EXHIBIT 99.2

TEST REPORT

on

Out of Plane Bending of LEEP Structural Core Panels

by

Jim Pedersen and John Stanton

Department of Civil Engineering
University of Washington
Seattle WA 98195

May 2000

Scope

      To establish the structural properties LEEP Structural Core sandwich panels subjected to out of plane bending. The loading in the tests simulates wind loading on a wall panel or gravity load on a floor. Three nominally identical panels were tested. All testing was conducted in the Structural Research Laboratory of the University of Washington Civil Engineering Department.

Referenced Documents

	ASTM E 72, “Standard Methods of Conducting Strength Tests of Panels for Building Construction”
	Allen, H. G., Analysis and Design of Sandwich Panels, 1969.

Significance and Use

      LEEP Structural Core panels consist of steel face sheet bonded to a foam core using sandwich construction. They are intended for use in building as wall, floor and roof panels. Their ability to carry out-of-plane loads is therefore critical to the proper functioning of the building. The tests described here were conducted to obtain the properties of the panels that are needed for design. The tests were conducted at the ambient temperature and humidity in the laboratory.

Test Set-up

      The panels were tested by placing them horizontally on supports at each end, and loading them at midspan. The configuration is shown schematically in Fig. 1. The test panels were 12 ft long x 5 ft wide.

      The supports were made from two steel beams, and were placed 11 ft apart. A 1" dia. round rod was placed on each support in order to allow free end rotation of the panel as it deflected, and a 3/8" steel plate was placed between the rod and the LEEP panel in order to avoid local stress concentrations.

      Load was applied through an inflatable air bladder, in order to achieve uniform pressure on the panel. The air bladder reacted against a timber frame that in turn reacted against a load cell beneath the head of the large Baldwin test machine in the lab. The test machine therefore served the purpose only of a fixed frame. The air bladder was 2 ft x 5 ft and was placed across the full panel width. Thus only the central 2 ft of the panel’s length was loaded. Calculations are provided that permit the panel strength to be computed from the results of these tests when it is subjected to full length loading.

Instrumentation

      Load Measurements.

      Load was measured in three ways. First, the pressure in the bladder was measured using a 15psi non-amplified pressure transducer. Second, a 20 kip load cell was positioned to

measure the total load being applied by the air bladder. Last, the load was measured by the pressure cell built into the large Baldwin test machine. This latter measurement was expected to be the least accurate of the three, because the applied load was only 1% of the machine’s capacity. However, it was readily available and served as a check.

      All load-measuring instruments were calibrated using load cells directly traceable to the National Institute of Standards and Technology. Copies of the most recent calibrations are available.

      Transverse loads in the range of 3000 pounds were expected, corresponding to pressure in the air bladder of 2.5 pounds per square inch. These loads are around 15% of the capacity of the pressure transducer and the load cell used, and can be considered accurate to approximately 1%.

      Deflection Measurements

      Three types of deflection were measured. At midspan, vertical potentiometers read the vertical deflection of the panel. One was placed at each side, and one at mid width of the panel. This arrangement allowed vertical deflection, transverse bending and twist to be recorded. The edge potentiometers measured the deflection of the top surface of the panel while the mid-width one measured the displacement of the underside.

      Potentiometers also measured any change in panel thickness. This was done by drilling a small hole in the top steel face sheet and the foam core. A bolt was attached to the bottom face sheet and projected up through the hole in the core. The vertical movement of the top of the bolt was then monitored by a potentiometer attached to the top face sheet. These measurements were made at mid span and at one support.

      Differential horizontal movement of the two face sheets was also measured using potentiometers, at one end of the panel. The purpose was to establish the magnitude of any shear deflections caused by deformation of the foam core.

      All potentiometers were calibrated prior to use and their readings are accurate to 0.001".

Test Specimens

      The test specimens consisted of rectangular panels nominally 5' wide by 12' long. The span was 11', so a 6" overhang existed at each end. Each specimen was comprised of two whole LEEP Structural Core panels, (each 2' by 12') bonded together using a two-part epoxy along the central tongue and groove joint. A 5" wide strip was then attached to each outer edge to complete the outer tongue and groove joints. These strips were obtained by cutting them from a third panel with a circular saw.

Test Procedure

      The length and width of the specimens were measured to the nearest 0.01-inch. In addition, centerlines of each specimen were properly identified as well as the location of the supports.

      The steel support beams were accurately positioned and leveled, and were then attached to the floor. Then the steel rod and steel plate were placed on the top flange of each beam. The panel specimens were then set on top of these plates, with the trough side down. The air bladder was placed on the panel, followed by the timber spreader frame and the load cell. The head of the Baldwin test machine was then lowered so that the load train had no slack in it.

      Zero values were taken on all instruments prior to loading. The load was then applied through the regulated inflation of the air bladder. The loading rate was such that loading to failure took approximately 15 minutes.

      Load-deflection curves were recorded to determine the initial stiffness, proportional limit, maximum load, and deflection at failure for each specimen.

Test Results.

      The primary test results and the properties computed from them are given in the table.

      The load-deflection plots for all three specimens are shown in the figure. The reported deflection was the average of the three midspan potentiometers’ readings.

      Throughout most of the test, the system behavior was essentially linear, as shown in the plots. The only events that were noticed were settling noises coming from the timber framing and visual observations of slight movements of the face sheets moving with respect to one another. Just before failure, elastic buckling was seen to be starting on the upper side of the trough in both panels comprising the specimen. Failure was sudden and was caused by buckling of the top face sheet at midspan. The buckle propagated immediately across the whole panel width.

      The behavior of all three specimens was similar. The stiffnesses and failure loads of the individual specimens lay within 5% of the mean values. The mode of failure was also the same in all three specimens.

Discussion.

      It is possible to compute from these test results the uniform wind loading that would cause failure on a full twelve-foot panel. The moment can be derived from:

      M = (1/4) * Q * (L - a/2),

where: M = Moment.
      Q = Total load
      a = Loaded length.
      L = Span length.

      Average test values are:
      L = 11 ft
      a = 1.833 ft
      Q = 3332 lbs at failure
      Thus M = 8,400 ft-lbs

      For a 12 ft span, uniformly loaded, the same width as the test panel,

      a = L = 12 ft
      M = 8400 ft-lbs
      Q = 8*8400/12 = 5600 lbs

      This load is applied over the area of the panel,
      w = Q/(bL) = 5600/(5'x 12') = 93 lb/ft²

      The wind speed that would produce this loading can be obtained from

      p = 0.00256 V²

where

      p = pressure (lb/ft²)
      V = wind speed (mph)

Thus in this case
      V = Square Root of(p/0.00256) = 191 mph

This is the wind speed corresponding to failure. A safety factor should be used for design purposes.

Figure 1. Test Set up