This article was originally published in the July/August 1998 issue of Home Energy Magazine. Some formatting inconsistencies may be evident in older archive content.
| Back to Contents Page | Home Energy Index | About Home Energy |
Home Energy Magazine Online July/August 1998
Can Duct Tape Take the Heat?
by Max Sherman and Iain Walker
Popular culture abounds with uses for duct tape: duct tape calendars, books like 101 Uses for Duct Tape, and more. But lab experiments have finally proved that duct tape, as it is generally used, should not be used to seal ducts.
The major conclusion we can draw so far is that one can use anything but duct tape--if we define duct tape as fabric-backed tape with rubber adhesive--to seal ducts. Under challenging (but realistic) conditions, duct tapes fail. Other kinds of tape and other sealant methods have good longevity when installed properly (see So Many Sealants, So Many Failures). The tests have also shown that tapes do not have to be strong to have good longevity, and that none of the various ratings, including those from Underwriters' Laboratories (UL), addresses sealant longevity in realistic conditions.Durability Is the Key Today, taping with duct tape is the most common method of sealing ducts. Field crews dislike mastics because they tend to be messy. Foil tapes are used on ductboard, but duct tape is most popular on the most common duct materials--flex duct and metal. Each sealant has its advantages and disadvantages, but with reasonably careful application, any of them can seal well--initially.
Longevity is another story. Today, houses are said to be designed to last 30 years. Flex duct systems are often rated for a 15-year life. Duct seals ought to last at least as long. But it appears that the physical properties of some of the sealants may cause seals to fail within just a few years.
While some sealants are UL rated, no UL rating addresses longevity (see Standards for Sealants). If people choosing duct sealants had relative ratings of longevity, they could make a more informed decision.Three Test Rigs We developed three test procedures--baking, cycling, and aging--to stress standard duct joints and their sealants in different environmental conditions. The baking test uses just a simple oven. The cycling apparatus was funded over three years ago by the U.S. Environmental Protection Agency to measure the longevity of aerosol duct sealant under accelerated conditions. The aging apparatus was built last year with funding from the California Institute for Energy Efficiency. In these testing rigs, we periodically measure duct leakage. We declare that a sealant has failed when it leaks more than 10% of the air that the joint leaked before being sealed. These tests measure the sealant's endurance in the face of exacting environments, but they do not address installation issues.
The baking test is the simplest. We build a metal-to-metal stepped transition finger joint of standard 4-inch sheet metal duct, and support the duct with independent mechanical supports. This is one of the hardest joints to seal. Some standards require a clamp over duct tape at flex-to-collar connections, but there is no way to apply a clamp over the tape or sealant at a stepped transition, such as the duct-to-plenum joint. We apply sealant to the joint, following the sealant manufacturer's instructions, if applicable. We then place the duct section in an oven set to the temperature of a hot attic or heating system supply air, in the range of 140°F-180°F. Temperatures are kept below 200°F because some of the tapes are rated to that temperature. Duct leakage is measured before baking and at various intervals during baking. When testing the leakage, we also look at the sealant and note obvious failures. The sections are baked as long as 4 months.
In the cycling test, we add temperature and pressure changes. We blow hot and ambient air through the duct at pressures between ambient air pressure and 200 Pascals (Pa) to simulate HVAC cycling. This test has its limitations. Cycles take a long time--20 minutes--due to the need to warm up and cool down the duct. And the cycling apparatus cannot subject the test sample to the cold temperatures that might be expected in the winter or even in air conditioning supply ducts.
Only the aerosol sealant has been put through the cycling test. A few aerosol-sealed leaks were sealed over two years ago and have cycled between hot and ambient air every 20 minutes ever since. There has been no significant change in duct tightness.
The aging test was designed to overcome the limitations of the cycling apparatus, and may be a useful prototype for conducting standardized tests on duct sealant longevity. The testing rig has a hot-air source and a cold-air source (see Figure 1). Duct sections in the rig have hot air running through them for five minutes, followed by cold air for five minutes. We have tested 19 tapes and sealants in the aging device.
A wide range of products claim to be suitable for duct sealing, but there is often little in the specs or product literature to differentiate them. For example, one major manufacturer lists 16 different duct tapes, available in a range of colors, and 8 foil tapes. Some have product codes printed on the tape, some have codes printed on the hub, and some have no product code on them. All the duct tapes are rated by UL Standard 723, Test for Surface Burning Characteristics of Building Materials, but only some of the metal foil ones are so rated. Some tapes are labeled as Code Approved by BOCA, but one tape with nearly the same characteristics as the Code Approved ones does not indicate that it is Code Approved.
All the products we tested all sold for use on HVAC ducts. Several companies have recently come out with UL 181B-FX tapes (see definitions in Standards for Sealants), which are UL-approved for use on flex duct systems when installed with metal clamps over the tape. Generally, these are not yet listed in product catalogs. While we have not investigated mastics as thoroughly as tapes, there seem to be fewer grades of mastic. Few mastics are currently UL 181B approved, although many are approved by UL 181A. This situation may change in the future.Quick Catastrophic Failure When we began the aging experiments, we expected it to take weeks to begin to see degradation in performance. We were surprised to find some duct tapes failing in a matter of days. Most failed catastrophically rather than gradually. This made it less necessary for us to use arbitrary numerical criteria in deciding that a sample had failed. Rapid failures have only occured for cloth duct tapes with rubber adhesives.
Of the 19 samples we have aged and 13 samples we have baked, many have failed; eight are still running. The only ducts that have become leaky have been sealed with duct tape (see Table 1). Most of them showed visible signs of failure within about three days of the start of the test. The tests give us no indication of time to failure in the real world. But they do allow us to see which sealants last relatively better than others.
In the baking test, only tapes with rubber-based adhesives have shown degradation. The duct tapes tend to be leakier than the other tapes. Some are approaching failure in the aging test, as well. The other sealants are all leaking less than 2% of the unsealed flow.
After the test samples had spent three days in the test rigs, we measured their joint leakage. The duct tapes had 10%-20% of the unsealed leakage. The premium grade tape had failed completely, falling off the test section. Such complete failure was due to delamination--separation of the cloth backing from the adhesive. The other failed tapes had just started to delaminate. We believe that at elevated temperatures, the rubber-based adhesives in duct tapes change their properties and tend to separate either from the cloth backing or from the surface. We tried a second sample of the premium grade tape; it lasted about seven days before complete failure. The metal-backed tapes with acrylic adhesive, the aerosol, and the mastic showed no visible or measurable signs of degradation after two weeks of testing.
Although our failure criterion was 10%, we continued to monitor most of the samples until their leakage was more than 50% of the unsealed flow. In most of these, leakage continued to increase rapidly, often ending with a catastrophic failure.
A visual inspection of the baked duct sections revealed that in most of the duct tape samples, the rubber adhesive had changed properties and the tape had delaminated. Some samples appeared to have baked on in such a way as to maintain their seal. However, the adhesive baked on without air pressure from the leaks pushing against the tape; such permanence is unlikely in the field.
In the aging test, we occasionally saw some duct tapes begin to separate from the duct and then get resealed when an overlapping piece of tape failed in such a way as to plug the first leak, leaving a bubble. We have observed this same phenomenon in the field. This behavior may explain why some duct tapes last longer; we did not observe it on any other type of sealant. We consider such failing and resealing to be unacceptable, but we did not fail samples on this basis.
There appears to be little difference in performance among duct tapes, as compared to the difference between duct tape and the other sealants. Different grades of duct tape have different strengths, but the differences do not affect longevity.Heat Exhaustion Although our testing cannot differentiate among the mastics and the aerosol sealant, the data show that duct tape is not a good sealant for use in ducts that operate at much above ambient temperature. We believe this is due to the rubber adhesive, but we cannot say so definitively. For the most part, cloth backing and rubber adhesives go hand in hand. The other sealant products have not demonstrated any of the failure modes we have seen in the duct tapes.
There are a few products that use rubber adhesives with a backing that is not made of cloth. We intend to test these products in the future. Although the current crop of duct tapes fails our longevity tests, there is no reason to believe that the adhesive cannot be reformulated to work better at the higher temperatures found in attics or heating systems.
We have found that clear, unreinforced plastic-backed tape--which we call packing tape--holds up well. At least one version of this tape has been UL 181B-FX rated and is commercially available. We have tested the UL rated version for over one month and the nonrated version for over three months, and there is no significant leakage.
Foil tape products with 181B-FX ratings are now available. The ore we have tested has held up fine for a month in the aging rig.
Packing tape has a low tensile strength. Because the purpose of a duct sealant is only to reduce leakage, we did not test strength. Some field users dislike using weaker tapes, perhaps because they like to hang ducts with tape, but duct systems are not supposed to be mechanically supported by sealants.Installation Matters Our testing focused on the properties of the sealants themselves. We made sure we got good initial seals for our test section by following good practice and the manufacturer's instructions. For example, the test section was clean and dry. We applied the sealant with meticulous care, and we checked for a good seal before beginning any of the tests.
In a normal application, such care is not practical. Access to the ducts may be limited, and ducts may be dirty. These problems make it difficult to install tapes. Thus some tape jobs may perform poorly because they were poorly installed, not because of any intrinsic fault in the tape. Field experience shows that mastics and aerosol sealant often seal better than tape in dirty or inaccessible locations.
The best choice of duct sealant will vary by climate, construction type, and local experience. Our recommendation? Consider installation issues, but use anything but duct tape.
For a more detailed description of the testing apparatus itself as well as the testing protocol, refer to the project report, Leakage Diagnostics, Sealant Longevity, Sizing and Technology Transfer in Residential Thermal Distribution Systems, Lawrence Berkeley National Laboratory Report No. 41118. Tel:(510)486-4022; Web site: www.lbl.gov.
Max Sherman and Iain Walker are staff scientists in the Energy Performance of Buildings Group at Lawrence Berkeley National Laboratory in Berkeley, California.
Home Energy can be reached at: email@example.com
- FIRST PAGE
- PREVIOUS PAGE