Box Culvert Guardrail – Phase I (405160-5)

See Phase II

Final Report Phase I:  Crash Testing and Evaluation of W-Beam Guardrail Box Culvert
Executive Summary — Phase I

Final Report Phase II:  MASH Test 3-11 of the W-Beam Guardrail on Low-Fill Box Culvert
Executive Summary – Phase II

TTI Research Supervisor:
William F. Williams, P.E.
Texas Transportation Institute
Texas A&M University System
TAMU 3135
College Station, Texas 77843-3135
(979) 862-2297
[email protected]
Pooled Fund Technical Representative:
Dave Olson, P.E.
Design Policy, Standards and Research Manager
Washington Department of Transportation
P.O. Box 47329
Olympia, WA 98504-7329
[email protected]


Guardrail installations frequently must pass over reinforced concrete box culverts used for transverse drainage under highways. In many cases, the depth of fill over the box culvert is very shallow and will not allow the proper embedment of steel or timber guardrail posts. A typical detail in these cases is a shortened W6×9 steel post attached to a steel base plate bolted to the top of the box culvert. An epoxy adhesive anchoring system is sometimes preferred to permit installation of the post without the need to enter the culvert and install a bolt-thru anchoring system.

In Phase I of this pooled fund study, NCHRP Report 350 test 3-11 was performed to evaluate a guardrail system with standard post spacing (6 ft-3 inches) across a low-fill culvert (see Phase I report). During this test, the W-beam rail element ruptured. The adhesive anchoring system worked as designed with the new W6×9 post and welded baseplate detail. No damage to the deck or failure of the adhesive anchors was observed.

In Phase II, the height of the W-beam guardrail system was raised from 27 inches to 31 inches above finished grade. The posts were spaced on 6 ft‑3 inch centers; however, the W-beam rail splices were relocated to the midspan of the 6 ft‑3 inch post spacing. Also, the crash test performed on the W-beam guardrail on low-fill box culvert was in accordance with test 3-11 of the new crash test guidelines, AASHTO MASH, which involves the 2270P vehicle (a 5000 lb (1/2 ton) Quad Cab Pickup) impacting the guardrail at a speed of 62 mi/h and an angle of 25 degrees.

Test Installation and Test Vehicle for Phase II Testing

In Phase II, the box culvert guardrail installation consisted of a 12 gauge W-beam guardrail system supported by W6×9 steel posts anchored to a simulated box culvert. Standard 6-inch × 8‑inch × 14‑inch long wood blockouts were used to block out the W-beam guardrail from the steel posts. The height of the W-beam guardrail system was 31 inches above finished grade. The posts were spaced on 6 ft-3 inches centers. The W-beam rail splices were located at the midspan of the 6 ft‑3 inch post spacing. The posts were anchored to the top of a simulated box culvert slab using the Hilti RE500 epoxy anchoring system. For this test installation, 9 inches of compacted standard soil material was constructed on top of the simulated box culvert slab. The total length of the simulated concrete box culvert slab was 105 ft. The W-beam guardrail system was anchored on each end using ET Plus end terminals.

The W6×9 steel posts were welded to 12-inch × 12-inch × ⅞-inch thick base plates. The total length of the posts was 40⅛ inches. Each steel post with base plate was anchored to the 9‑inch thick simulated box culvert slab using four adhesive anchor bolts anchoring the posts to the box culvert.

The W-Beam Guardrail on Low-Fill Box Culvert contained and redirected the 2270P vehicle. The vehicle did not penetrate, underride, or override the installation, however, the rail element was torn half the width at post 12. Maximum dynamic deflection during the test was 45.1 inches. Several blockouts separated from the installation. However, none of these detached elements penetrated or showed potential for penetrating the occupant compartment, nor to present undue hazard to others in the area. No occupant compartment deformation occurred. The 2270P vehicle remained upright during and after the collision event. Maximum roll and pitch angles were 9 degrees and 11 degrees, respectively. Occupant risk factors were within the limits specified in MASH. The 2270P vehicle exited within the exit box specifications. The W-Beam Guardrail on Low-Fill Box Culvert performed acceptably according to the specifications for MASH test 3‑11.

Test Installation and Test Vehicle after Phase II Testing


Not applicable.



Prior to the Phase I crash test, low-speed pendulum tests were also performed on a prototype guardrail post on a simulated box culvert as a surrogate for full-scale crash testing.

Pendulum Testing of Guardrail Posts on Box Culverts