ThirdRail7 wrote:F-line to Dudley via Park wrote:MattW wrote:F-line to Dudley via Park wrote:Never be Hudson service. Empire Tunnel enters Penn facing the wrong direction for continuing further. And even if Gateway, Penn South, etc. etc. gets built to ease some of the Amtrak/NJT/LIRR crunch Empire trains are still locked into a very limited number of platform slots by the Empire Tunnel alignment. Trying a ham-fisted reverse for the sake of 1 stop into New Jersey is a futile effort. For same reason Amtrak's never going to try to force-fit run-thru trains to Albany.
I don't understand this. Why is the Empire Tunnel facing the "wrong" direction for service? It enters Penn Station right? I'm not even sure what it has to do with the Bergen Loops proposed in the article.
http://i.imgur.com/21TaeHC.jpg?1" onclick="window.open(this.href);return false;
The pink line that reverses on itself is the Empire Tunnel. Trains entering Penn from the Empire Tunnel face northbound...towards New Haven. You cannot get to or from New Jersey via the Hudson Line without changing ends on-platform at Penn. With such a limited and unexpandable selection of platforms the Empire Tunnel can conceivably cross over onto, there's not nearly enough space for changing ends in the middle of an ongoing revenue trip.
That specific question viz-a-viz Hudson to NJ was asked 6 posts up in the thread, so don't shoot the messenger.
Why not F-Line? NJT does it every day, all day. I think you're missing the intent (unless I am). I don't think this is a matter of 1 stop. I'm thinking Andegold is talking about run through service. If you time it correctly, it is not that much of an issue. A Hudson train can arrive from the Empire Connection and assume the slot of another train to Port Jervis via the Secaucus Loop on the platform in the same manner as a NJT Midtown Direct becoming a NJCL train right on the platform. If you want a gap between the runs you can loop it through SSYD. For the record, the finalized plans already exist for expanding the reach of the Empire Tunnel. If someone wanted to fund it, you can reach up to 14 track probably within a year of the go ahead.
This idea is not a far fetched as it seems.
There is an enormous degree-of-difficulty difference between scheduling terminating commuter trains around a reverse at a one of the world's busiest terminals, and thru-run commuter trains that have to keep schedule on both sides of reverse at one of the world's busiest terminals. Every current commuter train has either 100% of its passengers boarding or exiting at Penn, necessitating a layover longer than the minimum time it would take for the engineer to change ends because that's the only physical way to clear that many bodies off the train. Those tight margins still have lots of margin for error built into them, because you can only herd cattle through Penn so quickly. And if Train #A arrives late, there are plenty of alternatives for making sure Train #B doesn't leave late and take Train #C and #D's slots through the North River tunnels with it. Like redirecting the late arrival to a different platform, scrambling a run-as-directed from Sunnyside to backstop the equipment before the dominoes start falling, etc. As well as options to scramble if there's an equipment problem when they change ends. That dance is done every day of the week, and it's done because the train schedules aren't joined at the hip in revenue service. Even if there were some sort of cross-train tix honoring through the terminal, passenger has no expectation that they won't have to scramble and move to a different platform if that's the way the equipment gets shuffled.
If you predicate a Poughkeepsie-Secaucus train on holding its time into Secaucus on one schedule with Penn as a non-terminal stop, you lose most of that margin for error and the dominoes are regularly going to start falling affecting other schedules. 95% of the arriving passengers are still going to be disembarking at Penn, meaning the layover time for a commuter train isn't shortened much at all (in contrast to Amtrak where the overturn on a run-thru is 80%, 75%, 60% or less). Any commuter rail train...including New Haven's and LIRR's that don't have to do the reverse. You lose most of the ability to juggle platform assignments around late arrivals solely because of that 5% staying onboard for the run-thru. You lose the ability to keep a late Poughkeepsie train from fouling someone else's slot through the North River tunnels because that late train zooms to the top of the dispatch priority pile. If it were late at the end of its run the the Penn layover and non-revenue shuffles elsewhere correct the disruption. If it were late at the beginning of its run it can make up for lost time late. If it's late in the middle of its run...something else has to take the hit to create a slot for it around the North River tunnels.
Yeah, it can be done...in some sort of Transit OCD abstract world. But it's damn ugly and would introduce all kinds of flow problems. Especially when that set of Empire-accessible platforms is locked into the North River Tunnels alignment. At least the NJT/New Haven and NJT/LIRR run-thru options have 1) a little more tunnel redundancy to choose from in the Gateway era, 2) don't have to hold their breath on completing a glitch-free reverse mid-trip on both ends of the trip, 3) can use the more spacious Gateway/Penn South side of the terminal to reduce dwell times, and 4) have a little more potential for less-total passenger overturn (esp. to New Haven) at Penn to shave an insignificant additional amount of time off dwells. I don't see how the ops awkwardness of it would ever permit that to happen on the Hudson side. Since those platforms aren't going to get any less crowded post-Gateway, and I don't see any plausible data that travel demand from the Hudson to New Jersey is so great that you can keep enough people onboard through Penn to shorten those dwells.
Transfers honored on a single cross-ticket...yeah, I could envision that. One-seat ride...forget it. Not even Amtrak thinks they have plausible run-thru options from the Empire to points south with Albany Regionals. And I suspect they have decades worth of traffic models--both Gateway-modeled and non- Gateway-modeled--starkly showing the ops infeasibility of that. If they can't swing it with a far lower ridership turnover at Penn, I don't see how commuter rail can do it.