Go to Forum Home › Materials and products › Rooflight critique wanted
- This topic has 10 replies, 3 voices, and was last updated 13 years ago by Anonymous.
-
AuthorPosts
-
-
6 March 2011 at 8:37 pm #31461Anonymous
Hi all.
Going back some time I was posting some questions about rooflights. I have now found a supplier who will adapt their design to allow me to add insulation outside the upstand and hopefully reduce the heat losses.
I understand that rooflights are always going to be poor performers but this is the best I have come up with so far. Even having said that there is one obvious issue with the design, where there is a single pane of glass resting on the wooden frame. I don't think I'll be able to persuade the manufacturer to do anything about this and I'm wondering whether it is a significant thermal bridge or not.
Details are shown in the attached diagram and I welcome feedback on whether this will be good enough.
Thanks,
Mark. -
6 March 2011 at 10:32 pm #37870Anonymous
there is one obvious issue with the design, where there is a single pane of glass resting on the wooden frame
It looks like an obvious application for a strip of aerogel tape. Or did I misinterpret the diagram?
It seems like it would just need a simple planing operation, or a change to the cutting schedule, so I'm not sure why the manufacturer would be reluctant, if they're willing to make other changes.
Cheers, Dave
-
7 March 2011 at 1:26 pm #37871Anonymous
there is one obvious issue with the design, where there is a single pane of glass resting on the wooden frame
It looks like an obvious application for a strip of aerogel tape. Or did I misinterpret the diagram?
It seems like it would just need a simple planing operation, or a change to the cutting schedule, so I'm not sure why the manufacturer would be reluctant, if they're willing to make other changes.
Cheers, Dave
The problem is that I think they have the upstand manufactured for them so it comes in as a standard design, already coated and painted and all they have to do is cut it to length with mitred corners and bond it to the glass.
Your suggestion of aerogel is an interesting one, but I don't think it works. Something needs to ensure the top pane doesn't lift off in strong winds, and I suspect that the spacer and sealant holding the three panes together isn't strong enough, hence the wood/glass bond directly to the top pane.
I had thought about pushing the second and third panes out across the whole wood upstand but then I'm purely reliant on the inter-pane sealant to hold the top sheet of glass down.
I'm now starting to think that I'm better off finding a local joiner who would put the frame together and source the glazing. I think a new diagram will be called for.
-
7 March 2011 at 3:09 pm #37872Anonymous
OK, so tackling this from the opposite direction.
The attached indicates a design that I think would work and would just about eliminate any thermal bridge and I can't see how to make it any better.
Any feedback on whether this is likely to work, or fall apart and whether I'm likely to be able to find someone to make it?
-
7 March 2011 at 10:14 pm #37873Anonymous
Hmm, I think I made the problem more complicated than it is.
The problem is that I think they have the upstand manufactured for them so it comes in as a standard design, already coated and painted and all they have to do is cut it to length with mitred corners and bond it to the glass.
So instead of planing a bit off the top of the timber, just cut the PIR a little bit taller?
(and as an aside, the least likely thing I'd do to a piece of timber that I was expecting to bond something to would be to paint it. So I guess I'm still not understanding everything?)
Something needs to ensure the top pane doesn't lift off in strong winds, and I suspect that the spacer and sealant holding the three panes together isn't strong enough, hence the wood/glass bond directly to the top pane.
There could equally be a wood-glass bond to the underside of the bottom pane. The pressure only needs to be resisted once. Or at the expense of aesthetics, some kind of hold-down strap or clips over the top pane.
I take it that this absolutely needs to be transparent, otherwise Kalwall or some other form of translucent insulation panel, Okalux etc, would be an obvious alternative.
-
8 March 2011 at 10:48 am #37874Anonymous
Hmm, I think I made the problem more complicated than it is.
The problem is that I think they have the upstand manufactured for them so it comes in as a standard design, already coated and painted and all they have to do is cut it to length with mitred corners and bond it to the glass.
So instead of planing a bit off the top of the timber, just cut the PIR a little bit taller?
(and as an aside, the least likely thing I'd do to a piece of timber that I was expecting to bond something to would be to paint it. So I guess I'm still not understanding everything?)
I understand where you are coming from, but in the original diagram the PIR is not fitted by the window manufacturer – they just supply the glass and the timber upstand already bonded together (everything inside the red box). If I were to ask them to fix the aerogel strip in between the timber and the glass then there wouldn't be the bond to the top sheet of glass any more, only the presumed bond at the bottom.
I accept that we could then try to bond the top sheet down using the PIR/external timber ourselves, but I'd be worried about the basic unit getting damaged in shipping.
The upstand is described as: “Solid resin impregnated timber and black plastikote finish”. The latter is effectively a paint but whether this is applied to the top of the upstand isn't obvious. The top of the timber appears black through the glass, but this could just be the sealant used to bond the two together.
Something needs to ensure the top pane doesn't lift off in strong winds, and I suspect that the spacer and sealant holding the three panes together isn't strong enough, hence the wood/glass bond directly to the top pane.
There could equally be a wood-glass bond to the underside of the bottom pane. The pressure only needs to be resisted once. Or at the expense of aesthetics, some kind of hold-down strap or clips over the top pane.
If we were to rely on the wood-glass bond on the bottom pane then we're implicitly relying on the spacers and the sealant around the edge of the frame to hold all the panes together. I wouldn't have thought that the sealant would be designed or strong enough for this.
I understand about possible straps or clips, but thought I had found an attractive and workable solution. Adding these introduces a further complexity.
I take it that this absolutely needs to be transparent, otherwise Kalwall or some other form of translucent insulation panel, Okalux etc, would be an obvious alternative.
Preferably transparent, but not mandatory as long as light transmission was high. However, I don't see how Okalux or Kalwall would help. Their thermal performance is no better, and probably worse, than the 3G panel I'm proposing, and I still need some insulated way of fitting it to the roof so that I don't get leaks. I guess it's my turn to be missing something.
-
8 March 2011 at 5:23 pm #37875
The first .pdf design looks no worse than any Uw1.1 Scandinavian-style 3G window – which ain't bad by rooflight standards. To improve on that, why not line the room face of the 70mm struct timber with insulation faced internally with pbd?
-
9 March 2011 at 9:24 pm #37876Anonymous
If we were to rely on the wood-glass bond on the bottom pane then we're implicitly relying on the spacers and the sealant around the edge of the frame to hold all the panes together. I wouldn't have thought that the sealant would be designed or strong enough for this.
I'd think it would be, but I don't know. Certainly best to check!
Preferably transparent, but not mandatory as long as light transmission was high. However, I don't see how Okalux or Kalwall would help. Their thermal performance is no better, and probably worse, than the 3G panel I'm proposing, and I still need some insulated way of fitting it to the roof so that I don't get leaks. I guess it's my turn to be missing something.
I was thinking of Kalwall Nanogel which has a stated U value of 0.3 W/m2K. And I was thinking of some Okalux product that doesn't exist – I misremembered it's U value, sorry.
-
9 March 2011 at 10:57 pm #37877Anonymous
Preferably transparent, but not mandatory as long as light transmission was high. However, I don't see how Okalux or Kalwall would help. Their thermal performance is no better, and probably worse, than the 3G panel I'm proposing, and I still need some insulated way of fitting it to the roof so that I don't get leaks. I guess it's my turn to be missing something.
I was thinking of Kalwall Nanogel which has a stated U value of 0.3 W/m2K. And I was thinking of some Okalux product that doesn't exist – I misremembered it's U value, sorry.
I just looked again at Kalwall nanogel. Yes it has a U-value of approximately half that of a 3G panel, but has a light transmission of between 12 and 20%, which is much worse. For the equivalent light transmission I'd need much bigger panels and the heat losses would also probably be worse as a result. I recall looking at this a year or so ago, and dismissing it then for the same reasons.
It still wouldn't get around the need to have a well insulated upstand with minimal thermal bridges.
-
9 March 2011 at 11:00 pm #37878Anonymous
The first .pdf design looks no worse than any Uw1.1 Scandinavian-style 3G window – which ain't bad by rooflight standards. To improve on that, why not line the room face of the 70mm struct timber with insulation faced internally with pbd?
This would be an option, but it means the insulation isn't “joined-up” and doesn't feel a very elegant solution. Still leaves the thermal bridge through the timber, but it is admittedly a long path.
-
11 March 2011 at 11:26 pm #37879Anonymous
For the equivalent light transmission I'd need much bigger panels and the heat losses would also probably be worse as a result. I recall looking at this a year or so ago, and dismissing it then for the same reasons
Ditto, but sadly I didn't remember before posting. Oh well. Are lumens per watt the appropriate figure of merit, the same as for lamps?
Tom's idea of a bit of insulation inside the timber does seem to make sense, though.
-
-
AuthorPosts
- You must be logged in to reply to this topic.