
This full-scale mock-up is ready for window installation and testing according to ASTM E-2112 standards.
In this post, the president of our company, David W. Boyer, an architect by training, exposes a bald and uncomfortable truth about the way we build — one that in my opinion there is simply no arguing with.
It’s a truth that must be faced before we can ever achieve true sustainability in construction. Why we must rethink the way we build
by David W. Boyer
Construction industry professionals – in an effort to take the guesswork out of how our buildings perform ‐collaborate in hopes of coming up with “best practices” that can be embraced by the entire construction industry.
One such effort is ASTM E‐2112, passed in 2007 as a “Standard Practice for Installation of Exterior Windows, Doors and Skylights.”
One of the articles announcing the publication of E‐2112 described it as containing “Instructions for complete integration of external wall components.”
The committee members who worked on E-2112 shared a common, praiseworthy goal – to create a product that promotes:
Energy savings
Reduced water leakage
Enhanced durability of our building inventory
Reduced maintenance
Improved occupant comfort
Enhanced value for our structures
E-2112 outlines and illustrates step-by-step installation procedures for exterior windows, doors and skylights. The following photo-sequence shows a mock-up constructed according to E-2112′s 21 steps, and then being tested in a Design Verification Test Chamber.

It starts with installation of a waterproof membrane into and below the window opening.

2. Install sheet metal tray.

3. Install sheet membrane into window opening.

4. Install shims.

5. Install window.

6. Install sheet metal head flashing.

7. Install Peel-n-Stick above from substrate to head flashing.

8. Begin building installation paper with drip edge flashing.

9. Install building paper, first layer, first row.

10. Install building paper first layer, second row.

11. Install building paper first layer, third row.

12. Install building paper first layer, fourth row.

13. Building paper installation first layer fifth row.

14. Install Building paper second layer, first row (stagger laps).

15. Install building paper second layer second row (stagger laps).

16. Install building paper second layer, third row (stagger laps).

17. Install building paper second layer, fourth row (stagger laps).

18. Install building paper second layer, fifth row (stagger laps).

19. Install building paper second layer, sixth row (stagger laps).

20. Lift second layer, fifth row at window to prepare for peel-n-stick.

21. Install peel-n-stick as shown. DONE! Next window.

The Design Verification Test Chamber is fitted with monitoring sensors to show at what point installed window and wall assembly mock-ups leak under various simulated weather conditions.

Water that contains a fluorescent dye is run through the calibrated spray racks and then viewed under a black light to help us locate any leaks that may occur. The assembly is being tested with 1.44 inches of water column at a pressure of 7.5 pounds per square foot, per AAMA 502.

Water becomes evident around the base of the window, about a minute into the test.

Two minutes into the test.

Severe leakage at 5 minutes in. Bettter to find out in a test on a mock-up than to let an unhappy building owner bring it to your attention after you've moved on to the next project.
There are two problems with having this many steps for window installation. First, each step creates multiple interfaces of material in the assembly. These interfaces of different layers and/or dissimilar materials are where air and water can leak through.
How many interfaces are there in this 21-step assembly? We counted — there are 74 places where air and water can leak through. And that’s if everything is installed perfectly, and in the precise order specified by E-2112.
Multiply that by, say, 300 windows, in a big building, and you’ve got 22,200 opportunities for air and water to leak through, carrying energy dollars away, stressing the HVAC system, creating conditions for mold, just to name a few of the problems leaky building envelopes offer.
Now I believe the American construction workforce is the best on the planet. But human nature being what it is, it’s only natural for even the most dedicated installer to daydream about the big ball game sometimes, and miss a step or two. How much more likely is that to happen in a process that’s both complicated and repetitive?
No wonder a five-year (1994-98) Environmental Protection Agency indoor air quality study of 100 randomly selected U.S. office buildings found 45 percent had ongoing water leaks.
But that’s what E-2112 was designed to remedy. Let’s imagine a perfect world for a moment, where workers never get distracted and details are always constructed precisely according to specification. That’s the world our mock-up comes from.
So now, of course, we test the mock-up, just to make sure it does what it’s supposed to do, before we actually incorporate the design and procedure into the building.
Because if we put an assembly into a multi-million dollar building without having tested it to the conditions it’s likely to face, then we’re just going on faith that it will stand up to those conditions.
And because that’s how so many otherwise prudent and responsible architects and contractors operate, there is a multi- billion dollar construction litigation industry that feeds on them and their buildings in which window and wall assemblies fail to resist the ravages of air and water infiltration.
Though it costs time and money to test and refine construction mock-ups, that cost is only a tiny fraction of a fraction of what it costs to fix after the building is up.
So, test how?
E-2112 does not call out a specific testing requirement.
That’s left to people like AAMA (the American Architectural Manufacturing Association – the window and door industry association) to spell out in their own test method 502.
AAMA 502 calls for testing at 1.44 inches of water column, which is equal to just less than a 55 mile per hour wind or 7.5 pounds per square foot.
So we install the mock-up in a Design Verification Test Chamber. The chamber, equipped with monitoring sensors, can simulate weather conditions from moderate to a Category 5 hurricane.
The chamber is first pressurized to 1.44 inches of water column or about a 55 mile per hour wind, per AAMA 502.
Within one minute of starting the rainfall, water becomes evident around the base of the window.
After 5 minutes leakage is severe. This was an unwelcome surprise to the people who commissioned this study. They were confident this method – one that they had already used on countless windows in the recent past, would stand up to this kind of pressure.
The testing highlighted just how difficult a “proper” installation is to carry out.
Even after taking pains to ensure the proper sequencing, lapping and overlapping, the system failed.
It allowed water leakage under what are fairly typical weather conditions all across the country.
Even more frustrating was trying to determine just where the leak began.
Most people involved in waterproofing realize that leaks typically originate at interfaces between different layers or dissimilar materials. But when you go back to de-construct this 21 step process, you’re left facing substantially more than 21 interfaces — 74, in fact, by our count.
So where does the failure, first noticed at the interior base of the installed window truly originate?
And how, once the window is in place, do you go about investigating and repairing that?
These questions point up some uncomfortable truths.
1. Many construction products and procedures — even some of our best, like ASTM E-2112 — are not designed for “real world” conditions. Twenty-one steps to install a window that, even when done perfectly, still fails in minutes under typical weather conditions is not real-world. And that’s just in a lab. We all know the real world can be a lot more brutal.
2. Many, if not most buildings, go up without their designers, builders or owners having a clue as to how well the building will perform when weather hits. Sometimes they’re lucky. Sometimes they’re not. Luck is no basis for sustainable design and construction.
3. The remedy is a simple formula:
Design
Test
Refine
Build
I’m not saying it’s easy. But those who follow it will surely have an easier time than those who guess and then get caught.
When our materials and methods take real world conditions into account, we will have taken a giant step toward sustainability.
When we know for a fact, from testing, that our buildings can withstand their environments, we will have arrived.
# # #
(That’s the end of David’s article. The installation photos keep going, and the test photos follow them – gary)
Read Full Post »