If you’re shopping for a garage door, the door’s energy performance may not matter — especially if you don’t heat your garage. However, there are a few reasons why you might be looking for a well-insulated, draft-free garage door:
- A good overhead door on an attached garage can keep the garage — and therefore the house — a little warmer than a leaky door.
- Since cars can be hard to start in sub-zero weather, homeowners in very cold climates — even those with unheated garages — may want a garage door that limits heat loss.
- If the garage is used for vehicle maintenance or woodworking projects, it may occasionally be heated.
So, how do you tell a high-performance garage door from a lemon?
“We sell high R-value doors!”
Many garage-door manufacturers advertise the R-values of their doors:
- advertises doors with R-17.09 polyurethane insulation.
- advertises that some of its doors are R-17.2.
- advertises that “a 17.5 R-value makes the 490 Series among the most thermally efficient doors you can buy.”
- advertises “R-values from 12.0 – 18.0.”
- advertises garage doors with R-15 insulation.
Unfortunately, these advertised R-values are almost meaningless.
Advertised R-values are inaccurate, irrelevant — or both
To determine the thermal performance of a garage door, you need to know two things:
The R-values that are trumpeted by garage-door manufacturers are measured at the center of one of the door panels. No manufacturer, as far as I can determine, reports the R-value of the entire door assembly (including the panel edges, the seams between panels, and the perimeter of the door) in their promotional materials. Moreover, manufacturers’ reported R-values tell us nothing about air leakage.
Most garage-door manufacturers are reluctant to share actual laboratory reports showing the results of R-value testing. When I asked Mike Willstead, a technical representative for Raynor, if I could see a copy of Raynor’s test results, he suggested I send him an e-mail. He later e-mailed his response: “I apologize if I misled you. I was informed that this is proprietary information that will not be disclosed.”
The window industry does a much better job
More than a decade ago, responsible window manufacturers realized that the reputation of their industry was being damaged by misleading R-value and U-factor claims. (U-factor is the inverse of R-value; in other words, U=1/R and R=1/U). To address these problems, industry leaders developed a method for testing and reporting whole-window U-factors. The U-factor reported on an NFRC label accurately describes the U-factor of the entire window, including the sash frame and the window frame — not just the center-of-glass U-factor.
When it comes to accurate reporting of U-factors or R-values, however, the garage door industry is years behind the window industry.
There’s nothing to prevent garage-door manufacturers from using the NFRC testing and labeling protocol — a protocol that yields a more honest and useful result than the center-of-panel numbers trumpeted by garage-door marketers. Alternatively, garage-door manufacturers could use the voluntary consensus standard (ANSI/DASMA 105) for reporting whole-door U-factors adopted by the Door and Access System Manufacturers Association (DASMA). A technical data sheet (DASMA TDS #163) describes this testing protocol, dubbed the “tested installed door” protocol by DASMA.
“For marketing purposes, the garage door people get a measurement on the center of panel,” said David Yarbrough, a research engineer and insulation expert at R&D Services in Cookeville, Tennessee. “The overall R-value of the entire door might be quite a bit less — in extreme cases, it may be half — of the R-value of the center of the panel. Not everyone approves of this kind of marketing. It’s been a hot debate in recent years.”
In fact, the percentage turns out to be much less than half.
Actual R-values are one-third the advertised values
Although it’s hard to obtain actual test results that report the whole-door U-factors of “tested installed doors,” I managed to obtain one report on a garage door from Clopay, and another on a garage door from Overhead Door.
Clopay provided test results for their model 3720 five-panel garage door. According to Mischel Schonberg, Clopay’s public relations manager, the door is insulated with 2 inches of polyurethane foam. Schonberg wrote, “This model is the commercial version of our residential model 9200 and has the same construction.”
While Clopay advertises that the 9200 door is R-17.2 — presumably, a claim based on a center-of-panel measurement — the test report for the installed door shows R-6.14.
While Overhead Door advertises that their model 494/495 Thermacore door has an R-value of 17.5 — a claim that, like competitors’ claims, is presumably based on a center-of-panel measurement — the test report for the installed door shows a U-factor of 0.16, equivalent to R-6.25.
Based on the only two test reports that I was able to track down, it seems logical to conclude that the R-value of a garage door is about one-third of the R-value claimed in a manufacturer’s brochure.
All over the map
Mike Thoman, the director of thermal testing and simulation at Architectural Testing Incorporated, a Pennsylvania laboratory, has tested many garage doors.
“The assembly R-values are not going to be nearly as good as the R-value of the material would indicate,” Thoman told me recently. “When you compare the assembly R-value to the material R-value, the percentages are all over the map. The percentage is a function of how the joints in the panels are made, and whether any attempt was made to provide for thermal breaks at panel edges — a lot of different things. Some products have a lot of insulation in the panel but have everything else wrong. We’ve also seen doors that do everything right. There’s really a wide, wide range.”
Are the reported R-values even accurate?
There’s another potential problem with the R-values reported by garage-door manufacturers: even if one accepts the fact that the advertised R-values represent center-of-panel values rather than whole-door values, the numbers are still higher than most insulation experts believe are possible.
Several manufacturers report that their polyurethane-insulated door panels have R-values between R-8.6 and R-9.0 per inch — values that are highly unlikely if not technically impossible, even for the center of a door panel.
“The R-value of polyurethane decreases with age,” said Yarbrough. “When it is absolutely fresh you might get R-7.5 per inch, but a realistic aged R-value would be lower — perhaps about R-6.5 per inch would be on the high end. I’m not sure I can explain these reported test results. I have seen labs make mistakes before. I think it’s an error.”
One garage-door distributor who doubts the accuracy of manufacturer’s R-value claims is Bill Feder, the president of Door Services Incorporated of Portland, Maine. On his own initiative, Feder sent a garage-door panel (Overhead Door model 194) to Yarbrough’s lab, R&D Services. The ASTM C518 test conducted by Yarbrough came up with a value of only R-7.83 for the 1 3/8-inch-thick panel. Yet Overhead Door advertises that the door is R-12.76 — or R-9.28 per inch.
Feder’s R-value challenge
“If anyone calls me about a door, I tell them about my R-value challenge,” Feder told me. “I will give anyone a check for $250 if they can bring in a document that shows that a 1 3/8-inch-thick garage door has an R-value of 12. They can’t do it.”
Unfortunately, Feder’s admirable challenge has not yet shamed the garage-door industry into correcting the numerous exaggerations in their product specifications.
What about air leakage?
If the day ever comes when garage-door manufacturers follow the path blazed by their more honest brothers and sisters in the window industry — that is, if they ever decide to report whole-door U-factors or whole-door R-values — an important piece of the door-rating puzzle will still be missing. The reason: when it comes to the thermal performance of garage doors, air leakage matters much more than R-value.
“Garage doors are so leaky that they are difficult to test,” Thoman said. “Their leaks exceed the capabilities of the available testing apparatus.”
When he needed to buy a garage door for his own house, Thoman ignored advertised R-values. “I find it almost offensive that garage-door manufacturers even publish the R-value of the insulation material,” Thoman told me. “I hate it when I see that, because it’s not a representation of the door’s performance. Air leakage is a much more important issue than the R-value of the door.”
The bottom line
Although some garage-door manufacturers have measured the whole-door U-factor and air-leakage characteristics of their doors, most won’t release the data. Until they do, purchasers of garage doors have to select their doors based on anecdotes.
If any readers have identified a garage door that limits air leakage, please tell us the brand. When you post your garage door recommendations, provide your full name — so we can tell the garage-door salespeople from the architects, builders, and homeowners.
“I tell customers that the R-value of the door should be the last thing you should think about,” Bill Feder told me. “Instead, look at the seals and the hardware. On my own garage I just have a raised-panel cedar door.”
Last week’s blog: “It’s OK to Skimp On Insulation, Icynene Says.”