Highways Agency

Safety fence provision

Design Of The Bridge Parapet And Safety Fences

9.1 As discussed earlier in this report, the designs for the Little Heck Railway Bridge were formulated between October 1971 and June 1972.

9.2 At the time of design, guidance on parapet provision appeared in the "Ministry of Transport Technical Memorandum on the Design of Highway Bridge Parapets" (BE 5), 1970. BE 5 was in its second reprint in September 1970 after it was first published in January 1967.

9.3 The 1970 revision of BE 5 (and indeed, its previous versions) specified that a P5-type parapet was required for use over all railways. In the specific case of a bridge carrying a motorway over a railway, BE 5 furthers this by stating that:

"205c(i) On bridges carrying a motorway over a railway, vehicle parapets shall comply with all the requirements for a P1 parapet."

9.4 P1 parapets were devised to be capable of containing a 1500 kg vehicle travelling at 120 km/h, impacting at a 20° angle of incidence. BE 5 provides examples of such P1 designs within an Appendix.

9.5 The BE 5 detail for a P5 parapet differs from that of a P1 parapet in that the bottom 600 mm of a P5 parapet should be constructed "so as to prevent stones from being projected onto the railway below by passing vehicles".

9.6 Another difference to note is that BE 5 also stipulates that, for bridges over non-electrified lines (such as at Little Heck at the time of the bridge's design and construction), the minimum height of the parapet above the adjoining paved surface shall be 1.25 metres.

9.7 The as-built parapet drawings for Little Heck Railway Bridge show that the parapet was designed in accordance with the criteria in Paragraphs 9.4 to 9.6 above. As-built drawings also designate the material parapet as being P5-type. Hence, it is felt that, at the time of design, the parapet at the material location met the design requirements of BE 5 (1970).

9.8 With reference to the provision of the safety fences BE 5 (1970) also states that:

"205a(ii) P5 parapets over railways: Where no effective alternative arrangements exist to prevent a vehicle from leaving the bridge approach and falling onto the railway below, a safety fence shall be provided on each approach towards the bridge, as in the cases of Group P1 and P2 parapets."

9.9 With regard to the provision of safety fences in advance of P1 parapets BE 5 (1970) also states:

"201 P1 Parapets, d (i) To prevent direct impact between a vehicle and the end of the vehicle parapet facing the traffic on the nearside, a safety fence shall be provided on each approach towards the bridge. The safety fence shall be at least 30m (100ft) long and shall continue the line of the traffic face of the parapet."

9.10 An as-built drawing from April 1972 (original Drawing Number: 2NE/M62/FP/12/M62-0RLY/1A) shows that an approximately 40 metre length of open box beam safety fence was provided in advance of the parapet at the time of the bridge's construction. This 40 metre length was not all at full height, i.e. it would have included the ramped end down to a buried anchorage. This 40 metre length is termed approximate here, as it has been scaled from the as-built drawings.

9.11 The as-built drawing shows that no flare was originally provided at the anchorage end of the safety fence. Such a flare was not required in BE 5 (1970), but was subsequently specified in the July 1982 revision of that document. It is considered that this may have been one of the factors leading to the apparent replacement of the safety fence during major maintenance work on the westbound carriageway in 1993.

9.12 Drawings issued prior to the maintenance work in 1993 indicate the inclusion of a flare to the end of the safety fence, and it is considered that the drawing is consistent with the safety fence layout currently at the material location.

9.13 A large area of concrete before the anchorage of the safety fence, identified during the post-accident investigation by TRL is also probably a result of the safety fence being replaced during the major maintenance work in 1993. The position of this concrete is consistent with the position of the original anchorage shown in the as-built drawing.

9.14 Safety fence provision at the material location, both at the time of original construction and present at the material time, appear to comply with the pertinent requirements of BE 5 (1970).

9.15 Hence, it is considered that both parapet and safety fence provision at Little Heck Bridge complied with the requirements of the "Ministry of Transport Technical Memorandum on the Design of Highway Bridge Parapets" (BE 5), 1970 at the time of its design and its construction.

9.16 The current standard regarding the layout of safety fences and barriers, "Safety Fences and Barriers", TD 19/85, was published in June 1985. The standard superseded previous publications Memorandum H 9/71 and H 9/73.

9.17 TD 19/85 requires the provision of at least 30 metres of safety fence to protect roadside obstructions. TD 19/85 also reinforces the need for the ends of safety fences to be flared away from oncoming traffic:

"5.1.3 The ends of safety fences facing on-coming traffic shall be ramped down to an anchorage at ground level and flared away from the carriageway by setting back the anchorage behind the line of the fence as detailed in the Specification for Road and Bridge Works, the RM/F series of Standard Drawings and TRRL Drawings"

9.18 Measurements taken at the material location show that a full height length of 33.5 metres is currently provided in advance of the parapet, plus an additional 9.2 metres of ramp down to a concrete terminal, i.e. a total length of 42.7 metres.

9.19 Hence, the requirements of TD 19/85 with respect to the advance length of the safety fence have been satisfied at the material location.

9.20 Technical standard BD 52/93 "The Design of Highway Bridge Parapets" was published in April 1993. The standard details the design of parapets, and supersedes BE 5.

9.21 BD 52/93 states:

"2.21 To prevent direct impact between a vehicle and the end of the vehicle parapet or vehicle/pedestrian parapet facing the traffic on the nearside, a safety fence shall be provided on each approach end of the parapet. A safety fence shall be provided on the departure end where considered necessary. The safety fence shall be at least 30m long and shall continue the line of the traffic face of the parapet."

9.22 As the detail of the safety fence previously outlined in Paragraph 9.18 has shown, the provision at the material location complies with the above requirement.

9.23 The standard (BD 52/93) also states:

"8.5 On bridges over railways with existing or programmed overhead electrification, parapets consisting of a plinth surmounted by metal posts and horizontal rails shall be provided with solid or open infill panels immediately above the top of the plinth up to the full height of the parapet"

9.24 At the time of design, the railway was neither powered by overhead electrification, nor was this planned. As a result, infill panels were not provided on the parapet until the application of aluminium sheets at the material location in May 1987, after notification by British Rail that the railway was planned for electrification.

9.25 When BD 52/93 came into effect in 1993, the parapets at Little Heck Bridge were already equipped with infill panels and therefore met these requirements.

9.26 TRL is informed that the Highways Agency has an ongoing programme of parapet upgrading which includes consideration of this site. This may result in this site being equipped with high containment parapets at some future date.

9.27 However, the provision of a high containment parapet at Little Heck Bridge would not require any greater length of safety fence in advance of it than is currently provided.

9.28 Parapet containment criteria per se are not an issue in relation to the accident involving Mr Hart, as his vehicle did not strike any part of the parapet or safety fence in advance of it. The main issue relating to Mr Hart's accident appears to be the length of safety fence provided. Therefore, it is considered that parapet containment issues are not relevant to these investigations and are not discussed further in this report.

Summary

9.29 In summary, the relevant issue to this report relating to safety fence provision at Little Heck Bridge is the length of safety fence on the nearside verge of the westbound carriageway in advance of the bridge. The measured length on site was found to comply with current national standards and their predecessors.

Performance Criteria For Safety Fences And Parapets

9.30 The controlled testing of safety fences and parapets has been carried out for many years.

9.31 The testing of safety fences has, until recently, been to the guidelines of Series 400 of the Highways Agency's "Specification for Highway Works". This states that safety fences shall be impact tested to the following criteria:

(Please note that types of safety fence not pertaining to the material location have been omitted)

9.32 The two required OBB tests have been successfully carried out in controlled impact test conditions. In both of the tests, the test vehicle was contained and redirected safely, and the barrier performed to designer's specifications.

9.33 Parapet testing is carried out following the guidelines within BS 6779, Part 3, 1994 "Specification for vehicle containment parapets of combined metal and concrete construction".

Vehicle Impact Test Criteria, as defined in BS 6779: 3: 1994

9.34 The containment levels contained in BS 6779: Part 3, 1994 can be related to the Parapet Group Designations given in BD 52/93, "The Design of Highway Bridge Parapets", 1993:

• P1; P2 (113km/h): Normal level of containment
• P5 (excluding footbridges): Normal level of containment
• P6: High level of containment

9.35 P5 parapets are designed for use over railways only and differ from P1 and P2 parapets in relation to their design. These differences are outlined in Paragraphs 9.5 and 9.6. The parapet at the material location is of P5 designation, and this is verified by the as-built drawings.

Possible Interaction Between The Material Vehicle And The Safety Fence

9.36 There are two possible ways in which the material vehicle and/or its trailer may have impacted with a longer safety fence at the material site:

• Striking the end terminal
• Full engagement with the safety fence

9.37 The vehicle involved in the material accident was a Land Rover County. This vehicle has a greater mass than that specified for approval testing in Series 400 of the "Specification for Highway Works" (with the material vehicle's mass being approximately 1800 kg). A vehicle complying with the 1500 kg mass requirement would be a larger saloon motor vehicle such as a Ford Granada or a Vauxhall Omega.

9.38 The Land Rover County has larger wheel dimensions than either of these typical test vehicles and, due to the height of the material vehicle (approximately 2050 mm), the vehicle's centre of gravity will also be higher than that for saloon motor vehicles.

9.39 Due to the differences between the material vehicle and a typical test vehicle, it is difficult to predict what interaction may have occurred if the material vehicle had contacted with a full-height safety fence.

9.40 Past impact tests at TRL have shown that vehicles striking the end concrete terminals of safety fences can, in some cases, become unstable. As the centre of gravity of the Land Rover may be higher than that of a saloon motor vehicle, it is considered that this mechanism may have occurred had the Land Rover struck an end terminal. This could have been to the extent that the vehicle may have rolled onto its side.

9.41 This may or may not have caused injuries to the vehicle occupant. However, it is considered possible that such an impact would have prevented the errant material vehicle from reaching the railway line.

9.42 In order for the entirety of the vehicle to have engaged with a full height section of safety fence, the length of the safety fencing present at the material location would have been required to have been in excess of twice its current length. This can be seen by the material vehicle's tyre tracks shown on Drawing No: 03719/106/2/001, to the rear of this report.

9.43 It is difficult to estimate with any certainty whether a longer safety fence would have successfully contained an impact with the entirety of the material vehicle. As previously stated, the higher centre of gravity of the material vehicle would have increased its propensity to roll in an impact. However, it is considered probable that even partial containment by the safety fence would have directed the errant vehicle away from the railway line.

9.44 The influence of a loaded double axle trailer unit on the path of a Land Rover, impacting fully or partially with a safety fence, is unknown. Due to the pivoting action of the trailer to Land Rover connection, it can be assumed that the trailer would have limited effect on the redirection characteristics of the safety fence on the Land Rover. However, the additional weight of the trailer, and its content, may have influenced the containment performance of the safety fence.