Why is my commuter rail so slow?

I was wondering when I approved this post how this discussion would go, but it's been very informative.

Otto is right; BRT is the service whose name is not spoken here. Go on over to New England Railfan, there's a topic on the proposed New Britain - Hartford BRT proposal, where we all spit blood at the thought of it.

Don't know much about Chicagoland transit, but this is an education, and thanks to those who are able to illuminate this topic. And Tuna, don't mind the sarcasm, it's just indicative of the passion for rail here.

If you hate the "waste" of railroads taking up space to offer "slow" service, then you must really hate the old Illinois Central line... They have exactly the system you're talking about with dedicated tracks for passenger and freight trains. It's about six tracks wide.

I think your confusion over relative speed is that you are comparing your ride on a train to your ride in your own car. The train is what we call "mass transit." It is designed for moving a large amount of people all going to the same place at the same time in the same conveyance. Your automobile is "personal transport." Two different modes of transportation, not even remotely related to one another. As far as Bus Rapid Transit and dedicated busways go, where do you think you live? Some sort of third-world country like Pittsburgh? You've got one of the best mass transit facilities there in Chicago when you combine the commuter rail, the rapid transit El and subway lines, and the bus transit lines. They all work in concert and one mode would not be able to survive without the support of the others.

-otto-

tuna wrote:As I said there are no freight trains on the areas I am talking about. I don't know if there are freight tracks somewhere else that are also called the Union Pacific North Line, but the Metra tracks are dedicated through all the stops I've ever been at, the vast majority of the stops. Also my idea for train timing involves smaller trains which probably means that they can stop faster. If you have trains that can come from full speed to a full stop in 1 minute, which isn't even that fast (as in it wouldn't be uncomfortable for passengers, I don't know how feasible it is for train technology), I don't see why you would need anywhere near 10 minute headways. By your own description, Metra is relying on these "traffic lights" and the human work of conductors to avoid crashing into each other. There should be a more efficient way that takes away human error and could lead to lower headways.

I'm not saying run on exactly the same infrastructure with exactly the same trains. What I was trying to get at in my OP is that the UP North line has these 2 tracks dedicated for commuter rail service, with no freight interference, no complicated junctions with other lines, what if the federal government wanted to invest a lot of money into developing the future of commuter rail, something that would be much faster and more competitive with cars for getting from the suburbs to the city center. So you take the existing land, stations, and tracks of the UP North line and lines like it that have continuous dedicated right of way, and you develop a system to be much faster... are there any systems like this that exist?

About the only commuter rail example that I can think of with this sort of an arrangement is the LIRR in NYC between Penn Station and Jamaica, a major transfer point. Within this corridor, it's a four-track railroad, with locals on the outside tracks serving intermediate stations at Woodside and Kew Gardens, while the expresses run on the inner two tracks. But there is a reason the LIRR has the distinction of being the busiest commuter railroad in the country, and a lot of it has to do with the fact that it serves the largest mass transit market in America. However, there simply isn't enough demand for that kind of a sophisticated commuter rail network like that here in the Chicago area, and as stated earlier, the volume of freight traffic that our region contends with on a daily basis just wouldn't allow it. With that said though, there are other solutions that would better serve our needs, to wit:

Since you ride the UP/N line, then you've probably heard about the bridge renewal project within the city portion of the route that will be starting up again soon. As you may or may not now, this line was once owned by the CNW (Chicago & North Western), and at one time, it was a three-track ROW between Chicago and Winnetka, with the third track used mainly by intercity trains to pass slower commuter runs. About 40 years ago, after leaving the intercity passenger business to Amtrak, CNW removed one of the tracks (along the western part of the ROW), and it's been a two-track railroad ever since. Today, the bridge renewal project calls for the restoration of this track (Metra refers to it as "New Track 1, it was originally CNW track 1, but I digress...), and eventually, all northbound traffic will be shifted here. The plans also call for the removal of the easternmost track (Track 2) when the project is totally done, but I've heard this isn't final, and if so, I'm hoping UP and/or Metra can at least consider retaining the track somehow; it'll only cost more to restore it later on if it's removed now. If it is retained, then extra capacity will be available to run more express trains during peak periods. Whether it would be enough to sufficiently speed up service depends on how well the schedules are written to make full and efficient use of the new capacity. At best, some peak trains may see up to 10 minute improvements in running time in isolated cases, but the average time savings will likely be less.

Another solution on UP/N could be something similar to what is currently in use on MED: The use of BOTH tracks in peak direction, a style of operation used for main line service between Kensington and Matteson, where rush hour trains are divided into three "zones" (Harvey, Flossmoor, and University Park). If applied to UP/N, you could run both tracks in the peak direction between Waukegan and say, Highland Park, for instance. But to really make it work, you'd still need that third express track along the city portion of the line. Either way, it would still be a better option than creating 2 "passing tracks" for each station, not to mention a helluva lot cheaper.

Great thread, BTW...

Tadman wrote:@Larry, that's one of the best ideas I've heard for a while and it would balance things well. UPW has 30k riders/day and MD has 45k total. It would not be an even trade, but I cannot figure out why 100% of MD and UPW trains must cross each other. Makes little sense other than "That's how we used to do it".

@Eruns, I don't remember exactly what's in CREATE, but it was designed to solve problems like this by heavily upgrading to local rail infrastructure. The railroads would've got freight through town faster, which creates jobs. The commuters would've got a faster ride and higher capacity, which opens up more living choices and less expensive living choices.

Instead, the Gov has his mitts in the cookie jar and also robbed RTA blind with the "seniors ride free" program. Ergo, we slog along on legacy arrangements.

Yup, I'm familiar with CREATE and the excruciating slow pace of it's implementation but even without it, existing systems can be improved as lstone19 pointed out. UP only operates trains for Metra, why can't Metra dictate the re-routing of the UP-W line into Union Station? I thought it had something to do with the old CNW station but Metra owns Ogilvie now so unless there's something in the operational/legacy contract I'm not aware of there seems to be no good reason NOT to do this. I doubt it would matter much to Amtrak, who owns CUS.

@tuna Well, there used to be a thrid track on the UP-W but it was torn out years ago. This is also the line that was to begin it's bridge rehab project months ago and seem to recall a neighborhood up in arms aboubt losing it's trackside garden during the necessary track relignment. In other words, above and beyond techonolgy and infrastructure limitations, what appear as simple fixes usually are not.

@Jeff, I agree. At first I figured the OP would get flammed but this has been englightening and informative. We should always challenge and look to make things better rather than crutching along with "what's always worked". I don't post a whole lot but really appreciate the usual thoughtful discussion that occurs here.

@otto While Chicago does have a great mass transit system they do not always work in concert and if they do it's usually by happenstance or progressive planning which is not the norm. The CTA, Pace, and Metra are all seperate entities who compete for federal dollars and customers. While they all fall under the authority of the RTA, it's not reagionaliztion in the truest sense but now we're veering off into politics. I'll just say there not being a universal fare card that can be used on all 3 systems best exemplifies the dichotomy that exists. Don't get me wrong, I'm glad we have a great system and utilize it often, but we shouldn't be satiated with the status quo.

tuna wrote:I'm saying just build switches at/around each station that would allow a "stopping" track in each direction so that there would be a total of 4 lines at each station. An inner 2 lines for passing through and an outer 2 lines for picking up passengers.

Hopefully you read and understood Larry's reply to this ... I'll expound a little.
Physical track layout. Switches are expensive. It isn't as easy as paving a bus bay next to a road. Putting in switches, remote controls and sensors to make sure the switches are FULLY in the correct position (especially important on the facing point switch used to go from one track to two) and then keeping it working is not cheap. Straight track is obviously much cheaper - and safer.
Signaling. Such a "platform siding" would require signaling entering the switches (from both directions) and for exiting the siding. If one wanted to they could make the express track one long control point. But the switch behind the train must be cleared by the train being passed before the route for the express could be set. Waiting for the express to pass just slows down the local even more.

There are much better ways to speed up a railroad than putting in short sidings that serve only to slow down trains even more.

doepack wrote:Another solution on UP/N could be something similar to what is currently in use on MED: The use of BOTH tracks in peak direction, a style of operation used for main line service between Kensington and Matteson, where rush hour trains are divided into three "zones" (Harvey, Flossmoor, and University Park).

The MED also runs three tracks in the peak direction from Kensington to 67th. Continuous four track service (or at least three with the center track being peak direction) would be an improvement for UP-N. But expensive. Simple crossovers on a two track line get a lot more complicated on a three track line.

One thing that would speed up the line is to run less trains. But that's not why the rails are there. Lowering the capacity of the service would not be a good thing.

Or don't worry about the legacy railroad distinctions and eliminate many of the crosses there by sending UP-W trains to Union while sending one of the MD lines to Ogilvie.

While I wouldn't say that this idea is without merit - I've previously pondered it myself - there are additional logistical factors that would likely negate the surface-level benefits:

• Crews on the Milwaukee District and NCS lines are Metra employees, whereas the crews on the three former C&NW lines are actually Union Pacific employees, part of the agreement in place between Metra and the UP for service. Mixing and matching which lines go into each terminal would require additional crew training and qualification on the different territories. Not out of the question, but a process that would take a number of months to implement.
• We cannot overlook the servicing of trains mid-day at the Western Avenue yard for Milwaukee District and NCS trains, and at the California Avenue yard for Union Pacific trains. All of these deadhead moves must pass through the interlocking at A2. And unless a massive new service agreement were put in place for UP employed maintainers to service Milwaukee District/NCS trains and engines from Union Station at California Ave./M19A diesel shop, as well as for Metra employed maintainers to service UP Northwest and UP North trains from Ogilvie Station at Western Ave. shops, these will still have to cross at A2.

As an aside, if you ever have the opportunity to observe a full rush hour period from the station platform at Western Avenue on the Milwaukee District side, you will be amazed at the amount of traffic and quick intervals that the crews running A2 handle.

tuna wrote: Hmm there are definitely no freight trains from the terminal to Lake Bluff, I can say from personal observation. Not on the two tracks that carry the Metra trains at least. Maybe Freight trains start at the Great Lakes Naval Station stop or farther north, but thats already the 22nd out of 27 stops.

While I admit I don't know all that much about the UP-North line, I believe the freight bypass I mentioned earlier joins the UP-North line at the Lake Bluff station. While there are no through freight trains south of Lake Bluff, there does appear to be some local freight service. That UP has retained ownership of the line rather than selling it to Metra suggests that UP sees a use for it.

Regarding your idea for station "pull-out" tracks, it sounds gosd but it just isn't that simple. Look at the Secauus Jct. station on the Amtrak mainline in New Jersey. The total distance from where the line first starts increasing from two tracks for the station until the station tracks are finished and it's back to two tracks is almost two miles. That's to allow a train to leave the main tracks at speed, slow for the station, and the get back up to speed before rejoining. And that's what you'd need to allow that express to stay at speed when passing. If the station tracks are just at the station, then the stopping train will need to be close to stopped before leaving the main track for the station and that express will need to be about three miles behind minimum to avoid seeing a yellow signal and having to slow down.

tuna wrote:I don't see why you would need anywhere near 10 minute headways. By your own description, Metra is relying on these "traffic lights" and the human work of conductors to avoid crashing into each other. There should be a more efficient way that takes away human error and could lead to lower headways.

Should be but so far there isn't. Since the system has to be designed for worst-case, I've yet to see any sort of automation or supplemental overlay system to insure separation that doesn't slow things down even more. A good engineer knows the performance of his train while automation tends to assume the train will always perform at its worst (e.g. braking performance as if the train is heavy and the rails are slippery). Even as a passenger sitting back in the train, I can feel the performance difference between a near-empty "light" train and the full "heavy" train.

One expedient requiring no additional infrastructure is grouping stations into zones. Without citing individual lines or stations, the evening rush-hour service could be structured by having a train run nonstop from A to N, then O. P. Q, and R; 3 minutes behind it a train runs nonstop from A to I, then J, K, L, and M, possibly continuing to R with or without passengers, depending on facilities for terminating trains; 3 minutes behind that, a nonstop to D, then E, F, G, and H, and finally, 3 minutes behind that, a "cleanup" train making all stops to R. You would then have to wait until the cleanup train got far enough ahead to run the whole pattern again as many times as passenger demand would support. The New York Central and the Southern Pacific (SF to San Jose) were doing that 40 or 50 years ago, and Metro North (out of Grand Central) still does. (For many years SP had a departure from 3rd & Townsend every 3 minutes from 5:14 to 5:32 or :35--the 5:14 didn't make its first stop until almost San Jose--, and all they had was 1 track in each direction with 3-aspect automatic block signals.) Even an interurban like SEPTA's Norristown line (former Philadelphia & Western) does that today. No additional station tracks or turnouts (in fact, SEPTA was able to take out an intermediate center track (at Wynnewood Rd.) when the ridership pattern changed). Actually, I believe Metra does this to Aurora, and I think it was mentioned above that Metra Electric does also; I think the only constraining factor would be if total ridership couldn't support that many trains.

...I think the only constraining factor would be if total ridership couldn't support that many trains.

Also the lack of equipment to run that much service. During the morning rush on the BNSF, for example, 30 trains arrive at Chicago Union Station between 5:30am and 9:00am. Of those 30, 15 turn around and go right back out for a second inbound trip. Granted, two or three of them turn late enough that their second inbound trips would be classified as "mid-day service", but that entire line is run with only 18 sets of equipment.

E Runs makes a fantastic point. Chicago transit does not make any effort to network Metra, CTA, and Pace. By playing with Google maps, I've realized I can transfer off an inbound MDN train at Mayfair to blue CTA trains, or at Evanston off an inbound UPN train to purple/red CTA trains. If I'm inbound from a meeting in the suburbs, I don't want to ride downtown just to catch a CTA back to my near-north neighborhood. Further, there's got to be plenty of people that work in the city but not downtown what would love to bail off Metra short of downtown, ride a connecting CTA to their job, and save 20 minutes.

Tadman wrote:Further, there's got to be plenty of people that work in the city but not downtown what would love to bail off Metra short of downtown, ride a connecting CTA to their job, and save 20 minutes.

Nothing stopping them from doing that now. Integrated fares might make it easier to pay for the journeys. Guaranteed connections would only serve to slow down the system(s).

Zone Grouping: Though an interesting concept, the riders habits and destinations on the UP-N are quite varied, as Metra discovered last September. There are quite a few, 20% if I recall the RTAMS figures from 2006, that get off at Evanston-Davis, not to mention other part way stops (New Trier students, Ravinia season, cubs fans from the CTA Red/Purple line, etc.). In addition, large segment of reverse commuters now (major destinations, Evanston/Highland Park/Lake Bluff).

Freight Traffic: Union Pacific runs many coal trains through to the Kenosha Subdivision joining from the Lake Subdivision at Lake Bluff. According to the cars, destinations very from the ComEd plant in Waukegan to the Edgewater plant up in Sheboygan. These trains tend to hold in Lake Bluff until after the rush goes through.

Decreasing Travel Time: In my observations there are certain places in the current structure that would prevent decreasing the travel time in the short term. The reverse commuter trains are often behind schedule due to the crowded cars, called by Metra, heavy passenger loading. Since some of these have to turn around mid way, they slow down the normal commuter trains. This is increased if they have problems with the switch, or the dispatcher. In addition, due to the configuration of the cars, loading time takes longer and especially longer if there is need to use a lift. High level platforms would solve the problem but, I do not see that as practical.

Priorities: In my opinion, the greater short term priority needs to be on increasing equipment problems on the UP-N line. Delays of 50-70 minutes due to equipment problems occurring multiple times a week make any savings from increased speed mute. Even a minor delay is made worse if some of the trains are already behind schedule. As the equipment continues to age, this will only increase. Apparently, the locomotives on the trains affecting my commute have not yet been rebuilt.

As it is, even with the Tri-State rebuilding complete, my commute home by car would still be longer than by train, and far more stressful.

I don't think anyone is suggesting guaranteed connections, but such connections are not advertised enough, or at all. You have to obtain and read multiple timetables from multiple agencies. Then you have to either budget enough "layover" to account for tardiness, or not do and just hope that everything is on time. Most people don't want to go to all this trouble.

There is "Goroo" (ex-RTA Trip Planner), but it isn't very good.

Outlying Metra/Amtrak connections are pretty terribly advertised as well, or at least not kept up to date. A look in an Amtrak City of New Orleans/Illini/Saluki schedule shows "METRA/IC line" at Homewood, with no real indication that it's another rail service one can use. The Chicago-St. Louis/Texas Eagle schedules also show no connections at Summit or Joliet (Summit I can understand, but omitting Joliet? Really?). Same deal with the Chicago-Quincy (Illinois Zephyr, etc) trains and their corresponding timetable, La Grange Road and Naperville being the possible connections. The Southwest Chief shows no such information for Naperville, but the California Zephyr does ("METRA/BN line"). Hiawatha schedules don't show Glenview as having any connections, although the Empire Builder's shows "METRA/Milw. line" for that stop. Again, it's there but not terribly informative.

Of course, Amtrak isn't the only one to blame here. Metra has a "Connecting Services" section of each line's timetable, but there is not one mention of inter-city Amtrak connections that one could make at any of the outlying stops which both agencies use.

The purpose of Metra is "get to work commuter rail". While it would do little harm to note the outlying Amtrak stations, I don't see it doing a lot of good. It is beyond their purpose of "get to work commuter rail".

Such out of the box thinking can work ... NICTD South Shore "weekend trip to Chicago" services as well as special promotions (no discounts) for the auto show, Bears games and anything else happening along the lakefront works. Service to Notre Dame games is also popular (although there were no specific chartered trains for Notre Dame games in 2010 - regular weekend service and a two hour after the game ends express return extra served the extra passengers). 50-100 years ago the "get away to the Dunes or Hudson Lake" trips were common - but Chicago's improved lakefront and other attractions have reduced that business to near zero.

Selling Metra as a connector to Amtrak long distance trains at outlying stations is out of the box. Perhaps if NRPC wants to pay NIRC for advertising in their timetables it could be done. Otherwise the benefit to Metra is low.