Session: CLT Adhesive Tests in Support of Mass Timber Buildings 1 - PDF document

Session: CLT Adhesive Tests in Support of Mass Timber Buildings 1

5/30/2018

5/30/2018

5/30/2018

5/30/2018 4

5/30/2018 5

American Wood Council Engineered and Traditional Wood Products Cross laminated timber is manufactured by laying up kiln dried sawn boards or structural composite lumber in layers that are perpendicular to each other and glued. Each billet of solid wood can be from 4 to 18 inches thick and up to 11 feet wide by 65 feet in length. Exact sizes vary by manufacturer. The resulting panel of wood is dimensionally stable in plane because there is parallel to grain lumber resisting shrinkage in both panel directions. 6

7 Here’s a close -up photo of some 7- ply CLT. Of course, we’re looking at the edges here – so you can clearly see the alternating orientation of each successive layer. It is hard to get an idea of scale from this photo, but each lamination is often about 1- 3/8” -thick – so 7-ply CLT like this can easily be close to 10 inches thick! And, really, the main limit to the length of CLT panels is transportability. They can be manufactured to just about any length that would be required for a tall mass timber building.

The history of Cross Laminated Timber (CLT) is discussed in this slide. The first patent for CLT was obtained in France in 1985. The first constructed projects occurred in Switzerland and Germany in 1993. in 1995, development of new press technology made manufacturing easier and more precise. Multi-story applications came in 1998 with a residential building in Styria, Austria. Throughout the decade, starting in the year 2000, Europe expanded the use of CLT and Mass timber significantly. For example, due to regulatory regulations to promote green and sustainable construction, England has seen an explosion in the use of mass timber construction with over 500 buildings to date. The US and Canada have lagged behind the world in recognizing the benefits of Mass Timber construction. In 2016 the first tall wood building, known as the Brock Commons Residence Hall, was started at the University of British Columbia in Vancouver, Canada. At 18 stories, this was the tallest structure in North America utilizing mass timber products. The US building code, known as the International Building Code (IBC) adopted provisions recognizing CLT as a heavy timber product.

9 • So what is CLT? Like GLT, it utilizes conventionally sawn lumber pressed together through an adhesive bond. Unlike GLT, it alternates the direction of the adjacent layers similar to plywood. This results in a large panel configuration suitable for floors and walls. In the United States, GLT is traditionally used in multi-ply, single width beams and columns.

10 The components of CLT include laminations, layers and adhesives. As it pertains to CLT, a lamination is a single piece of sawn lumber or structural composite lumber. A layer is defined as an arrangement of laminations of the same thickness, grade and species combination, laid out essentially parallel to each other in one plane. So, CLT is made up of multiple layers – at least three. Adjacent layers are orthogonal to each other, and bonded together with a structural adhesive.

5/30/2018

5/30/2018 Now that Sam has reviewed with us the basics of CLT and how it’s made, I’m going to talk a little about some of the fire testing that has been done on CLT structures. 12

Several series of large-scale fire tests have been performed in order to guide and inform code development for tall mass timber buildings. The objective of this code development activity is to ensure safe construction practices and safe designs for tall mass timber buildings. Many of these fire tests have been what are referred to as “compartment fire tests”. A compartment fire test is one that is performed within a compartment that has at least one opening. These tests simulate a natural fire within an enclosed (or partially-enclosed) area, such as a room or apartment. As such, the fuel type used for compartment fire tests often consists of real furnishings and other contents that would be typical of the simulated occupancy. Other fuel types are sometimes used – such as wood cribs or gas – either in addition to, or as an alternative to, the real furnishings.

The fire growth phases of a typical compartment fire may be characterized by the time-temperature diagram shown on the upper- left corner of this slide. At first, they are usually slower to develop than a standard ASTM E119 fire exposure. However, once they reach flashover, growth typically exceeds that of a standard E119 exposure. Flashover, by the way, is the point at which all exposed combustible surfaces within the compartment suddenly become involved in the fire. This begins what is called the fully-developed phase of the fire, in which temperatures are often in the range of 1800-2200° F ( 1000- 1200° C). Once the combustible contents within the compartment have been mostly consumed, the fully-developed phase ends and the temperature starts to drop. This marks the beginning of the decay phase. Barring any fire re-growth resulting from the introduction or involvement of new fuel, the fire will eventually extinguish due to lack of fuel. By comparison, the standard ASTM E119 fire exposure grows fairly rapidly at first, but then levels off to a certain extent. As such, the

standard E119 exposure does not reach the same level as the typical peak temperature in a compartment fire until more than 4 hours into the test. Of course, by that time, most natural compartment fires would have cooled because they had already consumed the available fuel.

One of the biggest fire test series that was performed as part of the code development process was a set of 5 full-scale building tests that were performed by the US Forest Products Lab at the ATF Fire Research Lab in Beltsville, MD. The test plan for this series was developed by the Fire Work Group of the ICC Tall Wood Building Ad-Hoc Committee (TWB) and ratified by the full TWB. The purpose of this test series was to perform tests of realistic fire scenarios applicable to tall wood construction to evaluate occupant and firefighter tenability for egress and suppression efforts, and provide data necessary to guide further development of relevant code and standard provisions. The results of this test series are publically available at www.awc.org/tallwood.

Just to give some of the specifics of the mass timber test building that was constructed for the ATF Full-Scale building tests: • The floor plan simulated a multi-story condo configuration. The building was constructed with 2 stories because that was deemed sufficient to simulate multiple stories for the purpose of data collection. • The interior dimensions of the simulated apartment/ condo on each level was 30’x 30’ (900 ft 2 on each level). The building also included a corridor and stair shaft to the exterior of the apartments on each level, such that the overall footprint of the building was about 37’x 46’. The general floorplan for the first level is shown in this figure. The floorplan for the 2 nd level was the same. • The TWB Fire WG, which Sam just mentioned, specified a target fuel load of 550 MJ/m 2 . This value was established as corresponding to the mean fuel load plus 1 standard deviation, based on a survey of typical Group R occupancies.

Five fire tests were performed in the test structure. The first three represented unlikely scenarios in which the automatic fire sprinklers, which are mandated by code, completely fail to activate and the fire service is unable to respond at all throughout the duration of the fire. In ATF Test 1, all of the mass timber surfaces within the test structure were protected with 2 layers of 5/8” Type X GWB. Since the mass timber surfaces did not contribute to the fire, it served as a baseline to which the subsequent tests could be compared. In ATF Test 2, 30% of the CLT ceiling area was exposed in both the living room and the bedroom. In Test 3, the CLT walls were exposed in both the living room and the bedroom. Test 4 had a code-compliant automatic fire-sprinkler system. In Test 5, the sprinklers were manually charged after a 20-minute delay to simulate manual charging at the FDC by the fire service. In both Tests 4 and 5, all mass timber surfaces were exposed in both the living room and the bedroom. The right column of this table shows the duration of each fire test. Test 1, in which all mass timber surfaces were protected, went for 3 hours. By that time, the

furnishings and other contents were almost completely consumed. Tests 2 and 3 were allowed to continue until 4 hours, just so that it could be verified that no fire re-growth would take place.