A ‘bad hair day’ – a short phrase known to many that suggests that everything that can go wrong does go wrong. It hasn’t been in common usage for very long. It didn’t exist before about 1988 but then it rapidly inveigled itself into the US vocabulary followed soon after by global adoption. All this, of course, comes from that great fount of knowledge – the internet – so it must be true!
This magazine is about to coin a new, but related expression – a ‘bad wire day’. There! It’s now coined. Why? Because Network Rail has had several recently and so we thought we had better provide some background to how Overhead Line Equipment (OLE) works and what can go wrong. We’re not trying to be particularly ghoulish. It’s just that, as in any engineering system, there are risks, but there are also many mitigation measures in place.
Two particularly awful bad wire days stand out. One was at Hanslope Junction on the WCML and the other near St. Neots on the ECML. To make matters even more ‘bad’ the Hanslope Junction episode happened on a Friday morning – probably the very worst time to happen causing massive and sustained disruption to traffic. The national press had a great time and everyone waded in with their particular slant on the industry.
We’ve been looking at these and other incidents, and the whole issue of OLE inspection and maintenance, with Nigel Edwards who is Network Rail’s reliability improvement manager E&P (electrification and plant) working in the maintenance services reliability team. His main focus is OLE and signal power supplies, but he and his team of two cover other E&P assets like points heating and distribution equipment as well.
Right at the start we have to say that any diagnosis of the high profile failures has to be viewed as provisional. There are ongoing technical investigations and these may confirm initial views or throw up new aspects.
Your guide to all that fizzes and crackles
So, how do the period 12 dewirement figures compare with other years? (And dewirement is the technical term for when all the OLE knitting hits the deck and causes a bad wire day).
Well, as Nigel pointed out, period 12 in 2012 had the same number of dewirements. It’s just that they occurred in locations that were far less sensitive.
And what causes the wires to come down? This is where we try and give youa brief summary of the OLE system and its components. This is The Rail Engineer spotters’ guide to all that mass of inexplicable wiring that fizzes and crackles on a misty day as you wait for your morning train, the noises off being caused by electricity “leaking” across insulators due to dirt and moisture.
Starting from the top and working down is the catenary wire from which everything is suspended. Then there are droppers hanging off the catenary wire and at the bottom end of these are clips which clamp onto the contact wire. The contact wire is a copper extrusion which has grooves formed in the sides of the wire for the clips to fit into. Both wires are tensioned with the tensions either being constant (auto tension) or varying with temperature (fixed termination). Auto tension equipment is used on higher speed routes and has tensions of between 8.9 and 13.2kN depending on the OLE system used.
Registration arms set the horizontal position of the contact wire, the stagger, which is alternately either side of the centre line of the track by up to 230mm on straight track and 380mm on curves. The purpose of this is to ensure even wear of the pantograph carbon.
The contact wire height varies because of tunnels and bridges and at level crossings where road vehicle clearance is needed. So, overall it varies between 4.165 metres to 5.94 metres contact wire height.
Below the contact wire (hopefully!) runs the train pantograph which collects the electrical current and which normally exerts an upward force of 90N, but this can be as much as 250N.
There are three basic types of OLE on the system, although there are dozens of sub types. There’s the Mark I version (there were versions before Mark I, but they were for DC and have either been converted to AC as on the Great Eastern, or removed Manchester – Sheffield – Wath) which you’ll see on the West Coast from Euston to Crewe, around Glasgow and in East Anglia. The structures are cantilevers in two track areas and cross-track portal structures in multi-track areas, with brackets, called small part steelwork, from which all the wiring is suspended.
Then the next major version was the Mark III, principally in use on East Coast Main Line, East Anglia and the West Coast Main Line extension to Scotland. This uses a system of headspans – basically two big masts either side of the track with cross span wires and then the headspan wire holding the equipment. It’s cheaper, but the disadvantage is that each line is not mechanically independent. So whereas you might get isolated damage on a portal structure which doesn’t affect the other lines, with a headspan you affect the whole lot.
The third type is what is known as UK1 which is the design range developed for the West Coast route modernisation as an upgrade to the existing systems for higher train speeds.
There are also systems designed by Siemens around Glasgow and Furrer+Frey on the Great Eastern Main Line.
New designs, Series 1 and Series 2 which eliminate failure modes from all the preceding designs are being developed for new electrification schemes including the Great Western and North West Electrification projects.
Three failures modes cured
Nigel explains that the biggest problem has been catenary wire failure. If the catenary wire fails, the whole system’s going to fall because that’s what holds it all up. This used to be a real problem on older versions of the Mark I equipment because the catenary wire passed over the top boom
of the portals and was kept clear of the steelwork by being carried over a small pulley wheel. In turn, the mounting of this wheel was from a short bracket connected to an insulator mounted on top of the boom. The pulley wheel allowed the catenary wire to move longitudinally as it heats up and cools down. The problems arose from wear and fatigue at the point where the wire passed over the pulley, from problems with insulators and from bird strikes.
All three failure modes were eliminated when the inverted cantilever principle was adopted. In this arrangement the catenary wire is suspended from an arm attached to a vertical pole clamped to the portal. There’s no pulley, there’s no ceramic insulator and there are no bird strike issues because of the increased clearance between the structure and the inverted cantilever catenary wire position.
Nigel acknowledges that insulators are a known problem area although, keeping in mind the thousands of insulators on the network, failures are comparatively rare.
One of the recent high profile incidents was likely to have been caused by a failed insulator.
Porcelain insulators are susceptible to hair line cracks where the insulator is attached to the end castings. If water gets in and then freezes and expands it’ll just burst the insulator as porcelain is very brittle. The impact the failure has on the train service depends on where it is in the system. An insulator failing in a live drop vertical on a Mark III headspan will result in a major dewirement because everything will get tangled up with a pantograph.
Despite such failures being rare, Network Rail has conducted campaign changes in the past.
In the Preston and Carlisle areas, all the insulators were replaced with polymeric types and in Scotland about £1 million worth are being changed. It’s a question of identifying the highest risk locations and spending money to best effect.
Contact wire types
Contact wire can fail. The contact wire is nominally ‘thick’. It’s pure copper or a copper alloy and, as you would expect, it wears over a period of time with the passage of the pantographs although it normally lasts at 25 to 40 years. Side wear in crossover areas, where the pantograph on the main line hits the cross over contact wire, has caused a couple of dewirements in the past year and short circuits – such as a tree branch causing a short – can part the wire. At the point of contact there is high resistance and so the wire will heat up during the passage of fault current and fail.
There are two types of catenary wire. There’s the copper catenary on the Mk1 on west coast and there’s an AWAC catenary on Mk III, which is an aluminium conductor with two aluminium coated steel cores for strength. AWAC uses a different suspension arrangement at the pulley wheels where stainless steel bridles are deployed as the aluminium would wear rapidly if it ran over a pulley.
Rolling stock changes?
The introduction of Pendolinos was probably one of the biggest changes to rolling stock usage on west coast. When the Pendolinos first came on line, there were problems with chipped carbons on a regular basis. Because they were higher speed and they had higher uplift, it found out locations on the system which could cause problems and those had to be identified and corrected.
Before any new train is introduced it will go through a safety validation process as one would expect, but even then there are things that occur that hadn’t been anticipated.
Going forward with the Ten Point Plan
Nigel has outlined a broad view on where the main OLE risks occur. But this is not the end of the story by any means. In our next article, to be published as part of our June electrification and power issue, we will be looking at how all the known risks are being controlled along with many initiatives that are encapsulated in what is called the “Ten Point Plan”.
But in the meantime, you should now have some idea of what you are looking at as you gaze up at the knitting. And despite a series of really awful bad wire days, just take comfort that Network Rail really has got the system under control and remember, dewirements can be caused by ‘other factors’ such as defective pantographs – but perhaps that’s another story!