Implementing Streetcars Demands Consideration of the Way Traffic Works

» The District of Columbia is installing streetcar tracks along Benning Road, Northeast, a nice advance for the city’s transit options. But it’s doing so in a way that will limit the speed and accessibility of trains.

Secretary of Transportation Ray LaHood’s announcement that he would award $130 million to streetcar projects across the country early next year excited transit advocates, who hope that the mode will soon become a frequent sight in American cities. Indeed, there are plenty of towns seriously considering building streetcar lines — at least 45 in North America as of the most recent count. Since streetcars, like local buses, run in lanes shared with automobiles, implementing them efficiently requires studying traffic patterns and attempting to find ways to avoid problems that could impair the efficiency of transit service. When the first projects get off the ground in 2010, one hopes that they’ll be built right.

Among those cities thus far that have committed to spending some of their own money — rather than just hoping the federal government will pick up 100% of the tab — is Washington, D.C. Though it has yet to secure all of the necessary funds for the massive system it plans to build, the city has already begun installing track on H Street and Benning Road Northeast, a prime corridor for improved transit investments. Once it assembles the necessary financing, it will add overhead catenary and begin running the service, potentially in two years.

The District has posted some relevant photographs on its Facebook page, and the images are indicative of the problems this corridor will encounter once it begins carrying trams. Of course, Washington isn’t alone, because streetcars face some fundamental difficulties as a result of the fact that they share their rights-of-way with automobiles. But its decisions about how streetcar tracks will be installed will plague the service provided by the vehicles and ultimately limit their utility.

Below, I have documented some of the quotidian situations that will result in delayed traffic as a result of the design of the District’s streetcar system. None of the problems are unique to streetcars — in fact, they’re shared with any vehicle that must share its running way with automobiles, including buses. But streetcars are put in a particular predicament in each of the cases noted below because, unlike buses, they can’t change lanes. If systems are designed with major flaws, such as those illustrated below, this means that these trains will operate at significantly lower speeds than equivalent buses; the result: a big investment investment in public transportation could actually mean less mobility.

But take note, other cities: these structural issues can be resolved through better designed streets.

Entering Streetcar Lane

Entering streetcar lane: The most obvious problem with streetcars, of course, is that they run in the street. That makes their construction relatively cheap, since tracks can simply be laid in a preexisting roadway. But because the streets are shared, cars can weave in and out of lanes, including those used by trains. This will force streetcar drivers to operate their vehicles overly cautiously and at limited speeds. By the same measure, while a bus has the ability to change lanes to pass stopped or slow cars, a streetcar is forced to maintain its course, even if that means getting stuck. There is no way to solve this problem unless some of the streetcar’s path is made limited-access.

Blocked Turn Lane

Blocked turn lane: Because D.C.’s system is designed with left-turn lanes to the left of the median-operated streetcar, cars have to pass through the streetcar lane in order to take a left at intersections. When there are too many cars and streetcars in that lane heading straight rather than left, cars from lanes further to the left will have trouble making the turn to the left-most lane. This could be a problem especially if there is reticence among some drivers to operate in the streetcar lane; they will remain in the middle lane until the last possible moment, where they may get stuck. Cities with streetcars hoping to combat this problem could install signage encouraging drivers to stay out of the streetcar lane most of the time, while also encouraging people hoping to turn left to move over as quickly as possible.

Too Many Turners

Too many turners: If D.C. gives its left-turn drivers their own signals, it could mean streetcars are held up by the number of cars waiting for a green light because of the limited length of the left-turn lane, which would push some left-turners into the streetcar lane. Even if the traffic light for vehicles heading straight is green, streetcars would be blocked in the above situation; this could force streetcars to wait for the next light cycle. At intersections with high expected rates of left turns, D.C. would benefit from longer left-turn lanes than what is currently being built. Alternatively, it could eliminate some left-hand turns and encourage drivers to take alternate routes to reach their final destinations.

Blocked Turns

Blocked turns: On the other hand, if Washington’s system does not have separate left-turn lights, through streetcar traffic could be significantly affected by left-turn drivers, especially if their cars are in turn blocked by pedestrians crossing in the allowed travel direction. This problem will be most prevalent at intersections with significant pedestrian and left-turn traffic; there, a separate left-turn signal might be necessary — as long as left-turn lanes are adequately long and streetcars are given signal priority before left-turning cars.

Stuck at the Light

Stuck at the light: Because of the shortness of the left turn lanes, two streetcars in a row (possible on several of the corridors D.C. is studying for its system, since they’ll service multiple lines) could block the possibility of left turns by automobiles entirely when stopped at a red light. If left-turn lanes were lengthened to ensure they are longer than at least two streetcars plus a car, this would solve the problem.

Stuck at the Station

Stuck at the station: Whether stations are positioned before the signal or after, when they stop to pick up passengers, streetcars will slow down following traffic (including other streetcars), especially if surrounding traffic is heavy. This occurs with buses as well, but since their stations are in the lower-speed outside lanes, their effect 0n surrounding traffic is more limited than median-running streetcars. On the other hand, if streetcar stops are timed with red lights ahead, changing to green once passenger pick-up and drop-off has terminated, this problem could be eliminated.

Conclusions: The fundamental problems with all of these approaches is that they encourage car traffic to move more quickly, effectively limiting any attempt to provide streetcars with a time advantage. To get a streetcar to move more quickly, the traffic system must be designed to move all vehicles more quickly, because they share the same lanes.

This raises questions about the value of streetcars in general — wouldn’t it make more sense to operate these trains more like light rail?  Cities could do just that simply by installing cheap ground-level barriers between streetcar and car lanes, and eliminating the left-turn lanes to the left of the streetcars (or moving the trams to the outside lanes), all while instituting aggressive signal priority. These approaches would dramatically improve the efficiency and speed of streetcars and provide them a relative advantage over automobile traffic, which would be limited by fewer travel lanes than before, ultimately leading to more public transportation usage. That’s a valuable goal.

50 Comments | Leave a Reply »
  • Andrew S

    I feel like the last pargraph should’ve come first. If you’re not willing to give a lane over to streetcars, don’t build them. You can’t equivocate more than that, they’re simply not worth investment to run in mixed traffic.

    cities with long experience in running in mixed traffic (like Toronto) are now doing everything they can to separate streetcars from traffic, and all new construction is of the “light rail” variety. Let’s move on people

  • jon

    The Portland Streetcar on Harrison Street (PSU to Riverplace) operates like these examples mentioned with left lane operation in the center of street. It isnt much of a problem on Harrison St. but largely because Harrison isnt the busiest of streets, it would be a different story on a busy major arterial like those in DC. DC seems to have the space on their blvds for reserved right of way so IMO it would be crazy for them to not go with a reserved RoW.

  • EngineerScotty

    If its a busy street, with streetcars in the median–simply prohibit left turns.

    Or permit them, but from a combined through/turning lane to the right of the streetcar lane, with an electronic “NO LEFT TURN” sign (or similar) which lights whenever a streetcar is present.

    Even without the streetcar, left turns off a busy two-way street are major issues with traffic.

    Speaking of which, DC is famous for its large number of rotaries (traffic circles), especially on the avenues. How will streetcars navigate those, should they encounter one?

  • Couldn’t be more true in NOLA, where the city has large medians (we call them ‘neutral grounds’ because of the way in which the median on Canal Street used to divide the Creole and American halves of the city) in which to run streetcars (where they historically were located).

    Streetcars could absolutely approach light rail speeds and efficiencies in New Orleans if several simple measures were taken:
    1. Place all streetcar lines in the neutral ground
    2. Eliminate many intersections that cross streetcar lines at-grade
    3. Reduce the number of stops
    4. Give signal priority to transit
    5. Modernize box fare/collection systems
    6. Modernize fleet to include low-floor and ADA-accessible cars

  • John

    Srsly, streetcars in traffic are just the mobility-impaired version of buses.

    There reasons why they got rid of them in the 50’s, it wasn’t all just conspiracy by the auto companies. Nobody thought they were romantic. Just crappy.

  • Eliminating left turn lanes doesn’t mean eliminating left turns. Cars turn left without left turn lanes on boulevards in Manhattan, even excessively wide ones like Broadway and Park. All that’s necessary in that case is signal priority for streetcars, together with separate signaling that gives the streetcar a green light a few seconds before the cars on the same lane; the same signaling method is used in Manhattan to give pedestrians priority at some intersections.

  • Jonathan

    Phoenix’s system runs to the left of the left turn lanes, and the left-turn traffic light does not come on when transit cars are approaching the intersection (in either direction).

  • Ari

    There is a perfectly good remedy for left turns, but you have to go to Melbourne and get a mirror: the hook turn.

    Melbourne has (I believe) the largest tram network in the English-speaking world (by track, the largest in the world) and carries 500k people a day. And in the Central Business District, it operates mostly in mixed-traffic lanes in the center of the street.

    (I’ll translate this for those of us in the States driving on the left) To make a left turn, the turning traffic gets in the right-most lane, drives in to the center of the intersection and waits. The left lanes (including the tram lane) is strictly for straight-ahead traffic. When the light changes, the waiting queue of cars makes their turn, followed by the traffic from the other direction. Everyone gets to make their turns, and the general flow of traffic is unimpeded.

    This would, of course, be quite hard to implement in the states for a few reasons, the least of which is that it would be massively confusing. Also, in a city like Washington, many drivers are from elsewhere and the learning curve would be steep. Still, in a perfect world, it is a very efficient to allow turns but keep lanes clear.

    In DC, and with other streetcar and bus systems nationwide, allowing drivers to hold lights for a few seconds might help the problem. For instance, a streetcar driver seeing a traffic hold-up ahead could, presumably, push a button, which would create a green light and turn light to clear the lanes. This might tie up car traffic, but why should a few SOVs get priority over a streetcar carrying many times the passengers?


    While I don’t buy the wholesale “GM killed the streetcars!!1!” argument, there are a few advantages to streetcars, such as:

    • It does not have to pull in to and out of stops (like a bus) and hope that traffic yields
    • It can have dramatically higher capacity than a bus, reducing crowding without increasing staff costs
    • It can more easily board through multiple doors than a bus
    • It is cleaner (higher PMPG) and quieter and has no local emissions
    • It can go from street-running to private rights of way with relative ease, and attain significant speed advantages on private ROWs

  • RE: DC’s Streetcars:
    I’m not overly concerned with the left-side operation. Signal timing will be key, but for we don’t have any specifics on that yet. And since streetcars are at least 2 years off, there’s plenty of time to work that out.

    But left-side operation is probably best in DC’s case, especially on a neighborhood commercial corridor like H Street. In places where DC has right-side bus lanes both with and without parking even further to the right, drivers still block the bus lane when double parking. It’s quite frequent. I don’t think streetcar rails would make a difference to people.

    DC does have lots of circles. Of the 37 miles of proposed streetcar, only two segments involve circles. The Fourteenth Street Lines will have to go through Thomas Circle and the K Street Lines will have to go through Washington Circle.

    At Washington Circle, streetcars will use the K Street underpass, so they’ll go under the circle, not around it or through it.

    DDOT has not specified how streetcars will naviate Thomas Circle, however, in the past, they used “express” roadways (like Mass Ave at Dupont Circle) to go through Thomas Circle.

    Additonally, streetcars will somehow have to go around Mount Vernon Square. Again, DDOT has not released any information on that, but the square is extremely busy for cars, buses, and pedestrians.

    While DC does have a few streets with wide medans, none are currently proposed for streetcars. This is mainly due to their “monumental” nature and Congress’ ban on overhead wires in the L’Enfant City. Really, the only place with a median wide enough for a streetcar line is Pennsylvania Avenue between Barney Circle (Anacostia River) and the Capitol. This street used to have streetcars in the median. But since the Orange/Blue Line is under the street these days, it’s lower priority for surface transit.

  • AndyDuncan

    Another option is to run the street cars along the center of parallel one-way streets. That of course involves converting streets to one-way operation, something people don’t generally like, but they generally work better in urban environments anyway.

  • Andy: on the contrary, one-way pairs are worse for public transit ridership than two-way streets – they require walking more, and make transfers more difficult. In New York, bus ridership dropped on the relevant routes every time another avenue was made one-way. (Ridership also dropped in every US city every time a streetcar was converted to a bus – so no, it’s not that buses are more efficient than streetcars).

  • Andy K

    Most of the problems listed can be simply solved.

    Signal preemption – when a street car is coming, it gets a green light, or extends an already green light.

    Cars are not allowed to block an intersection while waiting for a pedestrian to cross the walk. I know – I got a ticket recently for “blocking” an intersection.

    Left turn pockets need to be designed to handle the traffic load – maybe not peak load, but the vast majority of the time there should be enough room in the turn pocket to hold the # of cars waiting for a green light.

  • James D

    They should either go for hook turns as already mentioned, or force drivers to go right-right-right around the block instead of permitting left turns.

    One-way streets aren’t a great idea. They lead to transit lines that are busier in one direction than the other and confuse irregular users.

  • david vartanoff

    Indeed, letting autos sabotage transit whether bus or rail is a mistake. San Francisco’s T line is a fine example of a so called LRV which is slower than the buses it replaced for$600 MILLION or so. Whether a simple curb or a raised pavement, the streetcars need dedicated lanes or they will be costly failures. More importantly, they need the signal preempts, and other traffic controls to guarantee priority. FWIW DC had a streetcar median on part of New York Ave which was part of the old Benning Road route.

  • Max Wyss

    Not being familiar with Washington DC’s streets, my reasonings could be off.

    As it has been stated before, signalling with priority to the streetcars is the key. I assume that, except where there is a left-turn lane, the streetcar lane is blocked to traffic (except buses and emergency vehicles).

    Signalling could be set up in the way that a first signal is at the height of where the turning left pocket starts. A second set of signals is at the intersection, of course.

    Now, when a streetcar approaches, the first set of signals turns red. The signals for the intersection turn red as well, and then, the signal for the left turners gets green, giving them top priority (in other words, no pedestrians etc). This clears the left-turning pocket, and the streetcar can proceed. only after the streetcar has cleared the intersection, the signals for the straight lanes turn green again.

    Together with that, the rules must be enforced … no big fines needed, but they must be enforced. This will educate drivers rather quickly…

    Another approach has proven to be astonishingly efficient (but for that, it probably would be too late in DC), namely moving both streetcar tracks to one side of the street, and controlling the intersections crossing the tracks with signals.

    In any case, it requires the political will to give a high priority to transit (and may involve giving priority to private traffic to clear a zone).

  • Peter Smith

    I feel like the last paragraph should’ve come first.

    word. the whole post reads like a screed against streetcars, and i’m wondering if that was intended? fine by me — just state that at the outset so there’s some natural progression in making your case.

    other than that, i think the whole worry about how cars are affected is misplaced. that’s just a personal opinion — if you love cars, you’ll probably be inclined to say things like, “The most obvious problem with streetcars…”. If, on the other hand, you are into cities for humans instead of cars, you’d probably think, “The most obvious problem with cars…”. When you think about pedestrians and cyclists and transit-users of deserving of living in the city and even using the city’s streets, and even deserving priority over cars, it changes the way you think, and it comes out in your writing.

    i’m down with the livable streets transportation hierarchy, so the only reason i care about cars is making sure they don’t mess up the city any more than they already do — otherwise, i’m all about the pedestrian experience, the biking experience, and the transit experience — in that order.

    in fairness, you did hint that cars are not perfect. i’m happy to see that, but everyone’s entitled to their own opinion anyways.

    and excellent drawings/images. super-clean, easy to understand. we could use a lot more of this type of thing. Portland cyclists have some issues with their streetcar tracks, so i hope other cities will learn.

    Speaking of which, DC is famous for its large number of rotaries (traffic circles), especially on the avenues. How will streetcars navigate those, should they encounter one?

    i just saw a picture of tracks around a circle today — looked like Thomas Circle in DC from back in the day, so it must be possible. ah, there it is.

    Streetcars could absolutely approach light rail speeds and efficiencies in New Orleans if several simple measures were taken

    i think speed matters somewhat, but it’s not overly important. when we clear the streets of some cars, everything will flow smoother, people will be happier, healthier, etc. people care about a dignified, ideally pleasant, way to get around — be that walking, biking, transit, or car. but we can’t clear the streets of cars until we actually provide people dignified alternatives. streetcars go a long way towards that end.

    Srsly, streetcars in traffic are just the mobility-impaired version of buses.

    i love that so many cities are doing streetcars and light rail and all sorts of dignified transit. with GM and crew totally weakened, they can hardly manage to put up a fight. the best they can do is rely on the oil and tire companies to fund BRT propaganda outfits to kill light rail wherever possible. they’ve had success there, no doubt, but this rail thing is a tsunami — can’t stop, won’t stop. :)

    Phoenix’s system runs to the left of the left turn lanes, and the left-turn traffic light does not come

    we advocates do have some work to do. we can’t let car drivers kamikaze our trains whenever it strikes their fancy. we can’t let them continue to do it with impunity. make their insurance cover damages, physical damage to the train and to the people riding it. fine them. send them to jail if appropriate for the harm/injury they’ve caused.

    While I don’t buy the wholesale “GM killed the streetcars!!1!” argument

    i do buy it, and it’s important from an advocacy perspective to help people understand why this awesome form of transit went away over such strong public opposition, but ultimately, rail projects will prove themselves. we just have to keep cars from kamikaze-ing them to death. i’m guessing the damage that cars have caused along Phoenix’s light rail line is by now an astronomical figure.

    i’m hoping the awesome Vancouver streetcars kick up rail advocacy even another notch. dignified transit — finally. we need decongestion pricing, though, because too many people will want to ride. people are absolutely better to get out of their cars — we’re seeing that all over America. we just have to make it possible for them to do so.

    San Francisco’s T line is a fine example of a so called LRV which is slower than the buses it replaced for$600 MILLION or so.

    The T line is slow, but it’s a miracle — it’s seeing increasing ridership gains every day. People love it. I’d like to see some music on the trains, though — not elevator Muzak music, though — something decent. i hang out at the Panera at 4th/King — amazing area. the only problem? way too many cars, speeding, running lights, threatening pedestrians and bikers and transit users at every moment. we’re going to fix that, too.

    And the hard part with the T-Third is done — we put tracks in the street, and we managed to make a lot of it separated ROW — reappropriating some space from cars. we’ll get the signal preemption in time. meanwhile, SF is starting to have what resembles a real transit network. we’re about to get blown out of the water by the BRT-heads. that might do a couple of generations of damage, but at least the buses-instead-of-rail choices will continue to drive up bicycle ridership, which is already exploding. that so many progressive towns are building rail projects will help me and others fight off the bus folks a bit more. let the best argument win! :)

  • jon

    i think the way to go with DC streetcars is more or less lite-light rail especially if space permits… center dedicated lanes/RoW with just cheaper, and smaller light weight track and cars and simpler stations than standard LRT.

    i’m with alon, i would avoid couplets for transit. maybe its just me, but a transit line split on two streets really irritates me unless its at a terminus. it complicates the route placing it on another street. many parallel streets are not equal in their traffic and commerical opposed to serving one street really well, it dilutes the service for both streets requiring walking an extra distance. yet couplets and giant loops are really popular with modern streetcar routes… and its all because you can squeeze out an extra few blocks of properties that contribute to the capital costs of construction because they measure distance from the tracks.

  • NMRguy

    Amsterdam also has a very efficient system of prioritized signaling and segregated transit lanes all throughout its city. The system works very well and trams (streetcars) blow by the slower moving traffic. Tram stops are often put just before busy intersections (in both directions) to allow for traffic signaling to change while people get on/off the tram. Left turns with cars are allowed, but signaling obviously does not permit left hand turns when trams are passing through the intersection. And of course, buses use the same segregated lanes and stops to their advantage.

    The main point here is that construction of these segregated lanes greatly improves all forms of public transport along those arteries, at the expense of automobile access and mobility. Unless cities are ready to make this sacrifice, they may as well stick with buses. Otherwise, they can’t make use of the great advantages of trams/streetcars.

  • Peter, the post isn’t against streetcars – it’s against streetcars done wrong. The congestion problems with streetcars are not so much that they slow down cars as that they slow down the streetcars.

  • Thanks for the comments, everyone. Perhaps I should have written what I wrote in the conclusions at the beginning, but I wanted to emphasize why I saw problems with D.C.’s streetcar implementation first instead of simply dismissing the program from the beginning.

    As Alon says, this post is not meant to be against streetcars (I have repeatedly argued in favor of streetcars in a number of posts on this site), but rather for using streetcar technology more effectively. Cities spend a lot of money putting these things on their streets — they might as well do them right.

  • Brian

    Streetcars aren’t rapid transit. They’re urban circulators. They should be reserved for dense corridors where a slow ride is tolerable because the convenience is not in time but energy and comfort.

    The best analogy I’ve ever heard referred to streetcars as “horizontal elevators.” If going a few blocks, you walk. Just as you can take the stairs, if going a few floors. But often out of convenience other than speed, people take the elevator.

  • Justin Bernard

    Cars must NOT be allowed in streetcar lanes. It’s time that transit planners stop catering to the automobile, and make it less convenient for drivers. Give streetcars their own lane, and keep automobiles out. At intersections with left turns, put the left turn to the right of the streetcar lane, and permit left turns AFTER The streetcar has cleared the intersection.

    I notice with many plans that the auto must be accomodated. This has to end.

  • Dan

    One small problem that seems to cause a lot of delay on the Portland Streetcar is the location and size of the streetcar platforms. Most of Portland’s platforms are identical in length to the streetcars and they are placed right at the intersection. So, if there is a car waiting at a red light in the streetcar lane, the streetcar has to wait for the car to move before the streetcar can move into the platform area and open its doors. Sometimes the wait can be a full red/green cycle. If the platforms were either longer and/or moved away from the intersections, the streetcar could shorten the time spent waiting for autos to clear out of the streetcar’s platform space.

  • Vin

    Just throwing this one out there – why not eliminate the parking lanes and run the streetcars in a separated lane (just raise a curb, the separation need not be fancy) on the right? I suppose you would still need to deal with motorists making right-hand turns, but just make it clear that streetcars have the right-of-way, like pedestrians.

    Banning, or at least limiting, left-hand turns also seems to be an effective solution.

  • AJ

    Sprague’s streetcars killed a lot of horses until the system was perfected. And it didn’t require moving the tracks, it was just the culture shifted around the tracks.

    Similarly, Portland’s on-time performance has increased, ridership is comparable to LRT lines in some places, and people are really in love with the thing. This is because the culture changed– fewer people drive on the tracks, stop in front of the Streetcar, park in its path, etcetera.

    Improving the technology or making environmentally-appropriate changes will happen in time.

  • Russell Warshay

    Yonah, excellent article!

  • Dozens of European cities operate extensive streetcar networks. For example, Vienna (pop 2m) has over 100 miles of streetcar tracks, in addition to buses (LPG fuel), subways, S-Bahn, regional and long-distance trains (all electric). Only a small fraction of the streetcar network is fully grade separated, a larger fraction features dedicated lanes between intersections.

    Every single one of them has to deal with traffic flow optimization on a daily basis. Perhaps the most productive approach would be to learn from them. For example, make sure the streetcar tracks in each direction are as close to one another as the loading gauge will permit, with no traffic lanes in-between.

    Also, European traffic laws require drivers to yield to streetcars, which have their own signaling system. It is synchronized with the one for vehicle traffic, such that streetcars get a green light before cars do. On network sections shared by multiple lines, it is common for two streetcars to be stopped at a stop at the same time. In that situation, the one in the rear must wait until the first is well clear of the stop to avoid blocking motor vehicles that need to cross the streetcar lane. Most European driving tests include sections on the rules of the road in the vicinity of streetcars.

    The traffic lanes in the downtown areas of old European cities are usually narrower than anywhere in the US and, there are far fewer of them. Sacrificing traffic lanes goes against the grain in the car-centric US, but it’s often the only affordable option for giving streetcars the rush hour line haul times they need to attract high ridership. Equally important is land use planning that gradually increases residential, retail and commercial density within easy walking distance of streetcar stops. Much the same applies to BRT concepts – linear public transportation corridors work best if economic activity is clustered in the same pattern.

    Final thought: European cities have stayed relatively compact not just because established street patterns are hard to change but also because taxes on gasoline and diesel fuel as well as parking fees are much higher than in the US. Even in Western Europe, plenty of people rely on public transportation and bicycles because they cannot or choose not to afford to own a car.

  • AJ’s comments are largely wishful thinking – why on earth would a current driver in a streetcar corridor WANT to stop driving and ride that streetcar he (and all the other drivers) are currently slowing down? If anything, his incentives are in the other direction.

  • kk

    I agree with the Benning Road-H Street problem but they could also go about it several ways but many are to late since the tracks are already there.

    Build a Bridge/Tunnel over/under the intersection for the streetcar.

    Put the cars on the right lanes of the street.

    Completely change the street and have one side for cars and one side for streetcars

    example ( for benning rd lets say the eastbound direction take the right lanes plus the lane beside it and have tracks laid there with enough room made for the platform and give the other half of the street to vehicles.)
    like this bad example


  • rg

    I agree that dedicated ROW for the streetcars with limited left turns is best. However, I live near the tracks in question and can report that there is not too much traffic turning left off of this stretch of Benning Road. The streets running south of Benning Road NE between 15th Street the Anacostia River are mostly residential and not too westbound many cars turn left onto them. Even fewer eastbound cars turn left onto streets north of this stretch of Benning Road: there are only a few streets running north of it because of large lots/institutions that break up the street grid (Hechinger Mall/Spingarn HS/Langston Golf Course). So, this design should work out pretty well on Benning.

  • Rafael: in France and Germany, new construction is of the light rail variety, often with its own dedicated right-of-way. In the 1950s France and Germany both tore up their streetcar networks to make life easier for drivers, just like the US; the main differences are that gas taxes were higher in Europe, going back to the 1920s, and that in France and Germany many of those streetcar networks were replaced by subways.

  • Scott

    Muni F train is a great demonstration of how it doesn’t work. Its super busy (market), gets totally screwed up by traffic, a ton of trolleybusses run in the same traffic, and theres a more useful set of trains a floor or two below the street. I thought the F train was cool when I got here, but now its just sort of ‘meh’ and I’d prefer even the slowish muni LRVs.

  • indiecognition

    Tank-style reactive armor on the Streetcar would solve everything here.

  • @ Alon –

    I think you’re generalizing a bit. Paris and Hamburg did indeed get rid of their streetcars, but Munich, Stuttgart, Dresden, even Berlin all have streetcar systems. So does Lyon, which just recently brought an express tram from downtown to the airport into service.

    Where possible, new systems are indeed given dedicated lanes, e.g. in Strasbourg. These are often covered with grass sod to reduce noise, green the place up and discourage pedestrians/cyclists/motorists from straying onto these lanes. However, that’s not the same thing as fully grade separated light rail (e.g. in a freeway median, cp. LA Metro Green Line). Typically, European streetcar operators still have to make do with signaling priority at grade crossings, e.g. in Bordeaux. Note the absence of an overhead catenary system.

  • Berlin’s streetcar system is a holdover from East Germany. West Berlin replaced the streetcars with subways.

    And no, light rail isn’t grade-separated, unless it’s built as an excuse to build another freeway in LA. But it’s based on modern LRVs with dedicated ROW, often cutting through parks and contraflow streets, rather than on 1930s-era streetcars.

  • Mike

    The system we have here in Switzerland works pretty well.

    (1) Streetcars (and buses and trolleys, for that matter) run on a separate signalling system from cars – looks more like a rail signal than a traffic light.
    (2) Most signals can be remote operated to change to “go” as the streetcar approaches
    (3) Stops are located just before a set of lights, so that the streetcar can set down and pick up passengers while the lights are red anyway (in your “stuck at the station example, the left-to-right line would have a stop on the other side of the interchange, and traffic stuck behind it would have had to wait at the lights anyway).

    Add this to street design that builds in “refuges” or passing points at regular intervals where streetcars run in mixed traffic lanes, and you have a system that allows clockface timetables to be rigidly adhered to – even in sections where stops are less than 2 minutes’ travelling time apart.

  • Max Wyss

    In addition to Mike’s comments concerning Switzerland:

    • “Trolleys” mean trolley buses in this context.

    • The system normally used to remote operate the signal is so that the transit vehicle bears a transponder, and when approaching the signal, it requests priority treatment from the signal control system.

    • This same transponder can also be coupled to the door locking switch in the cab (as it is implemented for example in Zürich). When the operator locks the doors, the transponder requests an open path from the signal control system, and by the time the doors are all closed, the signal has opened. Consequence is that the vehicle can more or less immediately leave the stop (or if it is a bus, leave the signal-protected bay).

    In another comment, lawn-covered right of way has been mentioned. This is indeed a measure to reduce noise, and it is amazingly efficient (can be between 10 and 20 dB). In some cases, where the right of way has to be usable for emergency vehicles, concrete grid “tiles” are used, where the grass can grow between the openings of the grid.

  • A Guthrie

    The conclusions section largely misses the point. The primary advantage streetcars have over busses is not speed. In terms of choice riders–the group with ridership growth potential–either will always be slower than driving. Even LRT is in most cases. In other words, most urban choice riders choose transit for reasons other than speed. These reasons include lower stress, lack of need to hassle with parking, etc. The advantage of streetcars over busses is that they are streetcars, not busses; they excel at providing the smooth ride and quiet, comfortable environment needed to attract more riders. In fact, the streetcar’s inability to change lanes can actually be something of an advantage–the vehicle tends not to make the violent swings from side to side that make the experience of straphangers as stressful as driving. Anyone who has ever actually ridden both a streetcar and a local bus will agree they’d be much more likely to voluntarily choose the former than the latter over driving. Also, assuming any of these rider-attraction strategies succeeds, the higher capacity of streetcars (especially if MU control is used as on many European systems) will be needed. (Of course, all of this also ignores the quantity of development attracted by permanent infrastructure.) In addition, the reason more LRT-like services are not planned for these corridors is that they are urban streets where LRT is not desirable. While LRT would speed through the corridor between small, island nodes–and form a barrier between sides of the street, mixed-traffic streetcars can be integrated with the street scene, and with the pedestrian traffic required to support it, creating a continuous corridor of high-quality, attractive transit. The two are not a speed-based hierarchy of a single mode; that is the same line of thinking that led to the dominance of the CEI in the New Starts program. They are different modes which accomplish different things. The rest of the post is quite valuable; I really would like to see more discussion of streetcar-oriented traffic engineering.

  • chris

    “The advantage of streetcars over busses is that they are streetcars, not busses; they excel at providing the smooth ride and quiet, comfortable environment needed to attract more riders.”

    What are you smoking? There is nothing smooth, quiet or comfortable about streetcars or light-rail any more than there is for buses. The rail cars fill up with the same crazies and underclass miscreants that make life miserable for the rest of the passengers. Even half of the passengers who aren’t blatantly antisocial don’t seem to understand the basic rules of common courtesy for riding public transit (e.g., don’t spread out your legs so that you leak over onto your neighbor’s seat, don’t bunch up around the doors, abandon the seat next to your neighbor if an entire row of nearby seats opens up, don’t hog adjacent seats by placing your belongings upon it, etc).

    Riding public transit sucks, and is generally an uncomfortable experience regardless of whether the vehicle happens to be a train, a tram or a bus. The best way to make the ride more bearable is to make it SHORTER, meaning FASTER. Make it rapid or don’t do it at all.

    BTW, I once defeated the Portland Streetcar in a footrace from the waterfront to the park blocks. How pathetic. If Portland had any foresight, it would never have built the streetcar or at-grade light rail. It should have built either an underground tunnel or a grade separated platform for the MAX, while only constructing half the number of stops as it currently has.

  • chris

    PS, I rode the Skytrain in Vancouver, BC, and the downtown segment is far superior to that of the MAX segment in downtown Portland (not to mention the pointless Streetcar). All underground, no competing with traffic, and only about three or four stops. Passengers can walk the rest of the way, as nothing is very far by foot in downtown Vancouver.

    And no surface level tracks to fall upon when riding a bike. I have to be very cautious when crossing the tracks in downtown Portland.

  • W. K. Lis

    In my experience with Toronto’s streetcars in mixed traffic, in general, there are no left turn lanes. The exception are whenever there is new right-of-way for the streetcars.

    Where there are no left turns, the streetcars create an “advance green” for the opposite moving traffic. When the streetcar stop for passengers and open their doors, the traffic to the right and behind must stop before the open doors. In doing so, any left turners in the opposite direction get the chance to complete their left turn.

  • For me, one of the most appealing streetcar alignment options is to allocate two lanes of a trunk to bus lane, with the car traffic ONLY running the opposite direction in the facing lane(s). Next street over is the one-way car traffic that way and the one way bus traffic the opposite way.

    Then put the streetcars on the express (passing) bus lanes and train the bus drivers how to share properly with streetcars. I presume streetcar stops will be to the right, in front of the bus stop, with stopped buses at the bus stop yielding to an oncoming streetcar.

  • Max Wyss

    To W. K. Lis, message 41:

    Are those stops where traffic is stopped behind the streetcar stops with a platform, or are passengers stepping out onto the traffic lane?

  • D. M. Johnston

    In Europe, the difference between LRT and a tram, is the quality of rights-of-ways, with light rail operating on a reserved rights-of-way, which can be as complex as a lawned boulevard or as simple as a HOV lane with rails. Once LRT is grade separated, it becomes a light-metro and loses most of the cost advantages of LRT.

    A streetcar is a tram that operates in mixed traffic with little or no reserved rights-of-ways, nor signal priority at intersections.

    In Europe the the term streetcar is not used; instead the term tram is used. A tram operates on a tramway and has a legal right to operate on the road.

    In North America there has been much effort by planners and transit engineers to morph LRT as a light-metro, by grade separating it, which makes it more expensive to build, while at the same time, less attractive to customers..

    We must strive to keep the light (light in costs) in light rail or we will make the mode to costly to build.

  • What North American cities are you thinking of? I’m asking because some cities, such as Calgary, Portland, and Jersey City, physically separated most of its LRT from other traffic but did not grade-separate it. Those cities’ light rail systems have short streetcar-mode segments downtown, but elsewhere run in freeway medians or on abandoned mainline rail.

  • Ocean Railroader

    The City of Richmond VA old streetcar ran down the ceters of the streets in the city and in the county side before the time of paved roads. When the roads were paved in the 1920’s many of the streets built four lane wide streets with cirb and gutters on each side of the streetcars sealing many of them running down the center of the street off from traffic in the 1930’s and 1940’s. When the streetcar lines were abonadoned many of them were left as grassy strips or turned into a extra set of lanes.

    If light rail returns then they could close the extra pair of lanes in the street and turn them back into streetcar lanes and open up the grassy centers of the streets for streetcars again.

    I would suport the idea of turning a tarffic lane into a streetcar lane consdering how rare streetcars are and how common car based things are.

  • Max Wyss

    To D.M. Johnston, message 44: It is more than common that the same line acts as streetcar on some sections and as LRT on others. Therefore, it is not really a way to make a difference. A good example would be line 10 in Zürich, which is pretty much a streetcar line between Hauptbahnhof and Leutschenbach, and LRT between Leutschenbach and the airport. So, what is it?

    The term “Tram” or “Tramway” is used in some areas in Europe, others are using (literally translated) “Street railroad”.

  • Juniperhouse

    The REAL answer is to slap a $4/gallon tax on liquid fuels. This will have multiple beneficial effects, and provide funding for myriad transportation improvements. Claims that such a tax would destroy the economy are hogwash–European tax on fuels is at that level, and Europe has a thriving public transport system and a thriving economy.

    We cannot fix our problems by simply allowing ‘demand’ to overwhelm practicality and rationality–as we have done for a century now. Street railways move people efficiently, and autos do not. The less auto travel the better for the environment, society and everybody.

  • Lucy

    The advantage of streetcars is not that they go faster than a car; it’s that you don’t need to deal with a car on either end of the trip. In urban environments, most of the time spent in a short distance trip is spent in getting the car out of one parking spot and putting it into another at the other end. A streetcar running in mixed traffic produces a shorter overall trip time because it eliminates car-homing on both ends.

    Intersections can be timed, designed, and controlled to manage all the worries expressed in this post. Light rail is wonderful for intermediate distances, but for really urban places, streetcars allow a park-once or even park-never mobility within the urban area. It’s a walk-extender, not a commuter car-replaement.

  • Lucy, those same advantages exist with the standard city bus – with the added advantage that a bus can manuever around an obstruction in traffic. See Jarrett Walker for more on the mobility myth in the mixed-traffic streetcar…

Leave a Reply




Comment preview below as you type. You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Prove you\'re not spam (required) Time limit is exhausted. Please reload CAPTCHA.

For help if you have trouble posting or your comment is marked as spam, please email:
info (at) | Comment Rules

The Site / The Fight

  • by Yonah Freemark
  • Twitter: @yfreemark
  • yfreemark (at) thetransportpolitic (dot) com
  • Le progrès ne vaut que s'il est partagé par tous.

Email newsletter


rss feed
comments feed
twitter feed