Cost analysis without export

Been thinking about this a fair bit lately, particularly relevant to those of us who aren't grid-tied at all.

For the narrowboat and the shepherd's hut, I've no export option whatsoever — so the usual "savings vs export tariff" metrics that Victron and various apps throw at you are essentially meaningless in my situation. What actually matters is rate avoidance: how much am I displacing from the generator or from shore power when I'm moored up?

The Victron VRM portal gives you consumption data, solar yield, that sort of thing, but I've not found a clean way to express "you avoided running the Honda EU22i for X hours, saving Y litres of fuel at Z pence per litre." That calculation exists in my head in a spreadsheet, but it's manual and clunky.

Has anyone found a decent way to model this properly? Even a rough framework would help. My thinking:

• Baseline: cost per kWh from generator (fuel + maintenance amortised)
• Solar displaced kWh × that rate = actual saving
• Factor in battery cycle degradation (Fogstar Drift cells, so hopefully manageable over the long term)

The complication is that the baseline cost isn't fixed — generator efficiency varies hugely depending on load, and partial-load running is expensive per kWh.

Curious whether anyone's built something in Home Assistant or even just a decent spreadsheet that handles this properly for off-grid scenarios rather than assuming grid export as the reference point.

SolarTom | 847 posts

Great thread @Spud79. The standard payback calculators are essentially useless for us off-gridders since they're built around export tariffs and grid displacement figures that simply don't apply.

The way I've always framed it for my setup is: what's my avoided cost per kWh? So if you're currently running a generator, what's your actual cost per unit including fuel, oil changes, wear and tear? That becomes your baseline. Battery storage that lets you capture excess solar instead of running the genny at 6pm suddenly looks very attractive indeed.

For the shepherd's hut scenario specifically, I'd also factor in the convenience premium — not having to fire up a noisy generator has genuine lifestyle value that doesn't show in a spreadsheet but absolutely influences whether a system pays for itself in practical terms.

What's your current backup situation when solar falls short?

The real question isn't payback period — it's cost per kWh over battery lifetime vs whatever miserable alternative you were using before (generator diesel, hook-up fees, etc.).

My shepherd's hut Fogstar cells have basically paid for themselves just in saved campsite electricity charges, and I never once filled in a spreadsheet to prove it.

Good point from @Renogy_Nerd — cost per kWh over lifetime is the correct metric. Worth being precise about it though.

For my narrowboat setup I use this framework:

Usable cost per kWh =
(Battery cost + inverter/BMS) ÷ (usable DoD × cycle count)

With Fogstar Drift cells I'm getting realistic 3,000+ cycles at 80% DoD. Run the numbers and you're often looking at £0.06–0.09/kWh — compare that against diesel generator fuel costs which I was previously tracking at around £0.38–0.55/kWh including maintenance and wear.

The comparison baseline matters enormously for off-grid analysis. You're not displacing grid electricity, you're displacing your actual alternative — whether that's a genny, shore power charges at marinas, or even running the engine.

Factor in that the alternative cost is also inflation-linked and the battery economics become quite compelling even at modest usage.

OldSailor | 2,341 posts

@PennineNomad is right to demand precision — the full formula is: (Battery cost + inverter + BMS + installation) ÷ (usable kWh × cycle life × DoD efficiency), and people consistently forget the parasitic losses from the BMS and inverter standby draw, which on a narrowboat running 24/7 can quietly murder your cost-per-kWh figure.

My Fogstar Drift 200Ah cells pencilled out at roughly £0.04/kWh over projected lifetime versus £0.28/kWh equivalent from a diesel genny — that's the number that matters when you've got no export tariff to muddy the waters.

Also factor in replacement chemistry risk — LFP prices have dropped 40% in three years, so your second bank will likely be cheaper, which improves the whole-of-life figure considerably.

Lived this exact calculation building out my tiny house setup. The metric everyone's missing is opportunity cost of the alternative — for me it was either LiFePO4 or running a genny every evening. Once I factored in petrol, wear, and the sheer misery of generator noise at 9pm, my Fogstar cells looked very different on paper.

@OldSailor's formula is solid but static — batteries also protect against fuel price volatility, which is genuinely hard to quantify but very real. Diesel went properly mental in 2022 and my cost-per-kWh stayed completely flat while neighbours were wincing at every fill-up.

No export? No problem. Your baseline comparison isn't the grid rate — it's whatever you'd otherwise be burning or buying.

@OldSailor's formula is solid but I'd add one more variable that catches people out: degradation curve non-linearity.

Most lifetime kWh calculations assume linear capacity loss down to the 80% EOL threshold, but lithium (particularly LFP) tends to hold capacity well then drop more sharply toward end of life. If you're sizing based on usable kWh at year one, you're slightly optimistic across the whole cycle count.

Running a Fogstar Drift 100Ah in the shepherd's hut — now into its third year — actual measured capacity still sits at ~96Ah, which is encouraging. But my Victron BMV data suggests I should budget for a steeper fall-off beyond year five rather than assuming the same gentle curve continues.

For a true no-export setup, conservative degradation assumptions matter more because you can't lean on grid arbitrage to compensate when the battery underperforms.

VanHolly | 847 posts

Something nobody's touched on yet — seasonal mismatch penalty. On my van setup I've got surplus I genuinely cannot use through June and July, yet I'm capacity-starved November through January. Your cost-per-kWh calculation shifts dramatically depending whether you're averaging annually or modelling month-by-month. A battery sized for winter adequacy is "oversized" in summer, but that surplus capacity cost you real money. I'd strongly suggest running your numbers against actual monthly consumption profiles rather than annual averages — the honest figure is often worse than people expect.

Really relevant to my shepherd's hut situation this. One thing worth adding — opportunity cost of oversizing vs undersizing shifts completely when there's no export valve. On my setup I ended up going bigger on the Fogstar cells precisely because stranded capacity costs me nothing in lost revenue, whereas being short on a cloudy January week has a real cost in either generator runtime or just doing without. The maths flips compared to grid-tied. Size for your worst-case window, not your average.