly more effective by automating circuit-level load management. Adding $4–$6K of smart-panel hardware unlocks $500–$1,000/year of additional optimization on average.
- Buying battery without solar. Battery without solar is just an expensive UPS. The full economic case requires both.
- Skipping battery entirely on NEM 3.0. Some installers will sell solar-only because the system cost is lower and the deal closes faster. The customer’s 25-year savings are dramatically lower. Get a second opinion if your installer recommends solar-only on NEM 3.0 without explaining the trade-off.
- Overspending on three-battery configurations. Some installers push three-battery setups on customers with 12-15 kWh/day usage. The third battery contributes minimal additional savings vs the second; payback on the third unit can stretch past 15 years. Match the size to the actual load.
The honest bottom line for 2026
If you’re installing solar in California in 2026 under NEM 3.0, batteries are practically mandatory for the math to work. The economic analysis isn’t solar vs no-solar — it’s solar+battery vs no-solar, because solar-only on NEM 3.0 is structurally suboptimal.
Right size: two Powerwall 3 batteries (27 kWh) for most households. Three Enphase 10P units (30 kWh) for Enphase-aligned systems. Larger configurations only for very high consumption households (pool + EV + heat pump + large home).
Right cost expectation: $25–$30K installed for a two-battery configuration alongside solar. Payback in 7–10 years. 15–18 years of pure savings after that.
The decision isn’t whether to install a battery — it’s how big and which brand. We can run the actual math for your house if you upload your bill.
How big a battery do you actually need?
Upload your most recent PG&E bill. Within 24 hours I’ll send back the right battery size for your specific usage pattern, the projected payback, and which brand fits your situation. Free. No follow-up unless you ask.
Get My Free Audit →