With more intensive services coming to the East Coast main line, upgrading the overhead line equipment is all about sustainability
As is now generally acknowledged, the poor reputation for reliability of the East Coast main line (ECML) electrification was more the result of nurture than nature. Only in the past decade has the British Rail Mk 3a and Mk 3b Overhead Line Equipment (OHLE) received the maintenance required for reliable operation.
However, in the 40 years since the Great Northern electrification to Hitchin, and 25 years on from when the main line scheme was completed, much has changed, and is continuing to change. As a result the OHLE is also part of the programme of enhancements and upgrades that are creating the 21st century high capacity ECML.
So what has, belatedly, been done to fix the minor weaknesses in the BR equipment that were causing the high failure rates? For example, at one stage, short circuit-related catenary failures were averaging one a month. Because the cause was not understood it was colloquially known as ‘catenary cancer’.
In the pursuit of lightness, aluminium wire cable with two steel reinforcing strands (AWAC) was used for the catenary. As a result, British Rail’s engineers were able to achieve 125mph running from a system with 11 kiloNewton (kN) tension in both catenary and the copper contact wire.
I had assumed that the issue with AWAC was corrosion of the steel wires. The two steel strands were protected from corrosion by an aluminium coating plus grease during manufacture. ‘Forensic’ analysis showed that after years in service the coating was indeed being compromised, but short circuits proved to be the root cause of failure.
If a flashover hit, and melted one of the steel wires, almost half the catenary’s mechanical strength was lost. Two short circuits and you could be left depending on the tensile strength of aluminium and the wires would come down.
Under the current renewals programme, and remember that the oldest catenary has been in use for up to 40 years, AWAC is being replaced by copper-tin cable. This has finer strands so less strength is lost with a single flashover. In an extreme case up to half the strands can be lost and the catenary stays up.
So far the fast lines between King’s Cross and Hitchin have been renewed with copper catenary. There are also some targeted slow lines renewals underway.
But with 2,000 wire runs on the London North Eastern route, AWAC is going to be in service for many years to come. Hence the work to reduce the impact of flashovers.
Causes of flashovers include encroaching vegetation, birds with one claw earthed and one on a live part, and the infamous children’s metallised party balloons which caused two flashovers in a fortnight a while back. The balloons’ metallised ‘string’ struck a long arc between a return conductor and the OHLE, one incident resulting in a dewirement.
ECML’s maintenance engineers place great store on investigating fault recordings after a flashover. The current and voltage waveform is measured for each fault and reviewed to ensure that the power supply protection system is optimised to detect the full range of potential faults and trip as quickly aspossible.
But despite being an ageing asset, the AWAC failure rate has dropped significantly and this is attributed to the configuration of the protection equipment. What causes the damage is the amount of energy the short circuit puts into the OHLE. The more energy, the greater the probability of failure: the faster the current is tripped, the less the energy. When you realise that the fault level on the ECML is 6,000 Amps and the voltage is 25,000V, that is a very high instantaneous rating.
With confidence that the protection equipment will minimise the energy transmitted to the catenary, and thus the damage to the cable, the engineers have moved on to remedial action to minimise the risk of flashover.
In the case of York station footbridge, for example, a flashover to the OHLE caused by a bird flying underneath would not enhance the travel experience for passengers on the platform or the bridge! Bounds Green depot had a similar problem. The OHLE passes through cut-outs in the shed’s end doors. Pigeons flying through the gap made the depot the section with the largest number of traction power supply trips on the LNE Route.
For both locations a simple application of insulating paint removed that risk of flashover. In the case of Bounds Green, pigeons inside the depot are now the problem.
Managing fault energy has had the biggest impact on AWAC failures. But as with so much in engineering, it is the obsessive pursuit of minor details that pays benefits.
Take the saddles that sit on the catenary to support the droppers – the short lengths of wire that support the contact wire. As installed, these were made of aluminium to avoid corrosion with the AWAC.
Each time a pantograph passes, the catenary, droppers and contact wire become a dynamic system. Over time this movement was causing the saddles to wear. Replacement with steel saddles has generally seen an improvement, although at the north end of the ECML you get sea mist and fog that encourages electrolytic corrosion between the two metals.
A typical 70m span of OHLE will have seven droppers. The dynamic forces in the catenary can also induce fatigue failure with time. So while they look like a piece of thick wire just sitting there, droppers also have a service life.
Failures have been analysed and locations where pantograph uplift is more severe identified. Most of the ECML uses steel droppers that eventually fatigue and break in the middle. Copper droppers harden over time and snap in the middle.
Since this is life-related failure, a programme of campaign renewals have been introduced. For use at high-uplift locations such as bridge approaches, the dropper itself has been redesigned so that if it does fail it can’t drop down and foul a pantograph.
These examples make the point that over several hundred miles of railway you need the flexibility to accommodate local conditions, rather than trying to impose standardisation. Another example of this is the phenomenon called ‘side wear’.
This occurs at switches and crossings, for example coming off the main line and changing from one contact wire to another. In this situation, it is the side of the pantograph that initially picks up the contact wire for the new line.
As the photo of a section of used contact wire (above) shows, this causes not only heavy wear compared with the as-new sample, but the wear is at an angle. In this particular case, much more wear and the groove for the dropper clamp would have been reached.
This sample was found by local teams during visual inspection five years ago. Today, local teams know where side wear is an issue and implement more frequent inspections, renewing the contact wire before wear can get that severe. ‘Thin wire’ dewirement is also now a thing of the past.
It became a sour joke that if the winds blew the wires came down on the ECML. The cause was ‘blow-off’, where a strong wind across the track displaces the contact wire sideways until it can be hooked by the pantograph and down comes the knitting.
This is a function of mast spacing plus local conditions and became apparent with the 1970s West Coast electrification north of Crewe. The aerodynamic effect of wind blowing across embankments accelerated the airflow – requiring supplementary stand-off masts to prevent blow-off.
On the ECML, remedial action has been to install wind speed monitors at vulnerable locations and impose graduated speed restrictions. This has virtually eliminated blow off incidents, but at a cost in operational resilience.
However, because the delay minutes have been greatly reduced, the financial business case for physical interventions, in the form of blow-off structures, is poor. And Network Rail argues that because the speed restrictions are limited to the local areas at risk, the delay is reduced.
With major storms, the emerging national policy is to shut the railway down completely, sort out any problems such as fallen trees, flooding and dewirement immediately, and get the railway running again. This eliminates delays due to trains trapped on the network, which then have to be recovered before the service can start.
Something I had not appreciated was that the UK is unusual in using short neutral sections, spliced into the contact wire, at speed. In Europe the norm is carrier-wire sections where the neutral section is created by overlapping parallel contact wires. The pantograph runs off the energised contact wire onto a ‘floating’ contact wire, then onto a second floating contact wire and finally onto the next electrified section.
Carrier wire neutral sections were on the first sections of the WCML electrification, but a combination of factors resulted in a poor performance. As a result BR adopted neutral sections with a long insulator carrying ceramic beads spliced into the contact wire under tension.
Over the years these short neutral sections have been used at increasing speeds. However, while involving more hardware, carrier wires are considered more reliable and are being used with the Series 1 OHLE for the Great Western electrification programme.
After years of satisfactory use on the ECML, there were two major dewirements at neutral sections, including the infamous Retford incident. Even worse was the absence of any clues to the cause. And, worst of all, the ECML has 106 neutral sections.
While Retford might have been an aberration, the second occurrence suggested that this could be a new failure mode. Laboratory examination provided little intelligence, as did inspecting other neutral sections across the network. While the cause was being investigated, the contact wire was replaced at the entry to each neutral section as a prophylactic measure.
Clearly something was going on inside the clamp that joined the contact wire at one end and the neutral section at the other. Using a borescope, a flexible fibre-optic probe used to look inside jet engines, Network Rail’s engineers looked inside the clamp and saw the start of copper corrosion where the steel teeth gripped the contact wire. In other cases the corrosion was spreading across the surface of the copper.
When the area of the corrosion was examined, a fatigue crack was discovered. So, corrosion fatigue? No, it was not as simple as that. The corrosion was an indicator but not a cause.
Borescope examination of clamps allowed the high-risk locations to be identified and rectified. Which still left the question of what had caused the sudden epidemic of failures? High-speed photography provided the answer.
When the pantograph passed under the clamp, the localised uplift resulted in bending followed by six or seven vibrations after the pantograph had passed. It was this resonance that was leading to the fatigue cracking.
Further investigation showed that the resonance was due to a modification made by the WCML electrification team to smooth the pantograph’s run-in to the neutral section. This did indeed reduce the risk of pantograph damage, but, it turned out, at the cost of changing the dynamic characteristics at the joint, creating the resonance causing the fatigue failures.
Since the ECML doesn’t have the same issue with pantograph wear, the modification was removed. And while not eliminated, fatigue at insulated sections is now part of routine inspection and maintence.
What is interesting is that the ECML team adopted the modified splice as part of sharing best practice. Under the WCML upgrade, OHLE had been renewed with thicker contact wire. As a result, the problem did not emerge until six months after the ECML failures.
Access for renewals is disruptive operationally, and thus expensive. To close the ECML for a weekend possession with access to all lines costs £500,000 to £1 million.
Hence economics determines the OHLE renewal rate; that means that the newer Mk 3b equipment now has to achieve at least a 40-year service life. Meanwhile, the ECML is becoming busier. Hence the focus on improving the AWAC-based catenary while seeking to reduce the cost of renewal.
Network Rail’s sustainable renewals programme for the ECML is based on using wiring trains, similar to the GWEP High Output Plant System (HOPS). Since 2015 ECML has been using an Overhead Condition Renewal (OCR) unit that was acquired by the alliance responsible for OHLE renewals under the West Coast Route Modernisation. For future use on the ECML, desirable modifications include the ability to recover and then pay out the new catenary under tension and also handle tension balance weights, avoiding the need for a road-rail vehicle.
As ever, the headspan construction on three- and four-track sections represents a challenge, since wiring trains, so far, can’t handle threading the catenary wire through the transverse supporting cables. Replacement of headspans with portal structures has already been demonstrated on the ECML.
Whether a campaign of change to portals will be affordable under the restricted funding likely to be available in Control Period 6 is unclear. However, whether LNER or BR’s Eastern Region, the ECML management never sat back and waited on events.
As you might imagine, the same ethos lives on at London North Eastern Route, which is actively investigating ways to fund more renewals in CP6. Network Rail’s devolution of power to the Routes can only help the cause.