Clearances test

Buckley Rumford Fireplaces
Clearances to Combustibles Test
The BIA Fireplace
3/16/00

Click on picture for larger image
Photos courtesy of Bruce Davis & Chuck Fisher, OMNI

Draft Report on clearances test
by Jim Buckley, 3/16/00
Summary:
A standard or "typical" masonry fireplace designed by the Brick Industry Association (BIA) was built and tested at OMNI Environmental Laboratories in accordance with the UL 127 fueling protocol in order to obtain information about heat transfer and clearances to combustible materials for the purpose of developing new code language.
Objectives:
1) To explore the heat transfer on either side of a typical exterior masonry fireplace as it passes through a combustible exterior wall. We wanted to know if any part of (or the entire combustible wall) could be in contact with the exterior brick firebox wall if the firebox wall was 8" thick or if the firebox wall was 12" thick.
2) To determine if smoke chamber walls can be constructed like masonry chimneys with 4" (nominal) of solid masonry surrounding a clay smoke chamber lining made from clay flue liners.
3) Similar temperature measurements were taken on the BIA fireplace and compared with temperature measurements taken on another masonry fireplace with insulated firebox walls during emissions tests to see what effect the insulation might have.
Fireplace Construction and Test Setup:
The fireplace designed by the Brick Industry Association (BIA) was built with a clay smoke chamber lining made from flue liners. One side of the firebox was 8" thick of solid masonry and the other side was 12" thick. See pictures of the fireplace construction at http://www.rumford.com/emissions/BIAbuild.html. The fireplace was tested with the gasketed doors closed.
The masonry fireplace used for comparison during the emissions test had 8" thick firebox walls but Zonolite insulation was placed between the firebrick and the exterior brick wall of the firebox.
A thick plywood strip 14" wide, simulating the combustible wall, was attached in contact with the masonry fireplace walls on each side of the BIA firebox. Behind the plywood strips thermocouples were attached in a six inch grid both imbedded in the surface of the masonry and attached to the plywood in contact with the masonry.
The smoke chamber was constructed with 5/8" thick clay flue liner sections surrounded by 4" (nominal) of solid masonry just as the chimney above was constructed of a clay flue liner surrounded by 4" (nominal) of solid masonry in accordance with current codes. More thermocouples were attached to measure these temperatures as well as internal flue gas temperatures.
The fireplace was fired for about 15 hours with brands added every seven minutes according to the UL 127 fueling protocol. The "official" test lasted for twelve hours but temperatures at each of the 48 thermocouples were recorded every ten minutes from about an hour before the official test began until about two hours after the test ended. In addition a hand-held infrared temperature gun was used to measure the temperatures at other areas on the masonry surface every ten minutes.
Results:
UL127 defines as passing, after equilibrium has been reached, if no temperature on a combustible surface has exceeded 90 degrees F above ambient temperature. When testing insulated metal fireplaces equilibrium is usually reached in a few hours whereas equilibrium takes much longer for a masonry fireplace. The rules, therefore, stipulate that the test will go on until equilibrium is reached - or twelve hours - whichever is first. Whether the UL 127 test is fair to masonry materials, which have substantial thermal mass but little insulation value, or can predict anything about fire safety in the real world, are open questions.
The left side (attached to the 8" thick side) combustible panel exceeded the limitation of 90 degrees F above ambient at 9 hours and 30 minutes into the test and rose to 10 degrees above the limit by the end of the test.
The right side (attached to the 12" thick side) combustible panel did not exceed the limitations, however, did exceed the limitation by as much as 18 degrees 3 hours and 20 minutes after the test had ended.
To put these results in perspective, the temperature on the header placed 1" away from the smoke chamber wall as permitted by code, exceeded temperature limitations by 22 degrees and temperatures 6" away from the fireplace opening where combustible mantles and trim are permitted to be in contact, exceeded by 20 degrees.
Temperature readings taken on the surface of the masonry at the middle of the smoke chamber were consistently lower than temperatures measured on the surface of the masonry in the middle of the first flue liner indicating that smoke chamber walls need not be constructed any thicker than chimney walls.
What it all means:
David Johnson, now working with Washington Gas but who did a lot of work with UL safety standards, recommended working with Bob Zimmerman of UL to study the tests run in the 1960s during the development of UL 907 (the standard for stoves inserted in masonry fireplaces).
Since masonry systems seem to never reach equilibrium and will fail the UL clearances tests if the test is run long enough, David said the approach in developing 907 was to determine when a condition "known to be safe" fails the test, and then test to see if it takes as long or longer for the proposed change to fail. For example, we could test a code-meeting fireplace with 8" thick walls and 2" of clearance from combustibles. Let's presume it fails the UL 127 test after 12 hours. Then we could build the masonry wall fireplace wall 12" thick with combustibles in contact with the thicker masonry wall and if it fails after 12 hours we might assume the proposed construction is as safe as the the "known-to-be-safe" code permitted system.