CHAPTER 6

Railing Design Examples with Proposed Updates

A set of railing design examples using the impact loads and analysis and design procedures from both the current and proposed Section 13 were undertaken to evaluate the potential impact on railing designs and details from the MASH-based impact loadings and geometry requirements.

Rail type, materials, and Test Level were specifically selected to evaluate the broadest range of potential impacts on railing inventories.

Design Examples and Results

Design examples are provided in the appendix, supplemental to this report. A summary of the results for each example is provided in this section.

The design examples follow the provisions of the proposed Article 13.7 and the current LRFD BDS, Appendix A13 (AASHTO 2020a). Both specifications are for the design of crash-test articles and are not intended for direct application to design rails or overhangs for in-service use.

The rails in the examples that were crash tested fully comply with the strength provisions of proposed Article 13.6.1.1.1 and can be used to evaluate other rails using the provisions of proposed Article 13.6.1.1.2 (AASHTO 2020a).

TL-5 Concrete Barrier

The TxDOT Type T80SS was selected for the TL-5 concrete barrier design example. The T80SS rail meets the rail crash-test article design provisions of the proposed Section 13, demonstrating a current concrete barrier rail that meets the recommended loading and design methodology. It has the added advantage of being in U.S. Customary units. TxDOT notes this rail has been evaluated to satisfy MASH TL-5 criteria and was not crash tested.

In the example using the current Section 13, the railing provides ample resistance to the loading, with the end region controlling the design. The interior and end regions provide 310 kips and 201 kips of resistance, respectively, to the specified transverse impact load of 124 kips.

In the example using the proposed Section 13, the railing’s resistance at the interior and end regions is 397 kips and 258 kips, respectively. The proposed method shows increased resistance over the existing method. The applied loading with the proposed method is 162 kips.

The increase in resistance with the proposed specifications can be attributed to using expected material strengths, which increased the resistance by approximately 10%, with the balance of the increase attributed to the load application location and the trapezoidal yield-line approach.

Punching shear is not addressed in the current provisions but is a listed criterion in the proposed specification. This check was readily passed and indicated the railing could be thinner and still pass the punching shear criteria.

This TxDOT barrier also meets the recommended maximum reinforcing detailing and spacing in Article 13.12 with the vertical reinforcing but not the horizontal reinforcing. To meet this recommendation, the longitudinal reinforcing could be changed from No. 6 bars to No. 5 bars with tighter spacing and maintain the same rail resistance.

Details of the TxDOT Types T80SS rail are available at https://www.txdot.gov/insdtdot/orgchart/cmd/cserve/standard/bridge-e.htm#BRIDGERAILINGSTANDARDS.

TL-4 Steel Post-and-Beam Railing

The steel, three-tube Illinois-Ohio side-mount railing is used for this example. This railing passed its MASH crash-test program, which included tests for MASH 4-10 (small car), 4-11 (pickup), and 4-12 (SUT). As the type of beam for the side mount can vary, the anchorage of the railing was not included in the calculations; it was assumed that the anchorage was sufficient to achieve the post’s plastic moment capacity. This would be especially important with this rail as its construction details indicate difficulty in replacing posts if the beam concrete is damaged.

Structural damage to the railing after the MASH 4-12 test included a plastic hinge at one post for a two-span collapse and minor concrete spalling where two posts connected to the concrete deck. Localized deformation of the steel rails (HSS sections) was also observed. Exposure of reinforcing steel was not noted.

The transverse loading with the current Section 13 is 54 kips; the proposed transverse load is 68 kips for the 36-in. height, which is a 26% increase. If this TL-4 rail’s height increased, the proposed loading would increase.

The interior regions of this railing provided adequate resistance for both current and proposed specifications. The end region failed with both specifications, but this railing is intended for its end region to be structurally continuous with a guardrail and transition, so the calculations and results are not indicative of the railing’s intended use.

The resistance of the railing with the proposed specifications is approximately 55% greater than that determined with the current specifications. The difference can be attributed to a lower transverse load application height and using expected material strengths. The capacity-to-demand ratios are better with the proposed specification than with the current version.

Details and the crash-test report of the Illinois-Ohio side-mount railing are available at https://mwrsf.unl.edu/reportResult.php?reportId=391&search-textbox=Development%20of%20a%20MASH%20Test%20Level%204%20Steel.

TL-3 Steel Post and Aluminum Rails Mounted to Concrete Curb

The TxDOT Type T1F Traffic Rail is used for this example. This railing passed its MASH 3-11 test (pickup); the MASH 3-10 test (small car) was not performed as this test was deemed not required based on rail geometry. The selection of this railing as an example was to include more materials that are used for railings. It also highlights the combination of dissimilar metals, which is noted in the proposed specification.

Structural damage to the railing was a 3.2-in. permanent deformation in one of the rails and minor spalling with no exposed reinforcing.

Transverse load with the current specification is 54 kips; this load is 70 kips with the proposed specification, a 30% increase.

This railing’s resistance is sufficient when determined with the new specifications, with interior and end region resistances equal to 110 kips and 86 kips, respectively.

The current specification results in inadequate resistance at the end region but is adequate at interior regions, with calculated resistances of 42 kips (end) and 58 kips (interior).

The resistances with the proposed specification are close to twice that of the values determined with the current specification. The difference can be attributed to the use of expected material strengths and reduction in load application height.

Details of the TxDOT Type T1F rail are available at: https://www.txdot.gov/insdtdot/orgchart/cmd/cserve/standard/bridge-e.htm#BRIDGERAILINGSTANDARDS.

Deck Overhang with TL-5 Concrete Barrier

The deck overhang for this example comes from the same railing used for the TL-5 concrete barrier example. TxDOT uses an overhang that was tested to MASH TL-5 but with another railing. This railing and overhang is for the TxDOT Type T224, and it passed the MASH 5-10 (small car), 5-11 (pickup), and 5-12 (tractor-semitrailer truck) tests. This rail is a curb-mounted concrete post-and-beam rail that puts a more concentrated demand on the overhang due to the discrete posts instead of a concrete barrier that allows more load distribution.

Again, a significant difference in the current and proposed specifications for this example is with the transverse loading. The proposed loading, 162 kips, is 31% greater than the current specified loading of 124 kips.

The existing specifications indicate the overhang is not adequate, with the lowest capacity-to-demand ratio being 0.50 (Design Case 1, Interior Region controlling).

The proposed specifications show the overhang to also have insufficient capacity, with the end region (Design Case 1 controls) being a capacity-to-demand ratio of 0.92. The proposed specification offers a better prediction than the current method.

Deck Overhang with Concrete Post-and-Beam Railing

The deck overhang for this example comes from the TxDOT Type T223 Traffic Rail, a 32-in.-tall concrete post-and-beam rail with posts mounted directly to the deck. Posts are 4 ft wide and spaced on 10-ft centers. The rail passed its MASH 3-11 tests, which were performed on a simulated 5-in.-thick cast-in-place concrete overlay on a prestressed box beam. It should be noted that the 5-in.-thick overlay prevents anchoring the rail with reinforcing embedded sufficiently to achieve its yield strength.

Bogie tests were performed with this rail/deck overhang combination for the TxDOT, with the bogie vehicle hitting the railing at 90°. The overhang thickness was 8 in., and the cantilever span was 2.5 ft. The 50-ms average impact force at an end post was 62 kips. Interior posts were impacted with a 50-ms average impact force of 89 kips. Observed overhang damage was cracks emanating from the post edges downward through the deck, indicative of punching shear. A second impact test was performed on one of the end posts, which had resisted 62 kips. The impact force resisted at this post, with unrepaired damage from its first test, was 55 kips (50-ms average).

The recommended transverse load for TL-3 with the current specification is 54 kips. The proposed specification lists 70 kips for the transverse load—a 30% increase in recommended loading.

The current specifications indicate the deck overhang to be inadequate at the end region in punching shear, with a capacity-to-demand ratio of 0.71.

The proposed specification predicts better performance, even with the larger loading. The lowest capacity-to-demand ratio is 0.99, which is for punching shear at an end post.

Summary

Overall, the proposed specifications provide generally improved results than the current specifications. It must be re-emphasized that these design provisions, from both the current LRFD BDS Section 13 and the proposed Section 13, are for the design of crash-test articles. In each of the examples where the item successfully passed its crash tests, the as-tested item provides adequate strength in accordance with the proposed Article 13.6.1.1.1 and can be used for in-service use. Each successfully crash-tested element can also be used for evaluating other rails and overhangs in accordance with the proposed Article 13.6.1.1.2.