I took the Q back from Manhattan to Brooklyn in the middle of the afternoon yesterday. Because it was an off-peak train and the MTA is doing some work north of 57th St., the Q train turned around at 57th St. I boarded a train on the downtown express tracks, and what happened next was something sort of silly.
After 57th St., the Q these days stops at 49th St. to make up for the lost W train. I had thought it made that stop only if it were running into and from Astoria, but apparently, I was wrong. After 57th St., we switched from the downtown express tracks to the downtown local track in order to stop at 47th St. We proceeded along the local track from 49th to Times Square, and after Times Square, we switched back to the express tracks before 34th St. to continue downtown.
Furthermore, before leaving 57th St., we had to wait for an N train to clear in front of us, and we moved in front of an R train, thus holding up another full train at 57th St. At each switch, the train crawled, and by the time we left 34th St., we had probably lost a minute or so of travel time. It struck me as operations planning at its worst when the Q just could have skipped 49th St. while avoiding two switches and creating delays.
In the grand scheme of the MTA, this is a Little Thing. It’s impact on people individually is rather negligible, but it’s an inefficiency. Eventually it might matter.
Now, frequently when I talk about the Second Ave. Subway, readers want to know how the MTA will re-route the BMT Broadway Line. The current plans, developed before we lost the W train, called for the Q to run north from 57th St. to 63rd and Lexington and then up Second Ave. Today, we no longer have the W train, and it’s unclear what the MTA will do. They can’t cut service to Astoria, but they’ll need to run trains to the Upper East Side. It’s a decision that’s at least five years away, but it’s a popular topic nonetheless.
In my opinion, because of the switch, the train that runs up Second Ave. should be an express. The express tracks run north of 57th St. directly to the 63rd and Lexington line, and there’s no reason to slow down anyone’s trip because of the need to switch. The MTA will have to revive some sort of local service to Astoria by then as well. The ideal routing then would include a Q train from Brooklyn to 96th and Second via the Broadway express, the N from Brooklyn to Astoria also via the express, another local — call it the W — to Astoria via the Broadway local and the current R train service.
This is, of course, planning very far ahead, but in the interim, the MTA should eliminate the double switch the off-peak Q makes in the span of three station stops. It’s just unnecessary.
58 comments
Q riders get screwed on the uptown side as well. Wrote to you about this previously and experienced it EVERY day this week, so it’s not a fluke. Uptown Q express & uptown N local both leave 14th St at same time. Every time the Q holds at 34th St and waits for the N, then waits again for the N to leave. Why? What purpose does this serve? Why are Q riders forced to sit at 34th, then switch to the N they left at 14th St? Also, why isn’t the Q switched to the local tracks before 57th St instead of after 34th? Inefficient dispatch/switching decisions make commutes longer than they should be.
That is why I usually never take the Q train because the N and R seem to get to 42nd street faster and more often.
And not only that, but some Q trains short-turn at 42nd once they use the express tracks as storage tracks between 42nd and 57th at the end of rush hour… without ever announcing that. Getting onto an uptown Q in Manhattan at rush hour is a gamble. You never know if the local will be faster, but usually it is.
Some morning rush hour N trains do this as well. I guess to keep uptown express service at a similar level pre-cuts (or maybe Astoria can’t turn around all the Q’s and N’s running), some morning N trains from Brooklyn run express and terminate at 57th Street, or at 42nd Street. I don’t know how common it is, but I see it pretty often if I’m riding before 9AM.
The important point to make is that the Broadway line always had it’s problems with the local and express condensing to two tracks heading to Queens, and the cuts have only made the problems worse.
Not enough room for every N, Q, and R train from Brooklyn on the two-track segment north of 57th.
If the Q is running ahead of schedule, it will be held at 34th to let the N go first. (Looks like the Q schedule may need some pruning.)
I read somewhere that the Astoria Q stops at 49th because that’s a particularly busy stop among Astoria residents. (Think Rockefeller Center – unlike Brooklyn residents, who also have the B and D, Astoria residents only have the N and Q to take them to Rock Center.)
Why? Isn’t the point of an express train that it should run *faster*?
Still has to run on schedule.
But the schedule should allow the express to run faster..
No, the point of an express train is that it skips some stops. Local riders are in as much of a hurry as express riders; they’re just going to or from local stops, or maybe the local pulled in first.
Remember that the Brighton station reconstruction project is nearing completion. When the Q resumes service at M and H, that might be enough for it to fall back into place on the schedule.
If the MTA doesn’t get its operating finances in better shape by then, I could easily imagine Astoria getting cut down to one line.
Train frequencies are determined by train loading. If ridership in Astoria stays as high as it is now, it needs about as much service as it’s getting now. True, the N alone could cover that, but that would lead to a major increase in service at the Brooklyn end, where it isn’t needed. Running some of the service as W’s to Whitehall is cheaper.
And we also got to Union Sq at the same time as that N train. Dear Ops Planning: Please fix. Thanks.
I definitely think that Broadway line riders were the biggest losers in the 2010 service cuts, at least as far as the subways are concerned. I used to commute daily between 42nd Street and Canal and the N/Q express combination was a dream. These days the wait for a Q is interminable, and the N and R poke along. I’m sure that in absolute terms the trip has only slowed by a few minutes, but it feels like an eternity.
Probably a minute at most. Between 42nd and Canal, take whichever train comes first, and you’ll get where you’re going faster than you used to if you insisted on waiting for the express (especially southbound, since northbound you have to pick a platform in advance).
This is what makes the MTA a maddening service. So he Q is express, except when it isn’t, but maybe not wholly, oh OK partly, and by the way at the whim of the MTA.
On top of that, on a trunk line of 3 lines that basically has 10 minute headways between each lines’ trains (an average of 3.33 minutes between trains), they are all bunched together! And, like their infernal bus service, 40 years later there is no fix!
What’s the point of having 2 lines serve as Broadway/BMT locals if they run right behind 1 another? Then there’s a 9 minute headway between 2 trains on 2 lines instead of a 10 minute headway between 2 trains on 1 line. Makes no sense.
The problem with the BMT express tracks has always been that unlike the IRT and BMT expresses, they aren’t through-running in midtown. You can run an express train to Astoria or Forest Hills, but unless the MTA opts to use the 63rd Street crossover, you always have to interfere with BMT local service to get it to Queens. The 49th Street switchover for the trains terminating at 57th makes the situation even worse.
Even after Second Avenue opens, unless you make the N a permanent local in Manhattan you’re going to have crossover problems south of 57th Street (and if you revive the W to serve Astoria and 49th Street, then you’ve still got the N switchover to the local tracks northbound and to the express southbound that can’t help but cause delays).
How is this any worse than any other merge in the system? There are dozens that take place in regular everyday service. This one’s far less innocuous than some (DeKalb Avenue, I’m looking at you).
Or Rogers Avenue Junction…
What I honestly don’t get is why they chose toe end the express tracks. Shouldn’t the express trains be the ones that travel longer distances? (Also looking at the New Lots Line here…)
Neither set of tracks specifically end – they simply merge before turning east. But the local tracks are on the outside, so without fancy flyovers (which the BMT didn’t provide here), locals can’t terminate without interfering with expresses.
In this case I don’t think it’s a major problem, even if it isn’t ideal – the express service is mostly of benefit to Brooklyn riders, and few Brooklyn riders need to go north of 57th.
It’s more of a problem at Brighton Beach, which also has expresses terminating on the middle tracks (once express service on the line resumes), but the line only continues for three more stops.
I’ve never understood Utica. The northbound and southbound tracks are on different levels, so it wouldn’t have been any harder to have the New Lots line feed into the express than into the local. Of course, now we’re stuck with what’s built.
The original plan for the Broadway line had it continuing up the west side.
Stop the human control of subway trains!
It’s insane we still use 1800s technology for a system never anticipated to be so big.
We need computer controlled subway trains to ensure maximum efficiency and minimum human error. Former drivers will then be required to roam the train assisting riders and cleaning up. Delish.
Maybe by 2100?
Technology is there. Rolling stock (only R143 on L, R160 on others compatible with CBTC) and funding isn’t.
CBTC and automatic train control are two different things.
I’m referring to the Siemens CBTC system in use on the L. Siemens’s CBTC system can go as far as ATO, ATS. I’ve read the Siemens stuff on it (Google it), the system has a lot of features that can be added on request. If you can think of it, it supports it. (Sorry if this sounds like an ad, delete if needed, Ben.)
But there’s a lot more to CBTC than the cars. The CBTC system has to be installed on the line as well.
It’s coming soon to Flushing. Then Queens Blvd.
I know, I know. And if they’re sticking to the original plan for the Culver viaduct rehab, it’s being installed between 4 Ave and Church Ave on the northbound express track on the IND Crosstown.
For vendor interoperability testing. Wayside signals will still be used for regular service.
Out of curiosity, why are they not installing it on the Lex – older trains that aren’t CBTC-compatible?
Wouldn’t that problem be bigger on the 7? As I remember it, the stated reason for the L and 7 was that they’re the only isolated lines on the system, so it makes sense to prove the concept there. (Any idea what they’re doing with the 7’s equipment?)
Maybe Queens Blvd could use the extra capacity a little more than Lexington Avenue. Lines from multiple trunks feed it, leaving more potential for conflict. The Lex is pretty straightforward: the 6 is about as operationally independent as the 7, and the 4/5 are fed by a few secondary feeder lines.
Plus, by fluke or by design, many of the most recent orders have ended up on Queens Blvd – and the G is off Queens Blvd.
Precisely. The R142 and R142A (soon to become R188 on the 7) are compatible. The R62’s are not. As Andrew said, you of course need to install the system in the tunnels, but the R62 are (like the R68/A) reactionary trains (after R46 P-Wire fiasco) using 1930s technology. They’re SMEE trains, installing signalling equipment on them wouldn’t be worth it.
No, newer wayside signals. Some of the signals and signal equipment on the Flushing line dates back to the 1920’s if not earlier, and the QBL signals are original to the 30’s. The signal system on the Lex were modernized in the 50’s or 60’s, so it doesn’t need to be replaced yet.
Remember – given the expense of installing a new signal system, the primary goal is to bring the system to a state of good repair. Anything else, such as increased capacity (which is generally fairly minor) and ATO, is a side benefit.
The cars have nothing to do with it (sorry, Alex). No current IRT cars have CBTC equipment installed already, but the entire Lex, as well as the 2 (which is frequently rerouted to the Lex), has new cars. Retrofitting those cars with CBTC would be a lot easier than what’s going on on the 7, which also requires transferring cars and turning 10-car trains into 11-car trains.
Yeah, have the Q remain as Broadway express to 96th and 2nd avenue. Might even be possible to increase its frequency. Have the N remain local and go over Manhattan Bridge. Have a few N’s turn around at Whitehall or re-institute W. Keep R as is. Might be a bit tight btwn Prince and Canal but leaves out any potential conflicts in Midtown.
I’m assuming they have their finances in order and actually receive adequate funding on 2016 and the W returns. But then again, that assumes that in 5 years the public and the politicians realize how important funding the subways is.
Run the Q Broadway express – 2nd Ave, the N Broadway express – 63rd St – Queens Blvd local (finally making full use of the 63rd St tunnel), and the R Broadway local – Astoria. Increase R service (with possible short-turns at Whitehall) as demand merits.
The W will need to make a comeback because the N can’t provide enough service alone to Astoria.
Why cut off Queens Blvd. access to 60th St.?
I have always wondered that about the switches. As far as the SAS goes, would it be stupid and/or impossible to send the F to 96 and have the Q replace the F on the QBL? Then the R and N could both go Astoria and maybe have the G replace the R to Forest Hills.
The the F and the Q would have to crossover each other.
Even if it were possible, it would be a net service cut to the QB line.
This would suck for rush hour commuters. Many Queens Boulevard commuters are going to the business district along 6th Avenue, so they need the F . Most rush hour commuters in Queens do not go to Brooklyn, so the G simply is not needed in Forest Hills. So it won’t be a Queens Boulevard local again…..
There are fewer tracks leading to upper Manhattan and Queens from midtown than there are in midtown – until they finish the 2nd Ave subway. At that point, it will be properly balanced – until they continue building it south of 63rd. To maximize capacity after phase 1 is finished, I’d run the routes as follows:
N & Q: Both run over the Manhattan Bridge, Broadway express, and Second Ave to 96th (eventually 125th). Note, no merging/switching north of DeKalb in Brooklyn.
R & W: Both run to Astoria and on the Broadway local. The R runs to Bay Ridge as it does now. The W continues down the 4th Ave local and the West End local (like the M used to), except at night and on weekends when it stops at Whitehall. Note, no merging/switching ever happens north of Whitehall.
H & V: The Queens Blvd locals. The V shares track with the F under 63rd St and 6th Ave all the way to Church Ave. It runs local in Brooklyn. The new H train runs on the 53rd St tunnel with the E, the 8th Ave express, Fulton St local in Brooklyn to Euclid, replacing the C in Brooklyn. Note, this takes advantage of all the IND flyover track connections and minimizes merging/switching while maximizing the # of trains.
B & D: No Change, except the D runs express on the West End Line in Brooklyn, peak direction, while the W is running to Brooklyn.
E & F: No change to the E. The F runs as it does now, except express in Brooklyn between Jay and Church.
A & C: No change to the A. The C stops at the WTC with the E.
JMZ: The M reverts to its previous route and all three services stop at Broad St all the times.
Instead of running routes that somehow conform to where people want to go, this minimizes switching and maximizes capacity (and the use of express tracks for speed) based on how the system is actually built. People would have to transfer more, but if all these routes are running at maximum capacity, the transfers would be quick and easy and people would probably get to their destinations faster on average. Ideally, some pedestrian connections are also built between Bowery & Grand, 63rd/lex & 59th, and 7th Ave & 57th St to prevent having to double transfer or exit the system.
By trying to simplify matters, you’ve ended up causing a whole bunch of new complications.
Running down your list:
You have way too much service up SAS.
Your new W is completely unnecessary in Brooklyn (the West End M was discontinued because it was very lightly used), and running the D express would seriously degrade service for most West End riders (i.e., the ones at local stops), who will have to transfer to the D to go over the bridge.
You’ve replaced Queens Blvd. access to 60th with 63rd for no good reason. I guess everybody who wants to transfer to the Lex has to go through 53rd, which is already overcrowded now.
The Queens Blvd. locals run 10 tph each. The Fulton St. local runs 6-8 tph. Linking the two leads to far too much service in Brooklyn, at high cost.
There’s no need for additional service between Jay and Church – the F isn’t overcrowded. (Yes, the first car is crowded. So try the middle of the train, even if it’s a slightly longer walk from your exit.)
The WTC terminal can’t possibly handle 21 tph (C+E). The Broad St. terminal can’t possibly handle 18 tph (J+M+Z) – not to mention that the “new” M has been extremely popular, and there’s no particular reason to discontinue it.
The current service pattern may not be perfect for everyone, but it does a pretty good job of providing appropriate levels of service on each line. To the extent that your pattern is workable, it increases service unnecessarily on several lines while at the same time degrading service for a lot of riders.
I am assuming a reality that the 2nd Ave line is eventually extended to 125th St around 2024, and the city (and subway ridership) has grown substantially in that time. Ridership on phase 1 of the SAS will probably be close to 100,000/day and more than that after phase 2, meaning there will not be “way too much service” at those stations. In this scenario, the MTA can’t provide some services like the M or R that use track on multiple lines because doing so decreases overall capacity in a system that has become overburdened. In this reality, if the F & V are coming every couple minutes, a transfer at Delancey/Essex is more efficient for the system than running the M in the Chrystie St connection.
Although some of these pairing might lead to too much capacity on a line by today’s ridership, having less crowded trains that come more often is a good thing! I also foresee much increased ridership on all Brooklyn lines, especially the northern Brooklyn ones such as the JMZ and Fulton local. With increased ridership, If the Broad St and WTC terminals can’t handle that many trains turning around, then they should be redesigned – a relatively cheap way to increase overall capacity. If the Nassau line isn’t using the Montague tunnel, those tracks could be used to create a 4 track terminal at Broad St. Isn’t it silly that we have all these tracks without the ability to use them at full capacity? That was one of the main reasons for replacing the South Ferry station. If we are spending so many billions to build the 2nd Ave line, can’t we spend a little more on some pedestrian connections and higher capacity terminals?
The Queens Blvd local wouldn’t have direct access to the 60th St line anymore, but there are many problems with that connection as it is now. First, there aren’t that many people that need it. Of all the trains that cross the river from Queens, the R is the least used. It often takes 5 minutes to get from Queens Plaza to Lexington because of the many turns and switches it must use and then it delays the N/Q at the merge. Second, the stops on the F are very close to the stops on the 60th St line already, making the services somewhat redundant at the expense of capacity. Third, if the F and V are running under 63rd, and a good escalator/moving walkway connection is built to the Lexington line, the transfer to the Lex line might even be easier & more convenient than at 53rd/Lex. Fourth, if there are a ton of Ns and Qs going up 2nd Ave, and a ton of Fs and Vs on 63rd St, the cross platform connection to the Broadway express should be very fast and easy, making it faster than the R train is now. If you can get to Union Sq from Jackson Heights in 7 quick stops, you might not want to transfer to the Lexington line anyway. The N & Q will skip 49th and will not be delayed as they are now by the constant switching back and forth Ben describes in this post.
I think having a direct route to your destination becomes much less important when you have trains reliably coming at 2.5 minute intervals that are not delayed by burdensome switching and merging. Transferring is much less of a burden.
I want to write in support of more frequent services (meaning, two services) to the SAS Phase 1. A big part of the whole point is to attract people from the Lex, and if SAS offers 6tph off-peak compared to 30tph (15tph of which are “express”) on the Lex, it is going to be harder to attract people. Certainly, people coming from right near the new stations or east of 2nd Ave would likely choose SAS anyway, but given the station spacing and that the densest parts are probably between 2nd and Lex, it should be an attractive service.
If today’s N and Q both go to SAS, I certainly agree that they should all be express through to 57th, with no switching in Manhattan. But then what to do with 60th St and Queens? While 63rd St. cross-platform will likely be quite attractive and useful, most Queens-to-Lex riders want Grand Central itself or Lower Manhattan, neither of which N+Q will serve directly.
Obviously Astoria can only be served from 60th St and is growing, but QB to Lex is really best served by 60th St too (barring speed/frequency issues with R), with a much better interchange than 53rd or 63rd (even with improvements), and the fact that the demand is uneven makes it harder. Even if it were possible to run 30tph through the 60th St tunnel, what would then happen to them? Certainly don’t need anywhere near that much to Brooklyn via Montague, and Whitehall can probably never turn more than maybe 10tph (less now, but thinking incredibly optimistically). Would there be any sense to reviving City Hall lower level? This is sort of the intended purpose of it anyway.
During Phase 1 and Phase 2, there needs to be at least 12 tph on SAS if you want to divert people from the Lex. That would be accomplished by running the N and Q express up to the UES, as proposed previously.
I would then add a W line running local between Astoria and Whitehall St at all times.
Astoria needs more than the W.
The only time there’s a need to attract people from the Lex is rush hours. Off-peak, there’s room for more people on the Lex, and if service needs to be increased, there’s room for more trains.
That’s true as long as you consider 3PM as the beginning of the evening rush which is when the northbound trains start becoming jammed.
Running way more service than ridership levels require is terribly irresponsible.
Rebuilding terminals and interlockings (which is never a “relatively cheap” proposition) to handle those unnecessary service levels is criminally irresponsible.
The West End line currently gets along just fine with 10 tph on the D; you want to increase that to 24 tph on the D express plus 24 tph on the W local? Aside from the fact there is no possible way for a single track anywhere to handle 48 tph (the express track ends at 9th Avenue, so the D and W would have to merge there), how could you possibly justify the expense of increasing frequencies on the line nearly fivefold?
I must have missed where it was specifically proposed to run 48tph along the West End Line. That is obviously ridiculous. The issue here is how best to serve the Broadway Line and its branches. The line has always had an imbalance of heavy on the Brooklyn side (two separate two-track lines, each with multiple potential branches) and light on the Uptown/Queens side (a single two-track line with two branches).
For simple analysis purposes, let’s consider 1x supply/demand to be 6-10tph (closer to 6 in the off-peak and closer to 10 in the peak). At the moment that balance isn’t too bad capacity wise, as one of the two Uptown/Queens branches (Astoria) has higher demand, so that doubles up to create a 3×3 match (as the Brooklyn lines only serve 3 “branches” total – Brighton, Sea Beach, and Bay Ridge). However, there is an “operational” problem insofar as the trunk line is slow and clunky, largely due to the fact that demand is sort of 2-1 exp-lcl south of Canal, 1-2 exp-lcl in Midtown, and 0-3 exp-lcl north of 57th, requiring switching.
Fine – now we add a new Uptown/Queens branch into the mix – a very expensive one into a very dense and busy area that will likely have high demand, and that we want to encourage use of – so we likely then will have a 5×3 pairing of northside vs. southside demand. So, in the future how do we best serve this? Well, the good news is that the 2x demand matches up on each side of the exp, so that is good, but we have a 3-1 mismatch on northside lcl vs. southside lcl. At the moment, it seems to me that the best option is to create a 2x W, with half terminating at Whitehall St and the other half going through to Brooklyn and terminating at maybe 9th Avenue. Yes, this is too much capacity on the local south of Canal St, but we need it in Queens, and with growth in Downtown Brooklyn, Jay-Lawrence transfer, and growth around Park Slope it might be useful.
“Trains reliably coming at 2.5 minute intervals” implies 24 tph on each service. AlexB wants two West End services, so that’s 48 tph, if I’m understanding him correctly.
I don’t think SAS will require as much service as you claim – the Q will be plenty for Phases I and II.
Until recently, two trains were stored in the middle of the day at 63rd and Lex, on the tracks that will ultimately lead towards 2nd Ave.
Those tracks are out of service right now for the station rehab, so those trains are now stored instead on the express tracks between 57th and 42nd. (If the Q is shortlined to 57th, then there may be several more trains stored in that area, since there are fewer trains in service.)
That explains why your train ran local: nothing can operate on those express tracks while trains are stored on them. Everything has to be on the local tracks through 49th.
When SAS opens, I guarantee that the line serving it will be an express. The service plan you propose is the most likely possibility.
I don’t understand why 49th Street on the Broadway line requires THREE services when the 50th Street stop a block away has only 1 the vast majority of the time.
Why not use the Y-shaped ramps north of 57th Street on the Broadway line that connect to both the express and local tracks? Run the Q express to 57th, then merge and head to Astoria. The N and R would be otherwise unaffected during times the Q turns at 57th Street. Leave 49th Street completely out of the plan for Q trains.
Yes and this will alleviate the issue I brought up at the top with the stupid decision to constantly have the Q wait at 34th. Makes absolutely no sense why express train passengers are forced to sit for up to 3 minutes to wait for a local train they “met” at 14th St, then asked to walk over to the local because it will leave first. Run the Q express to/from 57th.
That just moves the conflict three stops up. There’s going to be a merge somewhere regardless of where you put it.
The Q has to merge with the N and R somewhere. Where exactly to make the merge should be done based on ridership concerns, since there’s no way to avoid the conflict. And 49th is a busy stop for Astoria residents, so it makes more sense to have the Q stop there than to skip it.
Also, some rush hour Q and N trains terminate at 57th. Keeping those trains on the express and running through trains on the local reduces the risk of a through train getting stuck behind a terminating train waiting for an open track at the terminal.
The 1 never shares trackage with the 2 or 3. Crossing the 2 or 3 to the local track would introduce unnecessary conflicts and would reduce capacity on the line.