RR503

6 hours ago, Union Tpke said:

@RR503 This could be a game changer. This is one of the things considered in the December 2019 study on speeds. This is some of the best news in transit I have heard in some time. I wonder how much travel times on the Culver express would be reduced.

Correct, and this was looked at previously in the 1990s when they set the current standards.
 

Culver Express wouldn’t gain much from this — its curves are either timed or on long uphills that make it impossible to even get to V4 — but on other parts of the system, where untimed geometric limits abound, this indeed could be great. IND Queens and IRT Broadway especially. 

Obviously there’s a difference between discussing ways to solve an issue and saying that said resolution should be a priority, but I would rank a  terminal swap pretty low among potential investments. You’d be *marginally* benefitting two stations (Stillwell has the , and you could potentially make a <D>) at huge infrastructure cost. If you want to increase terminal cap at Brighton Beach, just reconfigure the switches so trains off of A1 track can relay south of the station and come back on A2 — or essentially what Armandito proposed up thread 

17 minutes ago, R68OnBroadway said:

What's your view on sending the  down to Kings Highway and terminating on the middle there?

Not a great idea. KH middle gets a _lot_ of use as a short turn point for s and a location where layups can be cleared out of the way of mainline traffic. 

10 hours ago, R68OnBroadway said:

Seems like an idea that could make sense long term- it adds only a few minutes of runtime and can be done with simply adding some switches (IIRC 18th has crew quarters).

The switches are already there, but there aren't crew quarters AFAIK. You'd recrew on the northbound at Church a la 205/BPK on the . I'm not even sure this would actually add runtime; keep in mind that the move down to the relay is slow and when I've observed at Church, s regularly spend 15-20 down there. 

I know service cut arrangements are fun to tinker with, but the reality is that unless we expect depressed ridership to last for years, cuts won't save nearly enough money to have an impact on a hole of this magnitude; demand elasticities are a thing, and they mean that the more you cut, the less you earn in fare revenue. Now of course, I don't think the MTA will end up having to fill this full budget hole by itself, but even if they only have to cover 500m, simple cuts as proposed here are really not that effective at closing the gap -- not from a ridership loss perspective, and certainly not from an agency political capital perspective.

1 minute ago, Union Tpke said:

Do you think it is better to go further into debt to save existing frequencies, or cut service? They are both awful options.

I don’t think that’s the tradeoff at hand. I have confidence that the feds will pull through, the question is with how much money, and what to cut to fill the last x hundred million gap. 

Just now, Union Tpke said:

Any plans to replace your locos with GG1s?

Those went to heaven. We got the E60s.

1 minute ago, Union Tpke said:

Do they have steam trains there?

We sure do. Also some electric trackage on the core routes, with timed transfers to steam shuttles into the pits.

RE: OPTO, I'm no labor relations expert, but yes, that would seem like the conversation necessary to have. 

27 minutes ago, Deucey said:

Where in Hades have you been?

Walking on the coals, tending the furnaces.... 

Friendly reminder that the easiest, quickest and most impactful way to reform the MTA's cost structure is to get a handle on its productivity and process issues. Break down silos, do OPTO, reform maintenance procedures, etc. I would personally start with maintenance: NYCT spends more facility maintainer-hours per mile than does any US system outside NYC *by a factor of 5*.

It's nice to be back.

55 minutes ago, MHV9218 said:

Aren't the racked wires going to be complete FUBAR in both Rutgers/Canarsie if there's another tunnel flood (not unlikely)? Seems like they're uniquely susceptible to damage in a way that the bench walls were not.

On the contrary, they make damaged equipment hugely easier to repair. You don't have to demolish the bench wall to replace the conduit that got a little water in it. 

On 6/5/2020 at 11:00 AM, Union Tpke said:

Fixing what they screwed up at Forest Hills would be nice as well.

CBTC will eliminate the restrictions at FH and Roosevelt, as well as the DGT on 4 track north of the station (though you'll still have to take the switch a decent speed, of course)

On 7/19/2020 at 11:10 AM, Union Tpke said:

@RR503 Given that the signaling limits capacity on the line to 20 TPH, is it correct that not building this substation in now won't hurt capacity further? 

I had always assumed that SAS2 would include CBTC signals for the line. Everything was built CBTC ready. 

Broken record here, but prioritizing resource-intensive system expansion projects is not wise in a time when budgetary and, more importantly, human resources are stretched thin. Back to basics, and then build from there. There's also a ton of uncertainty about what ridership will look like post-COVID. Whether peaks (whose intensity on Lex were the driving force behind this entire effort) will be as pointy as before is an open question on

On 4/18/2020 at 10:07 AM, Late Clear said:

Yup and no remedy to easing congestion on the Lexington Line such as CBTC.  Even southbound gap fillers @ USQ are responsible for a tremendous amount of congestion. 

Most of Lex's current day issues are related to overlong dwell times, long control lines and issues closing train spacing via ST. Eminently fixable with CBTC, especially if NYCT fixes CBTC's interlocking compatibility issues.

FWIW, dwell times southbound at USQ aren't as bad as northbound. The GFs suck -- if for no other reason than ADA -- but they're not the biggest ops impediment to improved service on the corridor.

51 minutes ago, Theli11 said:

Running the  train the Bronx, is just asking to make 7th Ave-Broadway more crowded then it needs to be. The train runs ends in Manhattan to catch all of the Harlem-Uptown Riders, who can't fit on the train. trains run on Dyre Av to provide people on the Dyre Avenue line service to the East Side. The trains runs pretty light going downtown, until 96-72 St, the is usually crowded from 149 St to Franklin Av, the alleviates that. 

So you do realize that running more trains to the Bronx will reduce average load/train, right? While also allowing IRT West's express tracks to do more than their measly 22tph peak. This notion that you need to put in s because trains from the Bronx will always be full is just...wrong. Induced demand exists, yes, but a) it's weak during peak times, and b) this would allow a nontrivial increase in service level down the express there. There are also, I might add, ways to redistribute loads away from the , for ex by running more . That said, I agree that deinterlining 149 is a bad idea in the short run -- the platforms/stairs that'd absorb the transfer loads there are waaaay too narrow for the volumes that'd be transiting them. 

51 minutes ago, Theli11 said:

If we really want to fix Roger's Junction, we can do reconfigure it to look like something 59 St - Columbus Circle, that way, the merging doesn't have to be so bad. trains don't interrupt train service. and train service don't interrupt train service. This way we can remove the unnecessary elements on the line. The solution to everything doesn't always have to be deinterlining, especially when looking at other elements. As long as we don't ruin the good parts of the service we do have. 

I'm really amused by the suggestion that 59 St is a good junction. It isn't. It sucks -- so badly, in fact, that the entire IND is scheduled backwards from it. 

Regardless, you cannot easily reconfigure Rogers to look like 59. You'd basically have to reconfigure the entire Eastern Parkway line east of Franklin to do so, as preserving the stacked config makes it impossible to deconflict the interlocking (the best configurations the '90s and '00s studies on the issue could come up with were ones that put 3 services on one track through Franklin, the issues with which should be self evident). And, of course, I do not see any logic in spending an order of magnitude more money to build something with a _lower_ operational/travel time return than simply deinterlining the area. 

1 hour ago, Collin said:

I'm still not sure how yard access would work.  The main yard for the  is Livonia.  Some trains would still have to run to New Lots for trips to and from the yard.  That could be confusing for passengers, and would create additional merging.  One option would be to put sizeable tail tracks at Flatbush which would allow some trains to be stored there.

Tail tracks would be really nice, but it's worth keeping in mind that the majority (IINM 16/22) of AM  put ins come out of 148. Running a few early-AM put ins from New Lots wouldn't be that hard, nor would be skewing the balance of IRT West express service more towards s to allow more put ins from the Bronx's yards -- remember there'll be more space up there for put-ins given that the  would have some of its fleet transferred to Livonia.

29 minutes ago, CenSin said:

I don’t sweat it when I know I’m transferring to a service that is frequent. But I smack people out of the way if I know an extra second of dashing makes the difference between getting on a train and waiting who know how long for the next one.

And, of course, low service frequency is inextricable from problem of interlining.

11 hours ago, Collin said:

@R68OnBroadway People on this forum are quite knowledgeable about the operational side of NYC transit, but the human element definitely needs to be considered too.  Think of yourself commuting between various stations.  Yard access needs to be considered too.  For example, sending the  down to Flatbush with the  might seem like a good option, but that cuts off yard access for the .

Re: the  and yard access, much of the reason this was an issue in the past is because the rampdown in passenger volumes after the PM peak was extreme, so running trains north from Flatbush to yards in Manhattan/the Bx was financially inefficient. That's not the case today, and as has been noted in past discussions on the issue, it'd be trivial to run some s or s as put-ins from Livonia. 

2 hours ago, P3F said:

Just because it's there, doesn't mean it's logical to suddenly force thousands of people to use it. It's a long distance to walk with many stairs, and some of the passageways aren't exactly wide (so you can't direct infinite amounts of people into them).

To answer your pointless loaded question, I'd say that the Atlantic Avenue transfer is best suited for BMT - IRT connections.

Once again, the logic of Dekalb deinterlining is that said transfer really wouldn't have to be used all that much. If you're on the , you can use the  to almost exactly the same destinations in Midtown, or transfer to the  to get further east/west/closer to 14 St. Vice verse for Broadway: transfer to the  to move your final destination around if necessary. 

And, of course, Herald Square exists. 

15 hours ago, Trainmaster5 said:

I don’t have a dog in this fight but I have a rather simple question. Has anyone considered the average commuter in any of these ideas ? The basic idea is how much time is spent on a trip, bus, subway, railroad. Basically every transfer adds to the time spent on a trip. There’s a reason why people get upset when their one seat ride is screwed with. Just thought that point has been overlooked so far. Carry on.

I don't disagree that some of the proposals here are tending towards...the extreme, but I do think it's worth pointing out -- and I'm sure you understand this -- that transfers aren't the only contributor to trip time. Merge delays from poorly designed service pattern, dwell delays from unbalanced loads, etc, all lengthen trips as well, and generally have ramifications for the entire system vs people in some rel. limited submarket. This is about balance.

40 minutes ago, bobtehpanda said:

This is obviously not as good as fully removing Broadway from 11th. But what would be the problem with the following service pattern:

to 96, same southern terminals

Forest Hills to Whitehall, weekdays only

Astoria to Bay Ridge

- If you’re going to interline 11 St, you may as well give Bay Ridge the yard to simplify equipment moves 

- Phasing out Whitehall as a terminal should be a long term planning goal given its severe negative impacts on through service

- City Hall Curve can’t handle >21tph with existing signaling, which would leave you with just 6tph of WHL-CTL after filling Astoria

22 minutes ago, engineerboy6561 said:

Just out of curiosity, what would it take to get DeKalb able to process 90 tph? If you could do that then you'd presumably see the same sort of improvement in reliability, on-time performance and throughput you'd get from a reconstructed Rogers Junction?

Reforming punch treatment may get you +1tph, but beyond that deinterlining is needed. It alone is likely the most impactful step you could take towards upping capacity, but you’d probably run into other topological constraints if you tried to push tph above, say, 24. Not only is the signal system within the junction quite gnarly (lots of timers, long control lines, etc), but many of the corridors feeding the junction have constraints of their own, ex 36 St merge and long control lines on the NB express for the , the sharp curve entering Dekalb NB on the , Whitehall terminal, 95 St terminal and City Hall Curve on the , 59 St on the , etc etc etc. This isn’t to say deinterlining isn’t useful, you just have to take a system perspective when doing it. 

28 minutes ago, Lex said:

Ah, yes, let's separate Brighton from 14th Street. That's a great idea.

...Bleecker St, Barclays Center transfer to the ... Hardly separating them. 

FWIW, I think the most convincing argument against deinterlining Dekalb is that it'd make it difficult for West End riders to get to 8th Ave. You'd be looking at either an xfer to the  for Jay, a ride crosstown on the , or substitution using the  from Barclays for areas south of 42, at which point  are only a block from 8th. Do I think that that's sufficient reason not to deinterline Dekalb? No, both because 8th Avenue below 42 St has, by Manhattan standards, very low job density, and because Dekalb has huge operational ramifications for the system...including 8th Ave. But it's interesting food for thought. 

4 hours ago, shiznit1987 said:

As someone who (was) riding the QB Line (and getting on at 65st no less), TONS of people stay on the  past Roosevelt. When I moved here I thought I'd be on easy street thinking everyone got off at Roosevelt and I'd be guaranteed and empty train. Boy was I wrong...

People want to get to either Broadway or 53rd St, and those getting on between Forest Hills and Roosevelt will stay on a local if it takes them where they want to go. This is why I always am deadset against any local service going through 63rd st. It would be as much of a waste as the  if not moreso. 

I'm sure some do through ride. But the peak load point for the  is, last I checked, between Roosevelt and Elmhurst -- ie the sum of through riders and passengers boarding west of Roosevelt isn't enough to replicate loads east of Roosevelt. 

Re: O/D I'd strongly challenge the notion that the primary destinations are Broadway and 53.  trains gather most of their loads at 1 stop (Lex-59),  trains leave 47-50 packed, and SRO s north out of there in the PM are the norm. 53 is certainly a draw, but let's not forget that 6th Ave is damn near the median midtown job. 

9 hours ago, Collin said:

No one who is already on the express wants to ride the local from Jackson Heights so they can get to 53rd.  You'd have a tsunami of complaints if that happened.  There's already an incentive for customers who got on at local stations between Forest Hills and Jackson Heights to stay put that's the fact that they probably will have a seat vs having to pack in like sardines on the express.

Yeah, of course they don’t want to. But we aren’t getting new Manhattan<>Queens tunnels anytime soon, and the only way you can extract more capacity from Queens boulevard is by getting more people to ride locals west of Roosevelt, sooooo...

As anyone who rides QB can tell you, the incentives to stay on the local today are mighty weak. Ever been to Roosevelt during the AM rush? Wall of lining the express, that grows whenever a local pulls in.

1 hour ago, Around the Horn said:

So wait you're saying my original idea to run the as a WTC to Forest Hills local with as many trains per hour as physically possible (and the on the express) was actually a good idea?

I got roasted badly when I suggested that lol

Yes! This is exactly what I’d do. People underrate the capabilities of WTC (look at 8th  or SF ) as well as the operational convenience of having your QB-53 service be a short line.

16 hours ago, shiznit1987 said:

Among the ideas, this is my favorite. The  absolutely needs to be extended up 10th Ave and after that it's crazyness not to have it go across 57th St and do something...

LaGuardia, here's me piggybacking off your idea:

 up 10th Ave, across 57th st, turns up 2nd Ave and takes over SAS to 125th-Lex. 

 run down 63rd St to QB Express (179th St)

 run down 53rd St to QB Local ( to Parsons/Archer  to 179th St)

 Broadway Local/Astoria

 Extended under 34th St Manhattan as Crosstown service to Hudson Yards. 

As an aside, is there any possiblity of adding two more tracks to 42nd st to make the  4 tracks? I always wondered if that was possible....

This effectively limits 8th local and 6th local to a _combined_ 30tph. Have the decency to give the  its own trunk!

20 hours ago, Caelestor said:

I agree that the  should run local because the 63 St line doesn't stop at Queens Plaza like the  currently all do at 59 St. However, it's simpler to have only the  run on the local tracks between 207 St and WTC full-time, and move the to 53 St / 8 Ave express. Then all the B division trunk lines

• Concourse / CPW express / 6 Ave express
• 207 St / CPW local / 8 Ave local - 
• 53 St / 8 Ave express
• 63 St / 6 Ave local lines -
• Astoria / Broadway local  
• SAS / Broadway express  

can be deinterlined from each other in Manhattan. This also standardizes the headways at 4 minutes on the , 6 minutes on the , 8 minutes on the  so that continued interlining on QBL and at DeKalb Ave remain smooth.

It seems I'm fighting a losing battle here on convincing folks that interlining 36 St is a bad idea, but I'll make one last pass at it. 

As I mentioned upthread, 59 St -- whose merge configuration is exactly the same as 36 -- is a bad merge. It is, in fact, a _very_ bad merge. It causes a massive amount of runtime variability on the routes that pass through it, and is in fact so limiting of B division performance that the entire division is scheduled backwards from it. Here are variability charts for the  and  to help attach numbers to the issue:

The merge performs badly not because it's cursed with slow switch speeds or poor signalling, but because of its design and its position on the routes that transit it. 59 St is what I like to call a 'conflicting merge,' where a merge delay can ripple backwards through the pipeline and cause a delay on a different service (think: ). Here, this effect is especially pernicious because a  delaying a  or vice versa can further complicate things by messing up the merge of the second service -- if a  is delayed by an  and delays a  behind it, that  may end up delaying (or being delayed by) a . Worsening its impact is the fact that it's positioned immediately before a high-dwell station, meaning delays from a merge get followed by a nice dose of NYCT close-in fixed block ops, which are...bad. Finally, the merge in both the north and southbound directions is downstream of at least one other merge, making consistent operation through it that much more difficult. 

36 St would be all of this, but worse. While there isn't a high dwell station to aggravate things, switches at 36 are slower and train volume is higher. With _current_ throughputs (before you up service levels, which sorta is the whole point of this exercise) Jackson Heights (closest measure point with all services passing through) sees 4tph more trains (in both peaks) than 59 St does in the AM. Both charts show n/b train volumes: 

vs

Given that merge delays are proportional to throughput (see chart below of  runtimes through a day), you've got a problem. I am _extremely_ skeptical that you'd be able to hold the PM peak railroad together through 36 St, what with the lines having passed through merges and a bunch of high-dwell Midtown stops on their routes to 36. I think as throughputs went towards 50 or 60tph, you'd end up with trains stacking up through 63 St and into Queens Plaza, which really just...isn't a way to run the trains.

This operational logic alone should be enough to convince folks that this may not be the wisest of ideas. In case you don't feel this way, let's talk about O/D. Sure, QB local loses direct 6th local access, and QB express gets cut off from 53/8. But express riders can easily transfer at Jackson Heights to recapture that O/D, and local riders can get the  at 7-53 for 6th; neither of those losses should be dealbreakers. The O/D argument that _should_ give pause is that interlined 36 St would mean there is zero incentive to stay on the local past Roosevelt (or Woodhaven, if that gets built out). On the ? Take the . On the ? Take the . That would likely make the Queens Boulevard dwell time issue -- which already produces peak hour runtime increases as pronounced as this:

...to say nothing of the fact that it'd further reduce the efficiency of the Queens subway network by reducing loads on the one part of said network that has significant room for growth: QB local. 

I really don't suggest this course of action. 

12 minutes ago, R10 2952 said:

Been wondering for years; were yard put-ins during rush hours set up to be regularly scheduled for certain departures? I used to take the out of 207th weekday mornings in the mid-to-late 2000s, and the 07:22 departure was always a train of R38s that came in from the yard leads; the trains departing directly before and after that were always R44s coming in from the road (not the yard).  It's like they had everything timed and planned down to the smallest detail.

Yes, internal schedules schedule put ins, layups etc. They are not random operations. 

2 hours ago, Around the Horn said:

As a passenger, I notice that on the you either have an extremely skilled operator and a very smooth ride or an operator that's not up to par and a very rough ride and there's little to no in-between. Can't wait for Lex CBTC and some operational consistency especially in the peak of the peak where one slow train can throw all the rest behind it out of whack.

This. A bunch of new operators on a line with tons of gnarly GTs and poorly signed (if signed at all) STs is really not a recipe for consistent runtimes and service delivery. I do not mean to belittle the work those crews do, of course. It’s just they could benefit from better infrastructure, better training and a more nurturing supervisory environment.

22 hours ago, Trainmaster5 said:

The , aka " the Beast" originally ran from 241st St White Plains Road to New Lots Avenue in Brooklyn. Local stops in the Bronx and Brooklyn. The present day " Beast " also makes local stops in Manhattan overnight on the way to Flatbush-Brooklyn College. It's counterpart is the overnight from 207th Street Manhattan making all local stops to Far Rockaway. Either one is a guaranteed workhorse,  mainly for people with little seniority or gluttons for punishment,  IMO. My opinion.  Carry on. 

The other horrible one is the , especially during GOs or rush hours. A 2 hour round trip without leaving the cab PEL-BBR-PEL + adjacent track flagging or rush hour delays can really do a number on you. And if you're late, and your layover gets cut at Pelham...

I don't think it did...

2 hours ago, Union Tpke said:

@RR503

Do you know if this issue is being looked at? Asking other systems how their CBTC systems work with interlockings should be done, if it isn't already.

Again, do you know if this issue is being looked at?

1) Maybe the switches connecting City Hall yard with the mainline north of Fulton. The diverging move north of BB is awful. 

2&3) Yes.

1 hour ago, Deucey said:

I feel like if were eliminated, Franklin  were double-tracked and made 10-cars long and ran Brighton Local to Stillwell,  ran express on Brighton, and both the Jay St switches were upgraded to higher speed or had a siding installed along with adding a NB switch at B-way/Lafayette, you could make  a 6th Av Express and CPW Local (switching to Fulton Local at Jay St) and solve this QBL capacity issue without de-interlining CPW.

But that’s a lot of money to spend.

Deinterlining CPW is an extremely worthy policy objective in and of itself. 59 St is a terrible merge -- so bad, in fact, that the entire IND is scheduled backwards from it.