How much would a school roof near Splott generate?

Modelled at 908 kWh per kWp a year, a 320 kWp array on a Splott school models at about 290,600 kWh. What decides whether that is worth doing is how much of it the school uses itself, not the total. Source: EU PVGIS v5.2, SARAH3, 10 degree pitch, 14 percent system loss.

How does the school year affect a solar scheme in Splott?

It goes somewhere, just not into the timetable. 25.6 percent of the Splott generating year lands in July and August (EU PVGIS v5.2), and a closed school still runs servers, comms, refrigeration and ventilation. What that base load does not absorb is either stored, exported under an agreement with National Grid Electricity Distribution, or a sign the array is too big. We test which before sizing anything.

Which network operator handles the connection at a Splott school?

Applications go to National Grid Electricity Distribution, who run the network across Cardiff. An export-limited offer usually still works for a school, because the summer surplus is the part you were least likely to be paid much for anyway.

Specialist solar panels for schools in Splott

National Grid Electricity Distribution runs the network around Splott, and on a school site their answer on export shapes the design as much as the roof pitch does. The survey records sheet type, remaining life, purlin spacing and the state of the incoming supply, and the model runs off your own meter data rather than an average.

290,600 kWh a year, modelled for a 320 kWp array on roughly 950 sqm of clear roof. That is the size applied for at the nearest school scheme on record. Source: EU PVGIS v5.2, SARAH3, 10 degree pitch, 14 percent system loss.

Book a roof survey in Splott

1 teaching block
2 hall and kitchen
3 sports hall
4 1960s block, no array
first phase
later phase

Drawing Illustrative estate, not a Splott school

Blocks 4, 3 with array

Array about 345 kWp

Yield about 313,000 kWh/yr

Frontage about 90 m / Rev A / Cardiff

Illustrative layout for an estate of this size. The array is drawn at about 345 kWp, roughly 53 percent of the 650 kWp these three roofs would hold, because a school sizes to the load it can use in term time rather than to the roof it has: generation it cannot use is exported at a much lower rate than it pays to import. The hatched block carries no array because its structure has not been assessed, which on a school of that era is the usual starting point rather than an exception. Your own figures, and which of your blocks can take an array, come from the roof survey.

Splott / Cardiff

A two storey 1970s school block with a shallow pitched roof covered in solar panels, empty playground in front and a playing field beyond — School buildings of the kind we survey across Cardiff. Not a named school and not our work.

Solar PV for academy trusts across Cardiff

A one kilowatt-peak array on a shallow pitched roof in Splott models at 908 kWh a year, from modelled irradiation of 1,159 kWh per square metre in the plane of the array (EU PVGIS v5.2, SARAH3, 10 degree pitch, 14 percent system loss). That is roughly 2 percent above the mean across the towns we cover, so latitude is working slightly in your favour here.

Scaled up, a 320 kWp array, sized to what the nearest school scheme on record applied for, 13 miles away at Weston-super-Mare, models at about 290,600 kWh a year before shading, and needs roughly 950 square metres of clear roof. We take the size from the nearest real application rather than a round number, so the example is anchored to something a planning officer has already seen.

Spread across the year that is about 74,400 kWh landing in July and August and roughly 13,700 kWh across December and January. National Grid Electricity Distribution sets what may go back onto the network here, so the July figure is the one worth putting in front of a governing body first.

FIG. 1 Splott against the rest of Cardiff

Canton936
Pontcanna936
Cathays908
Splott908

Splott ranks 4 of 4 towns we cover in Cardiff on modelled yield, against a national mean of 886 kWh per kWp. The county runs from 908 to 936, a spread of 28 kWh per kWp. Within Cardiff that difference is small enough to ignore: where the building sits is not what decides this scheme, your consumption pattern is.

Bars are zero based, so length is proportional to the figure. Where a county is flat, that is the finding: latitude is not the lever, the timetable is. Source: EU PVGIS v5.2, modelled per town

The planning position for Splott schools

The Renewable Energy Planning Database holds no school scheme for Splott and the area around it. Read that as a gap in the record, not a gap in the roofs: the database is reliable above one megawatt and thin below it, and a school array is typically a fifth of a megawatt.

The closest recorded one is about 13 miles away at Weston-super-Mare: Weston College, a 300 kWp roof mounted array applied for by Weston College, which holds consent and is awaiting construction. We were not involved in it. We cite it because it is a public record of what has cleared planning in this part of Somerset. Source: Renewable Energy Planning Database, Q1 2026.

Why demand in Splott falls as the roof peaks

The generating year and the school year are out of step in Splott. July and August carry 25.6 percent of the modelled output (EU PVGIS v5.2), and the site is closed for most of those nine weeks. June is the strongest single month here at 14.2 percent of the annual total, which is still term time, and that works slightly in your favour against towns whose curve peaks in July.

At the other end of the year the mismatch flips. Only 4.7 percent of output falls in December and January, the two months when a Splott school is fully occupied and its heating and lighting are at their heaviest. The roof makes 6.5 times as much in June as it does in December, and August alone outproduces December by about 5.3 to one. 74 percent of the year arrives between April and September, and the point of that figure is not that the summer half goes unused, because April, May and June are full teaching months that draw hard. It is that the concentration peaks in the few weeks at the end of it when the building is shut.

Nationally this comes out at about 49 percent of annual generation falling on a session day. A business open every weekday sees roughly 71 percent. The 22 point gap is what a school appraisal has to absorb, and it is much the same in Splott as anywhere else.

The same mismatch repeats every day. Peak generation lands between eleven and three, the timetable ends shortly after, and by four the site is down to caretaking and cleaning. An annual consumption figure cannot show that, which is why we ask for the half hourly data instead.

The conclusion is not that a Splott school should do nothing. It is that the array should be sized to the load that survives the holidays, which is usually the server room, the catering refrigeration, the ventilation and the hot water, plus whatever the site lets out over the summer. Anything above that line needs storage or an export agreement with National Grid Electricity Distribution to earn its keep.

We set out the whole argument about term dates and the generation curve on the home page, and what it does to a payback figure under costs.

FIG. 2 A Splott roof against the Welsh school year

generated while the school is open
generated at a weekend or in the holidays

The same figures as a table

Monthly output for a 1 kWp array in Splott, split by whether the day is a school session day. Across the year 49 percent of generation lands while the school is open.
Month	Session days	Output, kWh per kWp	School open	School closed
Jan	20	23	15	8
Feb	15	39	21	18
Mar	21	73	49	24
Apr	12	106	42	64
May	17	127	70	57
Jun	21	129	90	39
Jul	11	127	45	82
Aug	0	105	0	105
Sep	20	82	55	27
Oct	17	49	27	22
Nov	21	28	20	8
Dec	15	20	10	10
Year	190	908	444	464

Per kWp installed, month by month, separating generation while the school is open from generation at a weekend or in the holidays. 74 percent of the Splott year sits between April and September, most of which is term time; it is the tail of that period, the last week of July and all of August, that the school is not there for. The daily split is a flat apportionment of each month, not a metered one. Session days follow a typical school year in England and Wales, 190 days on broadly the same dates. Your own authority or trust may differ by a few days. Source: EU PVGIS v5.2, session days from a typical school calendar

Grid connection through National Grid Electricity Distribution

National Grid Electricity Distribution is the network operator here, and their answer on export capacity shapes the design. Any commercial array above 3.68 kW per phase connects under G99 rather than G98, and the application fixes what you are allowed to push back onto the network. If the local network is tight, limiting export rarely breaks the case here: most of the generation is consumed on site while the building is working.

Book the survey

Send us the school postcode

The first survey answers three questions at once: what the roofs carry, what they generate against your own timetable, and how a Splott school pays for it. Send the postcode, an approximate roof area per block and half hourly meter data if the school can get it from its supplier.

Lenzie Consulting Ltd arranges the survey and passes your details to an MCS-certified installation partner so they can quote. We do not carry out the installation ourselves.

What Splott trusts ask before a survey

How much would a school roof near Splott generate?: Modelled at 908 kWh per kWp a year, a 320 kWp array on a Splott school models at about 290,600 kWh. What decides whether that is worth doing is how much of it the school uses itself, not the total. Source: EU PVGIS v5.2, SARAH3, 10 degree pitch, 14 percent system loss.
How does the school year affect a solar scheme in Splott?: It goes somewhere, just not into the timetable. 25.6 percent of the Splott generating year lands in July and August (EU PVGIS v5.2), and a closed school still runs servers, comms, refrigeration and ventilation. What that base load does not absorb is either stored, exported under an agreement with National Grid Electricity Distribution, or a sign the array is too big. We test which before sizing anything.
Which network operator handles the connection at a Splott school?: Applications go to National Grid Electricity Distribution, who run the network across Cardiff. An export-limited offer usually still works for a school, because the summer surplus is the part you were least likely to be paid much for anyway.