Decarbonarma

I've spent my career keeping systems up.
Then I pointed that at my own house.

I'm Gareth. I've lived in the same Oxfordshire house for over thirty years, and I've spent my working life in IT infrastructure — investment banks, telco, and more recently software startups. Everything on this site comes from one place: my own home, my own invoices, my own meter readings.

Decarbonarma is written by one UK homeowner who has spent four years decarbonising his own house — solar, battery storage, an air-to-air heat pump and a heat battery — and publishes what each part actually cost and how it actually performs. That £20-a-month direct debit covers everything the house uses, including charging two electric cars for 18,000 miles a year. The job isn't finished, though: there's still a gas boiler here, and I'll tell you what it costs to finish it. No installer sponsors this site, and no manufacturer has any say in what appears here.

How I think about all this

Let me be straight about what I am and what I'm not. I am not a heating engineer, an electrician or an MCS-certified installer. I'm a homeowner who has spent a career in IT infrastructure and then spent four years and a lot of his own money on one house. What I bring isn't trade qualification — it's the habit of thinking in systems, and a full set of invoices.

That career was built on lean thinking, the Toyota kind: maximise uptime, do the basics exceptionally well, make the money go as far as it can. Nobody congratulates you for a clever solution when you run infrastructure for a bank; they notice when it stops. Not all of that transfers — nothing in datacentre operations teaches you heat loss or flow temperatures. But the systems half of it transfers completely.

Your decarbonisation system has to be a team. Every part must slot into the next. Individually impressive components that don't talk to each other will disappoint you.

In my view this is the biggest thing the industry gets wrong. You're sold a heat pump. Then, separately, solar. Then, separately, a battery. Each vendor optimises their own box, nobody owns the join, and you end up with a house full of good equipment that doesn't cooperate — a hot water tank that heats at the wrong time of day, a battery that empties into the grid an hour before the cheap rate starts, an inverter too small for the loads it was always going to see.

I don't design for the brochure figure. I design for the system that still works on a wet Tuesday in February, and for the shortest payback I can honestly get to.

Why I did it

Two reasons, and they're usually treated as opposites.

The first is that I walk. Getting out into the countryside is one of the real pleasures of my life, and my instinct there is simple: leave it as you found it. No damage, no litter, nothing left behind. I can't tell you with certainty what the science says about global warming — I'm not a climate scientist and I won't pretend to be one. But I'd rather not leave the earth with extra CO2 on my account. That's an ethic, not an argument, and I'm comfortable with it being exactly that.

The second is money. I want the shortest payback I can get, and I want the same for anyone reading this. A system that doesn't pay for itself is a hobby. There's nothing wrong with hobbies — but they should be sold as hobbies, and too much of this industry sells them as investments.

Those two motives point the same direction far more often than people expect. Using less, wasting less, and paying less are usually the same project.

What's actually in the house

Not a demonstration home. Not a manufacturer's showcase. A lived-in house, worked on in stages over several years, with the real prices. One bit of vocabulary first, because the whole site turns on it: an air-to-water heat pump heats water and pushes it round your radiators and a hot-water cylinder — it's what most people mean by "heat pump", and what the government grant is built around. An air-to-air heat pump moves heat straight into the room, like air conditioning running in reverse. It's cheaper, it cools in summer, it gets no grant, and it doesn't make hot water.

£20/month
Whole-home Octopus direct debit: heating, hot water, electricity and 18,000 EV miles a year
~16 kWp
Solar across three arrays, with 15 kWh of battery storage
£10k vs £27k
My heating route against the air-to-water quote I turned down — but read the caveats below, they matter
InstallKitCost
Solar + battery
2022
12× Viridian 405W (~4.9 kWp), Sonnen 15 kWh battery, 3 kW backup circuit~£20,000
Re-roof + insulation
Oct 2025
SuperFOIL SF40BB breather membrane over the rafters, SF60 on the loft joistsPart of re-roof
Solar, second array
Oct 2025, +garage May 2026
18× Aiko 510W (~9.2 kWp), 9 north / 9 south, Enphase IQ8HC microinverters~£10,000
Air-to-air heat pump
Nov 2025
Daikin 3MXM52A9 multi-split, three indoor units — heating and cooling£5,300
0% VAT
Hot water
Feb 2026
Sunamp Thermino 300e heat battery (~12 kWh), replacing a limescale-killed cylinder£4,774
inc. 20% VAT

Prices are as invoiced, which means they sit on different VAT bases — the heat pump qualified for 0% VAT under the energy-saving materials relief, the Sunamp was billed at 20%. I've left them as they came, because that's what you'd actually have paid.

The honest version of "£10k vs £27k". That £27,000 quote was before the £7,500 Boiler Upgrade Scheme grant, so the fair comparison is nearer £10,000 against £19,500. And the two systems are not the same thing. Air-to-water would have heated every room and made my hot water. My route heats and cools three rooms, needed a separate £4,774 heat battery for hot water, leaves the gas boiler in place as a backstop, and — because air-to-air doesn't qualify — got no grant at all. In its favour, it also cools, which the £27,000 system wouldn't have. Judge it on that basis, not on the headline.

Two electric cars sit on the end of all this, covering about 18,000 miles a year between them. I count them as part of the household energy picture rather than as motoring, because that's what they are — another overnight load on the same tariff as the hot water and the battery. It's the reason the direct debit figure above is a fairer measure of this house than a bill for heating alone would be.

Across a summer the north-facing half of my roof produces about 70% of what the south side does — higher than most people expect, and the main argument for microinverters over a single string inverter. On a good June day the arrays make over 64 kWh. Those are the kinds of figures this site exists to put on the record, alongside the ones that look less flattering.

What it all cost — and the awkward question

Add the table up and it comes to a little over £40,000 across four years, plus the roof. I've never counted the roof as energy spending, because it needed replacing regardless; the insulation and the second solar array went on while the scaffolding was already up, which is exactly when that sort of work is cheapest.

Now the awkward part, because I've just told you that a system which doesn't pay for itself is a hobby, and £40,000 is a lot of hobby. So here's the honest accounting.

Not all of it was a free choice. The Sunamp replaced a hot water cylinder that hard water had killed after twenty years — that money was going to be spent on something whatever I decided. The 2022 solar and battery went in at 2022 prices, with the biggest inverter available to me at the time, and I would both size it differently and pay less for it today. What's left — the second array and the air-to-air system — is the part I'd defend purely on the numbers, and it's the part that took my whole-home bill down to where it is.

I'm not going to give you a single tidy payback figure for the house, because I don't think it would be an honest number. Four years of staged work, at four different price points, with two of the decisions forced by kit failing and one shaped by a re-roof, doesn't reduce to one percentage without hiding more than it reveals. What I can do is give you the per-project maths, which is what the case studies are for — and tell you plainly which of my own decisions would survive that test today and which wouldn't.

Where else this house has turned up

I approached Roger Bisby at Skill Builder — one of the UK's largest building channels, and not one that goes easy on heat pump hype. He came back to me, and the house has now featured twice.

Sunamp have also written the hot water system up as one of their own case studies. Over a 24-day monitored period the heat battery charged fully on solar alone on 12 days out of 24, and solar covered 93% of hot water demand even on the days that needed an overnight top-up.

You should be sceptical of that last one, so let me get ahead of it. Being featured on a manufacturer's website is a fair thing to raise on a page that argues for independence. So: I bought the Sunamp at full price, £4,774 including VAT, and the invoice is in the table above. They didn't pay me, discount it, or ask for approval over anything I write — they asked to write up numbers I had already published here. If that ever changes I'll say so. In the meantime, treat their page as marketing and mine as the data, and note that I've also written publicly about the parts of my setup I got wrong.

Left: the re-roof and solar install. Right: the air-to-air heat pump and Sunamp heat battery.

What I got wrong

Two expensive lessons, both worth more to you than any of my successes.

The inverter was too small. My 2022 system went in with a 3.6 kW inverter — the largest available to me at the time. It should have been 5 kW. A whole house pulls more than 3.6 kW more often than you'd think, and every time it does, you're buying from the grid while your own battery sits there unable to deliver fast enough. It isn't a fault. It's a ceiling I designed in and now live with. Size the inverter for the loads you'll actually have, not the ones in the sales brochure.

I bought a tank for a system I never built. Years ago I fitted a twin-coil Viessmann cylinder, sized and specified for solar thermal panels. The solar thermal never happened — solar PV improved so fast that it blew straight past it on both cost and flexibility. The tank sat there as a monument to a bet on the wrong technology, until hard water finally killed it after twenty years and a Sunamp heat battery took its place.

The lesson I'd pass on: don't buy hardware for a system you haven't committed to building yet. In a field moving this quickly, the option you're preserving usually expires before you use it.

Why I write it down

Because I watched too many people do this badly, on advice from salesmen whose interests weren't theirs.

That's the whole reason this site exists. Not because the technology is confusing — it isn't, particularly — but because almost everyone offering guidance is paid according to what you buy. Ask an air-to-water installer whether you need air-to-water and you already know the answer. I was quoted £27,000 before grant for a conventional air-to-water system. I heated and cooled my home a different way for around £10,000, and I've published the running figures ever since so you can judge whether that was the right call.

So here's my position on money, in full. No installer pays me. No manufacturer sponsors this site or has any say in what appears on it. Nobody has ever paid for a mention, and nothing here has been commissioned.

There are exactly two ways money reaches me, and you should hear them from me rather than find them yourself. The first is the Octopus Energy referral link on the home page: switch through it and you get £50 credit, and so do I. I use Octopus myself, and every running-cost figure on this site is based on their tariffs — I'd recommend them with or without the link, but you should know it's there. The second is a Buy Me a Coffee button that a handful of readers have used. That's the lot. If anything else ever comes along, I'll say so on this page before you read another word.

What's left to do

The house isn't finished, and I'd be suspicious of anyone who claimed theirs was.

The last piece of the puzzle is the gas boiler. It's still there, but these days it only appears in emergencies — a backstop rather than a heating system. The plan is to retire it properly: a heat pump solution that also brings cooling to two more rooms, which would take the whole house electric and finish the job the air-to-air system started.

When it happens, the costs and the metered results will go on this site like everything else — including the parts that don't go to plan.