From what I read somewhere else, I believe that for EWL and NSL, there's two modes of ATO operation; "normal mode" and "high mode"
In normal mode, the train will typically cruise at around the speed between 40 and 60km/h and can be observed pretty often during non-peak hour.
In high mode, the train will simply travel from one station to the other at the highest permitted speed and can be observed during peak hours.
So now come to the question, why need to have two mode when the purpose of train is to transport people from one place to another as quickly as possible? IHMO, train should only slow down IF;
-There's construction works going on/fire/oil spill not too far in front
-Train is approaching a curve which has speed limit lower than it's current speed
-Train is approaching station
-There's another train in front
Other than that, I don't see any logic in programming the train to run slower than normal. If train runs slower, it will means that;
-More train will be needed to maintain frequency
-Increased expense, more train on track consumes more electricity
-It'll eventually reach a point where many train wear out and requires maintenance all at the same time. But we know this is impossible because the facility only has that amount of space to service that many train. This may cause train to be used beyond their limit and eventually, you know what happen.
If we have the same
By having train running at their peak speed, we can have;
-Less train on revenue service to achieve the same frequency
-Less train, less electricity consumption
-Though train will need maintenance more often, you won't really hit the problem of many train requiring one at a time.
and at the end, passengers get what they want; Travel from A to B AFAP.
Please post your opinion regarding this and feel free to correct me if there's anything wrong. Thank you.
That's what getting on my nerves, everyday I must pray that I get onto the right train which will be using the fast profile mode or else, my journey to Khatib will take 2 mins more. I generally agree with you that one profile is enough and that should be the fast profile. When train travel at normal profile, they are extremely slow. Sometimes bus are even faster than train which totally defeat the purpose of MRT. Imagine that everyday you have to take a train to Admiralty, and the train will accelerate out of the station than coast all the way to Sembawang Station.
Originally posted by E5:From what I read somewhere else, I believe that for EWL and NSL, there's two modes of ATO operation; "normal mode" and "high mode"
In normal mode, the train will typically cruise at around the speed between 40 and 60km/h and can be observed pretty often during non-peak hour.
In high mode, the train will simply travel from one station to the other at the highest permitted speed and can be observed during peak hours.
So now come to the question, why need to have two mode when the purpose of train is to transport people from one place to another as quickly as possible?
yes you're right about this observation. but....
1. if every train is travelling at its fastest, then what's the need for scheduling?
2. if every train travels at fast mode, what can the train behind do to catch up with the one in front if it met with heavy pax flow at a station? nothing.
Originally posted by SBSTransport:That's what getting on my nerves, everyday I must pray that I get onto the right train which will be using the fast profile mode or else, my journey to Khatib will take 2 mins more. I generally agree with you that one profile is enough and that should be the fast profile. When train travel at normal profile, they are extremely slow. Sometimes bus are even faster than train which totally defeat the purpose of MRT. Imagine that everyday you have to take a train to Admiralty, and the train will accelerate out of the station than coast all the way to Sembawang Station.
it coasts due to the fact that the longer running time given along the stretch.
The ATO is mainly an automatic operation (as name suggests). The computer would do what it deems fit with what humans input. However for the peak hours as suggested, I would consider the fact that actually some trains are on CM mode.
Why so? A few times while on the first train out of Clementi, after leaving Buona Vista the train stops after the overspeed horn sounds (low pitch beep). The train would come to a rapid stop (using the highest normal braking power) then the operator would reset the controls and proceed again. This would likely not happen on Auto mode. Stops made at stations are also evident of CM mode; the brakes would be activated now and then instead of a smooth stop.
Driving in CM mode would seem logical; when peak hours you need better frequency, the best way to rely on is the human in the cab, provided that he is well trained!
The better system (for 'crush' hours/good frequency) would be the similar D-ATC deployed on the Yamanote Line and major private/JR lines in Tokyo, Japan. Signals are more or less digital, operation akin to ATO, but less the ATP system. Trains can have frequency of 30 seconds, and that is what happens during rush hours...
Originally posted by ^tamago^:
yes you're right about this observation. but....
1. if every train is travelling at its fastest, then what's the need for scheduling?
2. if every train travels at fast mode, what can the train behind do to catch up with the one in front if it met with heavy pax flow at a station? nothing.
1. There's definitely a need for schedulling as there's train going in and out of depot during operating hours. I don't see Japan having to resort to this kind of operation procedures (normal/cruising mode) during non-peak hours.
2. Blame it on the scheduler. When they plan the schedule needed to stop at each station, they would need to take into account of the amount of passenger load in each station at any given time. If passengers load are low, it totally makes no sense for the train to stop at a station for more than let's say, 10 seconds and to stop at a station with high passenger load for less than 20 to 30 seconds. Also, a maximum leeway of 20 seconds in the schedule is enough for train which are late to catch up to schedule, 1 to 3 minute leeway is simply too much. When they have finally achieved this and the punctuality part, they can proudly go and make one monthly train timetable magazines for the public and it can also serve as an additional personal collections for us train fans.
Originally posted by TIB1224Y:
The ATO is mainly an automatic operation (as name suggests). The computer would do what it deems fit with what humans input. However for the peak hours as suggested, I would consider the fact that actually some trains are on CM mode.
Why so? A few times while on the first train out of Clementi, after leaving Buona Vista the train stops after the overspeed horn sounds (low pitch beep). The train would come to a rapid stop (using the highest normal braking power) then the operator would reset the controls and proceed again. This would likely not happen on Auto mode. Stops made at stations are also evident of CM mode; the brakes would be activated now and then instead of a smooth stop.
Driving in CM mode would seem logical; when peak hours you need better frequency, the best way to rely on is the human in the cab, provided that he is well trained!
The better system (for 'crush' hours/good frequency) would be the similar D-ATC deployed on the Yamanote Line and major private/JR lines in Tokyo, Japan. Signals are more or less digital, operation akin to ATO, but less the ATP system. Trains can have frequency of 30 seconds, and that is what happens during rush hours...
This would means that our TO/CSO are not well train enough to brake smoothly when stopping. It would be good if we have ATO all struck off together and change it to D-ATC. As long as we have this ATO thing, 30 second frequency would be almost impossible. The closest it can go is probably 60 to 90 seconds for safety and failsafe purposes. Also, I am quite confused here, ATC is supposely to automatically brake if driver exceed the speed limit allowed back but in NSL and EWL case, ATP would kick in instead to put the train into complete halt. Then what's the point of ATC in CM mode when ATP overrides it? Can someone here explain this?
ATO is like a 'enhanced' version of ATC. Lets just say, it just makes humans more lazy. The operator will not have to operate the train every time. ATO would not work so well under crush periods and wet weather, where its pre-programmed data cannot keep up with actual situations (like wet rail, closer separation etc).
ATC systems are fully for the human, by the human. Humans have control over the train, but he cannot go over the speed limit. ATO, human have no control as long as its auto, except for doors and the emergency switch.
I would consider CM mode as a human-operated, human-centered ATO system and not an ATC system. Human would have control over stopping point, speed to go at, but not the separation distance, for all ATO goes is the safety issue, the 90 seconds separation (or rather one plus train block distance). ATP is accurately the system that works with ATO to ensure this, while you do not have ATP in full ATC systems.
It is exciting when you have rode on a full ATC system during crush hours... The speed would just go up up up, then down down down till you can see the train ahead. Literally, if anything goes wrong, the trains would be in a traffic jam, coupler to coupler (of course a distance like 5 meters apart).
Originally posted by E5:1. There's definitely a need for schedulling as there's train going in and out of depot during operating hours. I don't see Japan having to resort to this kind of operation procedures (normal/cruising mode) during non-peak hours.
2. Blame it on the scheduler. When they plan the schedule needed to stop at each station, they would need to take into account of the amount of passenger load in each station at any given time. If passengers load are low, it totally makes no sense for the train to stop at a station for more than let's say, 10 seconds and to stop at a station with high passenger load for less than 20 to 30 seconds. Also, a maximum leeway of 20 seconds in the schedule is enough for train which are late to catch up to schedule, 1 to 3 minute leeway is simply too much. When they have finally achieved this and the punctuality part, they can proudly go and make one monthly train timetable magazines for the public and it can also serve as an additional personal collections for us train fans.
What for? The scheduler plans according to the running time given. If a train dwells for a shorter period, it will eventually catch up with the train in front. That's the reason why there's a schedule in the first place. Do you wish for a situation where the driver just closes the door because he needs to follow the schedule irregardless of passenger flow? To give ample running time is important for the safety for all passengers. To ask for a ride faster by a minute in exchange for the safety of the passengers is selfish.
Timetables will not be useful here because of weather elements that affect the operations of our elevated trains.
In Japan, trains on urban lines are scheduled to dwell at major stations for around 45 to 90 seconds. They can accomplish good adherence because they have a lot of leeway in terms of extra dwelling time. And for intercity Shinkansen trains, they dwell at the multi-platform stations long enough for everyone to board before leaving on the dot.
Anyway, schedule adherence on our part is excellent during off-peak hours because every train is running at normal mode unless it is slower than scheduled. And the important of running normal mode is so that trains can maintain an even headway and be able to adjust itself back to scheduled time.
Originally posted by E5:This would means that our TO/CSO are not well train enough to brake smoothly when stopping. It would be good if we have ATO all struck off together and change it to D-ATC. As long as we have this ATO thing, 30 second frequency would be almost impossible. The closest it can go is probably 60 to 90 seconds for safety and failsafe purposes. Also, I am quite confused here, ATC is supposely to automatically brake if driver exceed the speed limit allowed back but in NSL and EWL case, ATP would kick in instead to put the train into complete halt. Then what's the point of ATC in CM mode when ATP overrides it? Can someone here explain this?
Have you even been a PTO before to criticise that our PTO's are bad?
You don't even know the design of our system to derive at the minimum frequency. Go count the number of seconds a train pulls in to a station, dwell and departs before the next train can pull in, and think how this affects the design of the operational minimum frequency of our system.
The main purpose of ATP is supposed to ensure a safety distance between two distance. That is not the main purpose of ATC, but both systems can change the behaviour of a moving train.
Originally posted by TIB1224Y:ATO is like a 'enhanced' version of ATC. Lets just say, it just makes humans more lazy. The operator will not have to operate the train every time. ATO would not work so well under crush periods and wet weather, where its pre-programmed data cannot keep up with actual situations (like wet rail, closer separation etc).
ATC systems are fully for the human, by the human. Humans have control over the train, but he cannot go over the speed limit. ATO, human have no control as long as its auto, except for doors and the emergency switch.
I would consider CM mode as a human-operated, human-centered ATO system and not an ATC system. Human would have control over stopping point, speed to go at, but not the separation distance, for all ATO goes is the safety issue, the 90 seconds separation (or rather one plus train block distance). ATP is accurately the system that works with ATO to ensure this, while you do not have ATP in full ATC systems.
It is exciting when you have rode on a full ATC system during crush hours... The speed would just go up up up, then down down down till you can see the train ahead. Literally, if anything goes wrong, the trains would be in a traffic jam, coupler to coupler (of course a distance like 5 meters apart).
I think the normal ATC can't "crush" the trains till 5m apart. The one on JR Yamanote Line runs on D-ATC and when the ATC speed limit is 0, TO can opt to overrun the red signal but drive at a speed of something similar to our restricted mode, 15km/h. Correct me if I am wrong anywhere.
Originally posted by ^tamago^:
What for? The scheduler plans according to the running time given. If a train dwells for a shorter period, it will eventually catch up with the train in front. That's the reason why there's a schedule in the first place. Do you wish for a situation where the driver just closes the door because he needs to follow the schedule irregardless of passenger flow? To give ample running time is important for the safety for all passengers. To ask for a ride faster by a minute in exchange for the safety of the passengers is selfish.
Timetables will not be useful here because of weather elements that affect the operations of our elevated trains.
In Japan, trains on urban lines are scheduled to dwell at major stations for around 45 to 90 seconds. They can accomplish good adherence because they have a lot of leeway in terms of extra dwelling time. And for intercity Shinkansen trains, they dwell at the multi-platform stations long enough for everyone to board before leaving on the dot.
Anyway, schedule adherence on our part is excellent during off-peak hours because every train is running at normal mode unless it is slower than scheduled. And the important of running normal mode is so that trains can maintain an even headway and be able to adjust itself back to scheduled time.
- I for one, definitely won't support the cause of safety over convienience
- Why not useful? Would you elaborate on this part?
- Hmm, I think I might be not clear enough on what I am conveying. Take for example a EB train is suppose to depart Lakeside at 10:15:20 and is suppose to arrive at JE station at 10:16:40. Calculating the leeway into the schedule in case the train is late and JE station is a high pax vol station, train will depart at around let's say 10:17:15. If a train entered the station early, it's entitiled to longer dwelling time. If it's late, let's say like 10:16:52, then TO will have to see how long it takes to load and unload the passengers. If passengers finished boarding and alighting before 10:17:15, train will leave station as per normal. Else, train will close it's door ASAP the moment all passengers are unloaded and loaded. And if you noticed, Japan driver's schedule is indicated to the seconds, the departure time shown to passengers is to the minutes. So I was wondering why they aren't showing us the departure time...
- This I agree with you but still, can be done better.
Originally posted by ^tamago^:
Have you even been a PTO before to criticise that our PTO's are bad?
You don't even know the design of our system to derive at the minimum frequency. Go count the number of seconds a train pulls in to a station, dwell and departs before the next train can pull in, and think how this affects the design of the operational minimum frequency of our system.
The main purpose of ATP is supposed to ensure a safety distance between two distance. That is not the main purpose of ATC, but both systems can change the behaviour of a moving train.
- May I ask but what is a PTO?
- Mind if you explain how the system works then?
- Thanks for clearing this up.
Originally posted by E5:- I for one, definitely won't support the cause of safety over convienience
- Why not useful? Would you elaborate on this part?
- Hmm, I think I might be not clear enough on what I am conveying. Take for example a EB train is suppose to depart Lakeside at 10:15:20 and is suppose to arrive at JE station at 10:16:40. Calculating the leeway into the schedule in case the train is late and JE station is a high pax vol station, train will depart at around let's say 10:17:15. If a train entered the station early, it's entitiled to longer dwelling time. If it's late, let's say like 10:16:52, then TO will have to see how long it takes to load and unload the passengers. If passengers finished boarding and alighting before 10:17:15, train will leave station as per normal. Else, train will close it's door ASAP the moment all passengers are unloaded and loaded. And if you noticed, Japan driver's schedule is indicated to the seconds, the departure time shown to passengers is to the minutes. So I was wondering why they aren't showing us the departure time...
- This I agree with you but still, can be done better.
Originally posted by E5:- May I ask but what is a PTO?
- Mind if you explain how the system works then?
- Thanks for clearing this up.
i can't believe you would place schedule adherence and getting to your destination faster at the expense of human lives. i dunno how many disruptions we will get every week and lives lost to needless accidents.
try googling or reading around for what PTO means.
a train takes 25s to decelerate into the station, and takes another 25s to accelerate out to the next block. add a minimal 20s of dwell time (pax flow is only 11s in this case including 7s of door bell) gives you 70s. a safety buffer of 10s makes it 80s. understood? it has nothing to do with whether the train is ahead or behind schedule.
and all the japanese schedules i've seen are indicated to nearest 15s, if you have ever taken trains there. same logic how our schedules are followed and also how ATO attempts to reschedule itself against OCC after a disruption occurs.
additional trivia: NEL schedules are indicated to nearest 5s, and tend to arrive at 25th and 55th second north of Kovan.
there's no point showing schedules in singapore because the trains come often enough and also that schedules are subject to adjustments throughout the day.