Network Rail update the Viaduct

Harringworth Viaduct is a major structure on the line between Manton Junction and Corby, set in glorious Northamptonshire countryside. It is a brick arch structure 1182m long with 82 spans, three of which cross roads and one of which crosses the River Welland. It was built between 1876 and 1878 using bricks made in three brick kilns on the site.

Brickwork repairs and other works have been required almost annually for a number of reasons.

Work in 2016/17 has been undertaken for Network Rail by AMCO, who also did some repair work to the viaduct in 2012, following previous repairs by May Gurney in 2009. This work has two aims: the first is a continuation of the kind of repairs which have been made regularly in the past. The second is the more complex one of strengthening the structure to allow heavy freight trains to pass over it at 60mph, the normal maximum for them on Network Rail’s infrastructure. Currently they are restricted to 20mph, and this is inhibiting the infrastructure operator from using the route as a diversionary one for freight trains serving the port of Felixstowe.

The first task was, on the surface, a straightforward one, something which might seem to require simple repetition of what has been done before. However, there are some changes in the remit, which have required some reappraisal of that attitude! The biggest change was in the scope. Instead of being asked to repair the seriously damaged areas on the viaduct, leaving less urgent damage to be dealt with at some future date, AMCO was  remitted to repair everything.

That made sense economically and in engineering terms, but the most important issue must be the economic one. The costs of mobilising a team on such a site are great in relation to the costs of the works, so once these have been paid for it makes sense to make the maximum use of this. The land under the viaduct is agricultural and a flood plain, so there were costs in obtaining access over the farmland, environmental protection costs and expense incurred in setting up site facilities and creating access tracks under and adjacent to the structure. All that and more before any work could be considered.

The other change is a technical one, Network Rail’s requirement to avoid the use of pattress plates in the way that they have been in the past here. Harringworth Viaduct is Grade II listed and a great deal of effort was given to maintaining its appearance unchanged.

The repairs undertaken were determined by employing Donaldson Associates, whose Derby office undertook the necessary investigations and prepared reports detailing the defects and recommending how to deal with them.

The piers, arches, spandrels and parapets all required significant work to ensure that they can continue to function safely to the present degree, before any consideration could be given to the freight traffic speed increase works.

The piers are consistently cracked vertically, as well as having areas of spalled or drummy brickwork and areas needing repointing.

Spalling and the occasional complete loss of a brick are features of the arches too. Drummy areas also feature, but the cracking is very much the significant issue affecting the arches. Some arches have transverse cracks at the third points suggestive of hinge formation, a serious concern. Most arches have longitudinal cracks, usually a short distance in from each face and apparently directly below the inner faces of the spandrel walls. Many also have a similar crack beneath the centre of the 6ft of the tracks above.

The spandrels had significant areas that were pushing outwards and/or separating from the arch rings, horizontal cracks and often water seepage.

Parapets are of substandard height above the ballast shoulder, many lengths are leaning outwards significantly, and there are many horizontal and vertical cracks.

AMCO and Donaldsons worked out appropriate repair methods based upon Donaldson’s reports and agreed with the client, Network Rail. These range from conventional repairs like repointing and brickwork repairs through relatively familiar methods like stitching and grouting to some more innovative approaches.

Of particular interest were the parapet repairs. The vertical cracking was being dealt with by replacing bricks that have cracked through, and by installing 6mm stitch bars within the brickwork across the cracks. The bases of the parapets were tied more securely to the rest of the structure by means of brackets or stitch bars. None of that is too unusual, but the approach to the horizontal cracks was more innovative. The concept of inserting vertical stitch bars is not new but the means of doing so is. Readers of Rail Engineer will have heard of Fulston Forge before in the context of innovative pieces of equipment. In this instance, they developed a special drilling rig for AMCO to mechanise and speed up the process of drilling the holes vertically down into the parapets.

The drilling rig is designed to be self stable, once it has been lowered onto the top of the parapet by an RRV. It consists of a frame that holds 3 drills at the correct centres along the wall, and once aligned, it can drill 3 holes down into the parapet within 3 minutes. 20mm bars are inserted and grouted into the holes, and failed bed joints are raked out and repointed to complete the repair.

All the foregoing works are scheduled to be completed by March 2017, at a cost of some £7.6m.

The question of strengthening for heavy freight is also being addressed by Donaldsons. Under the railway system of categorising track and structures by load capability, the Viaduct is currently rated as RA0, the lowest possible category, meaning that 25 ton axle freight can pass over at only 20mph. Network Rail require this to be increased to the maximum RA10 to permit the freight to run at 60mph.

Donaldsons used arch assessment software known as Archie-M, provided by Obvis. They assumed in the analysis that all of the structural defects they had identified in the Viaduct would have been rectified. Running the analysis confirmed the current RA0 rating and indicated that even in this repaired state the structure would remain at risk of collapse if RA10 loading was applied. The modelling indicated that the load path in the piers at certain spans would lie outside the pier structure, suggesting collapse would be likely. A Category 3 check of this work undertaken by Bill Harvey Associates confirmed the findings. Bill Harvey was the developer of Archie-M.

Donaldsons prepared a report suggesting options for the required strengthening to achieve RA10. These were threefold:

  • over arch reinforced concrete saddles
  • under arch strengthening by reinforced concrete arches supported on steel needles through the piers
  • increased “backing” to the arches, carried out by removing the random rubble fill either side of each arch over the piers and replacing it with mass concrete, up to the level of the arch tops

The first option would add the least dead weight to the structure and have practical advantages including simple incorporation new waterproofing and of works to tie in the parapets. However, it would be extremely expensive, require lengthy blockades, and because it would add depth over the tops of the arches, it would cause problems with vertical track alignment.

The second option would be even more expensive and would add 3 to 4 times the extra weight of the first, whilst also significantly altering the appearance of the Viaduct. Practical difficulties would result too, such as the need to construct scaffolding over the river, and the need to deal separately with the waterproofing and parapet works. These last would mean that the need for possessions would probably still be there, negating the advantage that under arch strengthening might otherwise have offered.

Option three would add less dead weight than option two, but still almost 3 times that of option one. Like that option it would allow the waterproofing and parapet works to be dealt with simultaneously, but would require similarly major possessions.

Both the second and third options might well require underpinning to strengthen the pier foundations to handle the extra dead weight and live loading combination.

The complexity of this is such that further work is required for option development and selection to proceed. As a result Network Rail has deferred the strengthening works to CP6. It is proposed to install a fibre optic monitoring system beneath the arches of the Viaduct using technology developed by the University of Cambridge Centre for Smart Infrastructure, led by Matthew DeJong. The outputs from this system will be validated by conventional deflection monitoring of a sample number of arches.

The results of the monitoring will be used by the project team, including Network Rail’s Senior Project Manager, Chris Chatfield, Senior RAM Mark Norman, Structures RAMs Chris Levett & Colin Watson and Assessment RAM Engineer Chris Etherington. A preferred strengthening option will be identified, developed and submitted for approval by the whole team.