Roof strap conundrum

[Roger440]

Im doing some work on the roof of my house.

 

However, its come to light that there appears to be nothing holding the roof down on the extension. There were a couple of straps nailed to a rafter and down into the wall plate. Which clearly isnt much use when the nails are in the vertical plane! They just pulled off by hand.

 

The work has introduced ventilation via the eaves that was previously blocked off. Which was causing some condensation isues unsurprisingly.

 

The house is reasonably exposed, ive introduced ventilation, therefore wind into the roof, and theres little holding it down. Which is sub optimal.

 

The front half of the house has 2 massive purlins that are embeded into 2 ft thick stone gables. So theres no realistic risk of the roof being pulled upwards.

 

The extension roof is at 90 degrees to the original part and all tied in and nailed together.

 

Ive been up into the gable end of the extension, put a strap on off the ridge beam onto the gable wall, and, addtionally, the end rafter has diaganols down to the ceiling joist. Ive bolted these to the gable wall too. 

 

So, im happy the gable end is pretty secure, the other end is well tied into the existing roof.

 

Its the bit in the middle thats troubling me. The wall plate is on the outer leaf with birdsmouth to clear the inner leaf. No straps to be seen. My current thougts for retrospective installation is on the inside wall and screw to rafter . The ceiling has the 45 degree slopy bit on the inside so can do all of it from inside.

 

However, as one might imagine, this involves making a mess of the bedroom, and so there is pressure to not do so.

 

Am i over worrying?

 

The roof is quite lightweight, 75mm rafter and composite roof tiles.

 

Thoughts please?

[Mr Punter]

Is the wall plate strapped down?  If so, you may only need a truss clip or structural screw to fix each rafter / truss to the wall plate.

[Roger440]

2 hours ago, Mr Punter said:

Is the wall plate strapped down?  If so, you may only need a truss clip or structural screw to fix each rafter / truss to the wall plate.

No, its not. Thats my issue.

 

 

[ETC]

Strap the wall plate down with 30mmx5mmx1000mm long wall plate straps at maximum 2.0m centres.

Install gable and ceiling straps at the same centres across three rafters/ceiling joists.

[Roger440]

1 hour ago, ETC said:

Strap the wall plate down with 30mmx5mmx1000mm long wall plate straps at maximum 2.0m centres.

Install gable and ceiling straps at the same centres across three rafters/ceiling joists.

 

I may not have explained myself well.

 

The house is already built, this bit 40 years ago. 

 

I cant access the wall plate, at least not without removing some roof, and even then, it would have to be on the outside of the wall.

 

Yes, its true i could see the top of the wall plate whilst doing the work, but that didnt really help as id still have the issue of the vertical part of the strap being visible.

[Nickfromwales]

Looks like you'll end up popping a few tiles off, and then drilling down with a 600mm sds bit at 10mm dia, and then resin bolting the wall pate down.

 

You know the only other option is to cut some slices out of the interior......

[Gus Potter]

On 13/09/2025 at 15:38, Roger440 said:

However, its come to light that there appears to be nothing holding the roof down on the extension.

 

On 13/09/2025 at 15:38, Roger440 said:

The house is reasonably exposed

Ok you say reasonably exposed. The vulnerable points are the eaves, verges and roof ridge. Say you don't live in the north of Scotland, Islands. Very quick sum and to cut a lot of the maths out. The wind load uplift will be about 100 - 150 kg/ m^2 (working load on a 50 year return period) at the exposed edges of the roof. But as you move towards the middle of the roof it will be less as an area average., the design codes call this roughly "non simulataneous action" as the wind is made up of vortices, big and small, that don't act on a roof evenly. 

 

The original part of the house is probably ok?  How old is it, if it's been ok up till now then is that ok? 

 

Think about this another way. On new build stuff as an SE I design to the codes but when assessing existing historic houses we need to be pragmatic. 

 

If the wind gets up it often blows the tiles / slates off at the edges of the roof; verges, eaves or around chimney stacks. There used to be guidance on this in the old building codes. 

 

On 13/09/2025 at 15:38, Roger440 said:

The front half of the house has 2 massive purlins that are embeded into 2 ft thick stone gables. So theres no realistic risk of the roof being pulled upwards.

Yes, but the quality of roof tiling workmanship would have been good. The key here is to not let tiles get blown off in the first place as as soon as that happens you lose the dead weight resisting the wind uplift. And here the quality of the extension roof needs to be examined.

 

This is obvious.. if you prevent the tiles / slates getting blown off at the edges of the roof and thus prevent progressive peal back of the tiles / slates then you, maintain the dead weight which resists the uplift. Your big purlins will not be contributing much unless the ends are well strapped down to the gables. 

 

Go back to basics and think.. what has been changed in terms of wind loading. The extension is the obvious thing. 

 

 

 

[Spinny]

Simpson strong-tie do a lot of ties and brackets. I found the MTS30 useful and capable of being covered by plaster. Services of a structural engineer useful for loadings though.

 

https://www.strongtie.co.uk/en-UK/products/medium-twist-strap-mts

[Roger440]

On 17/09/2025 at 00:21, Gus Potter said:

 

Ok you say reasonably exposed. The vulnerable points are the eaves, verges and roof ridge. Say you don't live in the north of Scotland, Islands. Very quick sum and to cut a lot of the maths out. The wind load uplift will be about 100 - 150 kg/ m^2 (working load on a 50 year return period) at the exposed edges of the roof. But as you move towards the middle of the roof it will be less as an area average., the design codes call this roughly "non simulataneous action" as the wind is made up of vortices, big and small, that don't act on a roof evenly. 

 

The original part of the house is probably ok?  How old is it, if it's been ok up till now then is that ok? 

 

Think about this another way. On new build stuff as an SE I design to the codes but when assessing existing historic houses we need to be pragmatic. 

 

If the wind gets up it often blows the tiles / slates off at the edges of the roof; verges, eaves or around chimney stacks. There used to be guidance on this in the old building codes. 

 

Yes, but the quality of roof tiling workmanship would have been good. The key here is to not let tiles get blown off in the first place as as soon as that happens you lose the dead weight resisting the wind uplift. And here the quality of the extension roof needs to be examined.

 

This is obvious.. if you prevent the tiles / slates getting blown off at the edges of the roof and thus prevent progressive peal back of the tiles / slates then you, maintain the dead weight which resists the uplift. Your big purlins will not be contributing much unless the ends are well strapped down to the gables. 

 

Go back to basics and think.. what has been changed in terms of wind loading. The extension is the obvious thing. 

 

 

 

 

Good info as always.

 

The thing that has changed is the introduction of ventilation into the roof space. It didnt have any before, and was pretty much sealed up all round. And the extension is, all in all, very lightweight for a roof. 75mm rafters with composite tiles. So wind uplift would, by all accounts be greater than before.

 

With a storm incoming, i decided to put some straps up the bedroom wall onto the rafters, in addtion to my other efforts. Every third rafter. Which is closer than required by BC as i understand it, but im onto the rafters at 45 degrees rather than a wall plate at 90. So obviously not as effective.

 

Looks horrendous, but as i will be doing further work to that room, including insulating, it will, in time, get covered up.

 

Storm came and went and i still have a oof, so thats a positive! But it was, in the end, only 55mph, so not much of a storm really.

 

 

[Gus Potter]

On 21/09/2025 at 13:58, Roger440 said:

With a storm incoming, i decided to put some straps up the bedroom wall onto the rafters, in addtion to my other efforts. Every third rafter. Which is closer than required by BC as i understand it, but im onto the rafters at 45 degrees rather than a wall plate at 90. So obviously not as effective.

Well done, bit of belt and braces works.. it would look a lot worse if the roof blows off!

 

You mention 45 deg roof slope, these in general tend to be less vulnerable than roofs with a 5 -10 degree pitch for wind uplift. Much depends on the wind direction. 

 

For a bit of fun and to cut short the wind calculation we first calculate the site wind speed and then convert that using factors for the height of the building and orientation and so on. The worst case wind load often comes from 240 degrees and in terms of roof orientation we look at how the roof faces and apply a direction factor. I screen shotted some tables from the British standard that lets you see how we reduce the wind load depending on which face of the building / roof we are considering and the wind direction. NOTE here I'm not considering a quatering wind like the Citicorp building as has featured on BH... and I'm assuming the topography is reasonable benign.. ie you are not at the top of a cliff or on the upperside of a hill. But it's an old house you've got so let keep things simple and qualitative.  

 

In the above you can see that a 240 degree wind gets no reduction but a North wind gets reduced by a factor of 0.78. 

 

Now if you have a look on U tube ( and I've have seen myself) at videos of roofs getting blown off, intially you often see some rippling movement at the edges / verges or ridge of the roof.. it's often tiny but here the high local suction lifts an edge, the wind gets into the roof which increases the roof internal pressure that cause the drama of the whole roof flying off. To stop that we need to keep the vulnerable edges and occasionally bits around chimneys well tied down. 

 

Roughly to calculate the wind load on small parts of the roof we take the wind pressure and multiply it by a local coefficient to give the local pressure, in this case we are interested in the local suction loads. 

 

Below is a diagram that shows the wind directions and the notation for a due pitched roof.

 

 

The above lets you understand the table below which shows the external coefficients that are applied to the wind loads for a duo pitch roof. 

 

In the table above the positive values indicate wind pressure.. a downwards load on the roof zone.. which is beneficial. The negative values are the suction coefficients. The wind load in the tables above is calculated as being applied perpendicular to the plane of the rafter (snow loads and roof access loads are on plan load so applied differently) thus the we need to calculate the vertical uplift component of load when designing the holding down straps. We take the perpendicular load and divide it by the square root of 2  = 1.41. 

 

One thing you can easily take away from the above table 10 is that flatter roofs are more onerously loaded in suction than a roof in the region of 45 degees. But the 45 degree roofs are still vulnerable when the wind is blowing on the gables or quartering to some extent. 

 

On 21/09/2025 at 13:58, Roger440 said:

The thing that has changed is the introduction of ventilation into the roof space.

If you have a study of this Roger it may hopefully inform and you can have a bit of fun looking at the values. The roof ventilation effective area is probably small, call it more trickle ventilation and thus is unlikely to dominate the internal roof wind pressure which gets added to the roof external suction. One word of caution again, if you have some significant topography then this can be significant. From time to time if say building near a big farm shed we may need to look at any wind funneling effects, but this mostly applies to walls. 

 

 

 

 

Edited September 22 by Gus Potter

[Roger440]

2 hours ago, Gus Potter said:

Well done, bit of belt and braces works.. it would look a lot worse if the roof blows off!

 

You mention 45 deg roof slope, these in general tend to be less vulnerable than roofs with a 5 -10 degree pitch for wind uplift. Much depends on the wind direction. 

 

For a bit of fun and to cut short the wind calculation we first calculate the site wind speed and then convert that using factors for the height of the building and orientation and so on. The worst case wind load often comes from 240 degrees and in terms of roof orientation we look at how the roof faces and apply a direction factor. I screen shotted some tables from the British standard that lets you see how we reduce the wind load depending on which face of the building / roof we are considering and the wind direction. NOTE here I'm not considering a quatering wind like the Citicorp building as has featured on BH... and I'm assuming the topography is reasonable benign.. ie you are not at the top of a cliff or on the upperside of a hill. But it's an old house you've got so let keep things simple and qualitative.  

 

In the above you can see that a 240 degree wind gets no reduction but a North wind gets reduced by a factor of 0.78. 

 

Now if you have a look on U tube ( and I've have seen myself) at videos of roofs getting blown off, intially you often see some rippling movement at the edges / verges or ridge of the roof.. it's often tiny but here the high local suction lifts an edge, the wind gets into the roof which increases the roof internal pressure that cause the drama of the whole roof flying off. To stop that we need to keep the vulnerable edges and occasionally bits around chimneys well tied down. 

 

Roughly to calculate the wind load on small parts of the roof we take the wind pressure and multiply it by a local coefficient to give the local pressure, in this case we are interested in the local suction loads. 

 

Below is a diagram that shows the wind directions and the notation for a due pitched roof.

 

 

The above lets you understand the table below which shows the external coefficients that are applied to the wind loads for a duo pitch roof. 

 

In the table above the positive values indicate wind pressure.. a downwards load on the roof zone.. which is beneficial. The negative values are the suction coefficients. The wind load in the tables above is calculated as being applied perpendicular to the plane of the rafter (snow loads and roof access loads are on plan load so applied differently) thus the we need to calculate the vertical uplift component of load when designing the holding down straps. We take the perpendicular load and divide it by the square root of 2  = 1.41. 

 

One thing you can easily take away from the above table 10 is that flatter roofs are more onerously loaded in suction than a roof in the region of 45 degees. But the 45 degree roofs are still vulnerable when the wind is blowing on the gables or quartering to some extent. 

 

If you have a study of this Roger it may hopefully inform and you can have a bit of fun looking at the values. The roof ventilation effective area is probably small, call it more trickle ventilation and thus is unlikely to dominate the internal roof wind pressure which gets added to the roof external suction. One word of caution again, if you have some significant topography then this can be significant. From time to time if say building near a big farm shed we may need to look at any wind funneling effects, but this mostly applies to walls. 

 

 

 

 

 

Thanks for the explanation.

 

If im honest, its probably beyond my comprehension from a maths perspective.

 

The wind table is that where the wind is coming from or going to?

 

The house is in a sizeable "hole" with big hills all around. Prevailing wind is, as normal (im in mid wales) from the south west, shielded by a massive bank of conifers, my barn, and behind that a 400ft high hill. Still 900ft up though.

 

But, im not really able to interprete your excellent information to any useful extent.

 

 

[saveasteading]

27 minutes ago, Roger440 said:

where the wind is coming from

From. so 180 is south, 270 is west .  as you say the prevailing wind is SW so 225.

 

29 minutes ago, Roger440 said:

in a sizeable "hole" with big hills all around.

That could be a bit sheltered or cause funnelling. But the design guides allow for that, ie not allowing any reduction for real or perceived shelter.

 

 

On 13/09/2025 at 15:38, Roger440 said:

Am i over worrying?

No, this is all very sensible.   If you keep to standard detailing all will be good.  As Gus says, the exposed edges are most vulnerable rather than the whole roof. After gales you will often see loose bits of roof hanging off, but seldom an area of roof.

On the steel buildings I worked with mostly. there were standard details for fixings and the eaves corners received twice as many screws as the mid sections.