People who seal ducts for a living are used to crawling in tight, dirty spaces, but even the lean and nimble technician will find sections of duct that are simply inaccessible or not worth the effort. A new method developed at Lawrence Berkeley Laboratory (LBL) is designed to reduce the frustrations of duct sealing, improve sealing capability, and reduce cost, time, and material requirements.

The method involves spraying a sealant as an aerosol into the ducts, where it seeks out and fills holes up to 0.2 in across. The device consists of a fan, a heater, and a sprayer. It can be used to measure the leakage of the duct system before and after sealing, so the crew doesn't need additional diagnostic equipment to take measurements.

Before the sealant is sprayed in, the technician blocks off the heat exchanger (to prevent the sealant from gumming up the works) and all the registers with modified pressure pans and connects the blower simultaneously to the return and supply sides of the duct system. It takes about 30-60 minutes of spraying to seal most of the leaks in a given house's duct system, although the volume of material injected is actually fairly small. About 10 to 20 ounces of solid material--roughly the equivalent of a cube 3 in on each side--is sent into the system.

The first prototype model cost more than $3,000 to make, but a second model is being designed that uses an airless sprayer and reduces the noise level of operation. A preliminary estimate by Mark Modera from LBL's Indoor Environment program, primary investigator for the project, is that the new design should cost about $1,000 to produce, with an ultimate selling price of two to three times that.

The aerosol is sprayed into the ducts with air, so there are no ozone-depleting fluorocarbons. The sealant itself is composed mainly of a vinyl acetate polymer (an ingredient sometimes used as a base in chewing gum) suspended in water and a small amount of solvent. The material was originally designed to help remove the smoky smell from ducts after a fire by coating the inside of the ducts and holding small fire particles in place so they are not delivered into the living space. The innovation by LBL was to modify and control the aerosol production and delivery so that the sealant fills the holes instead of lining the duct.

In safety tests at two of the houses where the sealant was used, researchers found little reason for concern about health and safety aspects of the procedure. The only element found in one of the houses was a minute concentration (0.5 parts per billion) of the solvent 2-ethyl 1-hexanol, which is not designated as hazardous by the Occupational Safety and Health Administration and was only detected during the sealing process itself. Moreover, the house in which the solvent was detected during sealing had ducts located in the living space and was not ventilated during the process. (Ventilation is now part of the sealing procedure.)

Mark Modera and others in LBL's Indoor Environment Program and Energy Performance of Buildings Group tested a prototype of the device in a laboratory on straight ducts, bends, wyes, and tees, and found that it could seal both holes in the ducts and leaks at the joints. They also field-tested it in three houses in early 1995 and found that it sealed about 60%-80% of the leakage area in the duct systems in as little as 15 minutes of spraying, and used $6-$10 worth of sealing material. When they went back to one of the houses ten months later, the leakage area had increased a little bit (from 35 cm² to 40 cm²).

Florida Power and Light (FP&L) is conducting a field test on 80 houses in April, May, and June, testing the sealing technique, the degree of sealing, and the time required for sealing. FP&L will also analyze utility bills from before and after the sealing is done. In addition to the Florida test, three utilities (Oklahoma Gas and Electric Company, Pennsylvania Electric Company in Pittsburgh, and Cinergy in Indiana) are each testing the device on five houses. They are doing detailed monitoring of air flows and noise levels (from friction due to the spray on the duct walls).

The field tests should help answer some of the questions posed by skeptics of the new method. How long will the seals last? Is the degree of sealing as high as expected? Are there circumstances under which it won't work as well? Will the method really be cost-effective?

If the aerosol sealing method proves workable, it should mean that more ducts can be sealed more completely and easily. This doesn't mean that we can throw away our mastics, though. It would still be necessary to seal large holes and disconnects by hand before using this technique to go after the smaller holes. In many systems, however, most of the leakage is through small holes or cracks at the joints, which are the hardest to find and take the most time to fix.