nope real problem is treack capacity in four tunnel tracks in Park avenue tunnel.
The interlocking at GCT can handle enough it spits into 8 at first split and to 40 or more on two levels at second split.
Metro- North considering Double Deckers
- Thread starter Dutchrailnut
- Start date
Help Support Amtrak Unlimited Discussion Forum:
The interlocking at GCT can handle enough it spits into 8 at first split and to 40 or more on two levels at second split.
Dutch,
I'm sorry but that simply doesn't make sense that the tunnel is the problem. The LIRR runs almost as many trains as MN, if not more, and the LIRR also competes for space in the East River tunnels with both Amtrak and NJT. MN competes with no one for the Park Avenue tunnel. Now I'll grant that the LIRR has the West End Yard which helps considerably. But then, NJT is taking up the slots being saved by the LIRR's ability to run into the West End Yard.
So with both "tunnels" each having 4 tracks, I don't understand how the tunnel is the restricting issue here. Yes, it might be nice if a way could be found to put 5 tracks under Park Ave, but for the moment the interlocking has to be the bigger problem. Penn has fewer problems with its interlocking plant, since there are fewer crossing movements thanks to how the 4 tracks connect to the station. With less conflicting movements, they can move more trains.
Green Maned Lion
Engineer
If I recall correctly, Alan, the LIRR has 3 terminals, only one of which is Penn Station. Metro-North only has two- GCT which is its exclusive territory, and the former Delaware, Lackawanna & Western terminal in Hoboken, NJ which it shares with NJ Transit. And it only sees a couple of MN trains an hour during peak periods.
Last edited by a moderator:
MNCR has about 150% more trains entering Manhattan than LIRR, this why ESA is being constructed to get more LIRR trains to were they are needed.
Dutch,
I of course know who you work for and what you do, and I hope that you know that I have the utmost respect for you when it comes to most things relating to RR's. But I'm sorry, I'm just not seeing it on this one.
A count of the inbound trains on MN during the 8 AM to 9 AM period, the busiest period, shows 49 trains. I'm assuming (and hoping) that MN makes no non-revenue movements in the peak direction, as that would make no sense. Counting up the LIRR movements during that same period shows 39 revenue trains inbound to Penn. That means that MN is has about 25% more train movements in the peak direction over the LIRR; not 50%.
And then MN doesn't have to contend with at least 2 Amtrak non-revenue moves during that time period and I'm guessing at least 2 or 3 NJT non-rev moves, as NJT loops some trains around at Sunnyside just to keep things flowing in the station. So while MN may still have the most peak direction train movements, at best they only have 6 more movements than the East River tunnels. And the major road block we keep being told at Penn is lack of platforms and crossing movements, not the actual tunnels. Conflicting moves at Harold interlocking doesn't help things either.
So again, I'm just not seeing how the Park Ave tunnels are the problem here. Not trying to be a pain, but I just don't get it.
Green Maned Lion
Engineer
Trains tend to move through both the North and East river tunnels at high velocity. I don't know the actual speed limit. However, I know that movement into the North River tunnels is in excess of 40 mph. Contarywise, speeds at Grand Central Terminal are slothlike going through the Park Avenue tunnels.
I'm going to guess this has to do with track maintenance, the general setup of the interlocking, and the general incompetence of the dispatchers and train masters at Grand Central Terminal and Metro North in general.
Generally speaking, I'd think that the nature of movement through Penn Station tends to result in good people keeping the dispatching jobs they had even way back when. Trains departing GCT always head north in one direction until after they leave Manhattan Island at the very least. There is only one peak direction for Metro North.
Penn Station has trains arriving at peak times from three entirely different train lines (East River, North River, Empire) as well as trains moving, during peak, into the West Side yards. Due to its flow through nature, and the mass of directional movements, any dispatcher that wasn't exceptional would be out of a job because the resulting backup from a simple inability to make order out of chaos would make the news.
So I continue to think that the basic simplicity of Metro-North's general system, and the fact that Metro-North is the only operator- remember, Amtrak dispatches Penn Station- results in a lower level of operations.
I'm going to guess this has to do with track maintenance, the general setup of the interlocking, and the general incompetence of the dispatchers and train masters at Grand Central Terminal and Metro North in general.
Generally speaking, I'd think that the nature of movement through Penn Station tends to result in good people keeping the dispatching jobs they had even way back when. Trains departing GCT always head north in one direction until after they leave Manhattan Island at the very least. There is only one peak direction for Metro North.
Penn Station has trains arriving at peak times from three entirely different train lines (East River, North River, Empire) as well as trains moving, during peak, into the West Side yards. Due to its flow through nature, and the mass of directional movements, any dispatcher that wasn't exceptional would be out of a job because the resulting backup from a simple inability to make order out of chaos would make the news.
So I continue to think that the basic simplicity of Metro-North's general system, and the fact that Metro-North is the only operator- remember, Amtrak dispatches Penn Station- results in a lower level of operations.
I don't believe that the actual dispatchers change, but Amtrak only controls PSCC 6 months out of the year. The LIRR is charge of PSCC for the other six months. That was part of the agreement when the LIRR permanently brought slots from Amtrak during one of its many cash crises.
I'm not sure of all the particulars as to how the sharing agreement works, although I believe that it's just management that changes.
between 8 am and 9 am just New Haven moves are outbound 9 trains (5 deadheads) and 20 inbounds
,,,,,,,,,,,,,,,,,,,,,,,,,,,Harlem moves are outbound 8 trains (4 deadheads) and 14 inbounds
,,,,,,,,,,,,,,,,,,,,,,,,,,,,Hudson moves are outbound 7 trains (5 deadheads) and 15 inbounds
I concur, I counted 49 inbounds myself via the TT's. Which again leaves MN with only 10 more trains than the LIRR. Toss in 2 or 3 NJT trains and 2 Amtrak trains and MN is barely ahead of the East River tunnels in terms of total inbound movements.
Outbound Amtrak pushes 3 trains, the LIRR moves 8 revenue trains outbound (don't know about deadheads), and while I don't have access to deadhead info if I had to hazzard a guess NJT probably moves at least 8 trains out and most likely that number is higher.
So again based upon the numbers that I'm seeing, it would appear to me that the issue is the interlocking plant and not so much the tunnels.
Outbound Amtrak pushes 3 trains, the LIRR moves 8 revenue trains outbound (don't know about deadheads), and while I don't have access to deadhead info if I had to hazzard a guess NJT probably moves at least 8 trains out and most likely that number is higher.
So again based upon the numbers that I'm seeing, it would appear to me that the issue is the interlocking plant and not so much the tunnels.
Nexis4Jersey
Service Attendant
How old is the Interlocking plant , there are so places in Japan with the same amount of Interlocks and operate on a Rapid Transit timetable for there Rail service. So upgrading would probably solve the congestion issues.
jis
Permanent Way Inspector
Staff member
Administator
Moderator
AU Supporting Member
Gathering Team Member
'Tis not the age of the interlocking plant that is at issue, It is the track layout and traffic flows. The original track layout was designed for a very different traffic flow. Changing it is a humongously disruptive and expensive proposition.
Devil's Advocate
⠀⠀⠀
Is there some place I can go to see where these tunnels and interlockings are located and how they're setup? I am having a heck of a time fully understanding this interesting but difficult to follow thread. Trying to envision the specific patterns, bottlenecks, and complications involved is making smoke come out of my ears.
http://www.richegreen.com/MNRRv6.pdf
Giant file, download not view.
It's the entire MNR system, find GCT. NYP is also in there, at the same scale. Speed limits are also a problem, into GCT it's 10MPH. I believe in NYP it's 15?
Green Maned Lion
Engineer
It isn't so much the interlocking layout as the very basic nature of the flow through Grand Central Terminal.
Like 30th Street Station in Philly, GCT has two levels, one designed primarily for suburban (meaning commuter) trains, and one designed for longer distance intercity service, such as the Twentieth Century Limited.
The lower level, with its 26 tracks, was intended to handle high-volume, quick-turnover commuter service trains. The upper level, with its 41 tracks, was intended to handle longer distance trains that would sit there for some time. Every train in that station, carrying passengers or not, in order to leave a platform, must pass through that choke point where the 67 tracks narrow down two four over the course of two interlockings.
Penn Station, is set up as a flow through station. The Long Island Railroad uses two tunnels, and Amtrak uses two tunnels. I don't remember the tunnel numbers, so I'll create an imaginary set of 4 tunnels, 1 through 4. 1 and 3 are for trains leaving Penn Station to Long Island, and 2 and 4 are for trains entering Penn Station from Long Island. All Amtrak trains leave via tunnel 1, and all LIRR trains leave via tunnel 3. All Amtrak trains arrive via tunnel 2, and all LIRR trains arrive via tunnel 4.
So imagine this as two parallel divided highways. Properly operated, and it is, an inbound Amtrak train should never have to wait for an outbound Amtrak train, let alone a Long Island train, to cross its path. Amtrak trains that enter Penn Station from New Jersey do not reverse direction in Penn Station. They enter through the east bound North River Tunnel, platform, discharge and receive passengers, and then continue in the same direction out either to Hell Gate on its way to Boston or Springfield, or to Sunnyside Yard.
Many LIRR trains in Penn will then continue from their track out to the West Side Yard, where they will be turned. LIRR trains generally don't have to wait for each other, either.
Grand Central Terminal is not set up that way. All of its tracks are dead ends. For a train heading southbound (and thus on one of the west two tracks on the ex-NYC main) to platform at a platform to its east, by its very nature, must cross over the northbound tracks. During the time it is crossing over the northbound tracks, northbound trains can't leave the station. That train will then be in a position to exit to the north without crosssing over.
But contrarywise, a train that pulled in from the north to one of the western platforms wouldn't have had to cross over on its entry, but it will have to do so on its exit. So with the exception of two trains on the middle tracks on (I believe) the upper level, all trains will have to foul the entire interlocking either on their entrance or on their departure. This is a problem inherent to a terminal station.
A ten track all-through-running station that is not a major terminal point for passengers could play host to about 500 trains an hour, because all the train has to do is stop, spend 30 seconds doing its business, and leave. With Penn Station, this can be achieved to a degree restricted only by its bottleneck of 4 tunnels to the east and 2 tunnels to the west. Operated purely as a run through station, and with a second set of tunnels (so that, for instance, all LIRR trains became NJT trains) Penn Station could play service to something like 150 trains an hour, or about 300 as equivalent to what we have today.
Grand Central can't do that because no matter what you do, you have to restrict movement of some trains in order to move others.
Like 30th Street Station in Philly, GCT has two levels, one designed primarily for suburban (meaning commuter) trains, and one designed for longer distance intercity service, such as the Twentieth Century Limited.
The lower level, with its 26 tracks, was intended to handle high-volume, quick-turnover commuter service trains. The upper level, with its 41 tracks, was intended to handle longer distance trains that would sit there for some time. Every train in that station, carrying passengers or not, in order to leave a platform, must pass through that choke point where the 67 tracks narrow down two four over the course of two interlockings.
Penn Station, is set up as a flow through station. The Long Island Railroad uses two tunnels, and Amtrak uses two tunnels. I don't remember the tunnel numbers, so I'll create an imaginary set of 4 tunnels, 1 through 4. 1 and 3 are for trains leaving Penn Station to Long Island, and 2 and 4 are for trains entering Penn Station from Long Island. All Amtrak trains leave via tunnel 1, and all LIRR trains leave via tunnel 3. All Amtrak trains arrive via tunnel 2, and all LIRR trains arrive via tunnel 4.
So imagine this as two parallel divided highways. Properly operated, and it is, an inbound Amtrak train should never have to wait for an outbound Amtrak train, let alone a Long Island train, to cross its path. Amtrak trains that enter Penn Station from New Jersey do not reverse direction in Penn Station. They enter through the east bound North River Tunnel, platform, discharge and receive passengers, and then continue in the same direction out either to Hell Gate on its way to Boston or Springfield, or to Sunnyside Yard.
Many LIRR trains in Penn will then continue from their track out to the West Side Yard, where they will be turned. LIRR trains generally don't have to wait for each other, either.
Grand Central Terminal is not set up that way. All of its tracks are dead ends. For a train heading southbound (and thus on one of the west two tracks on the ex-NYC main) to platform at a platform to its east, by its very nature, must cross over the northbound tracks. During the time it is crossing over the northbound tracks, northbound trains can't leave the station. That train will then be in a position to exit to the north without crosssing over.
But contrarywise, a train that pulled in from the north to one of the western platforms wouldn't have had to cross over on its entry, but it will have to do so on its exit. So with the exception of two trains on the middle tracks on (I believe) the upper level, all trains will have to foul the entire interlocking either on their entrance or on their departure. This is a problem inherent to a terminal station.
A ten track all-through-running station that is not a major terminal point for passengers could play host to about 500 trains an hour, because all the train has to do is stop, spend 30 seconds doing its business, and leave. With Penn Station, this can be achieved to a degree restricted only by its bottleneck of 4 tunnels to the east and 2 tunnels to the west. Operated purely as a run through station, and with a second set of tunnels (so that, for instance, all LIRR trains became NJT trains) Penn Station could play service to something like 150 trains an hour, or about 300 as equivalent to what we have today.
Grand Central can't do that because no matter what you do, you have to restrict movement of some trains in order to move others.
You got the tunnel numbers correct! The only correction is that during rush hour, LIRR trains do use the Amtrak tunnels. Otherwise, barring track/tunnel work or a problem with a train laying down, Amtrak does generally use 1 & 2 and the LIRR 3 & 4.
The only other thing that I would add to the description is that going from south to north, at Penn the tunnels run 1, 2, 3, & 4. However 2 & 3 effectively flip flop before they get to Queens, meaning that on the Queens side from south to north you'd find 1, 3, 2, & 4. This sets things up so that the two outbound tunnels are together in Queens and the two inbound tunnels are together.
jis
Permanent Way Inspector
Staff member
Administator
Moderator
AU Supporting Member
Gathering Team Member
I must admit this 150 or 300 train thing stumps me a bit. If you have two tunnels flowing each way feeding into Penn Station from both ends (NJ end and LI end), assuming that signal capacity of each track is 30 tph (120 sec headway, which is better than what we have today), my simple mind says that the max flow in each direction can be 60tph assuming no holdups in the station and free flowing congestion points. Which adds upto a total capacity of 120tph counting both directions. Are we getting the 300 number by suggesting that we will runs trains at less than 60 second headway through the tunnels? What am I missing?
Interestingly, when I saw this, the first thing that came to my mind is wow! This is exactly a current analysis problem in circuit theory with no resistors in the circuit! Ah those good old college days
Green Maned Lion
Engineer
I don't think you need 60 second headways to run trains through those tunnels. I think 30 would more than suffice. All you'd need to do, signal wise, is modify the PTC to automatically penalty brake trains in succession if one in the line dumped its brakes for any reason. 30 seconds would be more than adequate for the train behind it to not run into the one in front of it.
jis
Permanent Way Inspector
Staff member
Administator
Moderator
AU Supporting Member
Gathering Team Member
What will the max allowed speed be for running 30 second headways safely and one that FRA will accept? Speeds like on the Boston Green Line through the tunnels?
What will the block lengths need to be? Or are we talking CBTC? Not that it can do much about speeds restrictions much either. There is always so much length of track that you need to stop a train on a 1.8% downgrade from any given speed under the worst conditions. Just trying to check whether we are in la la land or not.I don;t know the answers to those questions and I don't have the time right now to do the back of the envelope calculations. Maybe someone like George can help figure this out.
Green Maned Lion
Engineer
Failure of logic, Jishnu. The reason I don't really need 250 feet following distance between myself and a car in front of me at 70 miles an hour is that it is highly unlikely (nay, impossible) for the car to suddenly crash into a brick wall, which is the only circumstance wherein said car would come to an immediate stop. Instead, it would also come to a stop in about 250 feet, assuming it is jamming on its brakes.What will the max allowed speed be for running 30 second headways safely and one that FRA will accept? Speeds like on the Boston Green Line through the tunnels?
I don;t know the answers to those questions and I don't have the time right now to do the back of the envelope calculations. Maybe someone like George can help figure this out.
I only need enough space to make up for the amount of time between which the driver in front of me applied his brakes and I followed suit.
If we developed a CTC/PTC/CBTC signaling system that would allow for the trains behind the one that initially stopped to go into full braking potential within a second of the leading trains initiating it, you don't need more than a 30 second headway. Thus, if the front train inline had some reason to suddenly go into emergency, or apply its brakes, all trains behind it would automatically also apply their brakes. Because of the near instant response, the trains would remain at 30 (or to allow for fudge, 25) second headways all the way to a full stop.
This system would be complicated and expensive, but it would also allow for massive capacity improvements at Penn Station such that multi-billion dollar deep cavern death traps would not be needed for many years to come.
jis
Permanent Way Inspector
Staff member
Administator
Moderator
AU Supporting Member
Gathering Team Member
Ah OK we are talking fantasy. In which case it is OK. Charge ahead
Last edited by a moderator:
Great, just what we need, more people in the death trap called Penn Station that can't even be evacuated nearly as fast as that never built station that you are criticizing as a death trap.
Besides, the limiting factor isn't the tunnel, it's A interlocking.
Green Maned Lion
Engineer
Green Maned Lion
Engineer
Saying there is one factor in fixing Penn's capacity issues is like suggesting there is one factor in fixing our country's financial situation.
Secondarily, Penn Station is far less of a death trap than the unbuilt ARC terminal for many many reasons that I am not going to debate with you because you have a highly unrealistic perspective on the entire project, and I am not in the mood to sit around banging my head against a brick wall.
I'm not saying that there isn't more than one factor in fixing Penn's capacity. I'm saying that fixing A interlocking is the first step. Fixing dwell times, capacity to push more trains out the east side, and adding some more tracks would be next. Then and only then would we want to start thinking about whether or not we can push more trains through the North River tunnels safely with closer headways.
My perspective on the entire project is quite realistic. I'm sure you don't think so; but that doesn't change the fact that I'm more in the center than you are on that project. You saw no good in it; I at least saw good and bad in it. And now I see that nothing will get done for probably another 50 years if we're lucky. As bad as that project was, it was still better than nothing, which is what we now have and what we'll maintain now for many more years. In fact, I'll probably be dead before we see a new real rail tunnel under the river now. And by "real" I mean a tunnel that doesn't carry a subway, which I'm also not holding my breath on either.
But a station that would have had more escalators and elevators than Penn station does, and fewer passengers at any given time than Penn does, is certainly not going to be more unsafe than Penn. And several fires at Penn have already proved just how unsafe it is. And I was there for one of the small fires too and I saw how quickly things smoked up. And that was just a fire on top of an AEM-7.
The fact that only NJ was building this doesn't mean that they didn't have to meet the very strick standards of the NYC's fire department for evacuations.
Green Maned Lion
Engineer
The project didn't have any good in it at all. Its potential for increased capacity was more puffery than reality, for one thing. For another, believe it or not as you will, it really, really would have increase commute times for certain types of passengers. Pascack/Main/Bergen commuters, for one, would have found their commute times increased as service to Hoboken ended.
The project would have increased commute times for M&E and M-B riders by several minutes as far as actually arriving into the city goes. Beyond that, the time needed to leave a station set up like that would have been impressive. NJT, in their unbridled optimism, said it would take only 7 minutes for passengers to reach the streets. It presently takes me 3 minutes to reach the street from the platforms if the train isn't too crowded.
But beyond that, Joe Clift told me he figured 7 minutes was an off peak estimate, and the nearest he could figure for actual track-to-street times would have been more along the lines of 10 minutes to 15 minutes. That would be double the longest I've ever spent getting to the street on an SRO NJT train. And I tend to wait for other people to depart, too, and get the back of the line. The project was flawed, at its very core, on the basis of a workable transportation improvement.
Penn Station has many more exits than 34th street would ever have. As a mechanism for actually moving people, escalators leave much to be desired. Which, as Joe explained to me, is why LIRRs concourse has a very limited number of them on platforms. Escalators move people slower than stairs do. 34th street, in a fire, would be a panicked mob of people attempting to go up escalators for 174 feet, or approximately 15 stories. And beyond that very simple item, there would be no way for firefighting crews to get into that station until it was fully evacuated- a time period that could have taken up to 45 minutes.
34th street was an ill planned death trap and a memorial to the ego of a very vain and very stupid man.
