Battery storage is the single most misunderstood investment on a commercial rooftop solar project. Oversized batteries kill payback. Undersized batteries miss the out-of-hours and time-of-use savings. The right sizing is driven entirely by your half-hourly consumption data and the tariff structure you are buying electricity under, not by the kWp rating of the PV array.
This is the FLD approach to battery sizing, chemistry selection and commercial economics for Welsh businesses in 2026.
When a battery actually pays back
A commercial battery storage system pays back through three mechanisms, in rough order of value:
- Self-consumption uplift on solar generation that would otherwise be exported at 5 to 15p per kWh but can instead be consumed at 28 to 32p per kWh
- Time-of-use arbitrage charging at off-peak rates (typically 10 to 14p per kWh) and discharging during peak rates (typically 28 to 38p per kWh)
- Capacity charge avoidance and demand shaving on sites above the HH-metered threshold where distribution and capacity charges apply
In practice, mechanism 1 dominates for most Welsh SMEs. Mechanism 2 matters where you have a TOU tariff signed. Mechanism 3 matters only for larger industrial sites on HH metering with meaningful capacity charge exposure.
If none of these mechanisms apply to your site, adding a battery to a commercial solar array will not pay back within a reasonable horizon. We will tell you so at survey stage rather than quote something that looks impressive on paper.
Sizing against half-hourly data
The correct way to size a commercial battery is to take 12 months of HH consumption data from your supplier portal, overlay modelled PV generation for your rooftop, and calculate the residual surplus generation that cannot be consumed in real time.
For a typical 100 kWp commercial rooftop in Swansea with a weekday daytime office load:
- Annual generation: approximately 95,000 kWh
- Without battery, self-consumption typically 60 to 70 percent, so 28,500 to 38,000 kWh exported at SEG rates
- Adding a 30 kWh battery shifts self-consumption to 75 to 85 percent, reducing export to 14,250 to 23,750 kWh
- Net benefit: 14,250 additional kWh retained at a spread of roughly 20p per kWh, worth £2,850 per year
- At a 30 kWh LFP battery cost of £12,000 to £14,000, simple payback is 4.2 to 4.9 years
The sensitivity is sharp. Double the battery capacity and the marginal kWh retained drops off quickly, extending payback past the 10-year mark. Halve the battery and the payback improves but the absolute savings shrink.
Chemistry: why LFP almost always wins for commercial
Three lithium chemistries are in commercial use for stationary storage:
| Chemistry | Energy density | Cycle life | Fire safety | Cost per kWh (2026) | Commercial fit |
|---|---|---|---|---|---|
| NMC (nickel-manganese-cobalt) | High | 3,000 to 5,000 | Poor | £380 to £450 | Niche |
| LFP (lithium iron phosphate) | Medium | 6,000 to 10,000 | Excellent | £320 to £400 | Default |
| LTO (lithium titanate) | Low | 15,000-plus | Excellent | £900-plus | Rare |
LFP is the default for commercial stationary storage in Wales for three reasons:
- Cycle life of 6,000 to 10,000 full-depth cycles means a 20-year service life at one cycle per day
- Thermal stability reduces insurance loading and, on some sites, removes the need for a dedicated battery enclosure
- Cost trajectory has been consistently downward and the supply chain is stable
NMC is lighter and slightly cheaper per installed kWh on paper, but the shorter cycle life means the total cost of ownership is higher over a 15 to 20-year horizon, and the insurance and fire-risk position is materially worse. We no longer recommend NMC for commercial rooftop-collocated storage.
Inverter topology: hybrid, AC-coupled, or DC-coupled
The battery inverter topology choice has meaningful implications for both cost and flexibility.
Hybrid inverter
A single inverter handles both PV DC input and battery DC input, and a single AC output. Pros: lowest component count, simplest commissioning, often 5 to 10 percent cheaper. Cons: if you need to replace the inverter, you take down both PV and battery at the same time.
DC-coupled battery
The battery sits on the DC bus alongside the PV array, sharing the inverter. Pros: highest round-trip efficiency at 92 to 95 percent because only one DC-AC conversion step. Cons: limited retrofit flexibility.
AC-coupled battery
The battery has its own inverter and sits on the AC bus alongside the PV inverter. Pros: maximum retrofit flexibility, can be added to an existing PV install without touching the PV inverter. Cons: round-trip efficiency 88 to 91 percent, higher component count.
For new-build projects with PV and battery designed together, hybrid or DC-coupled is usually the right answer. For retrofit to existing PV, AC-coupled is almost always the right answer unless the existing inverter is at end-of-life.
Time-of-use arbitrage: is it worth it?
TOU arbitrage means charging the battery from the grid at off-peak rates and discharging during peak rates. For commercial SME tariffs in Wales in 2026:
- Off-peak rates on typical economy tariffs: 10 to 14p per kWh during overnight windows
- Peak rates: 28 to 38p per kWh during afternoon and evening peaks
- Gross spread: 14 to 24p per kWh
Net of round-trip losses (88 to 95 percent) and battery cycle cost (roughly 3 to 5p per kWh on LFP at current capex), the net margin on a TOU arbitrage cycle is typically 6 to 15p per kWh. On a 30 kWh battery cycled once daily at an average 20 kWh useable capacity, that is £1,500 to £3,000 per year of additional revenue, over and above the solar self-consumption benefit.
The caveat is tariff eligibility. Many commercial supply contracts do not offer TOU structures, and those that do often have minimum usage thresholds that exclude smaller SMEs. Check your actual tariff structure before modelling TOU revenue.
Capacity charge avoidance for HH-metered sites
Sites on half-hourly metering above the 100 kW maximum demand threshold pay capacity charges based on monthly peak demand. A battery can shave that peak by discharging during the half-hour windows that set the capacity charge for the month.
For a Welsh industrial site with typical capacity charges of £5 to £10 per kW per month, shaving 50 kW of peak demand is worth £3,000 to £6,000 per year. For larger sites the numbers get bigger quickly. This is a specialist sizing exercise; it depends on the specific HH profile and the DUoS/TNUoS charge structure on your connection.
Safety, compliance and installation practice
Commercial battery storage installations in Wales are governed by:
- BS 7671 Amendment 2 for electrical installation
- IET Code of Practice for Electrical Energy Storage Systems for design and commissioning
- MCS Battery Scheme for grid-connected storage below 50 kWh
- G99 for the grid connection where the battery is part of a generation installation
- Local fire authority guidance on larger installations, typically triggered above 50 kWh or where the battery is inside a manned occupied building
Our installation practice:
- Dedicated battery enclosure or segregated plant room for installations above 20 kWh
- Smoke detection and emergency lockout isolation local to the battery
- All DC cable runs in metal conduit or SWA
- Commissioning to IEC 62619 and the manufacturer test protocol
- Full documentation pack on handover, including O and M manual, commissioning certificates and a disposal plan for end-of-life
Batteries on retrofit to existing solar
If you have an existing solar PV array and are considering adding a battery, the process is:
- Take 12 months of HH consumption data and export data from your existing system
- Model the shortfall in self-consumption that a battery could close
- Size the battery for the marginal kWh retained, not the nameplate PV capacity
- Choose AC-coupled topology unless the existing inverter is at end-of-life
- Apply for G99 amendment with the DNO to reflect the changed installation
For most retrofit projects this takes 8 to 12 weeks from survey to commissioning.
When we recommend against a battery
Cases where we will tell you a battery does not make sense:
- Site load profile is 24/7 industrial with high self-consumption (>85 percent) already
- Solar array is so small that export is minimal anyway
- Tariff structure offers no TOU spread and the site is not HH-metered
- Capital budget is tight and the same capex would extend the PV array to greater effect
Commercial battery storage is a tool, not a universal upgrade. Applied in the right cases, 5-year paybacks are routine. Applied in the wrong cases, paybacks push past 12 to 15 years.
Starting the conversation
If you want a battery sizing analysis against your actual HH data and tariff structure, send 12 months of HH export from your supplier portal to our office and we will turn around a sized proposal inside a week. Call Paul direct on 01792 321123 to start.