How to design a complete off-grid system

Spot on about the audit. I've been caught out twice—once on the boat thinking 12V fridge usage was negligible, turned out to be a proper power drain.

The bit nobody mentions: factor in vampire loads. Inverter standby, solar controller, monitoring kit all nibble away. On my tiny house setup I reckon about 2-3% daily loss just from stuff being switched on.

Also worth doing your audit across different seasons if you can. Winter usage on the boat is wildly different from summer—heating, shorter daylight hours, less solar generation. Designed my system around worst-case (Feb basically) and now I've got comfortable headroom rather than constantly sweating the battery voltage.

Spreadsheet method works dead well. Columns for each appliance, watts, hours per day. Chuck in some contingency too—real-world always uses more than theory.

Spot on thread. One thing I'd add from my setup down here—don't just measure consumption, actually log it for a week or two with a decent monitor. I used a Victron BMV and was shocked how much phantom draw stuff has, especially when you think everything's "off."

Also worth noting the seasonal swing. My winter consumption is nearly double summer once heating comes into play. Means you either oversize the whole system or accept that January-February you're running tighter margins. Both valid approaches depending on your priorities.

@CaddyProject's point about the 12V fridge is dead right—those things are silent energy vampires. If you're borderline on battery capacity, consider a compressor model or hybrid setup instead.

The other critical bit: know your peak loads, not just daily totals. Kettle + microwave + heater all going at once? That's what sizes your inverter, not your daily average. Caught me out initially when I spec'd too small an inverter thinking I was being efficient.

The audit's critical, but I'd push back slightly on one thing—don't get paralysed by perfection. I've seen folks measure every milliamp for months when a rough week's logging (as @DownsExplorer said) gives you 80% of what you need.

What actually matters: identify your peak loads separately from average consumption. A 3kW kettle for 10 minutes daily barely moves your daily Wh figure, but it'll size your inverter and battery. I made this mistake on my static caravan—undersized the inverter because I was averaging consumption, then nearly torched it first time someone boiled water while the immersion heater was on.

Second thing: build in 30% headroom for the stuff you'll inevitably add. Caravan heating, workshop compressor, electric lawn mower—always happens.

Once you've got those numbers, then you can properly spec battery capacity (usable Ah, not total), solar array size for your location, and charge controller ratings. Getting consumption wrong cascades into everything else being wrong.

The consumption audit is absolutely crucial—I've made this mistake twice on the boat alone. What's worth noting is that you need to log under realistic conditions, not just your intended usage patterns. I ran my first audit over a calm summer weekend; turned out winter cruising with heating and navigation gear running continuously painted a very different picture.

One thing @DownsExplorer's mentioned about week-long logging really matters. Your 12V battery monitor (I use a Victron BMV-712) becomes essential here. Plot the data and look for your actual peak draw, not the average—that's what sizes your battery bank and inverter properly.

Also worth separating consumption by voltage tier: your 12V essentials (fridge, navigation, comms) need different consideration than 230V loads you might run occasionally through an inverter. I've seen people massively overspec their entire system because they bundled kettles and microwaves into baseline calculations.

The mains equivalent method helps too—if your caravan runs 3kW peak demand, you're not sizing a 5kW inverter based on that alone; you account for surge and your battery

@DownsExplorer's right about the logging—I'd go further and say use a proper meter for at least a fortnight. A cheap kill-a-watt from Amazon (£15-20) gets you baseline figures, but you'll spot the oddities: that fridge drawing 40W constantly, or the water pump cycling at 3am.

The bit worth emphasising though is that consumption patterns matter as much as totals. I've got my narrowboat down to 40kWh/month average, but it swings wildly—winter heating adds 15-20kWh easy, summer's half that. That variation directly affects your battery size and generation capacity.

Also worth noting: measure what you're actually running, not what the spec sheet says. My Victron inverter idles at 15W—negligible on shore power, but on batteries that's 360Wh daily. My mate's Fogstar fridge pulls 80W under load, 5W standby. Those parasitic draws compound.

The real gotcha is loads you forget about—water pump, router, monitoring kit, chargers

Great thread, @BayTim. I'd add one thing that caught me out initially—seasonal variation matters more than you'd think.

When I first sized my system, I based it on summer consumption and nearly had a nightmare come November. Winter means shorter daylight, lower panel angles if you're fixed-mount, and increased heating demands. I'd suggest doing your audit across at least two seasons if possible, or at minimum factor in a 30-40% increase for winter months.

Also worth mentioning—don't just measure the big stuff like heating and pumps. Those phantom draws from chargers, controllers, and monitoring kit genuinely add up. I was shocked how much my system was bleeding just keeping things on standby.

@RetiredPlumber's spot on about avoiding analysis paralysis though. You'll never get it perfect, and that's fine. Start conservative, monitor everything meticulously once it's live, then tweak. I've adjusted my system three times now based on real-world data, and that's been far more useful than endless pre-build calculations.

Spot on about seasonal variation, @SarahLewis—winter consumption on my narrowboat is nearly double summer because of heating and shorter daylight. That's when folk realise their 5kW system isn't enough.

The logging point @RustySpanner mentioned is worth doing properly. I used a Victron Energy Meter for a month across different seasons before sizing anything. Real game-changer—showed me my fridge was pulling way more than the spec sheet suggested, and standby loads I'd completely missed.

One thing I'd add: factor in future load creep. You'll always add another device or upgrade something. Built my narrowboat system with 30% headroom and glad I did—added a small water heater last year without needing to overhaul everything.

Also depends what you're designing for. Emergency backup's totally different from full-time living. A 3-day blackout setup won't cut it for winter off-grid, but loads of people conflate the two.

@BayTim's got the fundamentals spot on, but I'd add something that nearly caught me out on my van conversion: don't just measure consumption—measure when you use it.

I logged power religiously for a month, got my daily average, sized everything accordingly. First winter trip? Nearly flat batteries by mid-afternoon because I'd sized for average usage spread evenly across daylight hours. Reality was different—heating loads clustered around dawn and dusk, solar generation peaked midday when I wasn't actually drawing much.

What saved me was tracking consumption patterns, not just totals. That's when I realised I needed either more battery capacity or different usage habits (spoiler: I did both). Added a second Fogstar lithium and adjusted when I ran the heating.

@SarahLewis and @PanelGraham are absolutely right about seasonal variation, but the timing of that variation matters as much as the magnitude. A spreadsheet with hourly breakdown costs nothing and reveals whether you're looking at a capacity problem or a power-matching problem—they need different solutions entirely.

Learned that the hard way at a Scottish campsite in January.

Spot on about the seasonal stuff. I learned this the hard way with my garden office setup—thought I'd sized everything fine until December rolled around and the heating load absolutely murdered my battery. Now I design for winter worst-case and call the summer surplus a bonus.

Also worth mentioning: don't just calculate your average consumption, look at your peak draw too. That kettle or power tool will spike your inverter requirements way beyond what you'd expect from the daily kWh figures. Caught me out when I added a small workshop to the setup.

One more thing—build in buffer capacity. Sounds wasteful but it's genuinely the difference between a system that works and one that's constantly limping along. I oversized my Victron battery by about 20% and it's transformed the whole experience. Less cycling, better longevity, and you're not living on a knife-edge if consumption runs slightly high.

Been trying to get my head round this for my shepherd's hut build, and I reckon I'm overthinking the consumption side. When @BayTim says "work backwards," does that mean I should be logging actual usage over a few weeks before sizing anything? Or is there a decent rule of thumb for a small permanent dwelling?

I'm particularly curious about the seasonal bit everyone's mentioning—I'm in Scotland so winters are brutal. Should I be designing for worst-case (December consumption) or some kind of average? Feels like sizing for peak winter would be massively expensive, but undersizing seems daft too.

Also, @MarinePhil—you mentioned getting caught out on your van. What specifically went wrong with the initial sizing? Was it phantom loads, inefficiency somewhere, or just wildly underestimating actual usage?

I've got a rough sketch of what I think I'll need (looking at Victron Multiplus for inverting, probably Renogy panels), but want to make sure I'm not building on dodgy assumptions from the start.

@RussScott, you're not overthinking it—consumption is actually the part where most folk go wrong. I made a right mess of my shepherds hut initially by guessing rather than measuring.

Here's what worked: live in the space for a week before you spec the system. Track everything—kettle, lights, fridge, laptop charging. Jot down when you use stuff. That pattern matters more than the raw numbers because your solar generation won't match your needs anyway.

For a hut, you've probably got decent roof space, which is your advantage. But if you're doing seasonal use (which most shepherd's huts are), you need to size for your worst month, not average out. I learned that one the hard way when I tried to run through October on summer calculations.

One thing that's saved me repeatedly: keep a spreadsheet. Amp-hours in, amp-hours out, battery state each day. Boring as it sounds, you'll spot patterns within a fortnight that'll tell you whether you've actually got enough capacity or if you need another panel.

What's your expected usage pattern looking like—constant occupation or weekends mostly

Right, consumption logging is where it gets real. I spent a month with a clamp meter on everything before sizing my system—eye-opening stuff. Most folk underestimate phantom loads and seasonal variance.

What helped me: stick a cheap smart meter on the mains first if you're still grid-connected, or use a Victron BMV to monitor actual draw once you're off-grid. That data beats any calculator.

@RussScott, shepherd's hut? You'll want to nail down heating separately from your 12/24V stuff—that's where the consumption math breaks down for folks. Electric heaters are brutal on batteries. Wood burner or gas is the move.

The seasonal bit @PanelKate mentioned is crucial too. Your winter baseline is what you actually size for, not your summer average. I'm in the North and my winter consumption is nearly 3x summer. Built for the worst case, everything else is a bonus.

Have you lot actually measured your consumption over different seasons though? That's what caught me out with my static caravan setup.

I thought 5kWh per day was reasonable, but soon realised winter usage was nearly double that—heating, shorter daylight hours, the lot. Summer was barely 2kWh. Makes a massive difference when you're sizing batteries.

@RetiredEngineer86's approach with the clamp meter sounds spot on. I'd add: don't just measure for a week. I logged everything for three months before committing to my Victron system, and I'm glad I did.

The other thing—be honest about what you actually want to run. If you're planning to work from a garden office like mine, your baseline consumption is completely different than someone just using the space weekends. The figures change everything downstream.

What's your typical usage looking like, @RussScott? Winter and summer?

@BayTim absolutely nailed it starting with consumption. The seasonal angle @OldSparky raises is critical—I made the mistake of sizing based on summer usage, then winter hit and I was scrambling.

What I'd add: log consumption over at least a full season, not just a week or two. Include your worst-case scenario—heating, hot water draw, whatever stresses your system most. I use a Victron BMV-712 to track this religiously; the data's invaluable for tuning later.

One thing that'll save headaches: separate your loads mentally. Essential circuits (fridge, heating, comms) versus discretionary (workshops, entertainment). Then size your battery bank to cover essential loads through your worst weather window—3-5 days typically for UK off-grid setups.

Also worth noting: actual consumption's rarely what the kit specs claim. My Fogstar fridge draws consistently less than rated, but my induction hob? Draw current's all over the place depending on what's cooking. Real monitoring beats spreadsheet assumptions every time.

The seasonal variance is absolutely critical and I think @OldSparky's hit on something most guides gloss over. I logged my static caravan for a full year before committing to kit—winter consumption was nearly triple summer figures, mainly heating and longer indoor hours.

What I'd add: don't just measure total kWh. Break it down by circuit or appliance category. You need to know your heating load separately from fridge-freezer baseline, because that changes how you approach battery sizing. A 10kWh battery might be fine for summer but useless for winter if you're running immersion heaters.

Also consider your profile, not just magnitude. A shepherds hut with woodburner and minimal evening cooking is totally different from a caravan where someone's working from home. The battery needs to handle your actual usage pattern—peaks matter as much as totals.

I used a Victron system-wide approach with the BMV-700 monitoring everything. Makes the data conversation evidence-based rather than guesswork. You can export the logs and actually see what's happening across seasons, which then feeds into your PV array sizing and battery capacity decisions properly