Talk:Tilting train
From Wikipedia, the free encyclopedia
 |
This article is within the scope of WikiProject Trains, an attempt to build a comprehensive and detailed guide to rail transport on Wikipedia. If you would like to participate, you can visit the project page, where you can join the project and/or contribute to the discussion. |
| See also: WikiProject Trains to do list | |
 |
This article lacks sufficient references and/or adequate inline citations. |
| B | This article has been rated as B-Class on the quality scale. (assessment comments) |
| High | This article has been rated as high-importance within the Trains WikiProject. |
 |
This article is maintained by the Passenger trains task force. |
Active tilting mechanisms are interesting, but I don't know how they actually work. One report I read suggested that something like a standard spirit level is used, and optical detectors are used to determine where the bubble is. In a train at rest, and not tilting, the bubble should be in the middle of the gauge. If a train travels round a corner without tilting, the bubble will move, so feedback can be used to recentre the bubble - by tilting the carriage body.
I have no idea whether this mechanism is the one actually used. Additionally, if this is the kind of mechanism used, being a feedback mechanism, it would require a disturbance before acting, and there would also be the possibility of oscillations, which might be worse than the original problem - at least from the passengers' point of view.
More sophisticated systems could use predictive or open loop control mechanisms as well as feed back systems, to start to tilt the train body as it enters a curve. Again, I've no idea whether this is done on any trains.
There is evidence that active tilting allows trains to run faster, and arguably reduces maintenance costs, while also improving (or at least not worsening!) passenger comfort.
Passive tilting appears to have no beneficial effect on the speed of the train, but may improve passenger comfort. David Martland 09:00, 1 Apr 2004 (UTC)
As I now added to the article, tilting trains traditionally used gyroscope sensors and an electronically controlled servo system. They were not perfect and they did cause nausea. Modern trains have Digital signal processors to read the track ahead and perfectly anticipate the required tilting action.
Details are very sketchy. The truth is that all the high speed tilting trains in use today are developed by the Italian Pendolino group, who also purchased the APT project when it flopped (I suspect in part that the did this to stop the technology falling into other hands). The Pendolino group was part of Fiat Ferrovia, which is now owned by Alstom but the Pendolino crew are still the same old people in the same old R&D departments, and they seem to keep very quiet about details....even history.
I have found that there is little detail and a lot of contradiction in the details about tilting trains. I am trying to find out more about the tilting mechanisms and the nausea problem and will probably add a "nausea" section to this page if I can get to the heart of the issue :-) There is no doubt that nausea has played a big role in the failure or at least bad publicity of many tilting train projects, and yet curiously the early and simple ETR 401 apparently did not suffer from this problem, indeed it remained in regular service till the end of the 90's being used primarily for "holiday charter" trains from the industrial north down to the south, and AFAIK is still not withdrawn. --Roger Irwin 22:27, 21 Sep 2004 (UTC)
BART trains had the first active tilting system, with the prototype cars in 1970. Each of the original cars sits on a triangular frame using three air bags, with inflation under electronic (probably analog) control. But I don't have a good cite for this. --John Nagle 04:25, 2 April 2006 (UTC)
I'd like to straighten out a few points here, especially with regard to the BR APT Project. In the 1970s I was the Tilt System Development Engineer for the APT-E, the gas turbine powered Experimental non-passenger carrying train, and also carried out some system development work on the APT-P, the 27 kV electric passenger carrying trains that entered service in the 1980s.
Firstly the APT did not follow the HST project, they were ongoing at the same time, and indeed HST benefited from the advanced wheel-rail technology from which sprung the APT. The APT was seen as a relatively long term project whereas the HST was seen as a shorter term solution to specific problems on the Western and North Eastern regions with their relatively straight main lines compared to the heavily curved West Coast Main line, which was the prime target of the APT project.
APT-E used an electro servo-hydraulic tilt system, of which the primary sensors were servo-accelerometers mounted on the centre line of each vehicle. These fed an analogue control system which in turn fed a servo-valve feeding pairs of vertically mounted hydraulic actuators at each end of the respective vehicle. The system was duplicated from the sensors down to the servo-valves and overall tilt performance was monitored by a spirit-level type sensor which switched control channels from the active to the stand-by channel if it detected lateral accelerations outside previously set limits. Each vehicle had it's own self contained hydraulic power supply mounted above floor level for development and maintenance purposes and these were powered by the train's 415 VAC system from an APU in each power car. (Originally this was one of the five gas turbines in each power car but later a diesel APU was fitted to free the fifth turbine for traction purposes)
The APT-E tilt system was inherantly unstable in that the vehicle would fall over onto it's bump stops (which were inside the tilt actuators) if the tilt system hydraulic pressure failed or a control system malfunction occured. This was a nominal maximum of some 9 degrees in each direction with a mechanical limit of 12 degrees. In this situation no part of the vehicle would foul the loading gauge, the 'hole' through which the train passed while running on the line and past lineside structures due to the tapered cross section of the vehicle bodies. Steps were taken during the test programme to minimise the consequences of such a failure by fitting various modifications to the system, culminating in hydraulic limiting valves fitted to the bogies which closed off the flow to the actuators.
Various different control systems were tried on APT-E but at all times during it's relatively short life the vehicles had separate tilt systems. No cascading controls were fitted, where one vehicle sent it's tilt sensing signals to the vehicle trailing it, however some tests were made with evaluating the use of such sigals in case of a failure.
APT-P similarly used an electro servo-hydraulic tilt system, but with the hydraulic power supply, control system and associated pipework all fitted under the floor of the vehicle in a removable crate. APT-P used horizontally mounted tilt actuators of considerably longer stroke because the system geometry was changed to make it more stable in the failure case. However this limited the ultimate tilt rate that was possible and I believe that APT-E had the highest performance tilt system ever built anywhere, despite it's obvious lack of sophistication and it's subsequent fragility.
Contrary to many items in the press, including that of the Economist of 21st Feb. 1988, the APT-P did go into service for much longer than the much stated 3 days. While it's true that the initial length of service was very short, it should be noted that this took place in a very cold and wintery December and the train's debut could have been better timed. Undoubtedly this timing was politically motivated but the weather was apparently not part of the considerations. I was on the inaugural soutbound run, as a real life paying passenger having left BR by then, and can attest to the severity of the conditions. The prime cause of the original withdrawal was icing of the overhead power lines which caused fluctuations in the train power, causing many other problems as a result.
Later on the train was put back into service and ran for a number of years, albeit not scheduled as such in the timetables. During this time many different tilt control systems were tried and developed on the three service trains, including some cascade controls and some predictive systems. Toward the end of the train's service I understand some trials took place with limited tilt angles to reduce the so-called 'tilt sickness' but these are undocumented in the puublic domain as far as I know.
Turning to non-APT tilting trains, the Canadian LRC had some considerable input from the APT project in that one of the prime engineers on the APT emigrated to Canada to join the LRC project.
On the other had it's difficult to support the statement that some aspects of the APT project were purchased by the Pendelino group. While it's possible that this took place in other areas, in the case of the tilt systems the two trains are not compatible in any way. Pendelinos use a different geometry, different control philosophy and different actuation systems. The history of this is somewhat muddied due to the various commercial changes that took place after BR was privatised.
ASEA was the group that purchased the part of BR which owned the tilt technology, and was then merged with Brown Boveri to form ABB. At some time this group developed the Swedish X2000 train which has a very similar tilt system to that used on APT-P and in the 'what goes around., comes around' scenario a similar system is used on current Virgin Super Voyager trains in the UK supplied by Bombardier, who took-over ABB, by then entitled ADtranz.
The Pendelinos, including the Virgin run trains in the UK, on the other hand, are supplied by Alstom, who bought out Fiat Ferroviaria who developed the Pendelino concept originally. Altsom purchased the vehicle manufacturing part of BR, rather than the R&D part, as far as I've been able to discover. While the Pendelino development team may well be 'the same old team', although I doubt it after this length of time, the BR team has long since been disbanded and relatively few of us are working in the same field any more.
The existence of at least two different types of tilt system in service, Pendelino and APT-P, in Europe at the present time, bears out these statements I hope.
The subject of 'tilt sickness' is often quoted as the reason why 'tilting trains don't work' but this is not universal and many people do not suffer, in the same way that some people are prone to sea sickness and some not. When on the inaugural run of the APT-P, on which I was a passenger as stated above, the start was in total darkness leaving Glasgow Central Station and it did not begin to get light until well south of Carstairs. At that time APT-P was running a control system that compensated wholly for any tilt deficiancy and under normal circumstances little or no lateral acceleration was felt during curving. The effect on that winter nmorning was that someone had straightened out the entire track south of Glasgow! I was very familair with that stretch of track and I was astonished how well it all seemed to work, after being away from tilting trains for some 2-3 years.
All this changed as the light came up, the horizon then became visible and at the very high speeds the train was running the tilt angles became all too obvious. Some of my fellow passengers did show some signs of illness, although it has to be said that the great majority of them were media and press representatives who had been entertained at BR's expense the previous night in the hotel near the station....
However others of us showed no signs of such sickness and this has been the situation ever since, some passengers feel ill and others don't. Such is human nature that only the complainants tend to get heard though, thus the myth that 'tilting trains make you sick'. Latterly tilt systems have been tuned to reduce this effect by leaving some of the required tilt for any given curve un-compensated, thus leaving the passenger with some feeling of the lateral acceleration that their eyes are telling them is required for the situation. Certainly the Swedish X2000 train that I travelled on a few years ago had such a system fitted and I'm told by members of the Virgin Pendelino development team that this is the type of system fitted to their trains.
On APT-E, with it's very high tilt rate, while almost no lateral accelerations could be felt, it was notable that seats next to the windows showed a high vertical acceleration as the vehicle rotated to it's correct tilt angle for the curve. The effect was similar to that felt in an aircraft entering a turn, a postive 'g' force pushing one down into the seat as the curve was entered, followed by a short period of negative 'g' as the circular portion of the curve was reached and the tilt angle reached a steady state, then a period of no acceleration while in the curve and then the opposite sensed acceleration periods as the curve was left.
It was found that those of us on the tilt development team could operate in two modes, one hyper-sensitive to tilt angles (to this day I can sense the angle of a surface I'm standing on to to with a degree) and one where we were totally oblivious to any surface angle. This did not make us good judges of how the systems would feel to 'Joe Passenger' and the BR Advanced Projects Division quite often used volunteer guinea pigs from the staff to report back on their feelings during test runs. My wife came into such a category more than once!
It must be remembered that these were pioneering days, no-one had done this anywhere until then and very little study had been made of such effects. Indeed, until such systems were built and put into service there was no way they could be studied, so adverse press comments could best be described as somewhat premature.
I'd welcome further comment and correspondance on this or any other associated issue.
(Regards, Kit Spackman 18:50, 1 January 2007 (UTC))
[edit] removed
- There is no foolproof cure, but, if possible, try and reserve an aisle seat as the centrifugal force has a less powerful control on you and your organs and coriolis effect has less takeover on balancing systems. If this is impossible, try and refrain from consumption of food during the trip. Eating food can cause nausea, but this depends on the person and the train. As you know, everyone is somehow different, so cures vary from person to person.
(inappropriate content/tone for an encyclopaedia) Stevage 11:39, 18 April 2006 (UTC)
