The Refurbishment of Union Chain Bridge: Part 1

Alison Church
Good afternoon everyone, and welcome. Thank you for joining us this afternoon. I am Alison Church, Senior Structural Engineer at Historic England, based in the North East, but today I am in sunny Portsmouth and there are blue skies out there, which is lovely. So welcome all. Without further ado, I’ll pass on to Simon and Ryan from Northumberland County Council.

Simon Rudman
Good afternoon everybody. Before Ryan kicks us off, we at NCC would just like to commemorate these two technical talks and the subsequent technical paper to the memory of Professor Roland Paxton, who sadly passed away earlier this month. Roland had quite a lot to play in getting this refurbishment finally across the line and was responsible for much of the international recognition that the project gained. To you, Ryan.

Ryan Convery
Alright, good afternoon everybody. As Simon said, my name is Ryan Convery and this is my colleague Simon Rudman. I'm a Structural Engineer for Northumberland County Council and I've been involved with the Union Chain Bridge project since 2017. During the refurbishment of the Union Chain Bridge, I was the NEC Supervisor and Design Team Lead for Northumberland.

Simon Rudman
I’m the Design Manager for Northumberland County Council and I effectively acted as the Client Manager during the refurbishment.

Ryan Convery
The bridge is located in Berwick‑upon‑Tweed, right on the border between Northumberland and the Scottish Borders. Interestingly, the Scottish end of the bridge actually sits further south than the English side. The bridge crosses the River Tweed, with the English‑Scottish border running directly through the centre of the structure.
The suspension chains span across the river, and the deck itself consists of timber supported by wrought iron chains, links and pins, cast iron hanger caps, wrought iron hangers, steel bottom chord angles, steel wire rope hangers, wrought iron handrails, masonry towers and the original anchorages.

Construction of the bridge was completed in 1820 and it was designed by Captain Samuel Brown. Brown was a naval ship captain, and he developed the idea for the suspension chains through his experience with ship rigging. It remains the oldest suspension bridge in the world that still carries vehicular traffic. Over its lifespan, it has undergone numerous repairs and refurbishment schemes.

To give a brief overview of the main components of the bridge: the bridge consists of wrought iron chains, links and pins; cast iron hanger caps; wrought iron hangers; a timber deck; steel bottom chord angles; steel wire rope and hangers; wrought iron handrails; masonry towers; and the original anchorages.

A typical chain node has the hangers passing through the cast iron hanger cap, which rests upon the link and pin connection between the chains.

Ryan Convery — History of the Bridge
The bridge was originally constructed on behalf of the Berwick and North Durham Turnpike Trust. Before the bridge existed, the river was crossed by a ford, but industrial growth in the early nineteenth century created demand for a permanent crossing. At the time of construction, it was the longest iron suspension bridge in the world and the first to carry vehicles. Captain Brown even demonstrated its strength by crossing it with a horse and cart in front of spectators.

Responsibility for the bridge has passed through several organisations. It was maintained by tolls until the late nineteenth century, then by the Tweed Bridges Trust until the 1980s. Since the dissolution of the Trust, Northumberland County Council and Scottish Borders Council have jointly been responsible.

In 2017, both councils signed a memorandum of understanding to conserve the bridge and work together in long‑term partnership.
Major maintenance over the decades included deck replacements, supplementary wire ropes added due to load concerns, and replacement of corroded links and pins. Paint analysis undertaken for this refurbishment identified many layers of historic paint, none of which had ever been removed prior to repainting.

In the 1990s, hanger failures began to appear, particularly on the shorter ones. Temporary repairs using threaded stainless bar and simple steel plates kept the bridge operational, but the repeated failures led to the development of a Conservation Management Plan. This document effectively started the journey that ultimately led to the full refurbishment of the bridge.

In 2013, the bridge was placed on the Heritage at Risk Register. Funding was difficult to secure, but a first‑round application to the National Lottery Heritage Fund was successful. After a development phase, a second‑round submission was awarded significant funding, which enabled both councils to bring the difference together. The construction contract was awarded in 2020.

Ryan Convery — Identified Challenges
Following the Conservation Management Plan and our own inspections, several challenges were identified, including issues with the anchorages, the deck, the hangers, the handrail, the masonry, the wire rope, and the chain components.

Ryan Convery — English Anchorage
The English anchorage sits behind the English tower within the rock face. The chains pass through a chamber at the top of the tower and are anchored with embedded plates slotted into the rock behind it. We explored how this arrangement could be investigated to determine whether it could be reused for a design life of over a century.

However, there was no way to investigate the anchorages without breaking out the rock face, which would undermine the strength of the system and potentially cause damage. Any load testing would provide information at only one moment in time and would more than likely disturb or damage the anchorage. They had been in service for over two hundred years and were not designed for modern usage.

The solution was to install new splayed rock anchors while keeping the original system in place. Inspection chambers were also added to aid future maintenance. This allowed us to guarantee the required design life and add redundancy to reduce the likelihood of major future works.
We developed the proposal for the new rock anchors, to be housed in a reinforced concrete interceptor beam, working closely with Historic England and our appointed Conservation Architect.

Ryan Convery — Scottish Anchorage

Similar to the English side, the Scottish anchorage relied heavily on archive information. Ballast plates were embedded into the ground, with the chains connected onto them. Again, investigations were infeasible as excavating would disturb the anchors, and there was no clear way to restrain the bridge temporarily while doing so.

Several refurbishment options were considered. A concrete pile solution was originally explored, as well as a splayed anchor solution like the English side. However, there was insufficient space. A value‑engineering exercise resulted in a concrete anchor block being selected — a pragmatic solution acting as a dead weight.

The block was installed beneath the carriageway with a shear key to resist sliding and included inspection chambers for future maintenance access.

Ryan Convery — The Deck
There have been several iterations of the deck over the years; however, throughout its lifespan, the deck is and always has been timber. A principal inspection undertaken in 2017 found substantial rot, which would be expected for a timber arrangement that had been in service for over forty years. The timbers had been creosoted, which explains why they lasted as long as they did, but they had reached the end of their design life.

The ACME panelling to the carriageway had also failed and had been reactively maintained simply to keep the surface usable.
The deck is arranged with carriageway and footway timbers spanning between the main transverse timber beams, which are fixed to the steel bottom chord angles. These angles transfer loads into the hangers and then into the suspension system. The transverse beams sit upon longitudinal timber beams, which are braced to stiffen the structure.

The new arrangement replicated the original, while replacing the bracing and bottom chord angles with modern steel equivalents. Analysis of the deck showed the existing steelwork sections would fail and could not be reused. They also needed to be made thicker to increase capacity to withstand the proposed three‑tonne weight limit, improving upon the unenforceable previous two‑tonne limit.

To reduce load eccentricity and discourage vehicles from trying to pass on the bridge, the walkways were widened to direct traffic to the centre. This also created a more pedestrian‑friendly environment.

Ryan Convery — The Deck Flap
Part of the deck is the deck flap, installed in the 1950s to provide articulation at the English end. Previously, the deck had been fixed at this end, slotted into the English abutment. At this end of the deck, the hangers are not fixed to a rigid point on the chain and are free to move. However, the deck itself was effectively fixed in place. The movement of the hangers induced significant stresses in the deck as they interacted with the bottom chord angle and the deck structure.

In the 1950s, a period of bad weather and high winds caused significant damage to the deck, which led to the decision to introduce the articulated flap. The arrangement was based on a welded round bar fixed to a steel angle running across the width of the deck, allowing rotation up to a maximum controlled by welded upstands.

A video showed how responsive this flap is, even on a mild windy day, demonstrating how lively the deck can be at this point.
As part of the refurbishment, the flap was to be investigated during the works to determine the best proposal for its renewal, with the initial intention of like‑for‑like replacement.

Ryan Convery — Masonry
The towers were in typical condition for their age and exposure: vegetation growth, cracking, staining, and substantial areas of erosion caused by long‑term exposure. Some historic repairs had actually created further damage, particularly cementitious mortar used throughout the towers. This material trapped water, leading to freeze‑thaw damage and further cracking.

The difference between the upstream and downstream elevations of the Scottish tower highlighted the contrasting conditions: the upstream elevation showed significantly more erosion, while the downstream side had extensive lichen growth due to being more sheltered.

Substantial inspections were undertaken, checking every stone to provide a comprehensive repair and replacement schedule. Stone samples were analysed, revealing the original stone type, and a suitable match was identified — Swinton Pink, a local sandstone.

Ryan Convery — Wire Rope
As mentioned earlier, wire rope added in the early twentieth century addressed concerns about load‑carrying capacity. The wire rope consisted of steel cables with steel hangers passing between the chains and connecting to the bottom chord angle.

However, finite element modelling showed the wire rope attracted no significant tension compared to the chains. Worse, its presence created further defects. Lateral instability caused the wire rope hangers to clash and rub against the chains.

After consultation with Historic England and the Conservation Architect, it was agreed the wire rope should be removed, returning the bridge to its original suspension arrangement.

Simon Rudman — The Hangers
The original hangers connected to the chains were experiencing failures. Though we were making temporary repairs, clearly something else was happening. We suspected it was due to the flexibility of the bridge.

The hangers are held solid inside the hanger cap. The hanger has a wedge end hammered onto it, sitting in a tapered pocket. This effectively makes the hangers fixed and unable to move. The bridge is flexible enough that we believed the hangers were being fatigued at the point where they connected to the hanger cap. This was supported by the failure surfaces we observed.

To investigate, we added microstrain gauges to several hangers and drove a vehicle up and down the bridge at varying speeds. The strain plots showed significant reversing fatigue, continuing for almost a minute after the vehicle left the bridge. The deck flap video shows there is essentially no damping, allowing oscillation to continue.

The original detail also had the hangers passing through the handrail to provide lateral support. Differential movement between the handrail and hanger caps caused deformation and bending.

This issue took considerable time to resolve with Historic England and the Conservation Architect. The solution was to remake the hangers in a modern high‑yield steel that could resist fatigue while maintaining the original appearance.

We also separated the handrail from the hangers by moving the handrail inboard, reducing strains and preventing further damage. Combined with widening the footways and narrowing the carriageway, the bridge became safer for pedestrians and discouraged cars from attempting to pass each other.

Ryan Convery — The Handrail
The handrail consists of wrought iron rails and posts. As mentioned earlier, the hangers also acted as posts intertwined with the handrail, contributing to fatigue damage. Moving the handrail inboard freed the hangers from this restraint.

Originally, the handrail stepped with the profile of the chains, meaning there was no formal handrail along the middle third of the deck where the chains acted as the parapet. With the parapet being moved inboard, a replacement section needed to be designed to integrate seamlessly with the original handrail.

Stainless steel backstays were added to provide restraint, contrasting visually with original components and highlighting what was historic and what was new.

The handrail showed typical defects such as corrosion and paint loss, along with cracking at connections caused by movement and interaction with the hangers.

A proposal was developed showing how the new central handrail section would join the original, maintaining a constant height until the handrail stepped at the far end.

Ryan Convery — Chain Components
This is an archive drawing of the node connection, showing the arrangement with the links and pins connecting the chains, with the hanger cap above. This arrangement has not changed since the bridge has been in service.

Here are some photos of the chain arrangement. Throughout, the condition was fairly typical, with significant paint loss and areas of surface corrosion. Within the connections, however, there were hidden faces that could not be inspected without dismantling the chains completely.

During targeted inspections at height, areas of corrosion at bearing locations on the chain rods were visible, highlighting the need for dismantling to fully understand the condition. At the top of the Scottish tower in particular, access was extremely tight and many components could not be visually inspected. Thick layers of paint also made it difficult to determine whether any hidden defects existed.

We undertook non‑destructive testing on the chains by conducting magnetic particle inspections. For every chain tested, a hairline crack was identified at the chain eye, further highlighting the need for more intrusive investigation — which we’ll cover in the next presentation

Ryan Convery — Closing of Presentation
We hope this has provided some insight into the challenges we faced during the development of the scheme. In the next presentation, we’ll look at the construction phase: how the refurbishment was carried out, how the proposals were implemented, and what further issues had to be resolved along the way.

It wasn’t just the engineering challenges we had to consider — access and the surrounding environment were also critical. An archive photo shows the level of flooding that can occur in the area, highlighting the importance of access, careful planning, and timing of the works.
Thank you for listening. Any questions?

Q&A Session
Alison Church
Thank you, Simon and Ryan. I have Martin Lowe, the Inspector of Buildings and Places for the North East, on the call as well, so he’s here to answer some questions. I believe Patrick Smith from Northumberland County Council is also here. If you have any questions, please pop them in the chat and we’ll go through as many as we can.

Simon Rudman
The first question is: What were the key factors in the successful National Lottery Heritage Fund application and award?
I think the key factor was the fact that we had gone through all the issues — each as a separate workstream — and developed them in detail with Historic England and our Conservation Architect, reaching workable solutions that everyone agreed on. When it came to evaluation, it was clear that the options were deliverable.

Simon Rudman
What does the vertical axis measure on the strain table?
That was microstrain — seconds on the bottom, microstrain up the side.

Audience Question
A comment rather than a question: “I visited the Union Chain Bridge just after the refurbishment. The ratio between the weight of the bridge and its maximum load struck me as quite large — I think it was 33:1. Amazingly inefficient, yet beautifully done.”
Simon Rudman
Good question. It’s not a particularly efficient structure at all. We’ll check the exact numbers and answer this again in the second presentation.

Audience Question
Was any further analysis of the paint layers undertaken?
Ryan Convery
This was undertaken by Crick‑Smith. Most layers were lead‑based. We’ll check the report for further details.
Simon Rudman
There’s a full Crick‑Smith report that goes through all the historic paint schemes. We can look into whether it can be published.

Audience Question
How did you find technical approval of the proposals?
Patrick Smith
As we deliver all services in‑house, we are our own technical approval authority. The hard work was in explaining and managing the defects we proposed to keep, and ensuring it was safe to do so. The easiest thing would have been to remove everything and start again, but that wouldn’t preserve anything. Simon and Ryan worked hard to convince me that retaining elements was safe, and it paid off.

Audience Question
Did movement and flexing make the corrosion issues worse?
Simon Rudman
For the handrail components, yes. The structure is long enough that individual components didn’t move massively, but bending in the chain rods was more significant.

Audience Question
How have you assessed the risk of scour with historic flooding?
Simon Rudman
Scour has never been a major issue. Although the area floods, the foundations are deep compared to any expected scour. The English tower is also set quite far back.

Audience Question
Is the bridge separately listed by Historic England and Historic Environment Scotland? Did this cause challenges?
Ryan Convery
Yes — it’s Grade I in England and Category A in Scotland. We had to complete all approvals on both sides. It created a lot of extra work, but the processes were similar.
Martin Lowe
The project involved several authorities, but all parties worked together well.

Audience Question
Was there originally a little house on the English side?
Simon Rudman
Yes — a toll house. A family of four lived there. If you visit the bridge, you’ll see an outline of its footprint in the paving. An archive photo in the deck flap section shows the family standing in front of it.

Audience Question
Do you have an idea of the additional dead load of the paint systems?
Ryan Convery
Negligible compared to the weight of the structure.
Simon Rudman
And less than when we started — after removing the many accumulated layers.

Audience Question
Was all the analysis and design done in‑house?
Ryan Convery
Yes — except for the Category 3 check by AECOM and design of the rock anchors by a specialist.

Audience Question
How was the potential for vehicle impact to the hangers or chain managed for ongoing operation?
Simon Rudman
A risk‑based approach. With the new arrangement, you would need to mount the curb, pass through the inboard parapet, and the bridge is designed to operate even with a hanger missing.

Audience Question
Did you need listed building consent on both sides, and did this allow for changes once parts were inspected offsite?
Simon Rudman
This leans into Part Two — but ultimately it was not an issue, and all parties worked together.

Audience Question
Did you take the paint off in situ?
Ryan Convery
Only small trial areas. The majority was removed offsite after dismantling.

Audience Question
At what point did you realise you needed full dismantling rather than in‑situ repairs, and why?
Simon Rudman
As we inspected more, the number of unknowns grew. We couldn’t see inside much of the structure. Smaller repairs would not achieve the required design life. Dismantling everything allowed us to inspect every component and ensure a known condition. This proved to be absolutely the right decision — and Part Two will show why.

Audience Question
Was the masonry different between the Scottish and English anchor areas?
Ryan Convery
No — same stone.

Audience Question
With hindsight, is there anything you would change?
Simon Rudman
That will be addressed in Part Two.

Audience Question
What does the future maintenance program look like?
Ryan Convery
There is a detailed plan including ongoing cleaning, painting, routine inspections, and anchor testing.
Simon Rudman
Apart from paint deterioration and cleaning, anchor testing is the main long‑term task.

Audience Question
Did paint build-up restrain movement between chain elements?
Simon Rudman
You would think so, but the chain rods are flexible in bending. Paint build-up didn’t appear to restrain movement significantly. There’s another issue caused by paint build-up, which we’ll discuss in Part Two.

Audience Question
What were your findings regarding lime mortar specifications?
Ryan Convery
An agreed mortar mix was developed based on analysis of surviving original mortar. Samples were produced and agreed upon by all conservation bodies.
Simon Rudman
Nothing unusual — quite standard for historic masonry repairs.

Audience Question
What non‑destructive testing did you carry out besides MPI?
Ryan Convery
Before dismantling, MPI on chains and dynamic stress monitoring on hangers. Once dismantled, further testing was performed — more on that in Part Two.

Audience Question
Why did you need an architect?
Simon Rudman
Specifically a Conservation Architect — essential to bridge the gap between engineering requirements and heritage preservation. They provided impartiality and helped reconcile differing priorities. We would use one again for any similar scheme.

Alison Church
That wraps up all the questions today. Thank you very much Patrick, Simon, Ryan, and Martin. And we’ve dropped enough cliffhangers that hopefully you’ll sign up for Part Two. We’ll see you in December.