Warm v's Ventilated Roof - Which Way to go ?

[gaschick]

Thanks @Bobthebuilder2 @Tempand others for this very useful thread. We have an alternative design for our ventilated roof which I was seeking some help with. We are also having a standing seam roof Greencoat PLX on a timber frame single storey extension. The 18mm plywood onto which the greatcoat is being fitted is bing supplied and fitted by the timber frame engineers, so we can't realistically have the extra sets of battens on top to form an extra ventilation cavity. Picture enclosed of the roof deign to which Roe are working.

I was hoping to be able to fill the space between the rafters with wood fibre insulation eg streico flex, rather than the standard celotex to maximise the thermal delay. We don't want to end up with a noisy caravan like roof!

 

No battens are shown in the design though, unlike  all of the others. Should these be fitted to the underside of the plywood to keep the cavity and stop insulation pushing up next to the plywood?

 

Anyone used a wood fibre insulation in this situation? Should we just stick to the celotex, realising that this will be much harder to fit tight between the rafters?

[JohnMo]

Not read the rest of the thread, but just took a look at the drawing, notice you have 200mm centres on the rafters, which would mean 153mm wide strips of insulation between.  Would the insulation value meet building regs?

 

Is this a typo?

 

[ETC]

Warm flat roof. More height, no ventilation required and will look great.

[gaschick]

@JohnMo I have no idea, I am not an expert in this area. I am a doctor, not a building control officer, architect or builder! This is why we are paying people to design the structure, and do the structural calcs etc for it. As a reasonably intelligent person though, I try to do due diligence as best as I can about the things we are paying s''t loads of money for (much at the annoyance of my husband). I have enclosed the timber frame section through the roof. 

[gaschick]

[Russell griffiths]

Your roof WILL be noisy. 
any vaulted roof without a loft space will be noisy. 
you literally have 300mm between you and outside, you would need some kind of miracle insulation for it to not be noisy. 
BUT you need to get this in context my roof is 300mm deep full of rock wool insulation with a pir board underneath, green coat standing seam on top when it rains I can here it without doubt. Can I here it over the noise of the radio, no I can’t. 
if I turn the radio off then most definitely there is a dull tapperty tap. 
but is it annoying ? Not to me, I quite like it. 
but is your vaulted metal roof going to be quite, NO. 

[gaschick]

53 minutes ago, Russell griffiths said:

Your roof WILL be noisy. 
any vaulted roof without a loft space will be noisy. 
you literally have 300mm between you and outside, you would need some kind of miracle insulation for it to not be noisy. 
BUT you need to get this in context my roof is 300mm deep full of rock wool insulation with a pir board underneath, green coat standing seam on top when it rains I can here it without doubt. Can I here it over the noise of the radio, no I can’t. 
if I turn the radio off then most definitely there is a dull tapperty tap. 
but is it annoying ? Not to me, I quite like it. 
but is your vaulted metal roof going to be quite, NO. 

@Russell griffiths It's a damn site thicker than the old roof we had on which was literally one layer of cedar shingles, a layer of silver foil, and a layer of plasterboard!

I am quite looking forward to the sound of the rain gently drumming on the roof as at least it shouldn't leak like the last one! 

[Temp]

4 hours ago, gaschick said:

Thanks @Bobthebuilder2 @Tempand others for this very useful thread. We have an alternative design for our ventilated roof which I was seeking some help with. We are also having a standing seam roof Greencoat PLX on a timber frame single storey extension. The 18mm plywood onto which the greatcoat is being fitted is bing supplied and fitted by the timber frame engineers, so we can't realistically have the extra sets of battens on top to form an extra ventilation cavity. Picture enclosed of the roof deign to which Roe are working.

I was hoping to be able to fill the space between the rafters with wood fibre insulation eg streico flex, rather than the standard celotex to maximise the thermal delay. We don't want to end up with a noisy caravan like roof!

 

No battens are shown in the design though, unlike  all of the others. Should these be fitted to the underside of the plywood to keep the cavity and stop insulation pushing up next to the plywood?

 

Anyone used a wood fibre insulation in this situation? Should we just stick to the celotex, realising that this will be much harder to fit tight between the rafters?

 

 

OK so that requires a ventilated void between the ply and insulation. The note top left refers to continuous ventilation at eaves and abutment (top) but the drawing doesn't provide any detail of that. I would expect there to be something like the following at the abutment but this particular solution may not be suitable if you have standing seams. The roof supplier really should recommend or make something suitable. 

 

 

[Temp]

The insulation is specified as 120+25=145mm of PIR. We have 150mm and wish we had more.

 

I don't understand why they specified 222mm deep rafters and only 120mm insulation between. You normally only need a 50mm ventilated void so could fit 222-50 = 170-175mm of insulation.

 

PIR is good thermal insulation however it's very difficult to fit PIR tightly between rafters so the ventilation in the void doesn't get into gaps. Its a mind numbing job cutting bits of insulation that builders do badly. Perhaps insist they foam gaps? It's why if I ever build again I'm probably going for a SIP or warm roof with the insulation above the rafters (eg design out the problem so builders can't get it wrong).

 

 

 

[gaschick]

36 minutes ago, Temp said:

The insulation is specified as 120+25=145mm of PIR. We have 150mm and wish we had more.

 

I don't understand why they specified 222mm deep rafters and only 120mm insulation between. You normally only need a 50mm ventilated void so could fit 222-50 = 170-175mm of insulation.

 

PIR is good thermal insulation however it's very difficult to fit PIR tightly between rafters so the ventilation in the void doesn't get into gaps. Its a mind numbing job cutting bits of insulation that builders do badly. Perhaps insist they foam gaps? It's why if I ever build again I'm probably going for a SIP or warm roof with the insulation above the rafters (eg design out the problem so builders can't get it wrong).

 

 

 

Thanks @Temp for the useful video and guidance. I don't know why only 120mm insultation. We have fired off emails this morning to various folks. We are indeed worried about the fit of the PIR in-between the rafters. If it's done badly with not a tight fit, you might as well have just stuck some bubble wrap up there I guess as at least this would be flush with the edges! Builder wanted to use Actis Hybris [tag line -'tomorrow's insulation today...'] to start with as the local builders merchant told him it was a great product specifically designed for timber frames, and easy to fit. Having entered another internet black hole on best forms of insulation, and wanting the whole room to not super heat, mention has always been made on the decrement delay of these products not being great. So, I have explored using flexible wood fibre insulation instead of either the PIR or the actis hybris. I have tried to get in contact with several suppliers of this to get some guidance about this use, and no-one has got back to me yet. We would probably need to use several layers of the stuff in order to get a decent amount of insulation. But, as it's a push fit and fits snuggly, this might be the way to go. 

[Temp]

One option might be to fit 50mm battens to the sides of the rafters. These would serve to preserve the ventilated gap and provide a face that PIR insulation could be sealed against. The idea is to make the cutting of PIR less critical. It would work with either all PIR or a combination of PIR and another type of insulation such as wood or wool fibre insulation.

 

I think this would work well where you just have straight rafters but roofs also have lots of triangular areas and odd shape bits where its harder to fit insulation without either leaving gaps or blocking the ventilation. 

 

 

[gaschick]

10 hours ago, Temp said:

One option might be to fit 50mm battens to the sides of the rafters. These would serve to preserve the ventilated gap and provide a face that PIR insulation could be sealed against. The idea is to make the cutting of PIR less critical. It would work with either all PIR or a combination of PIR and another type of insulation such as wood or wool fibre insulation.

 

I think this would work well where you just have straight rafters but roofs also have lots of triangular areas and odd shape bits where its harder to fit insulation without either leaving gaps or blocking the ventilation. 

 

 

@Temp you have literally just drawn a copy of the diagram that we drew yesterday! Great minds and all that! I have also done some calcs for the u value using the r values of just the Steico products in the first instance - flex in 160mm, their own general all purpose board, and then 40mm Steico flex in the service void. This suggested build up was from the Steico website https://www.steico.com/en/solutions/new-construction/roof-construction

 

I don’t know how practical and easy it is to insulate a service void - surely an oxymoron?! A void is a void? 
Anyway, doing this build up, my back of an envelope calculation gives a ball park u value of 0.16 even without including the r value of the outer 18mm plywood sheet. 
 

I will try running this again with different insulated plasterboards. 

[Iceverge]

On 19/03/2022 at 10:01, gaschick said:

Should we just stick to the celotex, realising that this will be much harder to fit tight between the rafters?

 

No. PIR isn't a wise choice. 

 

If I was to do this I would put, from the top down.  U value about 0.14

 

1.Standing seam metal roof.

2.Closely spaced battens 

3.Counter batten on the rafters to give a total 50mm air gap.

4.Roofing membrane taped at all joints.

5.Timber joists full filled with Rockwool, mineral wool, or blown cellulose if you can find it.

6. Airtight membrane. 

7. Battened 44mm service cavity with Rockwool. 

8. Plasterboard 

 

For added noise protection put a layer of OSB above the joists SE may require it anyway, or use thicker plasterboard. If you need better insulation the cheapest way is bumping the service cavity up to 75 or 100mm.  Or thicker joists or a layer of woodfiber sarking.

 

 

[gaschick]

9 hours ago, Iceverge said:

 

No. PIR isn't a wise choice. 

 

If I was to do this I would put, from the top down.  U value about 0.14

 

1.Standing seam metal roof.

2.Closely spaced battens 

3.Counter batten on the rafters to give a total 50mm air gap.

4.Roofing membrane taped at all joints.

5.Timber joists full filled with Rockwool, mineral wool, or blown cellulose if you can find it.

6. Airtight membrane. 

7. Battened 44mm service cavity with Rockwool. 

8. Plasterboard 

 

For added noise protection put a layer of OSB above the joists SE may require it anyway, or use thicker plasterboard. If you need better insulation the cheapest way is bumping the service cavity up to 75 or 100mm.  Or thicker joists or a layer of woodfiber sarking.

 

 

@Iceverge If you read the whole thread, then I am proposing to use wood fiber insulation tight between the rafters on our standing seam roof of our timber framed extensions (e.g. Steico flex or Pavoflex). As the timber frame is being made off site, then having the 18mm plywood applied on site to the design I have included in the thread, I think it will be difficult to put the battens in except underneath the ply to stop the insulation breaching the ventilation gap. Joists are 220, so we have some room to play with. My U back-of-a-fag-packet calculations (without including the ply and roof) were 0.16, which will be within building regs.

 

[gaschick]

https://www.womersleys.co.uk/techguides/productattachments_files_w_o_wood_fibre_insulation_u_and_r_values.pdf @Temp @Iceverge Here are the quoted values for the Steico flex products. Link to the suggested build up before. I realise that I have totally hijacked this thread, but it's the thread that keeps on giving!

[Patrick]

22 hours ago, Temp said:

 

 

PIR not the greatest choice if we talking about sound insulation ...

And that's what you partly want in the setup (I guess) .

 

[gaschick]

19 minutes ago, Patrick said:

PIR not the greatest choice if we talking about sound insulation ...

And that's what you partly want in the setup (I guess) .

 

Yes, also, we are keen to maximise heat decrement delay as well. Hence not going with something like the multi foil insulation materials. We need to stay cool in the summer, not super hot!

 

[SimonD]

1 hour ago, gaschick said:

Yes, also, we are keen to maximise heat decrement delay as well. Hence not going with something like the multi foil insulation materials. We need to stay cool in the summer, not super hot!

 

If you're intent on using the flex woodfibre between your rafters then I'd suggest it would be really useful to consider if you can increase the height of your roof buildup over the rafters. This is because if you're looking to maximise heat decrement delay then using the woodfibre sarking boards on top of the rafters is a better solution and is the typical detail for woodfibre roof buildups. Because the sarking board is denser and it sits outside the rafters it is more beneficial for reducing cold bridging as well as improving decrement delay than having insulation to the inside of your rafters within your service void.

 

In this sense, I'd recommend you seriously consider this buildup from the outside:

 

1, standing seam

2, iso-mat membrane

3, plywood (I'm not suggesting sarking board but if you prefer them, for a standing seam roof they should be between 18 -25mm thickness and fitted with a gap of between 3-5mm between)

4, battens to create ventilation space minimum 50mm

5, woodfibre sarking - you can opt for anything between 25 - 60mm, but if you full fill your rafters then about 40mm would be fine

6, rafters full fill with insulation

7, internals and you can then decide if you want a service void or additional woodfibre sheathing boards and then your plasterboard etc. etc.

 

With this buildup, you'll be around the 0.16 ish on your u-value with about 260 depth in. The important thing to consider with woodfribre is that natural insulation materials often perform much better in the real world than theoretical calcs and certain woodfibre builts benefit from improved reduction of thermal bridging and can therefore be designed slightly thinner than a basic u-value calculation might indicate.

 

This buildup is likely to give you a comfortable indoor environment and if you make sure you use all breathable paints etc. it'll be even better.

[gaschick]

12 hours ago, SimonD said:

 

If you're intent on using the flex woodfibre between your rafters then I'd suggest it would be really useful to consider if you can increase the height of your roof buildup over the rafters. This is because if you're looking to maximise heat decrement delay then using the woodfibre sarking boards on top of the rafters is a better solution and is the typical detail for woodfibre roof buildups. Because the sarking board is denser and it sits outside the rafters it is more beneficial for reducing cold bridging as well as improving decrement delay than having insulation to the inside of your rafters within your service void.

 

In this sense, I'd recommend you seriously consider this buildup from the outside:

 

1, standing seam

2, iso-mat membrane

3, plywood (I'm not suggesting sarking board but if you prefer them, for a standing seam roof they should be between 18 -25mm thickness and fitted with a gap of between 3-5mm between)

4, battens to create ventilation space minimum 50mm

5, woodfibre sarking - you can opt for anything between 25 - 60mm, but if you full fill your rafters then about 40mm would be fine

6, rafters full fill with insulation

7, internals and you can then decide if you want a service void or additional woodfibre sheathing boards and then your plasterboard etc. etc.

 

With this buildup, you'll be around the 0.16 ish on your u-value with about 260 depth in. The important thing to consider with woodfribre is that natural insulation materials often perform much better in the real world than theoretical calcs and certain woodfibre builts benefit from improved reduction of thermal bridging and can therefore be designed slightly thinner than a basic u-value calculation might indicate.

 

This buildup is likely to give you a comfortable indoor environment and if you make sure you use all breathable paints etc. it'll be even better.

 

On 19/03/2022 at 08:43, gaschick said:

@JohnMo I have no idea, I am not an expert in this area. I am a doctor, not a building control officer, architect or builder! This is why we are paying people to design the structure, and do the structural calcs etc for it. As a reasonably intelligent person though, I try to do due diligence as best as I can about the things we are paying s''t loads of money for (much at the annoyance of my husband). I have enclosed the timber frame section through the roof. 

Thanks @SimonD. The issue we have is that there isn't much space above where the new dining room roof joins the walls to add extra roof build up on top of the plywood (see drawing above on right). We have already had to remove a dormer window and replace with velux as we were concerned that we wouldn't have decent depth of flashing. The old roof leaked at the join due to lack of depth of flashing, and when it rained water pored down the internal walls. We don't want a repeat of this.

[Iceverge]

13 hours ago, SimonD said:

The important thing to consider with woodfribre is that natural insulation materials often perform much better in the real world than theoretical calcs

 

I agree but it is more of an anecdotal thing. Any further reading to point me towards. I'm fond cellulose. 

 

Good air sealing is vital too here. Although you have described nice breathable makeup its important to stop air blowing through the structure. I would include a membrane internally with a service cavity.

 

[SimonD]

53 minutes ago, gaschick said:

 

Thanks @SimonD. The issue we have is that there isn't much space above where the new dining room roof joins the walls to add extra roof build up on top of the plywood (see drawing above on right). We have already had to remove a dormer window and replace with velux as we were concerned that we wouldn't have decent depth of flashing. The old roof leaked at the join due to lack of depth of flashing, and when it rained water pored down the internal walls. We don't want a repeat of this.

 

I get it. It's one of those full of compromise. However, I assume the problem of height is just to do with section B-B? Can you confirm which flashing you're refering to here? Is it the flashing between roof and vertical wall, or something else? Assuming it's the one at the roof to wall junction, a standing seam roof requires a minimun upstand of 100mm for a roof of 13 degrees. Is there something not visible on the section drawing, like windows, causing the height issue?

 

Not knowing your design history, the other possibility is reducing the pitch? You've got a 13 degree pitch right now and standing seam can safely go down to 5 degrees even with penetrations for velux windows. I don't think a lot of adjustment is required to give you that extra space at your roof to wall junction.

 

I wonder whether it's worth a quick call to your designers to have a chat about your requirements as a client and how you'd like to revisit this design, in light of your updated needs? Or are you looking for something more definitive to go back with?

 

 

 

[SimonD]

1 hour ago, Iceverge said:

 

I agree but it is more of an anecdotal thing. Any further reading to point me towards. I'm fond cellulose. 

 

Good air sealing is vital too here. Although you have described nice breathable makeup its important to stop air blowing through the structure. I would include a membrane internally with a service cavity.

 

 

I like cellulose too. Having used cellulose, sheepswool and woodfibre, I like them all. Just the experience of working with them makes it a whole load more comfortable and pleasant. I've used one small sheet of PIR in my built at a point where I needed to reduce coldbridging near some steel and I hated it. I've got some papers somewhere re the natural insulation - I have to admit, my management of references has been useless while I've been building so it's a bit haphazardous - I'll have a look to see if I can find something. It might take a few days or so.

 

Yes, agree in good airtightness - totally essential - I just left those parts out as I assumed it was covered already and not the point in question. I'll be more thorough next time 😊

[gaschick]

2 hours ago, SimonD said:

 

I get it. It's one of those full of compromise. However, I assume the problem of height is just to do with section B-B? Can you confirm which flashing you're refering to here? Is it the flashing between roof and vertical wall, or something else? Assuming it's the one at the roof to wall junction, a standing seam roof requires a minimun upstand of 100mm for a roof of 13 degrees. Is there something not visible on the section drawing, like windows, causing the height issue?

 

Not knowing your design history, the other possibility is reducing the pitch? You've got a 13 degree pitch right now and standing seam can safely go down to 5 degrees even with penetrations for velux windows. I don't think a lot of adjustment is required to give you that extra space at your roof to wall junction.

 

I wonder whether it's worth a quick call to your designers to have a chat about your requirements as a client and how you'd like to revisit this design, in light of your updated needs? Or are you looking for something more definitive to go back with?

 

 

 

Don't get me started on the roof angle! It cant't be changed as the window in the end wall follows the roof line and this has already been built off site. The roof started at 15degress, but due to the massive structural steel needed to hold up the left and slide triple glazed patio door, the angle changed. The roof has caused a lot of agro and extra expense as we had to take out the existing dormer in our bedroom.

[SimonD]

2 hours ago, gaschick said:

Don't get me started on the roof angle! It cant't be changed as the window in the end wall follows the roof line and this has already been built off site. The roof started at 15degress, but due to the massive structural steel needed to hold up the left and slide triple glazed patio door, the angle changed. The roof has caused a lot of agro and extra expense as we had to take out the existing dormer in our bedroom.

 

Ah, okay, we've hit a sore spot - very common on here 🙂 😉, we all have them at several points of the build. Sorry to say this, but there's a few things that don't sit right with me here. If I've understood correctly, your roof is currently being designed by Roe, but the window to go into the timber frame has already been specified, ordered and made? Has the timber frame also been ordered and made? Because if that's the case the whole conversation re a change in buildup is moot and you're then limited to what you have. In that case you're limited in what you can add internally to the roof as you'll lose height where you've only got 2179mm height already, which I think is where you patio doors are going with the steel.

 

If that's the case, given the buildup, you can't full fill the rafters with woodfibre because you need the ventilation behind the metal roof and only putting in 170mm isn't really going to cut it for what you want to achieve.

 

If you're not completely stuck however, just for reference, when I refer to the pitch of the roof, I'm suggesting the change in pitch only to accomodate an alternative buildup using woodfibre with sarking boards. That would mean something less than about 2 degrees change if that. The finished roof level at the top where you'll install the standing seam would stay where it is. Depending on the size of your window, it might actually be possible to hide the difference in pitch, so nobody would be the wiser. In terms of measurement and to give you a better feel, this would mean that the height indicated on your sectiom B-B where the rafter meets the vertical wall reduces from 3586mm to about 3400mm, so really very little.

 

What has actually been finalised here, ordered and made and what is open for re-evaluation and re-design?

[gaschick]

11 minutes ago, SimonD said:

 

Ah, okay, we've hit a sore spot - very common on here 🙂 😉, we all have them at several points of the build. Sorry to say this, but there's a few things that don't sit right with me here. If I've understood correctly, your roof is currently being designed by Roe, but the window to go into the timber frame has already been specified, ordered and made? Has the timber frame also been ordered and made? Because if that's the case the whole conversation re a change in buildup is moot and you're then limited to what you have. In that case you're limited in what you can add internally to the roof as you'll lose height where you've only got 2179mm height already, which I think is where you patio doors are going with the steel.

 

If that's the case, given the buildup, you can't full fill the rafters with woodfibre because you need the ventilation behind the metal roof and only putting in 170mm isn't really going to cut it for what you want to achieve.

 

If you're not completely stuck however, just for reference, when I refer to the pitch of the roof, I'm suggesting the change in pitch only to accomodate an alternative buildup using woodfibre with sarking boards. That would mean something less than about 2 degrees change if that. The finished roof level at the top where you'll install the standing seam would stay where it is. Depending on the size of your window, it might actually be possible to hide the difference in pitch, so nobody would be the wiser. In terms of measurement and to give you a better feel, this would mean that the height indicated on your sectiom B-B where the rafter meets the vertical wall reduces from 3586mm to about 3400mm, so really very little.

 

What has actually been finalised here, ordered and made and what is open for re-evaluation and re-design?

SO, the timber frame is being made off site between 7-10th April. The windows were ordered from the timber frame design to fit into the timber frame in December, and are being delivered 8th April. This is a very common thing to do with the specifications on the timber frame being so accurate. Otherwise, we would maybe be waiting 3+months for windows to be delivered.  The top slope of the window is 13 degrees, but not shown terribly well in this plan! As you can see, there still isn't much space for the flashing at the point where the roof joins the exterior wall of the house. I hope that this drawing makes more sense.