After the publication in the April Journal of the article: “An introduction to the overhead electric traction system” by Garry Keenor, Paul Hooper and Peter Dearman, the following correspondence was received from PWI Member Anthony Foster:
Dear PWI,
Electrification Points
I’m pleased to see the beginning of a new article series ‘An introduction to the overhead electric traction system’ in the April Journal although I wish to question some technical points that it makes.
Within the article, it suggests that three-phase electrification is impractical and also that three overhead conductors are required, which I don’t believe is correct. Admittedly there was a 1901 project with railcars by Siemens & Haske that used three conductors, one above the other that ran for about 14 miles. It is the case that this project was over-complex and was discontinued (See Wikipedia “Experimental three-phase railcar”).
Nonetheless, before the days of semiconductor rectifiers, several Swiss and Italian mountain railways used three-phase traction using two overhead wires and the rail to give the three conductors required. They used induction motors and had fixed speeds set by pole-changing motors. The great benefit was the ability to regenerate when descending, much safer than friction braking. It seems that the original system lasted until 1963 but there seems to be a few surviving. Junctions are complex but they work.
The book “Electric Railways 1880 – 1990” by Michael C. Duffy published by the IEE as part of its History of Technology series, Chapter 8 and Wikipedia “Three-phase AC railway electrification” give details of this reasonably successful system.
Later in the article, the ‘Traction Return’ section talks about the aerial return conductor and mentions that ‘some’ return current can flow out of the rails to this conductor but no mention is made of how. The presence of the extra conductor would have very little effect and is only used where immunisation is not necessary (there is a length of the ECML North of York like this).
Older systems use transformers with their primary in series with the overhead line and secondary in series with the earth conductor which is connected to the rail at the mid-point of the between-transformer span. These transformers, called ‘Booster Transformers’ (BT) are mounted at distances of about 1-2 miles, determined by immunisation requirements. An ideal BT system shifts all of the return current from the rails to the return conductor.
Higher power modern systems don’t have a true return conductor, the same size of wire is energised at 25kV in anti-phase to the OLE. At intervals of about 10 miles an auto-transformer links the antiphase conductor, the rail and the OLE. This has the effect of reducing the average current in the rail to near-zero and makes the return conductor work for its existence. Power is supplied to the system at 50kV but the train sees 25kV. The main advantage is that Grid feeder points can be much further apart. (We used 40 route miles for the initial GWML calculations).
Yours, Anthony J Foster
The PWI thanks Anthony for his email and has been in touch with the authors, one of whom has provided the following response:
• 3-Phase OLE – “Anthony correctly notes that 3-phase traction has proved enduringly practicable for a small number of very specialised railway systems. However, the article’s broad conclusion that 3-phase OLE traction is impractical is undoubtedly correct in the context of modern standard-gauge mainline mixed-traffic and high speed railways, as evidenced by the overwhelmingly predominant modern single phase AC OLE systems used all over the world. The ready availability of software controlled rolling stock traction packages able to regenerate during braking and feed single phase AC power back into the OLE network, together with the use of Static Frequency Convertors at national grid power supply points (which enable traction power supplies to be taken from the grid “balanced” across the 3 grid phases) means it is very unlikely that 3-phase contact systems will be used in any future OLE schemes. The mention of 3 conductors rather than 2 was an unfortunate technical proofing oversight on the PWI’s part.”
• Return currents – “Anthony provides a useful expanded description of the legacy booster transformer system (now deprecated in favour of boosterless feeding) and the auto transformer system used in higher capacity AC OLE systems and I have nothing to add at this point. However, the topic of traction power supply (both AC and DC) is a large one and will be covered in greater depth in future Journal articles.”