How to size your battery bank for off-grid

Right, here's what actually works based on living off-grid for the past three years.

Step 1: Calculate your daily usage
Track everything for a week. Fridge, lights, laptop, kettle — the lot. Most folk underestimate by about 30%. I use an energy monitor to get real numbers, not guesses.

Step 2: Work out your autonomy days
How many grey days can you handle before your solar stops producing? I aim for 3-4 days minimum, especially through winter. Scottish winters are rough. Adjust based on your location and weather patterns.

Formula: (Daily kWh × Autonomy days) ÷ 0.8 = Battery capacity needed

That 0.8 is your usable depth of discharge. Don't size to 100% — lithium hates it, and lead-acid will die young.

Step 3: Check what you can actually afford and fit
Victron's LiFePO₄ batteries are brilliant but pricey. I started with good quality lead-acid, upgraded to lithium last year. Both work if you size properly.

Example: My shepherds hut uses about 8 kWh/day. With 4 autonomy days, I need 40 kWh ÷ 0.8 = 50 kWh. I run two 5.12 kWh Victron batteries plus a smaller backup.

Step 4: Match your inverter and charger
Battery size means nothing if your inverter can't handle peak loads. Size it for your biggest simultaneous draw, not your total daily usage.

Pro tip: Start conservative. It's easier to add battery capacity later than regret oversizing your bank and watching it sit

Spot on. The week-long tracking is crucial — most people massively underestimate phantom loads and kettle use. I've got 10kWh usable and still catch myself thinking "surely I didn't use that much today."

One thing worth adding: factor in seasonal variation. Winter'll absolutely hammer your winter capacity with shorter days and higher loads (heating, lights on longer). I size for winter worst-case, then you're golden the rest of the year.

Also don't forget to build in headroom for the BMS and DoD limits. If you're running LiFePO4 like most of us, you might have 15kWh installed but only 12kWh usable depending on your settings. Victron's battery monitor makes this transparent at least.

@PeakVanLifer's method is solid though — beats the hell out of guessing or copying someone else's setup.

The week-long tracking is absolutely essential, though I'd push it to 10-14 days if you can manage it. Seasonal variation matters more than people realise — winter consumption on a narrowboat is genuinely different from summer, especially heating and lighting.

Where most folk go wrong is the Peukert effect. Your lithium or LiFePO4 bank won't deliver nameplate capacity at high discharge rates. I've got a 10kWh Victron setup and it behaves nothing like the spec sheet under heavy kettle/induction hob load.

Also worth factoring in: usable capacity. You'll want to size for 80% DoD absolute maximum, 50% if you're running lead-acid. A 10kWh system that's only reliably usable to 4kWh changes your sums entirely.

The other variable people miss is your charge source capacity. Battery size without adequate solar/wind/shore power means you're chasing your tail through winter. I'd rather have a smaller bank with robust charging than an oversized battery that never fully cycles.

Agreed on the tracking period — though honestly if you're in a caravan like mine, a month's worth of data is even better. Winter usage is mental compared to summer, especially with heating and longer dark hours.

One thing I'd add: don't just total it up and call it done. Look at your peak draw separately. I nearly sized my bank wrong because I was averaging 2kWh/day but my morning routine (kettle, shower pump, fridge compressor all going) pulled 4kW simultaneously. That's what your inverter and battery can actually handle in real time.

Also worth noting — most of us use way less once we're actually paying attention. First week I tracked, thought I needed 10kWh. Ended up sizing for 6kWh and haven't regretted it. Victron's battery monitor is brilliant for this if you've already got one installed.

@PeakVanLifer's got the fundamentals spot on, but I'd add something I learned the hard way in my conversion — seasonal thinking.

I tracked religiously for two weeks in summer, sized accordingly, then nearly ran flat every November. Winter changes everything: shorter days mean less solar gain, but heating draws spike. If you're planning for year-round off-grid living, you need data from your worst season, not your best.

Also worth noting — and this catches people out constantly — there's a difference between average daily use and peak demand. Your kettle might draw 2kW for five minutes. That matters when you're sizing inverter capacity, even if it's peanuts on your daily Wh budget.

I run a Victron system in my setup and the monitoring data revealed I was using 30% more in winter than my summer calculations suggested. Don't make that mistake. Track through at least one full seasonal cycle if you can manage it.

Spot on about the tracking period. I'd also factor in your usable capacity — most folk forget that LiFePO4 systems shouldn't drop below 20% to avoid premature degradation, and lead-acid is worse.

So if you're sizing a battery bank, actual usable capacity is roughly 80% of nameplate. I've got a 10kWh Victron LiFePO4 setup in my garden office, but I'm only drawing from about 8kWh realistically.

Worth considering your worst-case scenario too — cloudy spell in winter with minimal solar generation. I aim for 3-4 days of autonomy depending on the season. If you're grid-connected or have a backup gen, you can get away with less, but off-grid proper? You'll thank yourself for oversizing slightly rather than undersizing.

Spot on lads, though after years of night shifts I'd add: don't just track usage, track your mood when the battery's at 15% on a grey February afternoon — that's your real minimum capacity right there.

@ZFS_OffGrid a month's brilliant if you've got the patience, but honestly a week in winter tells you more than a month in summer when you're giddy on solar gains. My Victron setup's taught me that most folk massively overestimate winter generation and underestimate their kettle habits.

Also, @QuietTrekker's spot about usable capacity — my LiFePO4 doesn't actually give me what the nameplate promises when it's cold, so I sized up accordingly. Learnt that lesson the hard way with my motorhome last December.

@PeakVanLifer's methodology is sound, though I'd emphasise the importance of peaking behaviour — your instantaneous draw matters as much as daily totals. My garden office setup taught me this the hard way when I assumed a 5kWh bank would handle my work setup, only to find the inverter cutting out when the kettle, microwave, and laptop charger fired up simultaneously.

This is where your BMS and inverter specs become critical. A Victron MultiPlus 48/5000 handles my peaks, but paired with inadequate battery capacity it just delayed the inevitable.

Also worth noting: DoD strategy varies by chemistry. LiFePO4 systems like Fogstar can comfortably cycle to 90%+ DoD repeatedly, but lead-acid setups demand you size for 50% maximum usable capacity. That changes your sizing calculation substantially — you might need double the rated capacity to stay within safe operating parameters.

The seasonal variation @MarinePhil mentioned is critical too. January solar generation in the UK is roughly 25% of summer levels.

Brilliant thread, @PeakVanLifer. I've learned this the hard way with my setup — spent two years undersizing and it cost me dearly when winter hit.

The bit everyone glosses over is thermal losses in your battery management. I run Victron lithium with a Fogstar BMS, and the difference between tracking theoretical usage and actual draw with temperature compensation is substantial. In December, my usable capacity drops noticeably. Worth logging ambient temps alongside your consumption data.

Also, @RetiredNurse49's point about mood is genuinely important — it's not sarcasm. Once you've hit 20% state of charge twice, you'll redesign your system. That psychological buffer pushes most people toward 25-30% oversizing, which isn't wasteful; it's realistic.

I'd add: size for your worst week, not your average week. Cold snap, cloudy weather, unexpected visitor eating into your reserves. I discovered this in February when I'd calculated for autumn averages. Now I keep detailed logs of my lowest production days paired with highest usage days — they're rarely the same season, which catches people

Right, so I've been living in a converted Sprinter for two years and can confirm: the best battery calculator is desperation at 11pm when your fridge starts making "that sound."

@PeakVanLifer's got it spot on — track everything. I initially thought my setup was fine until I realised I was running a Victron inverter like

The peaking point @TracyAllen mentions is crucial — I learned this the hard way with my narrowboat setup. I'd calculated daily Wh fine, but didn't account for what happens when the kettle, water pump, and fridge compressor all kick in simultaneously. That's when I realised my Victron BMV was showing massive voltage sags.

What actually helped: I added a separate sheet tracking peak draw (not just daily average) and sized the battery to handle that worst-case moment, not just the energy budget. For a narrowboat, that's typically kettle + other essentials running together.

Also worth noting — @PeakVanLifer's tracking methodology is spot on, but don't just do one week. I'd do it across seasons if you can. Winter consumption looks completely different to summer (heating, less solar gain, etc.).

One thing the guides rarely mention: factor in peukert effect and usable capacity. Your 5kWh LiFePO4 isn't actually 5kWh if you're pulling hard. A Fogstar or Renogy spec sheet will tell you the usable percentage under load

Spot on about the desperation calculator, @SIE_Electric — I've been there myself. The thing everyone misses is usable capacity vs rated capacity.

With my LiFePO4 setup (200Ah Fogstar cells), I only count 80% as actually usable. That 20% buffer keeps the BMS happy and extends cell life massively. On paper it looks like you've got loads, in reality you don't.

Also worth noting: when you use power matters as much as how much. I've got a 3kW inverter but my morning routine (kettle, shower pump, espresso machine all running) taught me that peak draw is different from average draw. Sized my battery bank for the peaks, not the daily average, and it's made all the difference between smug and stressed.

The week of tracking is essential — but do it in winter. That's when you'll actually see what your fridge and heating demand. Summer numbers are deceptive.

Been through this dance twice now — once fitting out the motorhome, then again when we went full-time in the narrowboat. The thing that actually shifted my perspective was realising I was sizing for worst case when I should've been sizing for realistic worst case.

@SIE_Electric's spot about desperation is painfully accurate. But here's what I found: most people nail their daily average, then panic and double it. That's how you end up with a £6k LiFePO4 bank you'll never fully utilise.

What worked for us was tracking three weeks instead of one — you catch the outliers (guests arrive, heating goes mad in winter) without designing around freak scenarios. Then apply the usable capacity rule: if you've got 10kWh, you're really looking at 7-8kWh you can actually draw.

For the narrowboat specifically, I sized around a 3-day autonomy window because that's realistic between shore power visits. The motorhome gets two days because we move more frequently.

The real trick is being honest about what "off-grid" means for your setup. Living in London suburbs off-grid? Different beast to actually remote.

The thing that caught me out was oversizing without accounting for charge time. I've got about 10kWh of LiFePO4 paired with 4.8kW of solar, and even on decent winter days I wasn't topping up fast enough because I'd sized the battery assuming I could drain it completely every day.

What actually matters is the gap between your worst-case usage day and what your solar can realistically recover in that season. Summer? You're laughing. January in Scotland? Different story entirely.

Worth noting too — and @PeakVanLifer's right about this — the usable capacity is what counts. Victron's monitoring makes this dead obvious once you've got it installed. Most people spec their banks based on nameplate capacity then wonder why they're hitting low voltage cutoffs earlier than expected.

If you're running any serious loads (immersion heaters, power tools), factor in your inverter's surge capacity as well. Seen plenty of folk buy the battery math perfectly then trip out because their 3000W inverter couldn't handle the kettle and microwave simultaneously.

The critical bit everyone glosses over is the discharge rate vs. usable capacity trade-off. You can have 20kWh on paper, but if you're pulling 5kW continuously, you're looking at thermal stress and voltage sag that'll shut your inverter down before you've actually drained the pack.

I've got 15kWh of LiFePO4 here (Victron Smart LiFePO4 units), and I size usable capacity at about 80% of nameplate—so effectively 12kWh available. That's not being conservative; it's accounting for BMS protection curves and the fact that charging/discharging near the edges kills cycle life.

The other trap @NickHughes mentioned deserves more emphasis: if your solar array can't replenish what you've used plus cover today's load before dusk, you're perpetually chasing your tail. Winter in the UK means you might get 4-5 genuinely productive hours. Work backwards from that.

For EV charging specifically, I'd argue battery sizing is almost secondary to when you charge. Better to have modest storage and time charging to peak solar hours than massive batteries and charger undersizing.

What's your actual load profile looking like?