How to properly size a solar + battery system to charge your EV — a real guide
By SolarSarah·5 replies·129 views
"How much solar do I need to charge my EV?" is the most common question I get. The answer is almost always "less than you think," but the right answer requires a few numbers. Let me walk through how to actually calculate this.
Step 1: Know your EV's consumption
Find your car's real-world efficiency in miles per kWh (not the EPA estimate — check your car's lifetime average in the energy screen). Common ranges:
- Economy EVs (Leaf, Bolt): 3.5–4.5 mi/kWh
- Midsize (Model 3 SR, Ioniq 5): 3.5–4.0 mi/kWh
- Larger/AWD (Model Y, Mach-E): 2.8–3.5 mi/kWh
- Trucks (F-150 Lightning, Cybertruck): 1.6–2.2 mi/kWh
Step 2: Calculate your daily charging need
If you drive 37 miles/day (national average) in a Model Y at 3.2 mi/kWh, you need about 11.6 kWh per day to replace what you use. Add 10–15% for charging losses (heat, AC-to-DC onboard conversion) and round up: call it 13 kWh/day.
Step 3: Know your sun hours
"Peak sun hours" is the key metric — it's not total daylight, it's equivalent hours at 1,000 W/m² irradiance. Varies significantly by location:
- Phoenix: ~5.5–6.5 PSH
- Denver: ~4.5–5.5 PSH
- Chicago: ~3.5–4.5 PSH
- Seattle: ~3.5–4.0 PSH
Look up your specific location on the NREL PVWatts calculator — it uses actual historical weather data.
Step 4: Size your array
Needed array size = Daily kWh need ÷ PSH × efficiency factor (typically 0.8 for system losses).
For our example: 13 kWh ÷ 4.5 PSH ÷ 0.8 = 3.6 kW of panels.
A 4kW system (8–10 standard panels at modern 400–500W ratings) will comfortably cover average EV charging for most drives in a mid-latitude location.
Step 5: Do you need batteries?
This is where it gets interesting. If you can charge during the day (work from home, Level 1 at home during daylight), you don't need batteries — just direct solar-to-charge. But if you charge at night, you need storage to bank the day's solar.
Rule of thumb: size your battery at 1–1.5× your daily charging need to handle a cloudy day. For our example: 13–20 kWh of storage.
The economics
A 4kW solar array + 15kWh battery system runs roughly $12,000–18,000 installed (post-federal tax credit, based on ~$2.50–$3.50/W for panels plus ~$500–$800/kWh installed for battery). At the 2024 national average of ~$0.165/kWh, charging 13 kWh/day saves about $783/year. Payback on EV charging alone: roughly 15–23 years. That's not a great standalone investment for EV charging alone.
But here's what changes the math: if you're also offsetting household electricity (which most people are), the system economics improve dramatically. Most homeowners with solar see 70–80% of their bill offset, which often means payback of 7–10 years on the whole system.
What size system are you running, and is your EV charging a significant part of it? I'm curious about people who sized specifically around EV charging vs. those who already had solar and just added EV charging.
Running a 6.4kW array (16 × 400W panels) in Denver with a 20kWh LFP battery. My Chevy Bolt uses about 9kWh/day average. The solar more than covers the Bolt plus the house — I export net 8–12 kWh/day in summer. In winter with snow loading and shorter days, I'm roughly net zero, pulling from the grid on extended overcast stretches.
The battery was absolutely worth it for time-of-use arbitrage independent of the EV. Xcel Energy Colorado's current TOU schedule runs about $0.21/kWh on-peak (5pm–9pm) vs $0.08/kWh off-peak — they shifted the peak window in 2025 away from the old 3pm–8pm window. The battery charges during the day on cheap solar and covers peak hours. That spread alone covers a meaningful chunk of the battery's cost over time.
Great guide. One thing I'd add for off-grid people: add a significant margin to the battery sizing. Grid-tied people can use the grid as a backstop on bad weeks. Off-grid, you're relying entirely on your stored energy and how much your panels can produce on a cloudy January day.
I run 9kWh of panel capacity and 30kWh of battery for my cabin + EV + well pump + refrigeration. It's bigger than the math suggests "necessary" but it means I haven't had a brown-out in 18 months even during a 5-day cloudy stretch in December.
What inverter are you recommending for setups like this? I've been looking at the Victron Multiplus and the SolarEdge StorEdge. Seems like Victron is the DIY community favorite but SolarEdge is what installers push. Is there a real difference beyond price?
@packrat Victron is hands-down the DIY/enthusiast choice and for good reason: open protocol (Modbus, CAN, VEConfigure), massive community, repairability, and modular expansion. If you ever want to expand or tweak your system, Victron lets you. SolarEdge StorEdge is a good integrated system for a set-it-and-forget-it homeowner who wants an installer to handle everything, but it's a closed ecosystem — you're dependent on SolarEdge's app, their monitoring, and their support. I've seen homeowners frustrated when SolarEdge's cloud monitoring had outages and they couldn't see their own system data.
For anyone doing it themselves: Victron. For someone who just wants a contractor to install and manage: SolarEdge is fine.
The step on efficiency factor (0.8) is doing a lot of work in that formula and I want to expand on it. That factor accounts for: inverter efficiency (~96%), wiring losses (~2%), temperature derating (panels produce less in heat than their rated power — relevant in Phoenix), soiling loss (dust, bird droppings — ~3–5%), and mismatch losses if panels aren't perfectly oriented. In practice I model 0.78 for ground-mounted systems in hot climates and 0.82 for well-oriented rooftop in temperate climates. Small difference but worth knowing why the factor exists.