As you say Richard, the "normal", or "received" wisdom is that people die in rail accidents when they leave the vehicle. However in two accidents that I can think one of the biggest sources of injury (and sadly, death) were "internal projectiles" - for example: items of luggage, other people and fist sized lumps of glass held together by the numerous sticky labels that adorn train windows these days. Another factor is that these trams have very few seats, thus a very high proportion of standing passengers and human beings don't like being dumped on their heads :-(

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

OK adding a little more information, now they have said which way the tram was going, it seems the slope was not a factor.

In this picture



The tram had come round the curve from the right of the picture and the front of the tram is nearest the camera.

Having ridden these trams many times, I too am a little surprised at the number of dead and injured. I suspect it may be due to quite a few people standing, I rarely sit down.

From looking at the topography on google maps, it seems as if the tram was climbing the slope from the old heavy rail line, so gradient is unlikely to have played a part.

In any event, trams/light rail were used for a long time in the hilly north of England (and Scotland): my nearest city of Bradford is built in a bowl almost entirely surrounded by hills, and all the main roads out have gradients of varying severity. The steepest - Church Bank, past the Cathedral - is still a struggle for modern buses, and was the scene of a major accident over 100 years ago.

Images and Board of Trade report

The centenary prompted reminiscences in the local paper, but I get the impression that it was largely remembered because it was such a rare event, and happened right at the entrance to one of the main public squares in the city centre. As the Telepragh & Argus (local paper) recalled on the centenary, nobody died - even though the flimsy wooden bodies of the early street cars were completely destroyed in even low-speed impacts.

It does seem as if the offspring of the modern marriage of street car with (relatively) high speed light rail may have failed to inherit some important lessons from its respective parents. While human control is (probably? - discuss) best in the shared urban traffic of the town centre, the high-speed segments on segregated/dedicated track perhaps require a different approach.

To which you can add Hatfield, where the human error involved was selling off track maintenance into the private sector.

Do we ban public transport, or humans being involved in it? Sometime the choice is made harder ..... The DLR seems fine.

I don't think you can consider banning public transport on safety grounds, unless you also look at doing the same for the main replacement - private cars. Wikipedia quotes a World Health Organisation estimate of 1.25 million deaths worldwide through road traffic injuries in the year 2010. I'm going from memory here, rather than statistical sources, but I think the statistical accident risk rate goes something like this:

• cycling (pedal and motor)
  
• private cars
  
• buses
  
• (I suppose trams might go here?)
  
• trains
  
• aircraft
  

- in decreasing order. And partly because of that, but also because of the larger number of potential victims in each vehicle, it's the least risky modes of transport which generate the biggest headlines when something does go wrong.

The DLR - Docklands Light Railway in London - is a special case because it runs on completely segregated track (according to that article, there's only one level crossing on the entire system - and that's on a private road in the depot), and the sections I've used seem to run at a relatively low speed. But having exclusive control of both track and trains has enabled them to deploy full automation, which should at least match speeds to track geometry in all cases.

Looking at Bernie's railcam images in the train thread (most notably Ashland), I've always been struck (if that's the right word, under the circumstances) by the 'openness' of the railroad infrastructure in the States. Here in the UK, railway tracks and stations are always fenced off, even in the remotest rural areas. That doesn't totally prevent fatal contact between trains and drivers/pedestrians, but it must help - at the expense of a much more cluttered and controlling local environment.

Be nice if you can refrain from adding your personal digs.

To clarify further, for those who use the system it is all London Underground. For those who/did work on it, the Underground is known as sub-surface while the Tubes are deep level underground.

The tubes asides from the stations are too filthy for true automation to work effectively as witnessed by the Victoria Line several years after opening.

Secondly with crews, mishaps can be dealt with quickly & efficiently. With no crews, what happens if a train had a trainline or mainline burst in a tunnel...

...oops!

On this if wrong, I trust someone more knowledgeable can correct me. While growing up I can recall a tram having a similar operating system to trains - a power up/down controller (with a deadman where if one fell asleep, deadman takes over) & brake control...

...what changed?

Every (modern UK) tram has at least one equivalent to the old "deadman's handle". These devices require the driver to make a control movement at some (random) time interval. However, given the performance of the modern tram (in particular its acceleration), it is not impossible for a tram tram to accelerate from a safe speed to a dangerous speed in that time interval, particularly when approaching a significant speed restriction. Now one can see a flaw in the logic here - if the driver falls asleep and in so doing moves the control to full power the vigilance system sees this as a control movement and resets the timer. The timer is typically in the range 30seconds to a couple of minutes....

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

More likely a case of "excessive speed in the wrong place". No measurement of speed alone (except the really exceptional case of 'greater than maximum system-wide line speed') can contribute to safety, unless the monitoring equipment has a sense of vehicle location.

If the trams have a recorded message system for "the next station is ...", then they have that sense of location already. Possibly beef up the precision a bit, and tie it in to the speed monitor - why not?

If the trams have a recorded message system for "the next station is ...", then they have that sense of location already. Possibly beef up the precision a bit, and tie it in to the speed monitor - why not?

Can't do that it would cut into their profits and that's been the bottom line for years

---

Life is what you make of it :-)

When i'm good i'm very good , but when i'm bad i'm shi#eloads better ;-) In't I " buttercups " p.m.s.l at authoritie !!;-)

From my experience as a passenger on the Croydon Tramlink sitting behind the driver, they use a joystick controller for the speed of the train. I assume that this would return to a nominal position like a dead mans handle, if let go by an incapacitated driver. Whether it would also put the brakes on, I do not know.

Neither do I but that is an interesting question.

Build the factories where there is a need for the goods .....

The factories are in China, the goods have already gone over the Pacific Ocean.

---

With no crews, what happens if a train had a trainline or mainline burst in a tunnel...

From my experience as a passenger on the Croydon Tramlink sitting behind the driver, they use a joystick controller for the speed of the train. I assume that this would return to a nominal position like a dead mans handle, if let go by an incapacitated driver. Whether it would also put the brakes on, I do not know.

Neither do I but that is an interesting question.

Your Underground trains still use air (or vacuum?) brakes? I can't speak for every city in North America, but in Chicago L trains use a combination of dynamic (and in the very newest cars, regenerative to the supply) brakes for major deceleration and electrically controlled, spring actuated friction brakes in the axles. This was first used on the L in the 1950s, having been adopted (by dint of reusing the parts from dismantled cars) from streetcars designed in the 1930s. I can say from personal experience that all such cars do go into emergency if the operator lets go of the controller in any power setting or even coast. I can also say that the L's last cars with air brakes, from the 1920s, merely cut off power if you let go.

(On the other hand, Illinois Terminal 415, sort of an overgrown streetcar / small interurban also from the 20s, will go into emergency if you aren't holding down the controller anytime the air brakes are released.)

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

The train in question was not an underground train, but a mixed segregated and street running tram. It has three braking systems, air, electric and hydraulic. Being a single side-stick controller makes the use of a deadman's handle difficult, so like many modern European trains and trams it uses a treadle that, in the event of holding a constant speed, or braking effort, both of which require the stick to be held in one position, has to be "kicked" when a "bell" goes off. One advantage of this sort of "I'm alive" detection is that it takes a positive action to kick the treadle, whereas the deadman's handle can be beaten if the driver collapses in such a way as to hold the handle in a valid position.



Going back to Richard's comment about using the passenger information system to locate the tram's position and so be able to interact with the train controls. While this sounds a good idea these systems are very simple and crude - On my journey to work I regularly see these displays giving the wrong station, or heading off in the wrong route, or being blank, none of which is of much use when trying to correctly locate the train and what speed it should be doing.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Going back to Richard's comment about using the passenger information system to locate the tram's position and so be able to interact with the train controls. While this sounds a good idea these systems are very simple and crude - On my journey to work I regularly see these displays giving the wrong station, or heading off in the wrong route, or being blank, none of which is of much use when trying to correctly locate the train and what speed it should be doing.

Then the system is not very good. Probably not enough people working on keeping the programming up to date.

North American Positive Train Control uses GPS to determine a train's position, along with logic in the software to tell it the train did not suddenly jump to the next track when the GPS inaccuracy says it did.

In the Chicago area, Metra uses GPS to trigger the automated on-train announcements of what the next stop is, but it would be nice if the conductor could turn it off when the train is full and runs express to downtown.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

The Passenger Information System is only intended to give INFORMATION to the passengers, and not do any train control. This has absolutely NOTHING to do with the software maintenance activities you talk about.

Systems like ERTMS, PTC, ATC and so on are designed to offer varying degrees of train control, thus have a far better "understanding" of where the train is , how fast it is going, and how fast it should be going.

To "bend" a PIS system into some form of train control system would involve a lot of re-engineering of the hardware on both the tram and the infrastructure and cost a lot of money. Indeed it would probably be less expensive to buy a new train control system than, wit all the required interfaces built into it than try to modify the PIS to do something it was never intended to do.

BTW. Using GPS as a trigger for anything "safety related" is totally crass - what happens if the USG decides to turn SA off - you suddenly loose the precision that we have grown used to, the same happens when the satellite coverage is poor, either due to local topography, or one of the regular "low satellite availability" events.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

Your Underground trains still use air (or vacuum?) brakes? I can't speak for every city in North America, but in Chicago L trains use a combination of dynamic (and in the very newest cars, regenerative to the supply) brakes for major deceleration and electrically controlled, spring actuated friction brakes in the axles.

Did some checking. The last rolling stock to use that system has been replaced with that replacement being completed by Dec 2016.

Have to say that the Bombardier "Movia" family of metro trains are a great improvement over previous LU rolling stock.

The Passenger Information System is only intended to give INFORMATION to the passengers, and not do any train control. This has absolutely NOTHING to do with the software maintenance activities you talk about.

My maintenance comment referred only to the accuracy of the info system for its own sake. On the contrary, it would be a better idea to have the info system be an offshoot of the safety system.

BTW. Using GPS as a trigger for anything "safety related" is totally crass - what happens if the USG decides to turn SA off - you suddenly loose the precision that we have grown used to, the same happens when the satellite coverage is poor, either due to local topography, or one of the regular "low satellite availability" events.

I don't know, but I assume that if PTC loses its GPS fix, it reverts to a "safe mode" of some sort.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

BTW. Using GPS as a trigger for anything "safety related" is totally crass - what happens if the USG decides to turn SA off - you suddenly loose the precision that we have grown used to, the same happens when the satellite coverage is poor, either due to local topography, or one of the regular "low satellite availability" events.

Switch to Glonass of course.

If HGV in the USA have a GPS with sat transponder that can report back to dispatch near real time information obviously it can be done for rail. Also very possible is linking the signal transponders to transmit serial numbers both ways so the database in the train knows which signal it just passed and the dispatch office is notified that train XYZ just passed. GPS can fill in speed information on long blocks to better predict arrival times or for those areas which do not connect back to the dispatch office.

Essentially all of this is done in the aviation industry and in three dimensions.

Oh as to GPS off, it isn't going to happen unless the NUKES are already in the air. FAA is now allowing pure GPS flight operations. Having airliners lost wandering around isn't in the cards.

---

Yep both part of Bombardier's "Movia" family :-) About time too as the A60's & A62's were getting a bit long in the tooth.

If the max speeds for any section were programmed in, it ought to be relatively simple to sound a warning bell in the cab for an over speed.

That implies the system knows the weather , track conditions etc , doubt any system used by TFL is that "clever"

---

Life is what you make of it :-)

When i'm good i'm very good , but when i'm bad i'm shi#eloads better ;-) In't I " buttercups " p.m.s.l at authoritie !!;-)

The Chicago Transit Authority's signal system, as I believe I've mentioned before, regulates train speeds in slow spots by keeping the next signal red until it senses that the approaching train is very close and going the proper speed. It also sends the maximum allowable speed in any given piece of track to the train, where it is displayed alongside the actual speed. All of this has been done since long before there were satellites, let alone GPS satellites. I'm not sure what the penalty is for overspeed, but passing a red signal trips something under the car which has to be reset manually, and a good talking to by the supervisor (at least; maybe suspension).

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.