Saturday, April 11, 2009

27> Designing For The U.S. Market

One thing I need to express is that, as an American, I cannot help but design for the U.S. market. Back in the 50s, then President Eisenhower spearheaded the construction of the U.S. interstate “super” highway system, and state and local governments quickly followed suit. By the late sixties suburban communities were popping up like weeds far from city centers, because without traffic lights and driveways greater distances were now commutable.

Fast-forward a generation and we have the great American suburban sprawl. Instead of a single “downtown” there might be ten. There are few pedestrians, because nothing is within walking distance. Traffic is not just confined to one direction. It is everywhere and may be worse thirty minutes from downtown than downtown itself.

Such a situation demands something much more ambitious than the little projects that the world has seen so far. It requires thousands of kilometers/miles of rail per city and speeds that are consistent with long commutes.

Obviously, no project can start on such a grand scale, but I believe any system that has a chance of adoption in the U.S. must be scalable to meet these needs. That means the cheapest possible track, the cheapest possible stations, and the fastest possible vehicles, designed for rides up to thirty minutes. That means a smooth ride is a must. The system, including stations, must have a minimal footprint. The station design, for example, promoted by the MISTER system is great, but only for about 10% of the stations, because, as I have said, there are no pedestrians. The sidewalks are empty but the streets are full.

Going fast means banking on corners or slowing for them, or clipping them (buying right-of –way) for larger radius turns. Banking track means more expense, complex engineering, and more specialization is required of the track builder. Perhaps there is a simple track design to do this but I have (so far)opted for the self-banking gondola design to address this issue. I am also inclined toward adaptable vehicle speed architecture rather than a set cruising speed. Empty vehicles should be very fast, traffic permitting, since there is nobody on board to get motion sick.

The U.S. transportation system is broken and needs to be fixed, and no little downtown “people mover” project is going to change that fact, just like light rail won’t nor will more buses, but that’s for a different post.

Finally, a question for my readers – In those systems with linear motors, what are the provisions for a power interruption? I was surprised to find that In the Taxi 2000 design they envisioned special “tow truck” type vehicles. Any thoughts on stranded passenger protocol?


Anonymous said...

By 'power interruption', i think you mean 'breakdown' since a power interruption in a system that uses a power rail instead of batteries is likely to be system-wide.

Further that was the context of Anderson's pusher for Taxi 2000.

Anderson, I believe, did not feel this problem was worth much thought because he believed that MTBF for his system was so extraordinarily high that you would virtually never see a failure on the track.

Not a student of statistics myself, I can only take his word for it. Anyone else shine a light?

Mr_Grant said...

Strictly speaking, Anderson believes a system as he designed it, with component redundancy, has such a high Mean Time Between Failure, that equipment failure is not much to worry about.

The math is easy. If component X has a MBTF of .03, then the likelihood it and a redundant X will both fail is .03^2, or .0009.

Continue this for a failure of both Xs, both Ys, both Zs, etc., then you see where Anderson was going.

If I recall power is redundant too in an Anderson system, a small battery backup enough to get a pod to the next station. I think the only thing NOT redundant on one of his designs is the snowplow.

Of course, component failure is one thing and external influences are another. Anderson probably would not take responsibility for an agency that installs his system but doesn't do basic things like check the condition of trees that could fall on the guideway.

Anonymous said...


I tried to ask him that very question on the PRT International blog, but have not received a response so far.

AndyChrist said...

There's nothing wrong with having ten downtowns. It works great in Japan, where the cities are really more like a collection of cities (each built around a subway or railway station) The problem is the suburbs often don't have any centralization. They've got a strip mall here and there, or more likely a mile-long stretch of road lined with them.

Dan said...

Dan The Blogger Responds-

Thanks for you observations A, Mr. G., I have little to add, except to say it’s great to see a little back and forth between readers. I, personally, am not an expert on Dr. Anderson’s work, other than that I felt the Taxi 2000 iteration of PRT was fatally flawed, at least as I understand it.

I am curious about how he deals with handicapped riders. Does he provide elevators? Does his system ever come down to earth? At the moment I’m on dial-up, (I travel a LOT) so I can’t just sit down and stream his DVD…

AC- thanks for dropping by…The problem isn’t the many city centers, it’s when the city develops along highways and their intersections, instead of mass transit routes. Once those highways can no longer support the traffic, city growth shifts few miles over and the process repeats…

Mr_Grant said...

Anderson's Taxi 2000 design was the first true PRT concept I studied. I don't know of any fatal flaws. I consider its hardware on par with ULTra and Vectus, and they have a control system which they have had the resources to test only in simulation and at scale.

Financial projections included elevated stations with hydraulic elevators, ADA compliance was part of the plan.

I can't think of any reason the guideway couldn't dip to the surface for at-grade stations.

Dan said...

Mr. Grant;
Have you seen the 2001 OKI loop study? I’m afraid I don’t have the link. It gives a very detailed option comparison and explanation for their rejection of Taxi 2000.

I do not believe that the linear motor propulsion is anywhere close to strong enough for steep grades. If the grades aren’t steep, they would block driveways, views of people’s businesses, etc. They would provide a climbing structure for kids and create a long corridor for possible collisions. My solutions to the collision problem include VERY slow speeds for the few feet where collisions might happen, something that would be intolerable for a long ramp. If they are walled off, this implies a very big footprint as far as land usage is concerned. The guideways would also be subject to filling with blowing dirt and trash, flooding, ice and snow, even vandalism or terrorism.

As I have mentioned before, the main reason I started this blog was in response to (what I perceive as) the shortcomings of the Taxi 2000 and similar designs. I was particularly struck by the massive track, which Ed Anderson obviously concedes was a bad flaw. In the streaming presentation on PRT International’s website, he himself points out that the Vectus track design is flawed and should replaced with a truss. In my August 9th post “First there is the track” I included a picture of the Taxi 2000 track to illustrate bad design. It is almost identical to the Vectus design. I am glad that Ed Anderson came around to something like my track design. Now if he would just turn it upside-down he’d really have something!