dissent of the week: trolley buses in wrong place vs motor buses in right one

DSC00153In a network redesign for Wellington, New Zealand that I led last year, we assumed that it was more important to design the network around customer needs than to maximize the use of trolley buses.  So we designed some routes that run partly under trolleywire and partly not, and that are therefore to be run (for now) by motorbuses.  A study of the future of trolley buses in Wellington is soon to begin, but as I said in my last post on the subject, they wanted their trolleybus policy to follow from the city's transport network, rather than preceding and constraining it.

One commenter said this was the wrong priority:

"This seems unwise to me. The benefits of trolleybuses are so great that it would often be better to run a trolleybus on a somewhat suboptimal route than a motorbus on a more optimal one. Thus any proposed routes or route changes should absolutely take into account the political and technical feasibility of operating the new routes with trolleybuses.

I am fairly familiar with the large trolleybus networks of San Francisco and Vancouver, and both have stretches of trolley wire that are no longer used much because the bus route underneath them had ceased to make sense as part of the larger network.  But I can also think of examples where buses still do pretty weird and possibly obsolete things so that they can stay on trolleywire.  Seattle's network, for example, is much more wire-constrained overall.  (A Seattle network designed without wire in mind, for example, would almost certainly make it easier to access Queen Anne Hill from the north.

Because trolleybuses boomed in the 1940s-60s but then stagnated, most trolleybus lines in North America (and New Zealand) have been unchanged for 50 years or more, while the city has grown and changed around them.  As a city's demand expands, and especially as it grows more multidirectional, routes need to either extend or be revised to fit a new structure that meets the new needs.  Trolley wire can become an obstacle to doing that.  

Vancouver's 41st Avenue the clearest example.  Today, almost all service at the west end of this busy transit corridor extends to the University of British Columbia, about 7 km beyond where the trolley wire ends.  UBC is now a superpower ridership source at almost all times of day.  Should trolleywire be extended the rest of the way?  Quite possibly, but right now, it would be absured to run 41st Avenue buses ending 7 km short of the university just because the wire ends there.

I certainly believe in the long term future of trolleybuses as part of both an emissions strategy and as protection against fossil fuel cost volatility, but sometimes the existing wire is just in the wrong place. Argue for more wire, yes, but meanwhile, should we just treat the trolleybuses as separate and inviolable, no matter how obsolete or problematic their routes become?  

42 Responses to dissent of the week: trolley buses in wrong place vs motor buses in right one

  1. EngineerScotty March 7, 2012 at 12:45 pm #

    Is there much salvage benefit to removing existing wiring, or does the cost of the work exceed the value of the recovered materials (wires, poles, substation equipment, etc.)?

  2. Jarrett at HumanTransit.org March 7, 2012 at 12:52 pm #

    Scotty.  I'm not an expert, but I understand there's a lot of salvage benefit.  Jarrett

  3. stino March 7, 2012 at 1:16 pm #

    I’m sure there are limitations to the technology, but can’t trolley uses be built with batteries built in to bridge in overhead wires? I doubt it’d work in every instance, though it seems like a decent fix, even if temporary.

  4. Jarrett at HumanTransit.org March 7, 2012 at 1:20 pm #

    stino … the situation is usually not a short gap in wire, but rather the need to incorporate quite long segments not under wire, such as the 7 km example below.

  5. Zoltán March 7, 2012 at 1:44 pm #

    It often seems a tempting idea, but I wonder if the cost of a) non-standard technology, and b) carrying around the weight of a very large battery might prove more expensive than more trolley wire.

  6. anonymouse March 7, 2012 at 2:03 pm #

    I don’t think trolleybus wire is as constraining as people are making it out to be. For example, SF just had no problem moving the southbound 30 and 45 from 4th Street to 5th Street, which required installing a bunch of new wire and supports on 5th Street, as well as a pile of new wire junctions at places like Market street, for a total of some 1.3 km of new wire. And this is just a temporary measure for subway construction. So given the motivation, it’s perfectly possible to add new wire, or reroute existing wire, and it’s not generally too absurdly expensive to do it.

  7. Zoltán March 7, 2012 at 2:04 pm #

    If you wish to see somewhere where previous generations left a well and truly ridiculous pattern of trolleybus service that the wires dictate must persist, you don’t get much better than Budapest. They are the thick, dark red lines here:
    There are sixteen routes over a small patch of the inner city, and none of them manage to run in vaguely straight lines, or flow to the logical end of what is a pretty logically gridded area of the city.
    To quote Jarrett ( https://www.humantransit.org/2010/05/seattle-the-end-of-trolleybuses.html )
    “Trolleybuses attract a particular kind of polarized NIMBYism: Replace them with regular buses, and neighbors scream becuase of the noise. Extend them, however, and different neighbors scream because you’re hanging wires on their street. ”
    The trolleybuses run in an old and pretty part of the city, and I suspect that if one proposed to do something about it, people would scream about diesel buses or about more trolley wire. While this prevents any changes, the best Budapest can do is small revisions to create a network of slightly skewed grid lines.
    Our grandfathers have often left us with networks that could have faced a major rethink if trolley wire weren’t an issue – what it leaves us with is a tradeoff between the problems of the status quo of transit routes, and the political difficulty of moving forward.
    It seems that in Wellington, they’re doing the latter. If the city ends up with the new trolley wire it needs to operate its new network, then it would support the position of not “treat(ing) the trolleybuses as separate and inviolable”, but designing the network as a whole, then selling the benefits of trolleybuses to that revised network.

  8. Zoltán March 7, 2012 at 2:40 pm #

    To speak specifically about Wellington, the consultants seem to have done a sound job of doing as they were asked, which is achieving improved mobility with existing resources, with a (very important) side order of legibility.
    If Wellington care about sustainability and wish to put their money where their mouth is, they can do one or a combination of two things:
    a) Invest some capital funds in hanging a small amount of new trolley wire.
    b) Maintain the spirit of the recommended network with some modifications to maintain more trolleybus routes, thus investing operational funds in more frequent service on some routes with trolley wire, and better diesel bus service to plug the gaps.
    Taking, for example, the C route operating along The Terrace (no wires) instead of the Golden Mile (wires) in order to stop a few hundred metres from Victoria University, either a) trolley wire could be hung along The Terrace, or b) the C could be routed along the Golden Mile and improvements made to diesel bus routes that go all the way up to Victoria University.

  9. EngineerScotty March 7, 2012 at 4:09 pm #

    For electric rail, at least, a big cost of electrification is reportedly substations. These don’t need to move if you only move a route by a short distance, but you don’t want to locate them too far from the actual route.
    Bus electrification has the advantage that the power demands are generally lower, as busses weigh considerably less than trains. OTOH, the fact you need dual-wire systems, and the closer integration of the wiring into the urban fabric, means that high-voltage systems (such as 12.5kV) commonly used with rail are generally not used with trolleybusses; meaning the I2R losses are higher, necessitating shorter distances between substations.

  10. Fried_brice March 7, 2012 at 4:19 pm #

    As late as 1992, the Los Angeles Rapid Transit District was toying with the idea of electrifying about a dozen bus routes, according to the Electric Railway Historical Association of Southern California [http://www.erha.org/rtdetb.html]. The buses would have utilized onboard batteries to allow for a limited amount of independent operation. The plans were cancelled due to NIMBY reactions against the needed overhead wires.

  11. Joel N. Weber II March 7, 2012 at 6:33 pm #

    How does the cost of the BYD eBus-12 described at http://www.byd.com/eBUS.html or the Proterra Ecoride BE35 decribed at http://www.proterra.com/index.php/products/productDetail/C22/ compare to the cost of trolleybuses?

  12. jfruh March 7, 2012 at 7:27 pm #

    The buses on Boston’s Silver Line are dual mode, aren’t they — running on trolley wires in the tunnel and then on CNG (I think) on the surface. Why not use buses like these? Are they super expensive?

  13. cph March 7, 2012 at 9:17 pm #

    Seattle has (had?) dual-mode buses, running on electric power in the downtown bus tunnel and diesel otherwise….The buses obviously carry two engines, so that adds cost and weight….

  14. Dexter Wong March 7, 2012 at 10:45 pm #

    I have heard that now Seattle Central Link is now running, most of the dual-mode buses have been converted to ordinary trolley buses.

  15. Christophe March 8, 2012 at 4:06 am #

    The city of Fribourg (Western Switzerland) used dual-mode trolleybuses before electrifying entirely 2 lines. We had also dual-mode trolleybuses in Lausanne, it was quite convenient to operate, using either electric or diesel mode in case of service disruption or important works on streets.

  16. Alex B. March 8, 2012 at 6:46 am #

    Without knowing the geography of the specific routes, I would also imagine that it’s not as simple as a conventional bus in the ‘right’ places and a trolley bus in the ‘wrong’ place, but rather a spectrum of potential locations on the right-to-wrong scale. E.g. the trolley bus location might not be ideal, but perhaps it’s not so bad as to be the ‘wrong’ location.

  17. Jonathan Hammond March 8, 2012 at 7:52 am #

    I think there are other reasons for the utility of trolley buses, not least for fossil fuel or emissions reduction, but also, pointedly, traction. Wellington, Seattle and San Francisco all have very difficult terrain as far as cities go, and dense grids that disregard that terrain at least somewhat.

  18. Wayne March 8, 2012 at 10:24 am #

    I use the 41 in Vancouver all the time. Early on it bothered me a lot that the buses weren’t trollies and I couldn’t make sense of it.
    It dawned on me eventually that several trolley routes going in and out of service, i.e. 3,8,20 wouldn’t be able to use 41st if the 41s were also trollies.
    I suppose this can be remedied in different ways but for now it’s an issue. I’d be all in favour of extending the service to UBC along S.W. Marine Dr.

  19. David M March 8, 2012 at 11:01 am #

    While the 41st Av trolley wires are not used for regular service anymore, they still serve a useful function. Many buses going into or out of service at the west end of the network will use 41st avenue wires.
    Extending to UBC would be expensive. The wires would be on a divided roadway with an 80km/h speedlimit. And there are no stops – it’s largely forest. but, given the ridership, might be a worthwhile expense.

  20. Jarrett at HumanTransit.org March 8, 2012 at 11:03 am #

    Alex.  Yes, apart from death and pregnancy, almost everything is a spectrum.

  21. Wesley Kirkman March 8, 2012 at 11:20 am #

    It seems to me like you are getting into the buses are flexible argument, an argument I’ve always abhorred. The idea that demand moves around a city and changes — popular here one day, not so much the next — is just silly. Sure, new demand arises in outlying, previously undeveloped or lightly developed areas. But existing activity centers and the most efficient routes between them rarely move, if ever. So if that’s the case, why do you need flexibility? Put the infrastructure in place (in a logical and efficient manner from the start) and leave it.
    Trolley buses seems like the easiest win (ignoring all the “but wires are ugly” NIMBYs) to be had in public transit. You get a fairly efficient and reliable system for fairly cheap and you don’t have to rip up the roadway to do so. Plus, the quality of life for residents is vastly improved by removing the terrible toxins of diesel fumes and extremely loud irritation of their engines. Win.

  22. Wesley Kirkman March 8, 2012 at 11:21 am #

    To think, the transit agency here in Seattle was actually considering getting rid of our trolley buses. That might have been the straw that sent me packing back home to San Francisco.

  23. Jarrett at HumanTransit.org March 8, 2012 at 11:28 am #

    You misunderstand me.  I've made your point many times.
    Changing routes according to really ephemeral demands is indeed foolish.  But trolley wire tends to prevent routes from changing at all, for decades, and on that timescale, cities and their demand patterns do change.  The growth and change of cities often produces predictable crises in network structure, where an older design (e.g. all-radial to downtown) ceases to be relevant to the city as it is, requiring the kind of redesign we did in Wellington, and that I've done many times.  

  24. Eric Doherty March 8, 2012 at 2:49 pm #

    One thing that should be considered here is that trolley buses are an intermediate capital cost technology between diesel buses and light rail. The cost problem of extending trolley lines translates to a much bigger problem for changing or extending a light rail line.
    Extending or changing trolley lines is also subject to economies of scale, like almost everything else. A 2 km extension will likely cost a lot more per unit than 80 km, especially if you do not have the capacity to do the 2 km work in house.
    But set aside diesel bus and trolley bus comparisons. Compare $1 billion in trolley bus rapid transit side by side with light metro (e.g. SkyTrain) and light rail rapid transit. Then think about how much each would do for the network effect and transit ridership in your region, and the impact on human health along the corridors. . . .
    (PS. I am thinking about the Surrey Langley rapid transit proposals here in Metro Vancouver Canada as an example. http://www.translink.ca/en/Be-Part-of-the-Plan/Rapid-Transit-Projects/Surrey-Rapid-Transit-Study.aspx Strangely, the Surrey/Langley rapid transit documents do not include a trolley bus option but the Broadway process in the City of Vancouver does.)

  25. Eric Doherty March 8, 2012 at 4:03 pm #

    Correction: Seems that trolley BRT is in the leaflet for Surrey/Langley rapid transit, but is not mentioned as a possibility in the survey I filled out on line or shown in the video.

  26. Aleks Bromfield March 8, 2012 at 5:02 pm #

    Re dual mode buses: Changing from trolley to battery is easy, but many systems are designed such that changing from battery to trolley requires some sort of manual intervention. This could clearly be made easier, but for whatever reason, this doesn’t seem to have been a priority for most trolley systems.
    Personally, I find this bewildering. Here in Seattle, where we have cheap hydroelectric power in abundance, wouldn’t we get a huge savings if every local/rapid bus route could use electric power through downtown? If that could be automated, e.g. by having a “attach to trolley wires” button inside the bus, wouldn’t the cost savings and environmental benefits be worth the brief stop?
    It’s particularly agonizing for routes like the 8, 11, and 48, which run almost entirely underneath trolley wire, and yet use diesel buses exclusively because of the non-wired segments. Even more so when you consider that the 8 and 11 routes have very hilly segments without wire. Going up, electric motors would be a huge boon; on the way down, regenerative braking could actually *produce* energy.

  27. Aleks Bromfield March 8, 2012 at 5:14 pm #

    Re flexibility: I think Jarrett’s point is well-taken. Here in Seattle, people have a habit of arguing out of both sides of their mouth with respect to bus route flexibility. One person will say that Link (our light rail system) was a waste because it can’t flexibly respond to changing demand… and they will turn around and argue against bus routing changes to meet Link, because they like their bus the way it is, thank you very much.
    It’s not wise to change routing (or even frequency) every month to meet the slightest changes in demand, any more than it’s worth it to change the bus fare every hour based on how many people are boarding. But many of our routes haven’t changed since the streetcar era.
    For example, consider the 12. From downtown, it runs along Madison, a major commercial corridor. Then it turns onto 19th Ave, which was a major corridor back in the early 20th century, but is now a quiet minor arterial with some multifamily buildings and small businesses, but not much else.
    By all accounts, the 12 would do better to stay on Madison until 23rd (a major transfer point). The 19th Ave service sees very little use, and there is more frequent parallel service 4 blocks to the east and 4 blocks to the west. But because it’s a trolleybus — and because there are no wires on Madison between 19th and 23rd — Metro has been very reluctant to change it at all.
    To cite another example, the 3 goes between downtown and the major public hospital/trauma center. It runs every 7-8 minutes during peak, making it one of the most frequent routes in all of Seattle. And yet the route it takes between downtown and the hospital is along a major I-5 entrance ramp. Thus, the bus often gets caught in highway traffic. This is an issue that simply wasn’t relevant when the 3 was first created. But now, the time it spends sitting in highway traffic costs Metro a lot of money, and causes its users a lot of aggravation.
    As Jarrett says, after 100 years, activity patterns really do change. Neighborhoods grow and contract; highways get built; new industries develop. And system planning has also advanced since then; we understand more about what it means to build efficient networks. Should we reject all change at all just because someone might like things the way they are?

  28. Keep Houston Houston March 8, 2012 at 7:51 pm #

    I don’t think you can be so sure that a lack of trolley wire on Queen Anne would have resulted in more through service.
    For instance, *why doesn’t Portland’s 51 bus go to Beaverton*? It’s been dieselized since the 50s.
    Yet the thing still splits into two useless, infrequent tails, just like the Council Crest streetcar did 80 years ago.

  29. Mikko March 9, 2012 at 12:32 am #

    A decision (possibly an implicit one) not to develop an electric system can be problematic for streetcars and light rail, too. The expansion of the Helsinki tram system essentially stopped for several decades in the 1950s, when the city decided to build a heavy rail metro system. The launch of the metro was delayed for 10+ years (opened 1982) and the construction turned out to be so expensive that the metro lines never reached many parts of the city they were supposed to. Meanwhile, the tram system was not expanded in anticipation of the metro, so many of the growing parts of the city ended up being served by buses.
    This lead to a situation where many bus routes overlap with tram lines to this day. Buses get in the way of the trams on major streets and both systems work suboptimally. It’s not that the lines are in the wrong place, but that the tram lines should have been extended by a few kilometers and operated with larger units and with better separation from other traffic, in which case the buses wouldn’t be needed at all on that route. This would have speeded up tram operations and improved service in general. The tram system extensions never happened, because “the metro was coming” and trams in general were considered an obsolote technology.
    Helsinki is now starting to rationalize the bus lines with the general concept of building frequently running trunk lines and minimizing the number of buses in the area served by the trams. There are several new residential districts being built near the center in areas freed from harbour use, and the tram system is being extended to those. An orbital BRT (of sorts) line is overcrowded and will likely be converted to light rail in the coming years. All of this has made it more likely that some of the tram system extensions that should have been built in the 60s will finally get done.

  30. dejv March 9, 2012 at 2:14 am #

    As a city’s demand expands, and especially as it grows more multidirectional, routes need to either extend or be revised to fit a new structure that meets the new needs. Trolley wire can become an obstacle to doing that.

    Brno, Czech Republic is a good example of what can be done with that – in early 1990s they realized that one such obsolete stretches actually followed half of one of city rings perpendicular to some 5 high-traffic radial routes and close to several university venues. Few years later, they accomplished to fil the gaps and created tangential lines 145/146 (25 and 26 since early 2000s) that were a large success and whose ridership has grown ever since.

  31. Zoltán March 9, 2012 at 3:00 pm #

    I have written a lot about the duplication in Helsinki and what could be done about it in two blog posts.
    It seems that often buses can perform a role of rapid transit more often, integrated with the Metro, where trams are providing local stops. The same applies to the Mannerheimintie, where there are both tram and bus lanes, particularly if bus lanes were to be moved to the middle of the road.
    I am interested to learn more about the plan to rationalise the bus lines – is there somewhere you can link me to with more information?

  32. Mikko March 9, 2012 at 11:52 pm #

    The study for the long-term transition to trunk bus lines is here:
    There’s a short summary in Swedish and English, otherwise it’s all Finnish. The recommendations in that don’t really cover the center area in detail as the study concentrates on the trunk radial and orbital lines (Jokeris 0, 2 and 3 in addition to the existing Jokeri 1). The particularly enlightened thing about that study is that it’s not hung up on modes, it talks about forming trunk lines and leaving the decision whether to make them bus or light rail for later.
    The plan for reorganizing the bus lines in central Helsinki this year is here:
    That one has been approved by the city with minor modifications, which are mentioned at the end of the summary. Some of the changes are to line 18 to Munkkivuori, which is one of the places that truly ought to be covered by a tram line extension. The plans for the extension exist in a preliminary form, but the planning machinery for new tram lines is currently clogged with other things (Jätkäsaari, Hernesaari, Kalasatama etc.).
    I’ve seen another paper concerning central/southern Helsinki that was more about long-term principles rather than this year’s changes to the lines, but I can’t find that one right now.

  33. Andrew March 10, 2012 at 10:25 pm #

    Lyon, France has been installing new trolleybus wire in places, adding to or extending its trolleybus network. Trolley wire could easily be extended to UBC in Vancouver for example at minimal cost.

  34. Max Wyss March 11, 2012 at 12:10 pm #

    @Aleks Bromfield: There are trolley poles on the market which have actuators to loer and raise. As stated, lowering is no big deal, and for the raising, it is possible, but not just anywhere. So, for a fixed “raising place”, such poles can be used. This system uses inverse-v-shaped guide boards, about 1.5 m long, which help guiding the shoes onto the wires. This also menas that the bus (driver) must stop pretty exactly at a specific place.
    These poles were in use in Zürich when a major square got completely rebuilt, and the trolley buses had bot operate a few hundred meters on auxiiary power (not batteries, but a (about) 50 kW diesel power unit).
    About auxiliary power: keep in mind that modern trolleybuses have a traction power rating of around 300 kW; now you can calculate the size and weight and price of a battery pack with sufficient capacity… and don’t forget the non-traction power requirements.

  35. Max Wyss March 11, 2012 at 12:21 pm #

    There are some criteria for trolley buses: steep grades and high capacity.
    Because of the electric drive, a trolley bus has maximum torque at lower speeds, and it can draw more power than rated for a short time. This allows to tackle steeper grades at reasonable speeds (where a diesel bus just runs out of “steam”. Modern trolleaubses also allow for regenerative braking, meaning that the energy cost are lowered again.
    High capacity is partly connected to the higher power installable into a trolley bus; vehicles like the SwissTrolley 3 are articulated and have two motors, rated at 150 kW or more driving the middle and rear axle. But it is also possible to build a double-articulated bus, about 25 m long. The same power train as the single articulated is still powerful enough to drive double-articulated beasts.
    The question whether to change electrification from one street to another depends, as someone else has stated, on the distances from the substations. The actual wiring costs are comparable to the lighting of a street from lamps hanging on wires between poles or buildings.

  36. Eric Doherty March 13, 2012 at 1:09 pm #

    Max wrote: “The same power train as the single articulated is still powerful enough to drive double-articulated beasts.”
    And when it snows? The ability to easily fit double drive units for snow performance is one of the key advantages of trolley buses over diesel. (If snow is an issue in the locale).

  37. Max Wyss March 14, 2012 at 11:18 am #

    @Eric Doherty: I was actually referring to the two-axle drive trains of the single articulated Swisstrolley 3, so, we already have that built in.
    Not very often, but it does snow in Zürich, and it may be bad enough that some bus lines are shut down (a notorious case has a 10% grade (or so), but its conversion to trolleybus is still in study phase. Problems occur more likely from automobilists blocking the streets because of inadequate tires; but normally, the VBZ trolley bus lines run on time, even with snow.

  38. Anon256 March 14, 2012 at 7:43 pm #

    My point was not that trolleybus routes should never be changed. Rather, it’s that politically it will often make sense to tie proposed route changes to the installation of new trolley wire, rather than trying to argue for each change separately.
    Having had a closer look at the proposed changes, they aren’t quite as bad as I thought when I wrote the quoted dissent. But I still think you should consider temporarily having route C follow the old route 11 until wires can be strung along the new route. (Ideally this should happen immediately, but if one change can wait then so can the other.)

  39. Nathanael March 31, 2012 at 3:42 am #

    What Anon256 said. You design a good system, but then you propose the *new trolley wire location*. And until that can be built, run on the old route.
    Never, ever dieselize. Not in this day and age with global warming and peak oil. Just don’t.
    “We were asked not to discuss that question” is a consultant’s excuse for a problem caused by politicians who actually want to get rid of trolleybuses. It’s a fair excuse if you have been ordered not to discuss that, but otherwise, you do have to talk about appropriate power sources. It’s not responsible to pretend that you can discuss a route network without thinking about what vehicles will operate the routes. Though again, if politicians are paying you not to think about that, then you can’t be faulted for taking the money.

  40. Nathanael April 18, 2012 at 9:13 am #

    P.S. Have you read these detailed criticisms of your Wellington review? Basically, they say you had bad data and misinterpreted the data you had….

  41. Nathanael April 18, 2012 at 9:15 am #

    Of course, he also lauds some of your advice — *and then says that Wellington Council is ignoring it*.

  42. Joel N. Weber II June 16, 2012 at 12:36 pm #

    New Flyer has a prototype battery powered transit bus that can go for four hours between charges on a 120 kWH battery pack (except during especially cold weather, when the battery can’t cover the full heating load), which may be in production in two years and is expected to be economically attractive.
    (It is not clear why they feel a need to stop at 120 kWH and use biodiesel for heat in the winter; the Tesla Model S is available with an 85 kWH battery pack for well under $100k, and so I don’t expect either weight or cost is an absolutely compelling reason a bus can’t have a battery pack bigger than 120 kWH.)
    http://www.winnipegfreepress.com/breakingnews/new-flyer-green-leader-156630045.html has details.