Leapfrogging
Eric Stoothoff is the chief engineer of the MBTA. Last month, he offered the following excuse for why the MBTA just deelectrified the trolleybuses in Cambridge, replacing them with diesel buses and hoping in a few years to obtain battery-electric buses (BEBs):
We want to leapfrog Europe, not play catch-up. If BEBs are the future, why not have the future now?
https://twitter.com/mbtaroc/status/1493768313154904073
Unfortunately for Stoothoff, BEB technology still does not work in freezing temperatures. The current state of it is buses that have diesel heaters – otherwise the battery drains too fast in winter, as it did three years ago when I reported it for CityLab.
The actual cutting edge of electric bus technology is in-motion charging (IMC), in which the bus spends part of the route under wire and then part under battery, with an off-wire range of about 10 km. IMC is especially valuable for Boston, which is unusual for an American city in having an unplanned street network in which the same trunk road splits into several farther out, and then the trunk can be wired. Cambridge’s now-defunct trolleybus network had a short trunk, but could still be an attractive IMC target. In Boston proper, Washington Street is a valuable trunk for wire, with routes splitting off-wire to destinations in Dorchester and Mattapan farther south.
Stoothoff seems unaware of this, because he is an insular, ignorant, incurious manager. He uses leapfrogging as an excuse not to learn. Other American agencies buy BEBs, and then find that they don’t work in winter without diesel heaters, and instead of seeing what Europe does, he talks of leapfrogging.
Leapfrogging means something completely different. It means skipping an intermediate tech that has been obsoleted by newer tech. A classic example of leapfrogging is China’s phone network: by the time China developed enough for mass use of phones, in the 2000s, cellular phones were ubiquitous and mature enough that China skipped wired phones entirely, and did not have to spend money on building phone cable infrastructure in rural areas. More recently, mobile payments are connecting rural areas in Africa between the Sahara and the Kalahari to banking without the need for physical branches.
On the level of infrastructure, it makes sense: there is no need to invest in intermediate technology if something better is available. In the realm of rail, there are a lot of technological dead-ends that nobody needs to develop anymore – superseded electrification standards, experimental jet- or nuclear-powered trains, obsolete track geometry standards, etc. Train stations today are designed differently from in the steam era: the train is not noxious to be nearby, so the train shed is integrated into the passenger concourse, and train turn times are short, permitting much smaller station footprints even in major cities.
But on the level of knowledge, it’s daft. Leapfrogging requires knowing what the cutting edge is. Chinese development experts know exactly what technology is used in developed countries and what they should imitate and what they can bypass. The PLA began its modernization process in 1991 after Desert Storm and only began innovating rather than implementing NATO standards a few years ago. African development experts are generally aware of trends in rich countries as well.
This knowledge is especially important in public transportation, because many legacy cities had higher ridership before WW2 than they do today and there’s a lot of nostalgia for that era. Understanding why the modern train station can be compact and platform-centric, without a waiting concourse and space for a telegraph operator and baggage handlers, is crucial in limiting the construction costs of stations on new lines. Without such understanding, it’s easy to imitate historic stations; even in Europe, where trains are integrated into train sheds without the separate waiting halls characteristic of North America, most major-city stations are historic and very big, because they’re inherited from when they needed to be and the land was at the edge of the city and therefore cheaper.
But what one does not do is tear up legacy infrastructure that is still useful. Europe’s great train terminals are almost all oversize, but there’s no point in blowing them up and shrinking them just because it’s more modern. Urban renewal projects at train stations are common, but they replace goods yards that left the cities alongside industry, not passenger circulation. And at least shrinking station footprints has redevelopment value in major city centers; deelectrifying trolleybuses has no such value.
So under no circumstances should cities with existing trolleys remove the tail electrification for IMC. This is not what IMC-using cities do – they use IMC to expand the network rather than shrink it. It may be too late for Boston, but San Francisco, Seattle, and Vancouver should keep what they have.
And it’s even worse, because Stoothoff wasn’t justifying deelectrifying on the way to the future. No: he misstated what the future is. His incuriosity is such that he assumes BEBs are the future, from a position of interacting with American agencies that think the same and find fixed wire infrastructure too hard. Peripheries that engage in leapfrogging are voracious consumer of the metropole’s learning in order to apply it to their own circumstances, but Stoothoff cannot even bring himself to admit that the United States is a periphery and needs to absorb this knowledge.
A better MBTA is one in which Stoothoff is replaced with a more competent chief engineer, perhaps hired from abroad. But it’s not just him. He’s a removable obstacle to progress, but there are many like him – many managers who assume the future is one thing when it’s the other, and use their wrong beliefs to justify not imitating best practices. They have an assortment of excuses, and misstating what technological leapfrogging is is among them.
Pedestrian Observations 
Do you think battery electric busses will become viable technology in the future? If not, then what will be the future of lower-frequency bus routes (say every 15 minutes) in a zero carbon world? Will they all get trolly wires? Also, what would you think about battery electric busses with a hydrogen based heating system (with hydrogen generated at bus depots from renewable energy)? Hydrogen may be inefficient, but using it for heat would seem easier than using it for motive power.
Honestly; the future is probably an in-motion charging setup; done that way you can electrify chunks of road that see a lot of service while leaving the stuff on the periphery as is. For Boston that would probably look like electrifying the Washington St/Warren St/Blue Hill Av corridor as a starting point, and then expanding to cover other spots that see at least ~5-6 buses an hour. Hydrogen-based heating systems could work, but the tech isn’t there, and hydrogen is both profoundly flammable and has a nasty tendency to leak through solid materials and embrittle metal (the hydrogen atom is so small that it can slip through gaps in the molecular structure of solid materials; the hydrogen that stays in the structure makes it brittle while the hydrogen that leaks through is lost). Honestly your better bet would be improved battery chemistries (a better energy-to-weight ratio plus in-motion charging infrastructure would get you lighter, more energy efficient vehicles.
BEBs are kind of OK for low-capacity, low-frequency routes.
OTOH, forget about hydrogen in any form for urban buses. The equipment needed is just adding more weight and little use.
There are barely any low capacity low frequency routes in Hong Kong, and even for those that are, the buses would usually get assigned to other routes in-between departures to boost the vehicle utilization. Most buses are also double deckers with a capacity of 100+, and those that still use single deckers are often due to physical limits of the terrain of the road they’re running.
That’s why a bus company with 4000+ buses in Hong Kong used government subsidy to acquire 10 single decker battery electric buses with the prospect of testing them in urban environment, yet after the buses delivered they just sat idly in depot as the bus company and operator failed to find appropriate urban routes that can accept these battery double deckers without disrupting operations, and then another company is now venturing toward hydrogen buses hoping it can solve the range problem.
As for extra weight, it also applies to battery buses.
Doubledeckers are the worst, weightwise, even if they have a twin axle device.
Hydrogen… the only environmentally acceptable hydrogen is the “green” hydrogen, produced with electrolysis fed from wind or photovoltaic power plants. Anything else is greenwashing, and proponents are spewing propaganda of the petrochemical industry. (a small exception can be if there is a (petrochemical) plant nearby where hydrogen is a by-product, and currently just burned off). So, they really have to look at trolleybus infrastructure (they would not be the first to re-open a trolleybus network; Prague is doing it right now, Berlin (Spandau) has extensive plans going through the planning procedure). And maybe they better get rid of doubledeckers and get articulateds (single or bi) instead. (note that the rearmost axle of a bi-articulated bus is steered, so that it follows the track of the third axle, making it behave the same way as a single-articulated one.)
Articulated buses are longer, take up more road space, and would only worsen the traffic condition which have already been quite congested along some corridor and making bus services unreliable.
And double decker buses can provide 100+ seats with 40+ standing space for a total of 150 passengers per bus, such figure seems difficult to attain on single deck vehicle even if they are articulated.
Trading number of seats for more standing space is undesirable either, since many people ride the bus over 1 or 2 hours each way each days, forcing passengers to stand this long would be unwelcoming.
And in fact, 20 years ago, a local bus company have put their own money into developing their own trolleybuses hoping to use it on the street, but government response have been far from positive, citing troubles of setting up catenary in dense city as well as overlaps in role with tram network that parallel most of their bus routes, that the vehicle ultimately just rotted in bus depot.
Some of the double deckers in Hong Kong already have rear axle steering, but in real world testing that doesn’t always help them clear street corners that a shorter regular double decker can do.
The fundamental problem with doubledeckers is that, because of the many seats, upstairs, and the little door space, they spend way too much time at stops for passenger exchange. This is not an issue with touristy lines, or strict point-to-point lines, but for urban lines, it is sheer poison. So, what an 18 m articulated bus may use more on (static) roadspace, a doubledecker more than uses up the road space dynamically, at stops, blocking all traffic. And if all-door boarding is not possible, it gets worse. Consider a doubledecker spending 2 minutes more at each stop, it explains the long travel time very well.
So, it is balancing between efficiency (and operation cost for a service), and “passenger experience”.
@Max Wyss
Even at the largest bus hub in the city, an empty double decker bus loading 100+ people onto the bus using only a single door, would still take less than 2 minutes. I have no idea how you came up with the figure.
Not to mention, roughly 80%+ urban bus routes in Hong Kong are loosely speaking point to point. They still have multiple stops, but most of them usually stop at like ~10-20 stops near start point, then run some kilometers of highways, then they would drop off the passengers at the ~10-20 bus stops in the destination district and reach terminal.
As for the few bus routes that indeed have linear passenger flow throughout the trip, I don’t recall them being a problem either, since origin and destinations are spread throughout the route, at most there are two or three dozens passengers boarding at each stops, and 30 seconds would be more than enough to let all these passengers board and deboard even at more popular stops.
BEBs work fine for high-frequency, normal-bus-route-capacity (i.e. doesn’t need a tram), SHORT, FREQUENT STOP routes.
You end up needing slightly more buses to provide a day’s service, and possibly more depot space, because after a while the bus has to go out of service at the depot and recharge. But if you’re doing a 3 mile loop, over and over, with frequent stops, it’s almost optimal for a BEB.
Currently, BEBs are no good when very long runs are needed.
Alon’s incorrect when claiming that they all need diesel heaters. This is nonsense. They work fine without diesel heaters if the route is SHORT ENOUGH so that there’s time to take the bus to the depot, recharge, and deploy the next bus out of the depot. The problem, as with all battery-operated vehicles, is range.
It is, of course, complete madness to remove an electrified trolleybus route. Viable frequent BEB routes look like Denver’s “Mall Ride”, constant start-stop on an extremely short route.
The laws of chemistry limit how good a battery can get. I don’t know enough chemistry, but surely someone does. What we don’t know is how close to the limits it is practical to get.
In any case the best possible batteries are very heavy for the amount of power you get. Either we need to plan on in motion charging or we need liquid fuels. Wishful thinking about leapfrogging won’t get around the laws of the universe.
Suburban areas have enough density for a bus every 10 minutes, but no transit system in America gives them good enough get where you want to go service to let anyone consider the bus for all but a minority of trips.
The Cambridge trolleybus system in Boston is not shut down until the 12th. It’s probably too late to make the MBTA change course, but the earlier they know they made a mistake, the better.
Was there some reason why it historically made sense to have waiting concourses in train stations, or was it always a mistake that just hadn’t yet been noticed?
It makes sense for intercity travel where passengers arrive in advance (or have long connections). You keep passengers in a central area, and then have them make their way to the platform a few minutes before the train arrives. Many HSR systems and non-US airports operate this way.
Why do passengers arrive in advance? Why do many airports still operate in the central concourse fashion?
Poor reliability, punctuality, and frequency. Waiting concourses are required because people pad their travel time to account for the poor reliability, punctuality, and frequency of their transportation to the station, and to account for the poor frequency of the intercity train. Therefore, people get to the station area well before their train, just in case there was some hiccup along the way, because if they miss the train they wanted, it’s a long wait or sometimes impossible to get on the next one.
In addition, the poor reliability and punctuality of the intercity train means the platform the intercity train will stop at isn’t known far in advance. The approach tracks are designed to be very flexible with train placement, and tons of extra platforms are built, to account for the unpredictability of trains.
In the case of airports: Security is an unreliable unpunctual process people have to account for, so they show up early, to avoid missing their plane since it’s very painful to do so. Airports in the modern day still deal with poor reliability and punctuality of planes, which is why even when an airport has you wait at the gate, the gate sometimes just changes on you.
Why would passengers wait in a concourse in the station instead of hanging around the station area?
Stations were traditionally at the edges of cities. There was presumably less to do back in the day.
Most airports are still far from city centers. In addition, you can’t leave the secure area, otherwise you have to go through the unreliable unpunctual security process again, so any waiting has to be done there, even if the airport is city center.
Nowadays, in the best case:
1. Stations are practically city center, since the city grew around them, and sometimes the city center even shifted to be adjacent to them.
2. Local transit is reliable, punctual, and frequent. You don’t have to arrive long in advance. Plan to show up right on time.
3. Intercity transit is frequent and never sells out. Even if you miss the train, it’s not the end of the world. Frequency is freedom.
Therefore, in the best case, waiting concourses can be minimized in modern train stations, even for intercity trains.
Ist viable to setup in-motion charging on highway/expressway, which account for most parts of journey of many bus routes in HK?
I’d say yes. I have seen videos of a BRT line in Beijing, just doing that.
Chinese Wiki article on Beijing BRT have no information on such sort of line.
Ordinary trolley buses are speed-restricted to about 60 km/h by the overhead wire, as trolley poles and trolley catenary for higher speeds is more complicated and expensive and such speeds are usually not needed for urban bus systems.
Germany, Sweden and the Los Angeles area have some pilot projects running with paired railway-style catenary atop the rightmost lane of a stretch of freeway; that should be usable for freeway speeds (about 100 km/h for trucks). The paired pantographs are automatically raised by the on-board computer when the truck enters the lane, and automatically lowered when the it veers off the centre of the lane or leaves it.
The buzzword used is “eHighway”, and a common search engine gives some english-language results if forced, like this one: eHighway – Solutions for electrified road freight transport (Siemens).
The current pilots look like they could be easily replaced by a short run-of-the-mill railway line with off-the-shelf rolling stock; so this looks like a solution in search of a problem. You might have found the problem. 🙂
Hummm… Such system seems easy to be rejected by Hong Kong government as those electric poles could become easy vandalization target, and the government have been in deep worries since year 2019 that street vandalization could disrupt the city’s normal operation and be used for “Western force” to hurt the city’s economy and consequentially damage the rise of China.
But on the other hand, this system seems similar to what JR Freight proposed in Japan 10 years ago. At the time, Shintomei expressway connecting Tokyo to Nagoya was still 4-lane both way combined with reserved median space, and they proposed using the median to lay dedicated tracks to carry trucks between Tokyo and Nagoya as well as Osaka with electrification. Unfortunately the project didn’t work out as they ultimately decided to expand the expressway to 6-lanes.
I guess the system is useful in places where there are too many roads and not enough tracks, as it’s cheaper to install catenary on existing road, than to convert roads into track before installing catenary.
Trucks with a few hundred liters of diesel in the tanks are much less of threat.
A truck terror incident would at most kill a few dozen people, which I am sure the government treat it as a more bearable consequence than having the roads blocked up and disrupting the society normal functioning for a few hours.
Stuffing wire into a glass bottle makes a truly ineffective Molotov cocktail. And trucks, diesel ones can drive over wires. Not so much through a 100 liters of burning fuel.
So te HK goverment is going to ban all trucks because someone might tip one over across the whole freeway to block it?
I don’t what the Hong Kong government is up to. Most places, where there is electricity, are fairly lousy with overhead wires. And fairly lousy with vehicles hauling around tanks of highly flammable fuel. I don’t see what the problem is with putting a few more wires over a lane or two of a highway is.
The level of incompetence of that Stoothoff dude is frightening. In fact, he should go back to his alma mater and return his engineering degree.
A statement of his equivalent in the Swiss City of St. Gallen: “Single articulated BEBs (with overnight charging) are simply not able to do a day’s work on the charge. But the BTBs with IMC have no problem at all.”
The core network of St. Gallen runs single articulated trolley or diesel buses, as well as double articulated trolleybuses every 15 minutes during the day. They are right now installing about 6 km worth of overhead wiring to decarbonise the diese bus lines, and use BTBs with IMC. These 6 km allow to extend the electric network by 30 km. (and there is potential for a couple of additional lines going beyond the city limits, and which are operated by another operator, but they did run the new vehicles in productive operation on those lines). Note that St. Gallen is at 800 m altitude, and has therefore not the warmest climate…
“but da fuuta” you may say. Fact is that the prices for transit-grade batteries do come down quite a bit. But what does not come down much is their specific weight (xx Wh/kg, which is around 200 Wh/kg nowadays); experts expect reaching 400 Wh/kg in maybe two decades. Now, weight is the limiting factor for the passenger capacity of a bus. The maximum legal weight of a single articulated bus is around 30 t, and of a bi-articulated one 40 t. Standard calculation is 12.5 passengers per t. FWIW, heating is a combination of infrared and heatpump air condition system.
So, a few numbers, based on the newest SwissTrolley 5 by Hess. Its deadweight is 18.8 t, and its maximum legal weight is 30 t. That means it can load 11.2 t of passengers, which translates to 36 seats, 129 standees, and a driver (this would be sardine can mode, but that’s about the limit. The traction battery has 66 kWh capacity, which allows for about 10 km unrestricted (all consumers active) off-wire range (and about 30 km restricted). A day’s work is around 300 km. A BEB of the same form factor has a unrestricted consumption of about 4 kW/km. Therefore needed energy 1200 kWh… This translates at 200 Wh/kg to a battery weight of 6 t. And that means it can only carry 6 t of passengers; instead of 166, it can only transport 80. Therefore, for the same line capacity, you will need twice as many BEBs…
TWICE AS MANY buses!! … That’s the “future” that Stoothoff dude wants to achieve.
I was wondering whether the battery improvement curve would solve the problems with BEBs – thanks for the clear explanation.
deelectrified the trolleybuses in Cambridge
Must be nice to live in a city where the air quality is so good adding diesel exhaust into it doesn’t make the air quality lower. There was a complicated environmental study to go with this decision wasn’t there?
SCR tech does reduce emissions to less than gasoline vehicle, manufactures claim. But, why bother with expensive to maintain particulate filters and DEF system when larger power plant can simply burn NG?
But NG is still carbon-based. So, nothing with decarbonisation…
NG power plants produce less CO2 than ICE engines, and are easier to replace with renewables in the future.
Not quite nothing, but not much (NG gets more energy per carbon emission because it has more hydrogen per carbon than diesel).
More because power plants are more efficient than ICEs than the intrinsic advantage of NG over diesel, but you’re still talking 10-20%, I believe.
NG isn’t as good as solar or wind, but better than diesel, gasoline, or coal. Although diesel cars obtain 25 to 35 percent better mileage and emit less carbon dioxide than similar gasoline cars, they can emit 25 to 400 times more mass of particulate black carbon and associated organic matter (“soot”) per kilometer. However, SCR eliminates nitrous oxide by conversion to nitrogen in exhaust manifold chemical reaction using aspirated urea based DEF fluid, and then exhaust stack includes a ceramic type filter that traps particulates. When computer sensor detects excessive exhaust back pressure, accumulated heat from engine is used to ignite and burn particulates into ash. Gasoline engine is less efficient (few BTUs) and frictional engine wear greater because of solvent properties vs diesel oily properties, causing more frequent oil changes and reduced engine life than Diesel engine, but catalytic converters help a little. Catalytic converters change harmful substances in a car’s exhaust gasses, such as carbon monoxide, nitric oxide, nitrogen dioxide and hydrocarbons, into less harmful substances like carbon dioxide and water vapour by means of chemical reactions. But, these processes in either engine is far from perfect, and CO2 is still emitted.
I don’t know. Combined cycle gas is much more efficient too. And the pollution is less… worrisome. It’s not at the curb either.
Massachusetts gets electricity from many sources. Nuclear, hydro and PV. There’s a tiny amount of wind too.
Barcelona has tram articulated busses powered by battery that operate on a fixed loop, stopping at an overhead charge station at end of loop. The only advantage I can see with this system over the city’s surface rail tram is that somewhat dangerous tracks are removed. Bicycles, wheel chairs, and pedestrians not paying attention can get injured crossing over rails. But, the energy efficiency superiority of hard rails over rubber tires on street pavement is pretty obvious in faster acceleration. Even busses with overhead wire power feed like in San Francisco don’t accelerate so quickly as hard rail electrical tram.
No one ever mentions how terrible dude quality is on a bus when compared to a tram. As such, all the getting equal, we get stuck with bouncy buses which make the potholes ever so bigger.
Busses have to contend with traffic and street lights, so it always seems that they are behind schedule. Plus, waiting for bus on city street curb is generally less pleasant than on rail station platform.
Yes they do.
Over and over and over and over and over and over and over for decade upon decade they do.
And a majority of them are at pains to point out that nobody has ever articulated the deep observation “bad bad bus, choo choo good” ever before.
Maybe the rail groove issue is outweighed by the fact that trains and their rails are easier to spot than buses, so collisions with vehicle are less likely?
There are special plastic plates used to level out surface to reduce risk where cyclists cross. Cyclists falling down shouldn’t be dismissed like often done in USA. The Barcelona rail trams have dedicated thruways, so cars better stay away 🙂
To me, this also reeks of leaving the MBTA more room to cut bus lines in the future. If there’s no investment in fixed infrastructure, all the easier to reduce service while saving face.
The other thing about this “leapfrog” idea probably fails to include cost of street pavement and tire maintenance. Welded steel rails with steel wheels must last longer with less maintenance than rubber pneumatic tire and paved streets. Street pavement with its underground sewer lines, gravel base, and weather prone asphalt must be more expensive than dedicated railway.
I attended the Feb 15th public meeting on this with similar outrage at this decision. Scott Hamwey lead the meeting and addressed the issue of in-motion charging (IMC) after an attendee asked about future plans for re-implementing catenary (Alon was actually mentioned in the question if I recall). Scott basically said that the MBTA wanted to have a homogenous fleet of BEBs for maintenance simplicity, a rather lame excuse. He also stated that the old supplier of trolley buses had gone out of business which apparently meant that purchasing new IMC BEBs would’ve required updating the existing catenary to a different configuration. If this is true then I suppose there’s some extra cost of IMC but I agree with the consensus here that this is more likely a nail in the coffin for IMC than a hiatus. The legal hassle of reimplementing catenary will deter any politician from pushing for it.
The meeting:
The fleet uniformity line is fraudulent given that buses operating in the Harvard tunnel need left-side doors, which means that deelectrifying the trolleys (which all operate in the tunnel) does not actually produce uniformity.
The suggestion that there would need to be a different design of catenary to operate trolleybuses with In Motion Charging is nonsense. In many cities, particularly in Europe, new IMC trolleybuses are introduced alongside traditional trolleybuses with no need for any technical changes. The only constraint I can imagine for Boston would be that sufficient wiring would have to be retained to be able to recharge the batteries. But that’s what all the other systems do: identify the best sections of route on which to retain wires to charge the trolleybuses sufficiently.
I’m not absolutely sure whether pernickety academics would agree with my use of the term, but saying, “we are abandoning trolleybuses because BEBs are the future” seems like a prime example of begging the question. Who says? And on what evidence? Answers:
A. various global vested interests plus front men like Musk;
B. patchy evidence that doesn’t prove the case.
There is still a fundamental flaw with BEBs for heavy duty applications and that is their extremely low energy density. While I am no fan of diesel buses, at least they can perform the duties that transit planners set. In contrast, the batteries on a BEB have an energy density of only 1% of diesel fuel. While engineers try to mitigate this great weakness, the fact remains that the BEB is vulnerable to external factors such as climatic conditions and traffic delays. This results in unpredictable, inconsistent performance and overall low productivity (at least 25% larger fleet is required). Meanwhile the issue of energy density does not arise at all for pure trolleybuses, which have an abundant external supply of energy. Even for IMC trolleybuses, density is hardly an issue, as their batteries are relatively low weight/mass.
This is kind of supported with the findings that BEBs (with Overnight Charging) can be used for low-capacity, low service level routes, where the total mileage per day is lower (essentially, what used to be done with a midi-bus).
“because he is an insular, ignorant, incurious manager. ”
You are being too kind. This may be the worst transit decision of the decade.
What if the “battery” in the battery electric bus is a radioisotope thermoelectric generator? Those produce enough “waste heat” to last thru the winter and there is a pretty big source of suitable radioisotopes (like Sr-90) in the US already, in the form of ~2000 metric tons per year of spent nuclear fuel. Of course that spent nuclear fuel is over 90% uranium, but Sr-90 still has a pretty high fission product yield…