High-speed SMART

"Unit 395008 at Ebbsfleet International" by Sunil060902 - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons.

The Javelin.

The trains SMART will use are relatively slow. While they have a top speed of 79 miles per hour, their federally-mandated tank-like structure means a very long acceleration and long deceleration. Each stop will be a significant time suck.

In 20-30 years, when SMART replaces its trains, it may have a chance to do things differently. How much speed could we realistically wring out of the SMART system? Quite a bit.

The fastest commuter trains on the market are the British Rail Class 395, nicknamed the Javelin. They operate around London and – for the nerds – have a maximum operating speed of 140mph, compared to 79mph for SMART’s trains.

With these trains, which would involve electrifying the tracks and upgrading them to 140mph for the low, low cost of $978 million or so, SMART will be able to make the trip from Cloverdale to Larkspur Landing in about 49 minutes, down from 93 minutes. Novato to Larkspur would be, of course, quite a bit less – just 11 minutes, down from 27 minutes. Exact times might vary based on dwell – how long the trains wait for people to get on and off.


High-speed SMART travel times

For Sonoma commuters, the Santa Rosa-Petaluma trip would be cut to 12 minutes.

If SMART soars over the Golden Gate, down Geary, and to the Transbay Terminal (for just $5-10 billion more!), travel times will be significantly cut there, too. From Transbay, it would be 6 minutes to Sausalito, 18 minutes to San Rafael, 26 minutes from Novato, and 68 minutes from Cloverdale. This includes local subway stops along Geary. Depending on how

A super-fast SMART, in other words, would fully integrate the North Bay into the rest of the Bay Area. That it would beat drive times along the entire 101 corridor would provide a powerful incentive to leave cars at home. It would transform the whole North Bay.

Despite that, as my statements about the high cost of this upgrade may betray, I’m not keen for this. The Bay Area has significant transit needs, such as BRT on El Camino Real, the evolution of Caltrain into a mass transit line, Dumbarton Rail, a second Transbay Tube, and, of course, the Geary Subway. Each one of these is huge and expensive, and each of them serve more people than SMART.

But it is interesting to imagine how transformative SMART could be with the right equipment, and the right rails.

Did you like this article? Become a patron of The Greater Marin today.


About David Edmondson
A native Marinite working in Washington, DC, I am fascinated by how one might apply smart-growth and urbanist thinking to the low-density towns of my home.

27 Responses to High-speed SMART

  1. ekoontz says:

    Sounds wonderful, hope I live to see this vision come true.

  2. Stephen N says:

    Yes it sounds wonderful. It will work too providing they don’t have to stop for passengers or cross streets. Opps. The chief limitations of the Smart system is the practical limitations imposed by its needs to operate in an urban environment. A far more efficient public transit solution would be BRT buses and van service. Of course nothing would be as efficient as driverless cars that could be daisychained together. This technology exists and is the future. Streetcars going 8 mph in downtown areas offer little for the modern commuter.

    • That includes 10 second dwell times, though more realistic dwell times would push the end-to-end to something like 77 minutes. Cloverdale-Larkspur would still be rather speedy at 53 minutes.

      But I think you’re taking this rather more seriously than I do.

    • I didn’t mention the needed viaduct through downtown San Rafael to accommodate longer trains, but grade crossings for the rather faster TGV are common: https://www.youtube.com/watch?v=L2jW-zqnOig

      Great, now I’m taking this too seriously, too.

    • Franz Listen says:


      Van services are not more operationally efficient than a train or bus in a corridor like the SMART corridor under any plausible scenario. The small vehicle would require lots of drivers to handle volumes, which would be a lot more expensive.

      True BRT service isn’t even an option in the 101 corridor.

      As for daisy-chained driverless cars… explain how they are a more efficient form of transit. Would I drive my own car into a daisy-chain so that I could be stuck on the freeway as part of the daisy-chain instead of being stuck by myself ? Whats the value added to being in the daisy-chain? Wouldn’t all the cars on the freeway need to be daisy-chained for there to be a capacity improvement in order to improve the flow of the highway ? Isn’t that a vastly more complicated endeavor than the already complex and expensive act of chaining a few cars together ?

      And if I bring my own car to the party to be chained with others , is that even transit? Or just a different form of driving ?

      If a public or private transit service ran their own driverless cars to pick me up, that would qualify as transit. If they took me anywhere I wanted to go that would be nice. However, how would that be more efficient than existing bus or rail service? If the technology and regulatory climate allow for a driverless car service to pick me up, couldnt buses and trains also be driverless? Wouldn’t a driverless bus on a fixed route and schedule be even more efficient than single, driverless transit taxis?

      • stephen nestel says:

        Ever heard of the “vanpool” program? The size of the vehicle and common destination make it more operationally efficient than metro trains or bus lines. You need to compare actual passenger miles per gallon instead of theoretical miles per gallon at optimum condtions. These programs work well for large employers. Really large employers, most notably Google and Genetech operate bus lines. From what I understand they are pretty nice. It is far more efficient to have full transit vehicles rather than shuttling around heavy pieces of iron up and down a train track. Other examples are the Airport shuttle services that operate in the private market AT A PROFIT. Many of these vanpools are run with green fleets. They are FAR more efficient than public transit and even the best hybrid cars because their are responsive to true ridership demand.

        • Franz Listen says:

          Vanpools have their niche. However they are not a more efficient means of carrying large numbers of people in major corridors. If GGT and SMART tried to move the same number of customers along the 101 in vans rather than trains and buses it would be vastly less operationally efficient.

          Tech buses are nice. And notice that they are using full-sized buses rather than running little vans or dispatching fleets of imaginary drone taxis. They deeply subsidize this bus service, which is fine with me, but it’s not available to the general public.

          Shuttle services like the airporters can indeed make a profit, but its not the vehicle type that makes the difference- it’s the choice of route. Private services can select only the choicest routes, while public services need to provide coverage through a network. Some pieces of public transit services make money too, but they cross subsidize other routes. You can argue against public transit- that’s fine. However, don’t argue for a public transit based exclusively on shuttles and vans. That’s not going to help.

          • Woofwoof2 says:

            Transit on major corridors can be achieved by buses like the Marin air porter or the two story megabus. The highway is the most efficient analog in a modern dispersed economy. The current issue is under capacity and lack of hov lanes. Narrow cars of the future could DOUBLE the freeway capacity without adding width to the freeway. Chained vehicles could use express lanes to major destinations . I’ve published some videos on http://www.savemarinwood.org that shows these concepts. Innovation is our future. Fixed rail is a 19th century solution created when factories and large employment centers were the norm.

          • I’m going to assume you’ll be posting under this name from here on, Stephen.

    • John Murphy says:

      Stephen – your vision implies replacing the entire US vehicle fleet. 254 Million cars. During the transition we’d be building expensive but useless toys.

      • stephen nestel says:

        The car fleet changes every five years or so. Hybrids became widely adopted through a combination of incentives, market demand and low operating costs. Believe me, you will LOVE a little car that gets 120 mpg like the new Volkswagen rather than standing at bus stops and wondering if you will get a ride home because of transit strikes. The free market doesn’t need coercion.

  3. SMART Guy says:

    Sadly, its not that easy…1) SMART shares much of its right-of-way (ROW) with heavy freight rail, thus the requirement that SMART’s vehicles also be of heavy construction so that they aren’t squashed like beer cans in the event of a collision; 2) SMART does not enjoy a grade-separated ROW. Several dozen at-grade intersections with area roadways are present along the ROW. These cannot be safely navigated at anything approaching 145 mph, or even 50 mph. In many cases, the current operational plan calls for speeds of well below 25 mph through these crossings to ensure that some knucklehead in a Fiat hasn’t tried to beat the train through the crossing.

    The real constraint on speed is not the rail vehicles that have been selected, though they aren’t exactly Lamborghini’s as you have pointed out. The real constraints are those imposed by constructing a system that will share space with freight trains and automobiles, with the latter being the most problematic of the two constraints. The solution, of course, is a dedicated, grade-separated system. This would cost many, many billions, and the ridership numbers will never support such an expense.

    • Come now, those concerns are regulatory, not real. If grade crossings are fine for TGV, they’re fine for SMART. Even under FRA regs trains can maintain 110mph with (relatively) standard crossing gear, and the fast acceleration would ensure it’s not much of a problem for travel times. With nonstandard gear, the FRA permits 125mph crossings.

      And the requirement for heavy trains that run on the same tracks as freight rail can be waived if freight only runs at times when the passenger rail isn’t running, and if the FRA approves. Since they’re moving in that direction anyway, I wouldn’t see a problem in this 2075 concept.

      • SMART GUY says:

        Based on my understanding of the TGV/LGV lines in France (I’m no expert, but the info on Wikipedia seems pretty solid), these lines are entirely grade-separated. They are also fenced against trespass. The tracks are also electrified, and have limitations on radii and grades. The trackbed is also substantially upgraded from conventional. None of these features are relevant to SMART, which occupies a ROW that was established in the 1800s for pokey steam locomotives. Upgrading the SMART line to anything approaching the TGV level would take many, many billions, and would require a lot of acquisition for grade separations, curve straightening, and what have you. Unless 10 million people move to Sonoma and Marin, the ridership numbers will never come close to justifying even a fraction of the expense. The cost-per-mile would be on par with the California HST, perhaps even more considering local conditions. I’m not sure why you wrote this post. I like most of your posts, and many are quite thought-provoking. This one is only humorous. I’d rather talk about things that are achievable.

        • I’ve seen video of a TGV grade crossing, but they go rather faster than the EMUs I’m thinking of here. I’m not sure of CPUC regulations, but FRA rules allow 120mph crossings, with the right crossing equipment.

          You’re right – there’s not enough people to even begin to justify the cost of actually upgrading. The point of this post is simply as a thought experiment about the relationship between distances and travel speed. I’ve always been a fan of fantastical and impossible what-ifs, which help me better understand the practical and possible. Don’t worry, though. There’s still plenty that’s thought-provoking and serious to write about, and I’m still writing plenty about that.

  4. Stephen N says:

    I sure hope you come back to Marin from Washington D.C. so you can pay your “fair share” for these billion dollar fantasies. Already people are leaving the state for Colorado, Texas, Nevada, and elsewhere where there isn’t crushing taxes and regional agencies making our lives “better”. Funny thing about the digital economy. It isn’t location dependent. That is why a friend of mind at a Gaming company finds it so difficult to retain employees in the Bay Area. They move to Austin.

    • John Murphy says:

      Interesting analysis. California’s population is growing, not shrinking. The Bay Area is home to Apple, Google, Ebay, Facebook, Intel. Austin is home to AMD, Dell, Freescale, and many other struggling enterprises.

      • stephen nestel says:

        I believe we have had several back to back net out migrations. Someone can look it up. Opportunity is great in Silicon Valley especially if you are young and unknown in need of VC capital. Mature companies seek better business climates. Why shouldn’t they?

        • Eric says:

          You do realize that most of the reason people leave is related to cost of living, i.e. housing costs?

          • Stephen N says:

            Yes, Eric. You do realize that was my original point. People (human capital) and Financial (capital) go to friendly environments. If the cost of housing were the only important factor, Detroit would be a boomtown. Companies benefit from being in Silicon valley when they are young. In the long term, the hostile business climate repels them. The ABAG analysis is ridiculous and history has proven it false. ABAG Demographer/Economic forecaster projects unlimited double digit growth into the future because “Silicon Valley is such a cool place and everyone wants to live here” Booms and busts and the emergence of high tech centers in Hong Kong, Ireland, Austin, Seattle all prove this fool wrong. This is why this currently fad for Smart Growth and light rail will end. It will seem like an embarrassing tattoo received during a wild drunken night in Vegas.

  5. Pingback: High-speed SMART | The Greater Marin | SmartphonesSmartphones

  6. Franz Listen says:

    I welcome regulatory changes that would make it easier for SMART to buy lighter equipment in the future. They are long overdue. However, I don’t think that a lighter vehicle alone would have much impact on travel time. With speeds capped at 79mph, the quicker acceleration of a lighter DMU might only shave off a couple of minutes.

    If a train could hit a top speed of 140 mph that would certainly make a big difference but, as noted, would require a better class of track, electrification and grade separations – which would all combine to be prohibitively expensive.

    However. here’s what would be realistic and could also speed up SMART in the future: aiming to spend as much time as possible keeping the speed as close to 79 mpg as possible. Things like additional sidings, brief and efficient dwell times, maybe some smoothed out curves, or a few strategically placed grade separations might make a meaningful difference.

    The next phase of SMART in 10-20 years may be about optimization, which could include a series of small scale improvements to trim travel times. Not a super sexy vision, I know, but it’s a way of contemplating how SMART could be realistically faster in our lifetimes.

    • John Murphy says:

      Nobody would want to have their station removed, but the answer IMHO is fewer stations and better feeder service. I would pull 3-4 of the current set of stations out.

  7. SFtoRosa says:

    We spent far too man decades without even Caltrain upgraded to electric trains. The Capitol Corridor is another obvious candidate. The real tragedy is that SMART with all our modern options did not choose to get lighter DMUs. Positive train control is now mandated, but nowhere in the US are we actually using it to run trains faster and closer. The first order of business is to upgrade the track at Penngrove near the Cotati grade. This is far too slow for commuter service.

  8. Pingback: Today’s Headlines | Streetsblog San Francisco

  9. John Murphy says:

    David – I think you’re a little less than fully educated as to what the right of way looks like. The tracks in Healdsburg run along a narrow strip between a road and a parking lot, with several grade crossings that would cost who knows how much to separate. This includes bisecting the middle of the 5 way intersection that is the gateway to town, and is currently slated for a rotary that will be bisected by the tracks. There are curves in the railways that would need to be straightened to hit such high speeds, and straightening the tracks through Healdsburg would mean eminent domaining a huge chunk of the town.

    Along Old Redwood Highway the tracks pass within a few yards of houses.

    It will be a long time and a lot of population growth before there will be a positive ROI on that sort of investment on the train from Cloverdale to Windsor. And the area won’t support that level of population growth within the water envelope.

    • As you might be able to tell, I didn’t do any real analysis of the ROW and was “not keen” on the idea, though apparently Streetsblog thought I was serious. I’m aware of what’s going on in Healdsburg and that there are tons of curves that would need straightening. The full upgrade plus extension would probably be in the $10-20B range, far more than should be spent on this.

      Grade separations aren’t necessary except for special circumstances, like that roundabout, or where the blocks around the station are shorter than the train. I didn’t look into the PUC’s ROW-width requirements because I didn’t want to dive quite so deeply into the minutiae of what is a thought exercise, but IIRC the only places where there is insufficient ROW for a double-track system are at the Puerto Suello and Cal Park tunnels. Feel free to correct me if I’m wrong.

      I guess I needed to put a big fat disclaimer at the start of this thing saying, “This is a thought exercise and should not be taken as a serious proposal by any means.” Will keep this in mind next time.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: