If you’re thinking of putting solar on a 4-bed with a NW/SE split roof, here’s what you can expect. I’ve got solar at home and I’ve been running panels on an SE and NW split for three years. The short version: it works, it isn’t hopeless, and the orientation penalty is smaller than the one decision that actually swings your ROI, whether or not you fit a battery.
My roof runs at 28° on the garage, five panels SE and four NW. On overcast days, of which the UK has plenty, the two strings produce almost the same output. On sunny days the SE wins by roughly a third. Across a full year the combined system pulls in about 75% of what a south-facing equivalent would, per PVGIS. That is smaller than most people expect, and smaller than the boost you get from adding a battery (see below).
Without a battery, your SE face dumps its midday surplus to the grid at 5p/kWh minimum SEG and you buy the same electricity back at 27-28p in the evening. That is the difference between an 8-year payback and a 4-year payback on this roof. Every other decision in this guide is secondary to that one.
What a NW/SE Roof Actually Does
Two principles to hold in mind before you pick hardware:
- Self-consumption beats export, by a lot. Every kWh you use directly saves 27.69p. Every kWh you export earns 5p on statutory SEG, or 12p on Octopus Outgoing Fixed. Even on the better tariff, grid displacement is still 2-3x more valuable than export. Size for self-consumption first.
- You need a dual-MPPT inverter. The two faces produce at different levels throughout the day. An inverter with two independent Maximum Power Point Trackers lets each string operate at its own optimal voltage. A single-MPPT inverter drags both to the weaker face. Power optimisers can solve this with a single-MPPT unit, but a dual-MPPT hybrid is cleaner and cheaper.
- Load more panels on the SE face. SE generates more annually. Weight the array 55-60% on SE, 40-45% on NW. The NW panels are supporting actors, not leads.
- Battery first, bigger array second. If you have to choose between more panels and adding a battery, take the battery every time. I would go further: on a really tight budget, buy the battery first and add solar later. That is not standard advice, but it reflects what the numbers actually say on this roof shape.
Why the Generation Curve Works Better Than the Total
A south-facing roof gives you a sharp midday peak. A NW/SE roof gives you a wider, flatter curve: the SE face lights up around 8am and peaks mid-morning, the NW face contributes from roughly 2pm and extends generation into the evening in summer. For a 4-bed family where someone is home, a heat pump is cycling, or a hot water tank is heating, that spread is actually better matched to load than a south-facing spike that you can’t use. The share of annual irradiance that arrives as diffuse light in the UK is high enough that the raw yield gap is narrower in real conditions than the PVGIS aspect penalty implies.
If you want to understand how this shape compares to other roof types, the panels-tilt guide has three years of my own metered data across pitched, flat, and vertical installations.
What a 4-Bed Actually Uses
Low to Mid: ~3,500 kWh/year
Gas central heating, no EV, standard appliances. Around £970/year on electricity at the current cap. Demand concentrates morning (7-9am) and evening (5-9pm), with a daytime dip unless someone works from home. An SE face already covers some of that morning peak directly.
High: 7,500+ kWh/year
Heat pump, or EV charging at home, or both. A heat pump alone adds 3,000-5,000 kWh/year depending on insulation. An EV adds 2,000-3,500 kWh depending on mileage. Your baseload is higher across the whole day, which is good news on this roof: you’ll self-consume a much higher share of what you generate because the loads overlap with SE generation instead of sitting idle waiting for evening. You’ll want at least a 5kW inverter to handle peak draw.
Budget Option: 3.6kW Hybrid, 12 Panels, 3.55kWh Battery
Solar Energy Concepts System Rating
System Specification
Inverter: Solis 3.6kW 5G Hybrid EH1 (£575-653). 3.6kW AC output keeps you under the G98 threshold (notification only, not a full G99 application). Dual MPPT, 90-520V range, 15A per MPPT, accepts up to 5.7kW DC total. Hybrid design so you can bolt a battery on later without replacing the inverter.
Overpanel target: 140-150% of the 3.6kW AC rating = 5,040-5,400W DC. On a NW/SE roof in UK conditions you will rarely clip, and the extra panels earn their keep in the shoulder hours (early morning, late afternoon, winter).
Panel Options
| Strategy | Panels | Total DC | Overpanel % | String Check |
|---|---|---|---|---|
| Best value | LONGi Hi-MO X6 455W × 12 (7 SE + 5 NW) (£73/panel) | 5,460W | 152% | SE string Voc: 7 × 39.15V = 274V. NW string: 5 × 39.15V = 196V. Both within 90-520V MPPT range. Isc 14.79A under 15A limit. |
| Smallest footprint | Aiko Neostar 490W × 11 (7 SE + 4 NW) (24.5% efficient) | 5,390W | 150% | SE: 7 × 40.98V = 287V. NW: 4 × 40.98V = 164V. Isc 14.93A under 15A. Saves one roof position. |
| Max overpanel | DMEGC 450W × 12 (7 SE + 5 NW) | 5,400W | 150% | SE: 7 × 35.5V = 249V. NW: 5 × 35.5V = 178V. Low Voc allows long strings. Warning: Isc 15.4A exceeds Solis 15A limit. Pair with Fox ESS H1-3.7 (16A) or SolaX X1 G4 3.7kW (16A) if using DMEGC. |
Battery
Pylontech US3000C 3.55kWh (£600-900). This is the non-negotiable part. On a NW/SE roof without storage, you export most of your SE surplus at 5-12p/kWh and buy it back at 27-28p in the evening. A 3.5kWh pack covers a typical 4-bed evening (kettle, cooking, lights, TV, a couple of hours of baseload). Compare options in the directory. The Solis EH1 is hybrid, so if cashflow forces you to defer the battery you can add it inside a few weeks without changing any other hardware.
Total System Cost
- DIY: £2,700-3,400 (panels £880 + inverter £615 + battery £600-900 + mounting/cables £500-1,000)
- Installed: £5,200-6,500
DIY Feasibility
This is one of the more DIY-friendly builds. At 3.6kW AC you only need to notify your DNO under G98, not apply. The notification is a 5-minute online form. You can mount the panels, run the DC side, and fit the inverter yourself. A qualified electrician handles the final AC tie-in to your consumer unit under Part P (£150-300). No MCS is required for the install, but without MCS you cannot claim SEG export payments. On a self-consumption-first system, the £60-100/year of foregone SEG is the price of admission for DIY.
Expected Performance
| Metric | Low-Mid (3,500 kWh) | High (7,500 kWh) |
|---|---|---|
| Annual yield | 4,232 kWh | |
| Self-consumption (with 3.55kWh battery) | 58% (2,459 kWh) | 88% (3,737 kWh) |
| Export | 42% (1,774 kWh) | 12% (495 kWh) |
| Grid savings | £681/year | £1,035/year |
| Export income (SEG 5p) | £89/year | £25/year |
| Total annual saving | £770/year | £1,060/year |
Savings rise if you move off statutory SEG. Octopus Outgoing Fixed at 12p/kWh would add roughly £125/year at low-mid consumption and £35/year at high (the high figure is smaller because you export less).
Premium Option: 5kW Hybrid, 16 Panels, 5.12kWh Battery
Solar Energy Concepts System Rating
System Specification
Inverter: Solis 5.0kW S5 Smart Hybrid (£707-770). 5kW AC crosses the G98 threshold so you need a full G99 application (see below). Dual MPPT, 90-520V, 15A per MPPT, accepts up to 8kW DC. That DC headroom is what makes 16-panel overpanelling easy.
Overpanel target: 140-150% of 5kW = 7,000-7,500W DC.
Panel Options
| Strategy | Panels | Total DC | Overpanel % | String Check |
|---|---|---|---|---|
| Best value | LONGi Hi-MO X6 455W × 16 (9 SE + 7 NW) (£73/panel) | 7,280W | 146% | SE: 9 × 39.15V = 352V. NW: 7 × 39.15V = 274V. Both in range. Isc 14.79A under 15A. |
| Smallest footprint | Aiko Neostar 490W × 15 (9 SE + 6 NW) (24.5% efficient) | 7,350W | 147% | SE: 9 × 40.98V = 369V. NW: 6 × 40.98V = 246V. Isc 14.93A OK. Saves ~1.7 m² of roof. |
| Max overpanel | DMEGC 450W × 16 (9 SE + 7 NW) | 7,200W | 144% | SE: 9 × 35.5V = 320V. NW: 7 × 35.5V = 249V. Low Voc maximises string length. Isc 15.4A exceeds Solis 15A. Use GivEnergy 5kW (15A but higher headroom at 580V) or Fox ESS H1-5.0 (16A). |
Battery
Sunsynk ECCO 5.12kWh (£1,200-1,500) or Pylontech US3000C 3.55kWh (£600-900) stacked. A 5kWh pack is the sweet spot for a heat-pump or EV household on this roof. Anything bigger than 10kWh is hard to justify on generation alone at this scale; it only pays off if you run overnight tariff arbitrage, which pushes your tariff exposure score up.
Total System Cost
- DIY: £3,200-4,200 (panels £1,168 + inverter £740 + battery £900-1,500 + mounting/cables £400-600)
- MCS installed: £6,500-8,500
G99 and MCS
At 5kW AC you are above the G98 threshold of 3.68kW. You must apply for G99 permission from your DNO before connecting (a formal application, not a notification). Most residential G99s are approved within the 45-working-day statutory limit set by ENA Engineering Recommendation G99. SSEN regions (Scotland, parts of the South) routinely take 60+ days when the local grid is constrained, per their published connections SLA. If you’re in an SSEN area, submit before the panels arrive.
MCS certification is worth it for the premium build because: you qualify for SEG export payments (meaningful at this generation level), you can access time-of-use tariffs that some suppliers restrict to MCS installs, and it protects your home insurance and resale value.
When This Is Worth It vs Overkill
Worth it if you have a heat pump, you charge an EV at home, or your annual consumption is over 5,000 kWh. At 7,500 kWh, the premium system pays back in under 3 years DIY and delivers £32,325 over 25 years (see the full ROI analysis). A heat pump run from the battery during peak rates is where the extra capacity earns out.
Overkill if your consumption is under 3,500 kWh and you have no heat pump or EV on the horizon. At that consumption the budget system already saves £770/year. The premium system adds £193/year. That is £3,800 of extra upfront hardware for less than £200/year of additional saving, which is not a good trade. Start budget, upgrade when your consumption does.
Expected Performance
| Metric | Low-Mid (3,500 kWh) | High (7,500 kWh) |
|---|---|---|
| Annual yield | 5,601 kWh | |
| Self-consumption (with 5.12kWh battery) | 54% (3,009 kWh) | 80% (4,462 kWh) |
| Export | 46% (2,593 kWh) | 20% (1,139 kWh) |
| Grid savings | £833/year | £1,236/year |
| Export income (SEG 5p) | £130/year | £57/year |
| Total annual saving | £963/year | £1,293/year |
Why the Battery Is the Real ROI Lever
Three years of metering on my own SE/NW setup taught me one thing that changed how I talk about solar: you do not earn your ROI from generating more kWh. You earn it from consuming more of what you generate. On a NW/SE roof, the compressed generation window makes this doubly true. The SE surplus between 10am and 2pm is either stored for your 6pm cooking peak or it becomes someone else’s 5p export.
In addition to shifting solar to evenings, a battery lets you do winter arbitrage: charge off-peak at 7.5p on Octopus Go or Intelligent, use it during the expensive morning peak. In 2022 the UK saw 400% energy price hikes. Having a home battery is a hedge against that happening again. On a NW/SE roof, the battery typically doubles your annual saving compared to a solar-only setup at the same panel count.
Tariff Strategy
Budget option: A flat-rate tariff is fine. Your savings come from direct self-consumption. If you want export income, Octopus Outgoing Fixed at 12p doubles the statutory 5p without requiring a battery. Skip Agile unless you enjoy watching half-hourly prices.
Premium option: Octopus Go or Intelligent Go is the natural fit. Overnight charge at 7.5p, carry through the morning on the battery, solar takes over by mid-morning, battery covers the evening peak. If consumption is very high, Octopus Flux beats Go for export (30p peak 4-7pm) but the self-consumption math with an EV often wins out on Intelligent Go.
ROI Comparison
| Budget DIY | Budget Installed | Premium DIY | Premium MCS | |
|---|---|---|---|---|
| Low-Mid Consumption (3,500 kWh/year) | ||||
| Upfront cost | £3,050 | £5,850 | £3,700 | £7,500 |
| Annual saving | £770 | £770 | £963 | £963 |
| Payback | 4.0 years | 7.6 years | 3.8 years | 7.8 years |
| 25-year return | £19,250 | £19,250 | £24,075 | £24,075 |
| High Consumption (7,500 kWh/year) | ||||
| Upfront cost | £3,050 | £5,850 | £3,700 | £7,500 |
| Annual saving | £1,060 | £1,060 | £1,293 | £1,293 |
| Payback | 2.9 years | 5.5 years | 2.9 years | 5.8 years |
| 25-year return | £26,500 | £26,500 | £32,325 | £32,325 |
Assumes 27.69p/kWh flat import, 5p/kWh SEG baseline, no energy inflation, and 0.5% annual panel degradation. Real returns typically run higher because electricity prices rise and because moving to Octopus Outgoing Fixed at 12p or Flux at up to 30p increases export value by 2-6x.
Generation Yield Source
All yield estimates here are derived from PVGIS, the European Commission’s free solar radiation database. Enter your postcode to generate location-specific figures.
| Parameter | Value |
|---|---|
| Location | 52.308°N, -0.717°W (Central England reference) |
| Roof slope | 28° |
| System loss | 14% |
| Database | PVGIS-SARAH3 |
| SE aspect | -45° → 931.46 kWh/kWp/year |
| NW aspect | 135° → 643.55 kWh/kWp/year |
| Clipping factor | 0.96 (applied to both systems) |
Key Takeaways
- NW/SE roofs produce about 75% of south-facing annual output (PVGIS), but the real-world gap is narrower because diffuse light dominates UK generation.
- A battery is the single biggest ROI lever. Without one, SE surplus exports at 5-12p and you buy it back at 28p.
- The budget build (Solis 3.6kW + 12 panels + 3.55kWh battery, £3,050 DIY) pays back in 4 years on a typical 4-bed.
- The premium 5kW build is for heat-pump or EV households above 5,000 kWh/year. Don’t pay the extra £3,800 otherwise.
- Use a dual-MPPT hybrid inverter. Each string needs to run at its own voltage.
Related Guides
- How roof tilt actually affects 10-year ROI. Three years of my metered data on SE, NW, flat, and wall mounts.
- Start here: your guide to home solar. Read this first if you haven’t decided whether solar is for you, read this first.
- Battery overview. Sizing, chemistry, and why storage is the ROI lever on most UK roofs.
- Seven solar myths. Includes the diffuse-light myth that matters most for this roof.
- Are solar panels worth it in 2026?. Full ROI breakdown.
- 3-bed east/west roof. The other “split roof” scenario, with a more favourable demand match.
- 4-bed south-facing roof. The same house size, the easier orientation.
