What is the biggest issue for the British railway system today? Is it the new high speed line, HS2? Is it capacity? Is it on-time performance? Is it the cost of running the railways?
What do you think?
Tuesday, 12 July 2011
Monday, 11 July 2011
The ATC Bump
Many metro systems use Automatic Train Control (ATC), where the train is driven automatically and has built-in collision prevention called Automatic Train Protection (ATP). The ATP part is the safety system, while the automated driving control, called Automatic Train Operation (ATO), sits on top of it and drives the train much like a driver would. What ATO won't do, unless it is specially programmed to, is fine tune the stopping to make it comfortable for the passengers. This leads to a phenomenon which I call "the ATC bump".
As one of my students pointed out to me this afternoon, the ATC bump is what you feel when the ATC controlled train stops. It stops hard with the brakes on and there is very little adjustment to reduce the jerk you feel at the stopping point. If you like, you could say that the jerk is not limited. Well, it should be. "Jerk limit" is incorporated in most traction and braking systems, the technology is available and, although it requires some effort to get it to work well, it can enhance passengers' comfort. It also helps to demonstrate to taxpayers' that their money has been well spent.
All the ATC operated lines in London (the Central, Jubilee and Victoria Lines) suffer from this problem. In the case of the Victoria Line, it's rather annoying, since the old fleet (the 1967 Tube Stock) has recently been replaced and, not many years before its replacement, it was given a modification to incorporate a smooth stop. The new fleet (the 2009 Tube Stock) that has replaced it has brought back the ATC bump. What a shame.
As one of my students pointed out to me this afternoon, the ATC bump is what you feel when the ATC controlled train stops. It stops hard with the brakes on and there is very little adjustment to reduce the jerk you feel at the stopping point. If you like, you could say that the jerk is not limited. Well, it should be. "Jerk limit" is incorporated in most traction and braking systems, the technology is available and, although it requires some effort to get it to work well, it can enhance passengers' comfort. It also helps to demonstrate to taxpayers' that their money has been well spent.
All the ATC operated lines in London (the Central, Jubilee and Victoria Lines) suffer from this problem. In the case of the Victoria Line, it's rather annoying, since the old fleet (the 1967 Tube Stock) has recently been replaced and, not many years before its replacement, it was given a modification to incorporate a smooth stop. The new fleet (the 2009 Tube Stock) that has replaced it has brought back the ATC bump. What a shame.
Sunday, 10 July 2011
Braking Curve
In railway engineering, you will sometimes hear talk of a "braking curve", often in connection with train performance, platform re-occupation times or signalling. A braking curve is used to calculate how long it will take a train to stop from a given speed. It can be used to determine both service and emergency braking distances and it can give braking times, if needed. Here we look briefly at the braking curve and what it means.
The braking curve is the shape formed on a speed/distance chart by a train as it slows down from normal speed to a stop. A typical curve looks like this:
The curve begins when the driver applies the brake. The brake system take a few seconds to build up to the required braking rate (the "feed up" time) and then the train begins to slow down.
With a constant brake demand, as selected by the driver, the train slows down more rapidly as the speed falls. This is because, at the lower speed, the train has less energy to dispose of. If the brake is left on at the same level all through the stop, eventually the curve will get steeper and steeper until it ends vertically at the stop. If this is allowed to happen, the train will stop with a sharp bump and a lot of coffee will be spilt. To prevent it, a skilled driver will ease off the brake as the speed falls and this will allow him to stop the train gently. The effect of this can be seen on our curve as it nears the stopping point.
Our drawing also shows an "equivalent straight line" curve. This is a simple way of showing the stopping distance that we can expect a train to cover, given an equivalent deceleration rate. It can be used to calculate stopping distances for rough signalling calculations, for example, although today, computer programs make accurate and detailed calculations simple.
The braking curve is the shape formed on a speed/distance chart by a train as it slows down from normal speed to a stop. A typical curve looks like this:
The curve begins when the driver applies the brake. The brake system take a few seconds to build up to the required braking rate (the "feed up" time) and then the train begins to slow down.
With a constant brake demand, as selected by the driver, the train slows down more rapidly as the speed falls. This is because, at the lower speed, the train has less energy to dispose of. If the brake is left on at the same level all through the stop, eventually the curve will get steeper and steeper until it ends vertically at the stop. If this is allowed to happen, the train will stop with a sharp bump and a lot of coffee will be spilt. To prevent it, a skilled driver will ease off the brake as the speed falls and this will allow him to stop the train gently. The effect of this can be seen on our curve as it nears the stopping point.
Our drawing also shows an "equivalent straight line" curve. This is a simple way of showing the stopping distance that we can expect a train to cover, given an equivalent deceleration rate. It can be used to calculate stopping distances for rough signalling calculations, for example, although today, computer programs make accurate and detailed calculations simple.
Saturday, 9 July 2011
The 3-minute Rule
It's time the railways in Britain got to grips with what passengers expect in timekeeping. The present 10-minute threshold described as "on time" for long distance trains and the 5-minute one for local trains is ridiculous. It is derided by passengers as a "lie" and "idoitic" and they're right. When a train is advertised as arriving at 10:07, it should get there at 10:07. 10:16 is not "on time", it's 9 minutes late and we travellers all know it, regardless of what the railway company or the Department for Transport might say.
If there has to be some leeway in train timekeeping for performance monitoring purposes, there should, logically, be a 3-minute rule for all trains. To understand why, we need to understand how a railway produces its product. We have to get, at least, the basics - what the railway offers as a service and how it makes that offer work. We should start by looking at the timetable.
A railway is a service provider. It provides a travel service. It provides the service by producing a stream of moving packages called trains, which are used to transport people along a production line called a track. The intervals between the the moving packages, the trains, are decided on (usually) according to the number of people who want to travel along that particular route at that particular time.
Since most railways don't have so many prospective passengers that they need to provide trains every few minutes they make it easier for passengers by writing a timetable¹. The timetable sets out the times trains are due to appear at each calling point.
Now, there are limitations on the number of trains you can run along one track in one direction. The limits are set by the trains' ability to stop safely. Since trains are heavy, long and often move at quite high speeds they need lots of room to stop. In fact, most trains need about 10 times the stopping distance needed by your car on the road. So the intervals between trains are set for safety reasons, in much the same way as the wise car driver does when he leaves enough room between himself and the car in front in case he has to stop quickly. Train drivers do the same but they are guided by signals.
Signals are provided so that trains drivers can keep their trains a safe distance apart. Signals provide guidance on the state of the line ahead so that the driver can adjust the speed of the train accordingly. On most major routes in Britain, the signals will allow trains to run at about 3 minutes apart, that's 20 trains an hour. In reality, it's slightly fewer trains per hour than this because a margin is allowed for different driving techniques, longer than normal station stops, different types of trains and variable local conditions.
So, our railway production line can send packages, our trains, along the line at 3-minute intervals and this is our basic performance criterion - a 3-minute service interval, or "headway", as we on the railway call it. If we are going to judge our performance properly, we should examine our production line, our train service interval, our so-called headway. If our railway is going to perform according to its capabilities, we should regard any deviation from our 3-minute headway, such as late running, as wrong, because it prevents our train "packages" from passing along our production line at the correct, 3-minute intervals. Put another way, late running reduces our production capability and therefore reduces our performance.
So, any train that runs more than 3 minutes late is occupying the time² belonging to another train and we have therfore lost a unit of production. This is how we should judge our performance - on our time within 3 minutes. Any other value judgement is false, as it ignores the simple unit of production that is the basis of any railway operation, the train headway.
Footnote 1: Generally, if a train service runs at less than 10 minute intervals, a full, published timetable isn't necessary. Of course the railway will produce its own "working" timetable to ensure that trains are run at the correct intervals but it will give the passengers a service frequency of, say, "a train every few minutes" or "a train every 7-8 minutes", so that passengers know what to expect.
Footnote 2: We call the time a train occupies a section of line as its "path". Trains are given a "timetabled path" and many railways use a graphical format for timetables to show how the train paths fit on a time plan.
If there has to be some leeway in train timekeeping for performance monitoring purposes, there should, logically, be a 3-minute rule for all trains. To understand why, we need to understand how a railway produces its product. We have to get, at least, the basics - what the railway offers as a service and how it makes that offer work. We should start by looking at the timetable.
A railway is a service provider. It provides a travel service. It provides the service by producing a stream of moving packages called trains, which are used to transport people along a production line called a track. The intervals between the the moving packages, the trains, are decided on (usually) according to the number of people who want to travel along that particular route at that particular time.
Since most railways don't have so many prospective passengers that they need to provide trains every few minutes they make it easier for passengers by writing a timetable¹. The timetable sets out the times trains are due to appear at each calling point.
Now, there are limitations on the number of trains you can run along one track in one direction. The limits are set by the trains' ability to stop safely. Since trains are heavy, long and often move at quite high speeds they need lots of room to stop. In fact, most trains need about 10 times the stopping distance needed by your car on the road. So the intervals between trains are set for safety reasons, in much the same way as the wise car driver does when he leaves enough room between himself and the car in front in case he has to stop quickly. Train drivers do the same but they are guided by signals.
Signals are provided so that trains drivers can keep their trains a safe distance apart. Signals provide guidance on the state of the line ahead so that the driver can adjust the speed of the train accordingly. On most major routes in Britain, the signals will allow trains to run at about 3 minutes apart, that's 20 trains an hour. In reality, it's slightly fewer trains per hour than this because a margin is allowed for different driving techniques, longer than normal station stops, different types of trains and variable local conditions.
So, our railway production line can send packages, our trains, along the line at 3-minute intervals and this is our basic performance criterion - a 3-minute service interval, or "headway", as we on the railway call it. If we are going to judge our performance properly, we should examine our production line, our train service interval, our so-called headway. If our railway is going to perform according to its capabilities, we should regard any deviation from our 3-minute headway, such as late running, as wrong, because it prevents our train "packages" from passing along our production line at the correct, 3-minute intervals. Put another way, late running reduces our production capability and therefore reduces our performance.
So, any train that runs more than 3 minutes late is occupying the time² belonging to another train and we have therfore lost a unit of production. This is how we should judge our performance - on our time within 3 minutes. Any other value judgement is false, as it ignores the simple unit of production that is the basis of any railway operation, the train headway.
Footnote 1: Generally, if a train service runs at less than 10 minute intervals, a full, published timetable isn't necessary. Of course the railway will produce its own "working" timetable to ensure that trains are run at the correct intervals but it will give the passengers a service frequency of, say, "a train every few minutes" or "a train every 7-8 minutes", so that passengers know what to expect.
Footnote 2: We call the time a train occupies a section of line as its "path". Trains are given a "timetabled path" and many railways use a graphical format for timetables to show how the train paths fit on a time plan.
Friday, 8 July 2011
Why did Bombardier lose Thameslink?
Basically because of their financing. The Thameslink deal was based on financing the order as well as building trains. Their train-building skills are as good as Siemens in most respects but they do, according to The Daily Telegraph, have a lower credit rating than Siemens. This means that the finance behind their offer will be more expensive than Siemens because their cost of borrowing is higher. The cost is so much more that it amounted to £700million over 30 years.
I know little about corporate finance but that seems a lot of money. I wonder also if Bombardier's own risk profile was skewing their contract terms, and that these made them less competitive than Siemens.
See more here:
http://www.telegraph.co.uk/finance/newsbysector/transport/8621498/Bombardier-had-little-chance-on-Thameslink-because-of-contract-terms.html
I know little about corporate finance but that seems a lot of money. I wonder also if Bombardier's own risk profile was skewing their contract terms, and that these made them less competitive than Siemens.
See more here:
http://www.telegraph.co.uk/finance/newsbysector/transport/8621498/Bombardier-had-little-chance-on-Thameslink-because-of-contract-terms.html
Thursday, 30 June 2011
"Disabled Toilets"
A woman said to be a representative of the disabled lobbly in the UK told the BBC Radio 2 presenter Jeremy Vine that "disabled toilets" were for the exclusive use of people with disabilities. She even responded to Jeremy's question about toilets on trains and said they were exclusive too.
Surely, toilets for disabled people are designed for "access" by such persons but they can be used by others. I often use them. On a train, they are usually the only one in the coach.
What do you think?
Surely, toilets for disabled people are designed for "access" by such persons but they can be used by others. I often use them. On a train, they are usually the only one in the coach.
What do you think?
Question about US Diesel Locomotives
Phillip wrote to RTWP:
I had read somewhere that Alco roadswitchers could be shut down and started again like an automobile, but that early EMD units such as the GP7 had to be left idling when not in use. Why is this, and how much fuel would have been wasted by this idling procedure? What was the first EMD locomotive design to not have to be left idling during downtime?
I had read somewhere that Alco roadswitchers could be shut down and started again like an automobile, but that early EMD units such as the GP7 had to be left idling when not in use. Why is this, and how much fuel would have been wasted by this idling procedure? What was the first EMD locomotive design to not have to be left idling during downtime?
Subscribe to:
Posts (Atom)