Nobody wants to touch my roof for solar and alternative location

[SimonD]

I had a friendly solar designer put together a design for me, but when he went out to his usual roofing people he didn't get any joy. Even the electrician he works with that does a lot of solar installs with others hasn't had joy with his contacts.

 

The problem is that I have a roof with 2 curves and it's a standing seam metal roof.

 

The issues are partly:

 

1. The curvature

2. Calculating wind uplift loads to see if they're an issue

 

I installed the roof and know my clipping distances which is fine so I'm going to ping the manufacturers a question and also get in touch with the se who original did the designs, including wind uplift calcs. 

 

The next issue is to deal with the curvature which nobody seems to be able (bothered) to model. And also the issue and cost of scaffolding they want - unfortunately nobody likes my kwikstage scaffolding.

 

If i just went ahead and did this myself, my thoughts are I could use the portion at the top of the roof where it's almost flat together with a flat portion of the roof between the curves.

 

However, I was actually thinking that the easiest solution would be to use the existing garage which has a standard tiled roof. The only problem with this is I've been told there's no point putting solar on a North facing roof (which would be half the installed area). The second is that the south facing roof is quite shaded by the house during the winter months. I'm not so sure this is as much of an issue any more?

 

I really don't know what the current understanding is about shaded and North facing walls for solar and would really like to understand this better. Can someone explain?

 

The advantage of this is that the batteries and infrastructure are going to go in the garage anyway so I think it might be better to have the panels there anyway.

 

Thoughts?

[JohnMo]

22 minutes ago, SimonD said:

North facing

I was looking at our almost North facing roof over the summer and was really surprised by how much sun it actually got.  North facing is seasonal producer, fine in summer, pants in winter - but all PV is pretty pants in winter anyway.  Best way is model it on https://pvgis.com/en

 

We have have loads of shading in general (loads of trees). The way I thought about is panels are cheap (£140 per kWp) so we have quite a few, yield isn't the best, but the number panels helps.

 

Can you do a solar fence or similar or ground array?

 

 

 

 

[Gus Potter]

3 hours ago, SimonD said:

The issues are partly:

 

1. The curvature

2. Calculating wind uplift loads to see if they're an issue

Good post. The following is a bit lengthy but hopefully it will give you some pointers and other BH members an insight into the structural design when fixing the panels to the roof. I've broken down the design into different elements and then recouple (kind of) at the end. 

 

Element 1: The PV panel itself. The general standard is that the panel should be able to with stand a static load of 2400 Pa for one hour. This is roughly 240 kg / square metre, then we apply a safety factor of 1.25 (called a material factor in eurocode design) so 2400 / 1.25 = 1920 Pa ~ 194 kg per square metre of panel. Glass has one at least funny feature in that (general) it is able to withstand a higher short duration load say one or a few seconds. The reason for this is partly due to crack propagation. Thus if your PV panel can withstand the one hour test it has a good chance of resisting a higher short duration load that you might associate with a 1 or 3 second gust of wind. 

 

In the UK there is the potential for the wind loading to exceed the panel strength. This is influenced by for example the wind direction, altitude, topography, shape of the roof, height of the roof, location of dormers and chimneys and importantly if the panels are close to the eaves, verges or ridge. Often neglected is snow drift loading or dumping of snow from a higher roof onto PV panels. Here the snow drift can often exceed the panel strength locally. 

 

Element 2: The fixings of the panel to the roof. Here we are interested in two things. As the fixings are perpendicular to the roof we can split the axial load in the leg of the fixings into two force components. One is perpendicular to the rafter, the other is parallel to the rafter. The latter force is the force that wants to cause the panel to slip down the roof. The former is required as it lets us check the underlying structure. 

 

Element 3: Calculation of the wind loading on PV panels so we can check items 1 and 2 above.

 

Element 4: Check the main roof structure. Here we are looking at two things in the main. Are we adding load to the roof, probably not if the main roof is tiled, but check anyway. PV panels are increasingly light weight cf tiles so we need to check the structure for uplift and check to see for example that any masonry below still works for a reduced dead weight. Brickwork / blockwork often benefits from a bit of dead weight so we need to be careful when we don't have it. If you have say a cantilevered floor that may be relying on dead weight from above, remove that and you can find yourself in trouble. 

 

Motto is, think of the consequences of making a building lighter!

 

 

3 hours ago, SimonD said:

The next issue is to deal with the curvature which nobody seems to be able (bothered) to model.

Simon, I've assumed you have a barrel shaped roof. 

 

The SE starting point here is to calculate the dynamic wind pressure for your roof. This is standard stuff so won't go into this. You might find in the original calcs for your house a value for this that relates to your roof. The interesting bit is how you calculate the wind loading on each of the panels on a barrel shaped roof as the panels are flat and the roof is curved, thus the gap between the panels and roof will vary.  

 

Reference wise the SE may look to BRE digest 489, Wind loads on roof-based photovoltaic systems. Eurocode 4, BS EN 1991-1-4:2005+A1:2010 and the older Bristish Standard BS 6399 part 2 for initial guidance. But then you find that there is no specific guidance on fitting PV to a barrel / curved roof. Which goes a long way towards explaining why no one wants to run with it. 

 

But the guidance gives you a handle on things that you can then do some bespoke design on. 

 

Ok so we have calculated the basic wind speed. What we need to do next is to arrive at the local pressure coefficient for each panel and this is the bit where folk back off. The maths is roughly qs (basic wind speed units are Pascals which we convert to kN/m^2) * local nett pressure coefficient ( non dimensional unit).

 

Example here for PV sitting 100 to 300 mm off the roof.  Take a middle England wind load of 800 Pa / 1000 = 0.8 kN/ sq metre ( about 80 kg / m sq). Now a typical nett pressure coefficient in the middle of the roof might be around -0.6 to -1.3, take the worst... 0.8 * 1.3 = 1.04 kN/m^2, a lot less than the PV rating so looks ok.

 

But move to Wales, parts of Scotland, maybe the Downs, close to the sea, add a bit of altitude, where the qs is 1.1 kN/m^2 plus and you have a PV panel  close to the edge of the roof (nett coefficient of potentially 2.2, here the sum is 1.1 * 2.2 = 2.42 ( 2420 Pa) so we check the panel rating. This is something your normal PV installer should be familiar with and that provides values for the fixing design and underlying roof checks. The high loaded wind zones are away from the roof perimeter so you can see why PV installers want to keep clear of these areas. 

 

But PV on a curved roof will have a varying gap under. They way to approach this is to understand what the guidance means in the above references, distill that down so a lay person can understand.

 

Here goes.

 

I would treat this as a hybrid PV roof for wind design. Start with an even gap under the PV and couple that with the guidance for canopy roofs. Think of a hay shed. In the autumn is gets filled with hay, the farmer takes the hay from the front, the back is blocked, called the blockage factor. The varying gap under the PV panels on a barrel roof is like a blockage factor, add to that the cross section area of the PV panel fixing legs (more blokage) and you can arrive at a reasoned Engineering justification for your selection of the pressure coeffient.

 

One you have that number everything else is pretty much standard design. There are a few extra checks as above and it may be that if you have your PV close to the roof edge or high wind local zones you need some  extra fixings. 

3 hours ago, SimonD said:

However, I was actually thinking that the easiest solution

Nah, give your SE a call, you might find they say, oh we do a lot of this so no problem!

 

Point of note. Once the gap gets over ~ 300mm then the approach might need to change / nuanced. Here we would look to the guidance part in BRE 489 that talks about mounting PV on stands, more like a solar field. 

 

Thanks again for posting. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Edited Sunday at 18:04 by Gus Potter

[SimonD]

4 hours ago, JohnMo said:

I was looking at our almost North facing roof over the summer and was really surprised by how much sun it actually got.  North facing is seasonal producer, fine in summer, pants in winter - but all PV is pretty pants in winter anyway.  Best way is model it on https://pvgis.com/en

 

We have have loads of shading in general (loads of trees). The way I thought about is panels are cheap (£140 per kWp) so we have quite a few, yield isn't the best, but the number panels helps.

 

Can you do a solar fence or similar or ground array?

 

 

 

 

 

Thanks for this. Really useful to know your experience of North facing panels. I'll check out the modelling.

 

We could do fencing. Would quite give us the surface area of the roof but would be directly south facing and without shading during low winter sun.

[Kelvin]

I decided not to fit PV panels to my standing seam roof for a few reasons one of which was aesthetic as it’s a nice looking thing and PV panels aren’t. I fitted 19 panels East/West to the garage roof instead. Likewise the batteries and inverter are in the garage so it made the install much easier. It works really well. Our North facing flat roof gets loads of sun in the summer months and I contemplated using that too but it would like awful so will ground mount any other panels. 

[SimonD]

1 hour ago, Gus Potter said:

The following is a bit lengthy but hopefully it will give you some pointers and other

 

Not at all. It's fascinating. This is another of your stunning posts Gus, so thank you for taking the time. I'm going to enjoy digesting it.

 

1 hour ago, Gus Potter said:

Motto is, think of the consequences of making a building lighter!

 

Ah yes, nobody mentioned that when we started our project, which was supposed to be removing the roof and adding an additional story, as soon as the existing roof with conceprete tiles was removed the walls could just be lifted apart, do I had to rebuild significant portions of it.

 

1 hour ago, Gus Potter said:

Simon, I've assumed you have a barrel shaped roof.

 

Yes, like this:

 

2350D 8 Detail Front Elevation Rev B.pdf

 

1 hour ago, Gus Potter said:

Which goes a long way towards explaining why no one wants to run with it. 

 

Thank you I'm quite relieved to read this. I raised the question during the initial solar design meeting as my experience from the standing seam roof design raised my worries a bit.

2 hours ago, Gus Potter said:

Nah, give your SE a call, you might find they say, oh we do a lot of this so no problem!

 

Haha, yes, will do!

2 hours ago, Gus Potter said:

Thanks again for posting

 No, thank you 🙏 

[SimonD]

41 minutes ago, Kelvin said:

I decided not to fit PV panels to my standing seam roof for a few reasons one of which was aesthetic as it’s a nice looking thing and PV panels aren’t. I fitted 19 panels East/West to the garage roof instead. Likewise the batteries and inverter are in the garage so it made the install much easier. It works really well. Our North facing flat roof gets loads of sun in the summer months and I contemplated using that too but it would like awful so will ground mount any other panels. 

 

That's a good point. The solar designer wanted to lather the entire roof with panels and I was less keen. Due to the aspect of the house sitting on a hill together with the curve of the roof, we can get away with some of the roof without it being visible. For solar gain, it's definitely the best place for it, but there's a long way to cable to the garage.

 

I've clearly got to have more of a think about it.

[Kelvin]

26 minutes ago, SimonD said:

 

That's a good point. The solar designer wanted to lather the entire roof with panels and I was less keen. Due to the aspect of the house sitting on a hill together with the curve of the roof, we can get away with some of the roof without it being visible. For solar gain, it's definitely the best place for it, but there's a long way to cable to the garage.

 

I've clearly got to have more of a think about it.

Took me over a year to decide what to do about our PV installation. I ignored it pretty much until May this year as I needed to make a decision and get it done to get the house signed off. I really wanted to ground mount it and have created a flat area to do that where it will be mostly hidden but it’s not straightforward hence why I put it on the garage roof instead. It was installed mid July and it’s generated 2MWh 

[Gus Potter]

29 minutes ago, SimonD said:

I've clearly got to have more of a think about it.

As a bit of encouragement. Hopefully you followed my last post.

 

The wind loads I started with were based on the design life of the house, in design terms for wind this is 50 years. But PV panels only last say 25 years so there is an argument to be made, why don't we design and check the house structure against the 50 year wind loading (that is safe) but design the pv panels for their life expectancy of 25 years. The fixings would get designed on 50 years as we don't want the panels flying off the roof and hitting someone. Simple solution is often just to add more fixings in the vulnerable areas. 

 

This is all about the probability of us getting the highest wind load. As SE's we design based on the probability of wind occurring and balance that against the risk. For wind calculations we call this the probability of occurrence. Simplistically, if we look at a 25 year wind  return period the basic wind load gets reduced and that can make the panel design fly.

 

Don't through in the towel just yet due to your curved / barrel roof!

 

 

 

 

[Gus Potter]

57 minutes ago, SimonD said:

Yes, like this:

That radius of roof looks quite large. Draw a line of your predicted panel lengths, the gap between on the roof side you want. Do you think you can achieve a minimum gap of 100mm and a maximum gap of 300mm between the roof surface and the underside of the panels? This would make calculation easier and pretty much in line with the digest BRE 489 with some consideration to the blocking factor due to the diminishing gap and cross section area of the fixings. The quasi parapet at the top is an advantage.

 

I would do a quick check on any snow drift but the parapet is not high so don't think it would prove significant. In any event snow drift is what we call an accidental load so the saftey factors are less onerous.

 

I think the main challenge is to make sure the panels look ok Architecturally and that you get the flashing details robust.

 

 

 

[saveasteading]

Wind wise as @Gus Potter says , a good SE won't be phased by this, or the snow buildup.

 

I have heard of bad installations through standing seam though.

Your average contractor doesn't know about SS.

Is yours fitted by snapping onto brackets that are fixed to the roof?

  Is it floating? Ie Allowed to expand and contract?

More later.

 

 

 

 

 

 

[Kelvin]

1 hour ago, saveasteading said:

Wind wise as @Gus Potter says , a good SE won't be phased by this, or the snow buildup.

 

I have heard of bad installations through standing seam though.

Your average contractor doesn't know about SS.

Is yours fitted by snapping onto brackets that are fixed to the roof?

  Is it floating? Ie Allowed to expand and contract?

More later.

 

 

 

 

 

 

Two companies I spoke with were a little reticent about fixing it to a standing seam roof when I spoke them. There are fixing  systems to do it. They were much more comfortable fixing to our metal workshop roof as it’s more bread and butter for them given all the industrial buildings they fix the panels to. I have zero penetrations through my house roof and glad of that. 
 

 

[Beelbeebub]

given how nice your roof looks, the various hurdles fitting to it and the fact you have to option of garage/ground mount. I would suggest spending the money you would save on calculations, specialists and scaffolding on the garage/ground option.  Panels are so cheap now that it's much less important to get the maximum efficiency out of their orientation, just stick more panels on to compensate.  

[JohnMo]

57 minutes ago, Beelbeebub said:

just stick more panels on to compensate

And optimisers where they are likely to get periods of shade over part of the array. Or the shade bits on their own MTTP. 

[saveasteading]

3 hours ago, Kelvin said:

have zero penetrations

Great . That's the entire point of standing seam, with the appearance becoming a sales thing later.

It would be essential to fix through to the structure.

Hence my concern.  Cladding flies off or holes in it.

 

So I agree that you retain the house roof as intended and forgo solar on it.

 

 

[Alan Ambrose]

FYI ArtisanElectrics have a video where they put PV on the curved metal roof of their new office. Probably trapezoidal rather than standing seam.

 

 

[saveasteading]

Yes that's trapezoidal. 

 

I have only flicked through this but it looks as if they are fixing only to the crowns of the panels.

The steel will be 1mm thick or so.

 

I hope they have an approved spec for the right self tappers into that panel, or it could end up on the car park, come Storm Sod.

[sgt_woulds]

On 27/10/2025 at 11:08, Kelvin said:

Two companies I spoke with were a little reticent about fixing it to a standing seam roof when I spoke them. There are fixing  systems to do it. They were much more comfortable fixing to our metal workshop roof as it’s more bread and butter for them given all the industrial buildings they fix the panels to. I have zero penetrations through my house roof and glad of that. 
 

 

 

 

You need to find a company that has experience with commercial installs where standing seem is more common. 

 

Unfortunately it is rare to find a commercial installer who will aslo do domestic as they are generally flat out with commercial installs and have no spare capacity or inclination to fit in small domestic jobs.

 

The company I used to work for could possibly still do this, depending upon location and install capacity - assuming I'm allowed to whisper their name here.

 

But last time I spoke to them they had a six month wait for a survey, let alone an install...

[saveasteading]

13 minutes ago, sgt_woulds said:

experience with commercial installs where standing seem is more common. 

Unfortunately it doesn't follow that they understand what standing seam is, and the measures required. The consequences of a bad installation could be dire.  But some good ones will know and do it appropriately.

[Gus Potter]

On 27/10/2025 at 11:23, Beelbeebub said:

I would suggest spending the money you would save on calculations,

But I posted a long rant on how you do it as an SE, it's not that difficult! If another SE reads my posts all they have to do is look up the references I give, make their own judgement on the geometry of the roof and off you go! It's 3 hours work checking / learning that I'm not talking shite (call that research that will add add to your portfolio) and looking at your drawings. Then you go back to the PV designer to check their brackets are ok, do the rest of the checks for due dilligence. 

 

The key is to identify the pressure coefficients. For the house design the SE has probably worked out all the wind loadings already!

 

9 hours ago, saveasteading said:

Unfortunately it doesn't follow that they understand what standing seam is, and the measures required.

This sums it up plus, as I said previously, the PV needs to look good Architecturally, you can't have a big unsightly gap at the ends for example. The detailing and weather proofing needs to be spot on. 

 

On 26/10/2025 at 14:16, SimonD said:

The problem is that I have a roof with 2 curves and it's a standing seam metal roof.

Keep asking your welcome questions so you can make an informed judgement. 

Edited Tuesday at 23:36 by Gus Potter

[saveasteading]

For others on here. I offered standing seam but always told the client the pitfalls and cost difference.

 

Ie that there must be zero penetrations. That fixing solar panels would be challenging and any future damage would be difficult to repair.

Advantages?  The absence of screws esp if badly fitted. Appearance. Curves and not possible with all trapezoidal cladding.

 

As it happened, every client (hundreds) chose screwed trapezoidal. But none was domestic.

So I never did one and can't advise on any surprise benefits or problems.

[-rick-]

On 28/10/2025 at 08:15, saveasteading said:

Yes that's trapezoidal. 

 

I have only flicked through this but it looks as if they are fixing only to the crowns of the panels.

The steel will be 1mm thick or so.

 

I hope they have an approved spec for the right self tappers into that panel, or it could end up on the car park, come Storm Sod.

 

Was looking at solar fixings recently and this seems to be the standard approach now for trapezoidal. Surprised me a bit as it seems weak but I guess good for preventing leaks. MCS approved.

[saveasteading]

44 minutes ago, -rick- said:

seems weak

It's easy for them , being the high points and in a plane.

If they use the right screws with a coarse thread then the pull-out force is good. And to avoid damaging the panel simply use more screws to more crowns, to spread the load.

 

I don't know if installers know all this, and you can be sure there are cowboys aplenty.

I guess though that the suppliers of the kit are learning better standard solutions through experience. 

 

44 minutes ago, -rick- said:

preventing leaks.

Reducing certainly. 

[-rick-]

12 minutes ago, saveasteading said:

It's easy for them , being the high points and in a plane.

If they use the right screws with a coarse thread then the pull-out force is good. And to avoid damaging the panel simply use more screws to more crowns, to spread the load.

 

Saw rivets mentioned as the preferred option with screws the more standard one.