The procedure for identifying design alternatives will be to study existing data and information from tests of long-span guardrail designs to determine cause(s) of failures, and then to identify modifications to the design to remedy problems experienced earlier.

    Two deigns were tested during 2006. One was tested under this pooled funds study and the other was tested by the Midwest Roadside Safety Facility.

    As mentioned above, in the first phase of the pooled funds study (405160-1), the rail element ruptured where the rail section changed from nested to a single layer.

    In the test performed at Midwest Roadside Safety Facility, a single layer of 12 guage W-beam was used across a 25 ft long span. The guardrail was configured similar to the Midwest Guardrail System (MGS) with a 31-inch rail mounting height, 12-inch deep blockouts, and splices at midspan between posts. Modified posts were used adjacent to either end of the long span. The guardrail performed acceptably under NCHRP Report 350 test 3-11 conditions at two different impact locations. The reports and video from these tests have been requested and will be reviewed under this phase of the study.

    Design/behavior features that may have contributed to rupture of the rail element in the pooled funds test include:

	1.	The higher deflection of the rail element in the long span area.
	2.	The high strength of posts adjacent to the downstream end of the long span.
	3.	The sharp curvature introduced into the rail element when the vehicle approached the posts at the downstream end of the long span.
	4.	The abrupt change in flexural stiffness of the rail element at the cross section where it changed from nested to single layer.

    Design changes to be considered that might improve performance of the long span guardrail include:

	1.	A longer nested section to move the change in cross section farther away from the ends of the long span.
	2.	Use a 10-gage W-beam rail element in the long span that extends some distance beyond the ends of the long span.
	3.	Provide reduced lateral stiffness of the posts adjacent to the ends of the long span to provided better transition of the lateral stiffness of the system.
	4.	Provide structural steel tubes behind the W-beam in the long span area and beyond to provide additional support/strength for the W-beam in that area.
	5.	Other ideas that might be developed during this phase.

Task 2 -- Perform Computer Simulations

    The finite element code, LS-DYNA, will be used in this phase to perform simulation of long-span guardrail.

    The first step in the computer simulation work will be to accurately simulate and capture behavior of the guardrail in the unsuccessful pooled funds test. This will help validate/calibrate the long-span guardrail model and be of benefit in identifying the basic cause(s) of rupture of the rail element and in suggesting possible remedies for the problem.

    The next step will be to simulate two of the most promising design alternatives identified in Task 1 of this phase. The results will be used to assess the impact performance of each design and evaluate their probability of meeting NCHRP Report 350 requirements.

Task 3 -- Perform Full-Scale Vehicle Crash Test

    The researchers will perform test 3-11 of NCHRP Report 350 (2000P vehicle, 100 km/h, 25 degrees) on the selected design from Task 2. It is argued that this is the critical test for this design and that test 3-10 (820C vehicle, 100 km/h, 20 degrees) is not needed.

 
QUARTERLY PROGRESS REPORTS:

December 2007 Progress Report
September 2007 Progress Report
June 2007 Progress Report
March 2007 Progress Report

____________________
* K.K. Mak, R.P. Bligh, D.J. Gripne and C.F. McDevitt, “Long-Span Nested W-Beam Guardrails over Low-Fill Culverts,” Transportation Research Record No. 1367, Transportation Research Board, National Research Council, Washington, D.C., 1992.