Merits of the TPWS system

By Felix Schmid

While I agree that TPWS is not as good a system as full Automatic Train Protection, I must challenge most of the assertions in the article:

1 While TPWS would not have prevented some accidents, it would have reduced their consequences, for example, at Watford. The Watford case is particularly interesting since the implementation of TPWS would have resulted in a review of the signalling arrangements.

2 Switching off the Automatic Warning System was seen as acceptable since the system is purely advisory. Switching off TPWS is far more serious and is covered by different rules and would have been covered by different rules had it been introduced earlier. TPWS forces the train to a stop. Southall would not have happened with TPWS in place. The driver and his management would have viewed its absence as far more serious than that of AWS.

3 TPWS incorporates the functionality of AWS. It is NOT reliant on working AWS train-borne equipment. TPWS replaces the AWS box on the train although it uses the track equipment.

4 TPWS is expected to be more reliable than AWS since it is largely solid state (electronic). It will therefore be switched off less frequently.

5 TPWS is being fitted in terminal stations to prevent buffer stop crashes of the Cannon Street type. Cannon Street happened at a speed where many ATP systems would have cut out. The consequences were so horrendous because the train was totally outdated.

6 Properly used, TPWS would have prevented Ladbroke Grove as effectively as ATP. If I remember correctly, HSE was saying that an operational ATP system would have prevented Ladbroke Grove.

7 Standard TPWS is suitable with normal overlap distances at speeds up to 70mph. TPWS+ introduces additional speed monitoring loops and is viable up to about 100mph. At greater speeds it still reduces the risks substantially and reduces the collision effects if a train travels beyond the overlap at low speed.

8 TPWS is being introduced rapidly because Professor D Davies recommended its use in his study. He states clearly that it is possible to introduce TPWS within three years and thus achieve major benefits. Introducing ATP, which is much more complex to install, would take 10 to 15 years to achieve the same level of protection. However, he encourages the longer term introduction of ATP. This is happening on the West Coast Main Line.

9 TPWSE, using Eurobalises (part of ERTMS), is being studied by a number of suppliers. This will provide most of the functionality of ATP at an affordable cost.

I feel that the Railwatch article is dangerous since it implies that TPWS is useless. It is not. The re-interpretation of Railtrack statistics is dubious since it relies on an hypothesis about the status of AWS at Cowden and assumes that TPWS would have been abused in the same way as AWS has been in the past.

Felix Schmid MSc is programme director of Rail Systems Engineering at Sheffield University.