Volume 42, Issue 2 - February 2007
| Testing 1 … 2 … 3 … 4
When it Comes to Mock-ups, Everyone Has Their Own Perspective
by Charles Cumpston
Remember the story about the blind men and the elephant? According to the tale, a group of blind men touch an elephant to learn what it is like. Each man touches one different part, such as its side or tusk, then they compare notes on what they felt. In the process, no one can agree on what the elephant is like. The story shows how reality is often viewed differently depending upon individual perspectives. If there was a glass industry parallel to this fable, it would probably have to be mock-up testing. All parties involved have a slightly different take on the process. Still, there’s some common ground. Here’s a glimpse at how a number of industry professionals view the testing process.
The Architect’s View
“Once the decision is made to incorporate a mock-up, the architect immediately sets about designing the system that the contractors will erect,” Sealy says. “The prudent architect works closely with the contractor in the construction process to gain first-hand knowledge of the challenges that the contractor may face. It is also an opportunity to readily observe if the design can be improved technically or made more simple.”
He continues, “In this respect, mock-ups have always proven to be time and money savers and they are invaluable design tools because they eliminate problems that may be encountered in the field. Nothing is more devastating than to have a building completed and occupied and then discover that a part of it is going to fail. In that respect, mock-ups are invaluable.”
If the test does fail, Sealy says, then the remedy begins with a forensics process and both the contractor and architect must dismantle the mock-up and find the reason for the failure. “It could be that a change in a particular material or component caused the failure or it could be that a particular component was misused by the designer or misapplied by the contractor,” Sealy says. “Therefore, the participation of the architect from beginning to end is essential.”
The Trouble Finder
“The origin and type of failure must be determined, which sometimes requires isolation testing and partial disassembly,” Warner says. “Once the exact location is determined, it is important to characterize the failure as either design-, material- or workmanship-related.”
Approximately 90 percent of all mock-ups experience some form of failure, with most related to water leakage, according to Warner. To decrease the chance of failure he says it’s a good idea to include the entire construction team (owner, architect, general contractor, wall designer and installer) in the process.
“It is reasonable to assume that a mock-up will have, not necessarily a failure, but some minor challenges in going through a perfect process,” he says. “For example, a curtainwall with a corner condition being tested for the normal or bare load minimum—static air, static water, dynamic water, structural design, seismic racking, several cycles—usually has some ‘hiccups’ in the air or water portion that have to be solved right then and there.”
When that “hiccup” happens two or three times, though, Heitmann says it’s disconcerting. “You go back and look at the mock-up drawings that were approved.”
If there is anything different at any stage, he explains, this is the time when it is caught.
“Your last resort, after it’s been done, is to tear it down and start all over again; you don’t really want to do that,” he says. “If it is a catastrophic failure, that [tearing down and rebuilding] is probably going to happen.”
The Veteran View
“I’d be surprised if there was any wall system that didn’t fail at the preliminary stage,” he says.
However, if due diligence has been done and no short cuts have been taken, the full testing generally goes well. “If you try to rush, then there will be problems.”
In mock-up testing, Carney says, the contract glazing companies are working with experts and they can see what works and what doesn’t, which is valuable.
Carney says the specifications he’s seen have called for mock-ups to make sure the materials supplied meet expectations. As GANA’s technical director, he says he is recommending mock-ups more on a visual, rather than a performance, basis.
“The purpose of the visual mock-up is to allow the architect to see what the design looks like,” he explains, adding that air, water and structural testing generally follow the visual mock-up.
“You want to distinguish between a mock-up for the purpose of performance and those for the purpose of aesthetics,” he says. “At the manufacturing level, we are very seldom involved with a performance mock-up. The glazing contractor is the primary party because they are the ones installing the [mock-up] system in accordance with the project conditions. The term ‘system’ is important because the test will involve the framing component, the glass component and, in some cases, other items such as sun shades, composite panels or even granite.”
System mock-ups can be very expensive, Kamber points out, especially when they do not meet the intended performance.
“The extent of the cost to take corrective action varies depending on the reason for failure,” Kamber says. “Obviously, the remedy to the issue depends on the components involved in the system and whether or not there is a problem with one or more components.”
What does failure mean in terms of time and money? In commercial construction, time is money. The longer a problem with a system mock-up goes on, the more it can cost.
“The worst case scenario is that progress on the project is halted pending the mock-up’s success,” Kamber says.
Who gets blamed in the case of failure? “They typically look to the manufacturers and fabricators as well as the installers as the responsible parties,” Kamber says.
All formal mock-ups should have a formal test procedure issued by the testing laboratory for approval, in the opinion of Duf Hudson, executive vice president of Accura Systems Inc., a Dallas-based manufacturer, distributor and producer of architectural building products.
“This procedure should be in accordance with either the job specifications or, if those are lacking, current industry-accepted standards,” Hudson says. “This latter part is somewhat subjective depending on the specific lab’s history and practice.”
While Hudson says there are basic test requirements that are undisputed, he makes the point that any lab, consultant or architect can manipulate each of those tests by requiring several repetitions of certain parts of the testing after completion of other parts.
“We seldom see just a basic test procedure anymore,” Hudson says. “Most consultants will introduce at least another round of static air and static water after the structural testing, and often repeat the cycle of air/water/structural several times. Into this mix, consultants and architects often add seismic (lateral deflection) testing and then return to another cycle of static air and water.”
Hudson agrees that most mock-ups fail at first, but if it’s a good design and has been installed in a good workman-like manner, the failure should be limited to a minor sealant problem, usually perimeter.
“My personal goal has always been to shoot for zero failures, even if it’s only a pre-test with time to analyze and correct, but having done at least 60 to 70 mock-ups, I have been able to achieve zero failures only a handful of times,” Hudson says.
In his experience, Hudson says a glazing contractor’s top employees are usually sent to install the mock-up. This crew may or may not include the lead employees who will be on the actual project.
“In dealing with these top mechanics and installers at mock-ups, several have never participated in a formal test, even after ten, 15 or more years in the industry,” Hudson says. “They have never experienced the intensity and difficulty of passing these pressure tests.”
What these novices experience in their first run through the testing process can be a surprise.
“Some sealant methods that have proven to be satisfactory in the field under normal situations will not meet the test in a formal pressure situation,” Hudson says. “Even a relatively common (for today’s high-performance window and curtainwalls) 15 pounds-per-square- foot static water test is equivalent to a 76 miles-per-hour wind. Remember, the nutshell description of a formal water test is sucking all the air out of the enclosed chamber while trying to force water into every nook and cranny.”
The Installer Side
“Failure frequently happens on pre-tests, but the issue is resolved before the formal test where the architect and owner are present,” he says. “Frequently—nine times out of ten—the failure is a matter of the sealant not being applied at the proper location.”
How does it affect the job?
“It takes time and money to fix it,” Heinaman says. “Usually, it is a minor fix and doesn’t take much time or money. If a major failure happens, as it occasionally does, it can hold up the project and be very costly.”
Lou Cerny, vice president of engineering and the project manager for MTH Industries, a Chicago-based contract glazing company, mainly sees three different types of mock-up tests.
Visual testing, he says, is generally of least concern, because presumably the drawings and details have been approved and the materials installed in accordance with the detail. “This type of mock-up is typically the basis for the rest of the installation, and work and materials are judged against the approved construction and referred to if there is a question regarding appearance or execution,” Cerny says.
For the most part, few are rejected, typically for color match or workmanship. If color is an issue, the mock-up may need to be completely redone. This is seldom the case, though, because color approval should have been obtained prior to the mock-up. “Even if specifications do not require a color range submittal, we consider this a ‘must’ to avoid discussion about color ranges [especially in anodized finishes] and re-fabrication of materials for a new mock-up,” he states.
Site mock-up for performance generally is required on all projects of any size, and may be part of the actual installation and left in place, Cerny explains.
“There is the visual review and a water test (hose spray) may be required and performed,” Cerney says. “This can be plain hose stream or a differential pressure test (vacuum chamber on interior). In any case, the hose test should be performed, even if it is not specified to assure that the crew is installing the system correctly and the system does not leak.”
Many times water will penetrate into the interior, not because of the glass system, but because of work done by other trades involved in the opening.
“It is to everyone’s benefit to test the system onsite to assure it is all watertight before completing any significant amount of the work,” Cerny says.
Generally the architect and general contractor witness this test. Off-site performance testing is done by a certified facility, based on the specifications and the customer’s instructions. Usually, the architect’s team, the glazing contractor and general contractor, at a minimum, witness the tests.
The Fix Is On
“The corrective action is generally written up and distributed to the team before action is taken,” Cerny says.
“When the correction is minor, this may be done verbally at the test facility, the repair performed and retesting done as soon as practical.” This occurs, more often than actual system failure, he adds.
It’s prudent to perform mock-up testing, including a water test onsite, with all the other work in place, even if off-site testing has been done and passed. “Usually a different [crew] or more than one crew performs the actual field installation and it is very possible that all details have not been accurately implemented by all trades involved,” Cerny says. “Because it is very difficult to determine the cause of leaks when the building is nearing completion (especially if it is a surrounding condition issue), it is recommended that a site mock-up be installed and tested before too much of the work is in place, even if specifications do not require one.”
Heitmann agrees with Cerny about the installation crew.
“One thing that we, as an independent wall consultant, have always suggested and encouraged is writing right into the specification that the crew of people who will be in the field—if not all of them at least part of them—should be at the mock-up,” Heitmann says. “The same crew that will be erecting the wall should erect the mock-up so they can see first-hand any nuances or challenges and try to minimize those in the field when they are doing it on a much larger repetitive nature.”
Time vs. Money
“Let’s say a failure occurs and it’s not an easy fix,” Heitmann says. “We need to go back, rethink, revisit and come back and make that remediation, go back into the testing—you might lose weeks or months, but ideally only days; and schedules are very aggressive today anyhow. There is not that luxury of ‘air’ built in. In terms of money, it could vary from $10,000s to $100,000s.”
In calculating the extra expense, Heitmann starts by looking at what the mock-up costs initially.
“This varies depending on many factors—the specifics of the team, the complexity of the construction—but if all goes well, it’ll probably add roughly eight to 12 weeks to the schedule at a minimum and between $100,000 to $300,000 to the cost,” he says.
“And then you run into a problem,” Heitmann says. “Think about all that is impacted when you have a failure that is not a simple problem. That fabricator is going to arrange for an independent testing laboratory. [Let’s say] that specimen is 25 feet by 35 feet and it has a corner condition. The chamber has to be prepared to receive it and then the specimen goes through the battery of tests. You enter into extra costs for team members, possibly the wall consultant and the architect, as they have to spend more time and increased costs to go to the laboratory to see the new test. So there really are a lot of things that are impacted.”
Heitmann adds that schedule is probably more impacted than costs.
“I say that for one reason. The subcontractor, if [they are working on] a unitized system and they have worked ahead to help themselves on the schedule, which is not uncommon, might have to change or alter all those units they have in their facility at that time. This has a tremendous impact on the design and engineering, manufacture and fabrication. So the time on that overall schedule is probably the most impacted, although no one likes the additional costs.”
According to ATI’s Warner, in laboratory testing extended chamber fees and retest fees can range from 10 percent to as much as 50 percent, or more, over the original base contract fee.
Like the elephant, mock-up testing is a big subject and it’s easy to see why different segments looking at it from their own perspective sometimes see things completely different.
Charles Cumpston is a contributing editor for USGlass magazine.