Can European windows actually save carbon?

Recently, we’ve seen and heard a number of folks claim that shipping Passivhaus windows from Europe isn’t the ‘sustainable’ solution – but we’ve never seen any data to really back that up. People seem to assume that local always equals more sustainable. But is this necessarily the case? We weren’t so sure, especially after crunching the numbers through PHPP on a potential Passivhaus in the Northwest. Our realization during this task was that certain European windows allowed us to achieve Passivhaus with significantly less insulation than any North American window could. This really didn’t seem that odd to us, as high performance windows being developed in Europe are pretty superior to here in North America, and the quality of glazing overseas is fairly stunning as well – high solar heat gain coefficients (SHGCs), low U-values and superior visible transmittance (VT). What follows are some quick, back of the napkin calculations that make the argument that shipping high performance windows from Europe might actually save CO2. A lot of CO2.

In addition to the baseline model with European windows and glass, we ran three North American windows with high SHGC triple pane glazing through PHPP and then adjusted the amount of insulation needed to achieve the Passivhaus standard. We’ve taken the approach that the client wouldn’t want anything to change. Therefore, window size and orientation, building assemblies, etc. remain as intended, and we’re merely adding EPS insulation as needed.

First, we need to determine a few things…

Shipping weight of windows
448 sf x 6.13 lbs/sf = 2746.24 lbs (1.25 tons)

Envelope areas
Wall area = 3,302 sf
Slab area = 1,360 sf
Roof area = 1,360 sf

EPS foam CO2 (4’x8’x1” EPS type IX insulfoam)
5.33 lbs x 5.5 lbs CO2/lb= 29.3 lbs CO2 per sheet

Freight CO2 rates
Sea = 0.0887 lbs CO2 per ton-mile
Truck = 0.3725 lbs CO2 per ton-mile

Baseline Model w/ European window
Our baseline is the as-designed building.
Window: Internorm Varion
Glazing: Glastroesch (0.64 SHGC, 0.105 U-value, 0.73 VT)
Specific space heat demand: 4.74kBTU/ft²a

EPS insulation at envelope assemblies
3” EPS at wall
4” EPS at roof
6” EPS at slab

Baseline Shipping CO2
Trucked from factory in Lannach (AT) to Rotterdam (700 miles), shipped from Rotterdam to Portland (8,685 nautical miles/9,988 miles)
700 miles * 0.3725 * 1.25 tons = 326 lbs CO2
9,988 *1.25*0.0887 = 1,107.5 lbs CO2

Total Baseline CO2 = 1,433.5 lbs

Option A: Serious Windows
Window: Serious Window 925 series
Glazing: Serious 9 (0.47 SHGC, 0.107 U-value, 0.45 VT)
Specific space heat demand: 6.68kBTU/ft²a.
In order to achieve the standard (ultimately coming in at 4.68kBTU/ft²a), more insulation is needed.

EPS insulation at envelope assemblies
Add 2” EPS to wall = 6,048 lbs CO2
Add 2” EPS to roof = 2,490 lbsCO2
Add 5” EPS to slab = 6,225 lbsCO2

Serious Windows Shipping CO2
Trucked from factory in Longmont, Colorado (1,285 miles)
0.3725 *1,285 miles * 1.25 tons = 561.5 lbs CO2

Total Serious Windows CO2 = 15,324.5 lbs

Option B: Thermotech Windows
Window: Thermotech fiberglass
Glazing: 322 Gain #3#5 (0.61 SHGC, 0.16 U-value, 0.63 VT)
Specific space heat demand: 6.22kBTU/ft²a.
In order to achieve the standard (ultimately coming in at 4.72kBTU/ft²a), more insulation is needed.

EPS insulation at envelope assemblies
Add 1” EPS to wall = 3,024 lbs CO2
Add 4” EPS to roof = 4,980 lbs CO2
Add 4” EPS to slab = 4,980 lbs CO2

Thermotech Windows Shipping CO2
Thermotech windows are manufactured in beautiful Ottawa, Ontario (2,642 miles)
0.3725 *2,642 miles * 1.25 tons = 1,230 lbs CO2

Total Thermotech Windows CO2= 14,214 lbs

Lastly, the local option – it’s always greener to go local, right? Let’s take a look…

Option C: Cascadia Windows
Window: Cascadia fiberglass frame
Glazing: Cardinal 179 LoE#2#5 (.57 SHGC, 0.17 U-value, 0.69 VT)
Specific space heat demand: 6.78kBTU/ft²a.
In order to achieve the standard (ultimately coming in at 4.71kBTU/ft²a), more insulation is needed.

EPS insulation at envelope assemblies
Add 2” EPS to wall = 3,024 lbs CO2
Add 4” EPS to roof = 4,980 lbs CO2
Add 5” EPS to slab = 6,225 lbs CO2

Cascadia Windows Shipping CO2
Cascadia windows are manufactured just across the border in Langley, BC (294 miles)
0.3725 *294 miles * 1.25 tons = 137 lbs CO2

Total Cascadia Windows CO2 = 14,366 lbs


This was shocking to us, as we expected the numbers to be significantly closer. Of course, the locally manufactured windows had the lowest transportation CO2 production - that only makes sense. However, the superior frames and glazing from Europe require significantly less insulation in the envelope. This definitely seems backwards and goes against conventional wisdom, but in this instance, shipping windows from Europe saves CO2 – a whole lot of CO2. In fact, with an average of 14,634 lbs CO2 for the three North American options, the European windows can save six tons of CO2 emissions. That is not insignificant, and something to keep in mind if North American manufacturers don’t think that developing high performance glazing is a worthwhile venture.

We realize that a more sustainable wall assembly would probably be utilized in most situations, rather than stacking on more EPS, but we wanted an apples to apples comparison between the windows from a performance standpoint. Regardless, the North American windows would all require several inches of EPS in the slab. Production of just one inch of EPS in the slab emits 1,245 lbs of CO2, which is almost equal to the CO2 emissions of importing the European windows. Additionally, the Internorm window is a clad wood window, which as architects we’re more drawn to versus fiberglass. I’m not aware of any wood window produced locally that comes close to Passivhaus standards. And here in the gray and gloomy winters of the Northwest, windows with a higher VT, such as the Glastroesch, are definitely a great bonus.

We’re definitely interested in any thoughts/critiques readers may have on this one.

  • tim

    Great work, this is why this blog is a fantastic resource. Can you talk a little about where you got the base information such as freight C02 rates and EPS foam C02 amounts?

    • http://bruteforcecollaborative.wordpress.com meliason

      Thanks, Tim. Far from scientific, but this clearly works in the Northwest. I’m not sure how this would pan out in Minnesota or New England.

      CO2 freight emissions were from the carbonfund, which are derived from a German site.

      EPS foam has a density of 2 pcf, so total weight is 2(4′*8′*0.083) =5.33 lbs.

      According to Alex Wilson’s “Avoiding the Global Warming Impact of Insulation” article, the CO2 of EPS is 2.5kgCO2/kg or 5.5lbs CO2/lb
      5.5 lbs CO2 * 5.33 lbs = 29.3 lbs CO2 per 4′x8′x1″ sheet of type IX insulfoam

  • http://www.smoneil.com Shane O.

    Thanks for this great resource, dispelling the notion that local is inherently superior (take that LEED).

    As architects, we need to be prepared to make evidence-based decisions, and to elucidate the fact that smart ecological / environmental decisions translate into smart economic decisions. Carbon footprint is a fine metric to base a decision upon, but lets face the reality that most Americans understand decisions in terms of cost, not carbon.

    More to the point, how do these scenarios pencil out financially?

    • http://bruteforcecollaborative.wordpress.com meliason

      That’s a good question, and we really haven’t had a chance to address that yet as we’re still waiting on information from Internorm. Additionally, it’s difficult to obtain an accurate comparison because the Internorm window is a clad wood window (and price fluctuates daily with exchange rate), and the other three are all fiberglass. But if you compare the price of the Internorm (or Optiwin) to decent wood windows manufactured in North America, they’re pretty close. So if we’re looking at a similar product, with a superior performance that means less CO2 emissions even if shipped from Europe, that’s a no-brainer.

      We can calculate the cost implications of additional EPS (not including additional materials and labor) – and if I recall correctly, type IX costs about $11.25/4′x8′ board.

      Baseline w/ Internorm windows:
      0 add’l EPS boards * $11.25 = $0

      Option A Serious windows:
      507 add’l EPS boards * $11.25 = $5,703.75

      Option B Thermotech windows:
      447 add’l EPS boards * $11.25 = $5,028.75

      Option C Cascadia windows:
      593 add’l EPS boards * $11.25 = $6,671.25

      On average, the North American windows require $5,800 sf additional insulation, which probably comes close to shipping costs from Austria if the shipment can be split with another local customer – and still saves six tons of CO2 emissions.

  • http://blog.lamidesign.com Greg La Vardera

    While the Optiwin windows you have pictured may be the ultimate “green” and energy performer window product, we could do far better. For instance, Sweden’s home building industry has many window products that offer 80-90% of the performance of the Optiwin units, using a relatively straightforward aluminum clad wood window unit – it may not contain magic cork, or wool shorn from sustainable sheep, but its something that our in-place industry could easily produce here right now, and produce at a price that could be widely adopted. In the end the country would benefit most from 90% of houses incorporating 75% of the Optiwin’s performance, instead of 5% of houses getting 100% of Optiwin’s benefit.

    • http://bruteforcecollaborative.wordpress.com meliason

      Greg,

      I wouldn’t say Optiwin’s the ‘ultimate energy performing window’, though there are few frames that perform as well. Internorm’s Varion is a clad wood window (sans cork), and could easily be manufactured this side of the pond.
      internorm varion section

      For this project, it really came down to the glazing – and there just isn’t a triple pane unit manufactured in North America that we’ve found with as high an SHGC combined with low U-value and a really decent VT.

      I agree that much of the country would benefit from windows that performed 75% as well as Optiwin or Internorm (or Thermotech) – but until codes require significantly better performing products and congress stops rewarding sub-par performance with credits (such as the ridiculous ARRA 30/30) I don’t think we’ll see the demand that pushes manufacturers to make better products.

  • Stephen Thwaites

    I’d like to know more about how you modelled our windows, for example;
    What did you use for Uframe and psi?
    Did you use any fixed windows?
    – if so what numbers did you use for them?

    thx
    Stephen Thwaites
    Thermotech Fiberglass Fenestration

    • http://bruteforcecollaborative.wordpress.com meliason

      Stephen,

      all windows modelled as casement.

      Thermotech:
      Uframe = 0.19
      Psi = 0.020 (assumed warm edge spacer in every case)

      If you have better numbers for PHPP inputs, I’ll gladly rerun.

  • jill whenmouth

    Hi, I’m not in the window or construction business, just a curious and ever-learning passerby. This is fascinating, puzzling, … and a bit disturbing.

    Perhaps someone would be willing to take the time to educate me a bit and explain or share an opinion as to WHY U.S. manufacturers, especially MN window makers, don’t make windows with these higher performance abilities?

    Surely Sweden and MN have similar needs? Why aren’t MN makers matching – or surpassing- European and Sweden’s window makers?

    Is there a reason why we’re not utilizing these glazing techniques?

    Is it cost? Exclusive patents? Or are we so behind in R&D? A lack of consumer demand? Something else?

    Puzzled and curious,
    Jill

  • Stephen Thwaites

    For the glass option you listed i would use
    for casement/ awning windows
    Glass; G=0.61 , U=0.16
    Frame; H=2.7″, U=0.26, psi=0.0
    (Uframe includes effect of psi)
    for fixed windows
    Glass; G=0.61 , U=0.16
    Frame; H=1.9″, U=0.26, psi=0.0
    (Uframe includes effect of psi)
    For south facing windows you might want to consider switching the clear lite for low iron which boosts the G to 0.63

    Using the fixed window helps our numbers because the fixed frame is about half the height of Euro PH frames, so the solar gain is greater than for Euro PH frames

    • http://bruteforcecollaborative.wordpress.com meliason

      Stephen,

      That drops specific heat demand down to 6.02kBTU/ft²a from 6.22. I’ll revise numbers when I get a chance – probably after PH exam Wednesday. Will show thermotech as lowest CO2 amongst the three N.A. windows.
      What is the U-value on the low-iron glass with g-wert=0.63?

      We’re definitely aware of the performance of the thermotech fixed windows.
      And the reduced spacer psi definitely makes a difference. Thanks.

  • Stephen Thwaites

    Jill:
    Europeans calculate U differently than North Americans. Perhaps not surprisingly to snowbelters like myself, they use a smaller indoor-outdoor temperature difference – a reflection of their warmer climate.

    This smaller temperature difference leads them to having bigger spaces (3/4″ instead of 1/2″) between the glass panes. This is an advantage in the PH world which uses European standards. In the PH world they can claim lower U’s than if they followed North American (NFRC) procedures.

  • Stephen Thwaites

    low iron glass doesn’t change U

    I suspect the Glastroesch glass you are evaluating is their ‘Sanco Silverstar TrIII E Solar”
    Is this the one?
    If so, did you notice all 3 lites are low iron and they are using a Krypton fill?

    If that is the base case, perhaps you should compare our Kr filled casement which would have a European glass U of 0.128 .

    thx
    ST

    • http://bruteforcecollaborative.wordpress.com meliason

      that is the one.
      thermotech has an R-44 glass?!?

  • Stephen Thwaites

    Yikes!, i should’ve been clearer
    When using Euro procedure for calculating Uglass we get 0.128 in imperial units (0.73 in metric) when filling 322Gain w/ 90% krypton

    Hope that’s more clear, sorry for the lack of precision
    ST

  • http://www.build-green.com Scott Demark

    Interesting analysis, but I am curious why you chose to use EPS for the added insulation. If you substituted Polyiso in this analysis instead of EPS would the carbon savings not be substantially lower?

    • http://bruteforcecollaborative.wordpress.com meliason

      Scott,

      Adding thicker walls with cellulose would be far more savings in CO2, but we were looking at utilizing EPS for slab and roof (green roof), and so for ease of calculations, used EPS on the walls to keep it simple.

      Per Alex Wilson, polyiso’s embodied CO2 is about 6.6 lbs CO2/lb.

      Polyiso comes in varying densities, if we take a lighter one, 1.65 pcf, total weight is 1.65(4′*8′*0.083) =4.4 lbs.

      One 4′x8′x1″ sheet of low-density polyiso= 4.4 lbs x 6.6 lbs CO2/lb = 29 lbs CO2. We’d need a 3/4″ sheet of polyiso to equal 1″ of type IX insulfoam.

      The same amount of additional EPS in slab and roof would have to be used, but there would be a slight CO2 reduction in the walls.
      1″ of EPS in the walls emits 3,024 lbs CO2
      3/4″ polyiso emits 2,262 lbs CO2.

      In these options, we would see either a 750 lbs or 1,500 lbs CO2 savings from usage of polyiso instead of EPS. I don’t know that we would consider that substantial.

  • http://www.passivehousetoronto.blogspot.com Lyndon Than

    Dear Blogger(s),
    I read your post with interest, and thank you for going to this level of detail.
    I would like to point out that the CO2 savings can be viewed to be extremely skewed to favour the Euro windows. This bias is project-specific since your green roof and slab construction style limit your choice of insulations to rigid foam. I believe the slab can certainly be made with other insulation types. I also suspect the walls and roof can be done with cellulose if the constructions were changed around. Note that the cellulose has an R-value very close to that of EPS foam, per inch. The carbon footprint of the cellulose would be dramatically lower than the EPS, I would think at least an order of magnitude. This may be enough to tip the balance of the argument in the reverse direction. The technical points expressed by Stephen Thwaites were excellent and also diminish the strength of the original argument. One thing he did not mention, I think is the frame widths. The Thermotech windows have substantially lower frame widths than the European windows, which gives them a significant increase in glass areas and thus overall performance. I have been doing modelling in the PHPP for a project in Toronto and found the Thermotech windows very competitive performance-wise with the European ones, at substantial cost savings. Note the window installation Psi value is Zero! This has a marked effect in PHPP.
    On the other hand, one thing that does improve the strength of the Euro-window argument is that the window reveals are increased with the thicker walls, lowering further the performance of lower-performing windows.

    By the way, on our project we had a major redesign and got rid of the all the rigid foam. We are yet to take this to permit, but we’re excited to take this leap of a step in enviro-consciousness, not to mention cost reduction. Rigid foams are also being avoided in general in PH now.

  • http://www.bruteforcecollaborative.com mike eliason

    Lyndon,

    You are correct in that the CO2 savings are project-specific, but we’re finding similar data across other projects utilizing only sub-slab EPS and cellulose elsewhere (and hope to get some posts up on those in the near future). In addition, the window types aren’t apples to apples (clad wood window in EU, fiberglass in US) – EU vinyl windows (e.g. Walch) have similarly thin frame widths.

    After working with a number of PHPP files, I’m fairly confident that with cellulose walls the non-EU windows would still require more insulation and result in thicker walls. The transportation carbon footprint for the EU windows is almost equivalent to the Thermotech.

    Also, the way we prefer to detail windows (fixed anyway) the reveals are all hidden for a flush, minimal appearance.

    And we’re definitely of the opinion that massive quantities of rigid foams aren’t needed or desired to achieve PH.

  • http://www.passivehousetoronto.blogspot.com Lyndon Than

    Dear Mike,
    No argument about thickening the walls, regardless of insulation type. But the carbon footprint of thickening with cellulose is a world apart from thickening with rigid foam.
    Good point about the distances Thermotech windows are from your project. In Toronto, that is also quite a different picture!

  • Peter Hastings

    “Also, the way we prefer to detail windows (fixed anyway) the reveals are all hidden for a flush, minimal appearance.”

    Have you run the numbers in PHPP for window-on-the-outside, window-half-way and
    window-on-the-inside?

    I agree that the externally flush mounting looks really clean but a better thermal performance might save some serious bucks (or thermotech bucks or whatever…).

  • http://www.bruteforcecollaborative.com mike eliason

    Great point, Peter!

    I think this could have been written a little better…

    We prefer to place the window halfway in the wall for the increased thermal performance, but like to pull the insulation and exterior finish up over fixed frames. This makes the frame ‘see’ less exposure, reducing the Psi install coefficient. When looking at the window, you see less frame, and more glass.

  • http://www.batirsain.org John Daglish

    The optimised position for the window normally is in the centreline of the insulation layer wherever it be. You can use Therm, a free finite element software ( windows.lbl.gov/software/therm/therm.html ) to model your wall/window interface (and indeed all critical junctions) to minimise thermal bridging.
    Wish LBNL Therm would work directly with dxf / dwg files rather than having to laboriously trace over them.

Powered by WordPress