Maximising a 30kW array

[allthatpebbledash]

Hello.

 

Part of my house plans is to construct whole new roofs. The back of the house faces almost directly south. I am at the edge of town. No shading anywhere in sight and most likely never will be. The solar potential is huge I’m told and don’t need any micro inverters, and can probably get away with installing one array on a single string reducing cable runs.

 

Winter sees the sun covering the entire rear of the house moving across the horizon nicely left to right.

 

Summer sees the sun rising at front left, moving around to the back and almost directly over the house, all way round to the front right as it starts to set.

 

I’d much rather cover my new roofs in PV panels than tiles.

 

Main roof facing south slope is 20 degrees.

Extension roof facing south slope is 12.5 degrees.

 

The other side of the main roof slope facing north is 35 degrees but I intend to keep that panel free for street appearance and it isn’t as large area wise.

 

I can comfortably fit 44 Viridian panels on the main, and 33 Viridian panels on the extension. This is as a starting point using 445W panels, can shop around but I do want an inline system.

 

I have prior DNO approval to have a 55kV 3 phase upgrade of incoming supply.

 

I am with Octopus on a basic tariff, can’t remember which, but told I can change later after I move in to the house as the property is currently vacant pending works so just standing charges at the minute.

 

I have mains gas. I want a system boiler setup to heat via wet UFH and provide DHW. No electric showers, they will run off cylinder/s.

 

I have a dedicated plant room to accommodate everything at ground floor level. Size approx 4x5m.

 

I do not want an ASHP.

 

I want some sort of battery. Can’t decide what size.

 

Most likely won’t go MCS certified route to install due to costs. And due to Octopus giving increased payments non MCS.

 

I will probably install 1 or 2 22kW EV chargers too. No EV yet, but post moving in I plan on upgrading the daily’s.

 

I’ve attached my PVGIS data, but I don’t know what to make of it.

 

I’m being advised the DNO will most likely allow a higher export limit beyond the 11.04kW of 3 phase I believe.

 

I’m being told I should oversize my water cylinder, dump excess solar into that. Forget batteries.

 

Im also being told I should simply export everything that’s not being used, and only have a small battery to keep power to run for 24 hours usage. It’s a waste to go any bigger.

 

I’m also advised I should forget getting 2 arrays, and focus on the main larger one only.

 

How would you make the most of this situation?

 

 

[Nickfromwales]

MCS is required to sell your excess, and you’ll have a LOT of it. 
 

You should defo consider an ASHP, as your winter generation would offset gas use.

 

Split roof elevations mean 2 strings.

 

You can only use so much hot water, and electricity, so where’s the maths to justify this enormous array?

 

Asking for a friend   

[JohnMo]

12 minutes ago, allthatpebbledash said:

I’m being told I should oversize my water cylinder, dump excess solar into that. Forget batteries

If you can get export via octopus, dumping to water makes zero sense. But you need to ensure you have all the correct paperwork to get octopus to approve system without MCS certificate.

 

Even if you did dump to water in summer you will have a hot cylinder by 9 to 10am with that size array.

 

14 minutes ago, allthatpebbledash said:

do not want an ASHP

So you will get loads of electrical energy from PV, then pay for gas to heat your house. Not really making sense.

 

11 minutes ago, Nickfromwales said:

Split roof elevations mean 2 strings

Yep, plus that many panels, will drive multiple strings on each elevation.

 

18 minutes ago, allthatpebbledash said:

Viridian

Not the cheapest system. 

 

You need to do some maths, cost of panels and mounting system?

 

edge uplift on roof for your wind load area, is full roof acceptable?

 

Massive oversized makes sense if you can actually sell it.

 

Gas or oil boiler makes zero sense, not what you want to hear, but...

 

25 minutes ago, allthatpebbledash said:

I’ve attached my PVGIS data, but I don’t know what to make of it

Blue graph is a monthly output average to expect in kWh. So some days will get very good output, wet rainy day almost nothing. Peak generation is in summer when you can use almost nothing you produce, so you export it, zero justification to have that many panels if you have no export agreement.

 

[-rick-]

Octopus offer the 15p export price but AFAIK they have a limit of 15kW of solar, so not sure they will offer that price to you if you go over that. That might make a 15kw export array more profitable than a 30kw one. Not sure about other providers and their pricing. Even if you do get the price then don't bank on it being around for the life of the array so don't base financial decisions on it.

 

Agree that a large hot water tank isn't worth much given the amount of electricity you plan to generate. If you do a lot of miles via EV then being able to use the solar for that would be a big win, but then you have to wonder if the cars will be at home plugged in when the sun is up. If you get the 15p export rate an can export as much as you want it's no big deal as you can export at 15p and import overnight at 7p. But if you are export limited then batteries would help you time shift, but that is going to be very expensive and it probably doesn't make financial or environmental sense (to have sufficient batteries to charge cars).

 

Your roof angles are also not particularly well optimised for winter generation, where excess panels can be really helpful. PVGIS is showing you that in winter you don't generate much. It may be worth considering if there is a location on your property that is not the roof that might better suit at least some of the panels. If you can angle at 45/50^o you could generate a lot more in winter, which would likely be more useful long term (especially if export pricing during summer months becomes less attractive).

[Dillsue]

Research the Octopus non MCS scheme as it was originally a trial with limited numbers so be sure you can get onto it before finalising plans. Octopus offer the same 15p non SEG export rate whether you're MCS or non MCS but that rate doesn't come with any guarantee of continuation. 

 

If you're set on gas CH/DHW then you're going to have alot of spare eleccy. If you get the Octopus 15p rate then I wouldn't bother with an immersion diverter but just let all spare eleccy go to export at 15p/unit and heat your water/charge your EVs overnight on an 8p/unit off peak rate.

 

Before you do anything, check with your DNO what your max export is going to be. No point in having a big array but having to throttle it because you can't export it. 

[allthatpebbledash]

17 hours ago, Nickfromwales said:

MCS is required to sell your excess, and you’ll have a LOT of it. 
 

You should defo consider an ASHP, as your winter generation would offset gas use.

 

Split roof elevations mean 2 strings.

 

You can only use so much hot water, and electricity, so where’s the maths to justify this enormous array?

 

Asking for a friend   

Maths is that PV panels are cheaper to cover a roof than tiles. According to Octopus, you don’t need MCS to export now. Not done a deep dive in their T&Cs but have read on another reputable forum it’s a case of fill in a form, pay a small fee with relevant certificates, and you’re good to export.

 

17 hours ago, JohnMo said:

If you can get export via octopus, dumping to water makes zero sense. But you need to ensure you have all the correct paperwork to get octopus to approve system without MCS certificate.

 

Even if you did dump to water in summer you will have a hot cylinder by 9 to 10am with that size array.

 

So you will get loads of electrical energy from PV, then pay for gas to heat your house. Not really making sense.

 

Yep, plus that many panels, will drive multiple strings on each elevation.

 

Not the cheapest system. 

 

You need to do some maths, cost of panels and mounting system?

 

edge uplift on roof for your wind load area, is full roof acceptable?

 

Massive oversized makes sense if you can actually sell it.

 

Gas or oil boiler makes zero sense, not what you want to hear, but...

 

Blue graph is a monthly output average to expect in kWh. So some days will get very good output, wet rainy day almost nothing. Peak generation is in summer when you can use almost nothing you produce, so you export it, zero justification to have that many panels if you have no export agreement.

 

 

Yes I agree, which is why the size of arrays would be finalised once I get DNO confirmation of what I can export.

 

According to Octopus, they have no export limit on their end either.

 

I get what you’re saying about loads of electric generated, but that’s during the summer when I’ll need an ASHP less. During the winter months, I need it to run most and that’s when I’ll produce less electric myself and have to import at extortion rates. It also makes very little sense unless you have an airtight and well insulated house to get an ASHP running efficiently with an MVHR system in tandem.

 

Summer. Use electric produced or stored in batteries overnight to heat DHW in well insulated cylinder or two. Rest of electric to run majority of appliances. No ASHP needed. Bills low.

 

Winter. Use electric produced or stored in batteries overnight to heat DHW in well insulated cylinder or two. Also run CH. If needed, gas boiler runs on demand to desired temps. No ASHP needed. Bills low.

 

IMO gas will never be more expensive than electric. Ergo, no ASHP needed.
 

17 hours ago, -rick- said:

Octopus offer the 15p export price but AFAIK they have a limit of 15kW of solar, so not sure they will offer that price to you if you go over that. That might make a 15kw export array more profitable than a 30kw one. Not sure about other providers and their pricing. Even if you do get the price then don't bank on it being around for the life of the array so don't base financial decisions on it.

 

Agree that a large hot water tank isn't worth much given the amount of electricity you plan to generate. If you do a lot of miles via EV then being able to use the solar for that would be a big win, but then you have to wonder if the cars will be at home plugged in when the sun is up. If you get the 15p export rate an can export as much as you want it's no big deal as you can export at 15p and import overnight at 7p. But if you are export limited then batteries would help you time shift, but that is going to be very expensive and it probably doesn't make financial or environmental sense (to have sufficient batteries to charge cars).

 

Your roof angles are also not particularly well optimised for winter generation, where excess panels can be really helpful. PVGIS is showing you that in winter you don't generate much. It may be worth considering if there is a location on your property that is not the roof that might better suit at least some of the panels. If you can angle at 45/50^o you could generate a lot more in winter, which would likely be more useful long term (especially if export pricing during summer months becomes less attractive).

Can’t do much about the roof pitches at the back, it was either a small slope at 35 degree angle or a much larger slope at 20 degrees.

 

14 hours ago, Dillsue said:

Research the Octopus non MCS scheme as it was originally a trial with limited numbers so be sure you can get onto it before finalising plans. Octopus offer the same 15p non SEG export rate whether you're MCS or non MCS but that rate doesn't come with any guarantee of continuation. 

 

If you're set on gas CH/DHW then you're going to have alot of spare eleccy. If you get the Octopus 15p rate then I wouldn't bother with an immersion diverter but just let all spare eleccy go to export at 15p/unit and heat your water/charge your EVs overnight on an 8p/unit off peak rate.

 

Before you do anything, check with your DNO what your max export is going to be. No point in having a big array but having to throttle it because you can't export it. 

 

Thanks. That sounds like the plan of action. Confirm max export. Then based on that size arrays accordingly. Use what I need, charge cars etc, then export at max rate. Import overnight for DHW and batteries, and gas on demand for CH during the colder days.

[JohnMo]

10 minutes ago, allthatpebbledash said:

also makes very little sense unless you have an airtight and well insulated house to get an ASHP running efficiently with an MVHR system in tandem.

Sorry that is the press speaking and you listening. Size emitters and heat pumps correctly you should be a seasonal efficiency of 4 on any building.

 

12 minutes ago, allthatpebbledash said:

import at extortion rates

I pay 15p to import. Get 3x windows a day, on octopus cosy. Fill battery as needed.

 

14 minutes ago, allthatpebbledash said:

IMO gas will never be more expensive than electric.

It already is CoP of 4. 15p per kWh, gives an effective cost of 3.75p. my effective rate last week was 10p per kWh so only 2.5p per kWh less than half the price of gas. A good solar week in winter it's even cheaper.

 

Gas is cost per kWh and a boiler efficiency of 85%, so 15% more than you actually pay, plus a stand charges.

[Dillsue]

30 minutes ago, allthatpebbledash said:

Summer. Use electric produced or stored in batteries overnight to heat DHW in well insulated cylinder or two. Rest of electric to run majority of appliances. No ASHP needed. Bills low.

 

Winter. Use electric produced or stored in batteries overnight to heat DHW in well insulated cylinder or two. Also run CH. If needed, gas boiler runs on demand to desired temps. No ASHP needed. Bills low.

If you're getting paid 15p to export and can import overnight at half that rate, you likely dont want to use any of your own electric to power things you can schedule overnight such as DHW.

[Beelbeebub]

From your OP i assume you haven't actually built (or modified) anything yet.

 

Just FYI: I have a carport with a 12degre pitch covered in solar (24 panels)

 

Initially I thought of using the panels as the rood covering but couldn't find any way to seal between the panels even for a carport standard of weatherproofing.  It was either silicone seals or rubber inserts.  There were a few Chinese systems that used extruded aluminum sections as the rafters and "gutters" to catch the water, but that would not be suitable for a house roof.

 

I then looked at the "in roof" tray systems (GSE trays).  The wooden roof structure needs lots of additional battens to support the trays and IIRC there is a minimum pitch requirement and I don't think 12 degrees was enough though 20degrees might be). The cost of each tray (or pairs as they have split system now) was not insignificant either.

 

I ended up using a trapezoidal metal roof (think out of town retail park roof/cladding) and then rails attached directly to the roof.  It worked really well for construction, 12 degrees is perfectly easy and safe to walk on and the metal roof is nice and robust. 

 

The result looks good and wasn't expensive.

 

I didn't worry about the sub optimal roof pitches, the few % loss of output is compensated for by just having more panels

 

I used a s SolaX IES (single phase) 8kw which could handle up to 16kw of panels.  They do make a 15kw 3 phase unit that will take 30kw of panels 

 

They also make various models that aren't integrated but do have a bit more flexibility on batteries 

 

With 30kwp of panels you will easily be able to provide "baseload" power for your house even on dark days.  My 11kwp panels provide for the 400w base load of my house even on pretty gloomy drizzly days which is about 3.5% of peak.  If you had 30kw you'd be making over 1kw on all but the worst days.  Given you have gas heating I can't think why your house would need more than that (unless you bitcoin mine or run an aluminum smelter as a hobby!)  

 

As for batteries, if you can export alot there is an argument that you don't need huge batteries - the grid acts as your battery.  Even if your import/export prices aren't perfectly aligned (and I don't know how long the Octopus 15p rate will hang about) and you only get (say) half the price for export as you pay for import, it doesn't really matter if you are exporting so much that your export earnings handily pass you import costs.

 

[allthatpebbledash]

2 hours ago, Beelbeebub said:

From your OP i assume you haven't actually built (or modified) anything yet.

 

Just FYI: I have a carport with a 12degre pitch covered in solar (24 panels)

 

Initially I thought of using the panels as the rood covering but couldn't find any way to seal between the panels even for a carport standard of weatherproofing.  It was either silicone seals or rubber inserts.  There were a few Chinese systems that used extruded aluminum sections as the rafters and "gutters" to catch the water, but that would not be suitable for a house roof.

 

I then looked at the "in roof" tray systems (GSE trays).  The wooden roof structure needs lots of additional battens to support the trays and IIRC there is a minimum pitch requirement and I don't think 12 degrees was enough though 20degrees might be). The cost of each tray (or pairs as they have split system now) was not insignificant either.

 

I ended up using a trapezoidal metal roof (think out of town retail park roof/cladding) and then rails attached directly to the roof.  It worked really well for construction, 12 degrees is perfectly easy and safe to walk on and the metal roof is nice and robust. 

 

The result looks good and wasn't expensive.

 

I didn't worry about the sub optimal roof pitches, the few % loss of output is compensated for by just having more panels

 

I used a s SolaX IES (single phase) 8kw which could handle up to 16kw of panels.  They do make a 15kw 3 phase unit that will take 30kw of panels 

 

They also make various models that aren't integrated but do have a bit more flexibility on batteries 

 

With 30kwp of panels you will easily be able to provide "baseload" power for your house even on dark days.  My 11kwp panels provide for the 400w base load of my house even on pretty gloomy drizzly days which is about 3.5% of peak.  If you had 30kw you'd be making over 1kw on all but the worst days.  Given you have gas heating I can't think why your house would need more than that (unless you bitcoin mine or run an aluminum smelter as a hobby!)  

 

As for batteries, if you can export alot there is an argument that you don't need huge batteries - the grid acts as your battery.  Even if your import/export prices aren't perfectly aligned (and I don't know how long the Octopus 15p rate will hang about) and you only get (say) half the price for export as you pay for import, it doesn't really matter if you are exporting so much that your export earnings handily pass you import costs.

 

Thanks for this, pretty informative. Can you share images of how to the sheet metal roof looks with panels fixed on top?

[allthatpebbledash]

On 04/11/2025 at 10:33, JohnMo said:

It already is CoP of 4. 15p per kWh, gives an effective cost of 3.75p. my effective rate last week was 10p per kWh so only 2.5p per kWh less than half the price of gas. A good solar week in winter it's even cheaper.

 

Gas is cost per kWh and a boiler efficiency of 85%, so 15% more than you actually pay, plus a stand charges.

This CoP thing is not really indicative of real world scenarios. It’s the only way to win the argument for ASHPs I seem to find online. Unless you’re looking at these technologies from an environmental or carbon emissions perspective, they make very little sense.

 

Most people I know do the following just before they wake up, boiler on for CH, 2-3 hours, then leave for work and turn off. For an ASHP to do this, it can’t.

 

It needs to be on for a lot longer, practically all the time. Hence why it isn’t really cheaper.
 

There’s the fact to have one installed properly and maintained costs a lot too. Whereas for gas, it’s a very competitive landscape to have installs, servicing, repairs etc.

 

On 04/11/2025 at 10:50, Dillsue said:

If you're getting paid 15p to export and can import overnight at half that rate, you likely dont want to use any of your own electric to power things you can schedule overnight such as DHW.

This makes sense. Thanks.

[JohnMo]

54 minutes ago, allthatpebbledash said:

This CoP thing is not really indicative of real world scenarios.

I monitor mine via certified heat meter and electricity meter. Heat pump has started twice today for central heating and actually ran at average CoP of 5.5. that is real world. Average for DHW cylinder heating today is 3.8.

 

So for a heating today's energy cost with zero pv input is 1.8p per kWh in the real world.

50 minutes ago, allthatpebbledash said:

Most people I know do the following just before they wake up, boiler on for CH, 2-3 hours, then leave for work and turn off. For an ASHP to do this, it can’t.

They don't need to nor do boilers. Blasting heat at a house does two things, it heats the whole airspace and buildings fabric quickly, it also has to make up the heat lost, while the heating has been off. It's also not that comfortable while it's doing it. You generally have to have a warmer space to compensate. Long on periods stabilise the building fabric so you can drip feed energy into the space. Your house can be a few degrees cooler and very comfortable. Overall running energy input on a like for like basis in kWh is similar. Gas boilers get an efficiency improvement of around 10 to 15% so a gas boiler is actually cheaper to run when run in a similar manner to heat pump.

 

Although my heating is on 24/7, it only actually runs when it needs to. Drip feeding heat into the floor to make up for lost energy. So in the last 24 hrs has only fired up 4 times running for about 4 hrs in total. So maybe less hours running than you. So in real world they do work.

 

 

 

 

[Beelbeebub]

3 hours ago, allthatpebbledash said:

Thanks for this, pretty informative. Can you share images of how to the sheet metal roof looks with panels fixed on top?