How good insulation can be achieved with window glass calculate insulation

[g00ey]

Sometimes I've seen pictures of really nice houses with sleek design and lots of glass. When I look closer, I realize that these houses are mainly intended for more southern latitudes and that such a house cannot be possible to construct in a climate like ours in the north while having reasonable heating costs.

But I still want to ask the question because technological development is advancing and maybe they've managed to create an amazing window that insulates better than insulation itself.

So how good insulation capability can be achieved with a window, be it 3-glass, 4-glass, or 35-glass, and how does such a window compare to a "normal" wall?

Regarding heat insulation, what is the biggest problem factor, is it diffusion heat, i.e., heat conducted away through physical contact, or is it radiation heat, that it glows or radiates out in the form of black body radiation? I guess Stefan-Boltzmann's laws might be helpful here.

When calculating insulation capability, how does one dimension windows and calculate the heating costs to be expected per square meter of window surface compared to wall area? Is there any difference between placing the window near the ceiling vs near the floor in this regard?

 

[huggan]

Everything is measured in U-value, which has the unit Watt per square meter Celsius. This means that a 1 square meter window with a U-value of 1 costs 1 watt to maintain the heat inside if the temperature outside is one degree colder, if it's 7 degrees colder it costs 7 watts.

The best triple-glazed window I've seen had a U-value of 0.7, while a wall is around 0.15.

 

[huggan]

In Sweden, we have requirements for building energy use, which makes it more difficult to have a lot of glass. But you can compensate for this with better roofs and floors.

 

[Stringfellow Hawke]

0.58 including the frame is probably the best in Sweden right now (Skaala).
The problem is that you also need good daylight in a room, and more glass with metal layers reduces the light that passes through.

 

[g00ey]

Let's say you have an enclosed porch that you might choose not to heat, resulting in, say, two separate glass panes between indoors and outdoors, and let's say these both have a U-value of 0.7, what would the combined U-value of these two glass panes be?

How do you compensate for large glass panes with better roofs and floors? In what ways can you make roofs and floors "better," whatever that means?

 

[klaskarlsson]

Let's say you have a glazed entrance area that you might choose not to heat up, so you have, say, two separate glass panes between indoors and outdoors, and suppose these two glass panes have a U-value of 0.7, what will the combined U-value of these two glass panes be?

How do you compensate for large glass panes with better roofs and floors? How can you make roofs and floors "better," whatever that means?

You can calculate U-values by adding the inverses and then dividing 1 by these:
1/u1+1/u2=1/utot

If you have a different temperature in the intermediate space - that is, the difference is not negligible - you have to go the longer route and calculate the heat losses into the entrance area and then out into the outdoor air.
This is how I calculated for my attic:
The floor has 150 mm kutterspån (27.3 sqm). U-value: 0.53 W/m2K
The walls have 95 mm isolina (13.3 sqm). U-value: 0.38 W/m2K
The roof has been constructed with 125 mm (75 + 50) isolina (35.3 sqm), with an 80 mm air gap that vents
into the attic spaces. U-value: 0.304 W/m2K

Heat balance:
At 20°C indoors and 10°C outdoors, the heat balance will be 15.7°C in the attic (obtained by iteratively calculating so that the losses to and from the attic are roughly equal):

Apartment to attic:
Calculate heat loss (20°C indoors, 15.7°C in the attic (Temp diff = 4.3°C)) over Area 27.3 m2.
0.53 * 27.3 * 4.3 = 62.2 W energy. Gives: 62.2 * 24 * 365 = 544.9 Kwh /year

Attic to outside:
Calculate heat loss (15.7°C attic, 10°C outside (Temp diff =5.7°C)) over total area 13.3+35.3 m2.
U-average: 0.22 W/m2K
0.22 * 48.6 * 5.7 = 60.9 W energy. Gives: 60.9 * 24 * 365 = 533.9 Kwh /year



By compensate, it means that with thicker insulation in, for example, the roof and existing walls, you get a better average value...the larger the glass sections, the less this will do for the averaging, so then the roof insulation goes towards infinity eventually to compensate for insulation that goes down towards zero for the glass
.
Here is a good link that helps calculate it:
http://www.energiberakning.se/U_medel/U_medel.aspx

/K

 

[g00ey]

This sounds too good to be true. What are the typical U-values for roofs and floors?

I made the following simple estimation calculation, let's say we have a 100 sqm single-story house with a U-value of 0.15 for all external surfaces in the room. It is 2.5 meters from floor to ceiling and let's assume the room is square, i.e., 10x10 meters.

Calculation: (2*100 + 4*2.5*10)*0.15 = 45 W/∆°C

If it's a 15-degree cold winter day and we want 20 degrees in the room, then it would require 45*35W = 1575W, which would mean a heating cost of perhaps 1700 SEK/month.

Let's be a bit extreme and assume that half of the walls are completely covered by glass windows, then we get the following:

Calculation (2*100 + 2*2.5*10)*0.15 + 2*2.5*10*0.7 = 72.5 W/∆°C

Under the same circumstances as above, it would require 2537.5W for the same heating, which would result in a total monthly heating cost of perhaps 2700 SEK. If we are even more extreme and assume that all external walls are completely made of glass, 3500W would be needed under the same circumstances.

Then maybe you wouldn't want to keep all areas in such a house heated, which should further reduce the costs.

Is this really reasonable? What are the requirements for energy consumption in this case?

 

[klaskarlsson]

This sounds too good to be true. What is the typical U-value for roofs and floors?

I did the following simple estimation, let's say we have a 100 m² single-story house with a U-value of 0.15 for all exterior surfaces in the room. It is 2.5 meters from floor to ceiling and let's assume the room is square, i.e., 10x10 meters.

Calculation: (2*100 + 4*2.5*10)*0.15 = 45 W/°C

If it's a 15-degree cold winter day and we want to have 20 degrees in the room, it would require 45*35W = 1575W, which becomes a heating cost of perhaps 1700kr/month.

Let's be a bit extreme and assume that half of the walls are completely covered with glass windows, then we get the following,

Calculation: (2*100 + 2*2.5*10)*0.15 + 2*2.5*10*0.7 = 72.5 W/°C

Under the same circumstances as above, it requires 2537.5W for the same heating, which results in a total monthly heating cost of perhaps 2700kr. If we are even more extreme and assume that all exterior walls are entirely made of glass, then 3500W is needed under the same circumstances.

Then perhaps you wouldn't want to keep all the spaces in such a house heated, which should reduce costs further.

Is this really reasonable? What are the actual energy consumption requirements in place?

I've heard that at most 55kwh/m2/year are the requirements in the Stockholm area at least...then I believe hot water is included in that, and it only applies to electric heating (heat pump, radiator, etc.), not bioheat, etc.

So none of your solutions would be approved for a 100 m² kitchen with direct electric heating...a heat pump or alternative is required.

When I calculate umedel in your second case with the link, I get 0.23 W/m2K...so the difference probably matches?

/K

 

[g00ey]

If we assume an annual average temperature of 9 degrees and use a geothermal heat pump, it seems that it might be possible to achieve that in a fully glazed house above.

But I might have been a bit too generous with the U-value of the roof and floor; they might be worse than the walls.

 

[Anders243]

Isn't it true that it is the total energy loss that should fit within, for example, 55 Kwh/m2/year? (I'm unsure about the figure, so take it with a grain of salt)

Then should losses from, for example, ventilation also be counted? (according to the norm)
Perhaps there are templates for how to separate hot water, ventilation, and the building envelope from each other in the "energy requirement."

Edit: Consider this a question, I am not sure that it is so.

 

[klaskarlsson]

Isn't it true that the total energy loss must fit within, for example, 55 kWh/m2/year? (I'm unsure about the figure, so take it with a grain of salt)

Then losses from, for example, ventilation should also be counted? (according to standard)
There might be templates for how to separate hot water, ventilation, and building envelope from each other in the "energy requirement"...

I also think so - that's probably why an exhaust air heat pump is often good for new builds (relatively cheap solution) ?

/K