Thursday, August 30, 2012

145> How fast is fast?

Let’s talk about speed. How fast should PRT ideally go and why? I suppose the first question that comes to mind is how far the routes extend. I have a bit of a problem with the slower, downtown-only type systems. First, of all, many people have to travel a long way, often in heavy traffic, just to get there. Then, unless the streets are just too narrow (and this is admittedly sometimes the case) traditional public transportation can often suffice. For example buses or light rail, even with the added wait time for passengers to get on and off at each stop, actually work fairly well for short hops. It is when you have to endure that stop-and- go for mile after mile that such transit starts to get increasingly impractical. Second, it seems to me that slower systems lack optimism. If PRT is so good, won’t it end up extending far and wide? And if that is the case, isn’t designing a slower system not just putting the brakes on speed, but putting the brakes on the extent of the system as well? I know, ideally people should live downtown and this should be encouraged. After all, giving people a means to commute actually encourages environmentally destructive suburban sprawl. But we are not talking about running PRT to lots of outlying neighborhoods. The shape of the urban center has changed. It is no longer a circle or square but rather shaped more like an octopus, with many essential goods and services being located along the arteries that extend outward. PRT, it seems to me, should go where the action is.   

When you talk about these longer trips, such as commuting, speed becomes much more important. Nobody is going to leave their car behind to take a 25mph vehicle for six miles that probably won’t take them exactly to their destination - unless, of course, the traffic is absolutely horrible. PRT has a “last mile” problem on both ends, and speed along the way can make up for that. Running parallel to highways, the vehicles are moving billboards for the system, since they will either be speeding past the cars or the cars will be speeding past them. I think it is vitally important that the former, and not the latter, be the case.   

For PRT, commuting and downtown circulation would seem to have a symbiotic relationship. Each makes the other much more viable by adding passengers that would otherwise use other means.  Often travel within cities is marred by certain chokepoints that make an otherwise speedy journey by car (or bus) into an exercise in frustration. PRT must serve more than that chokepoint area itself. It must extend to areas on either side that are easily accessible. This, by definition then, means that PRT should service areas with low traffic, as convenient entry points into the system. I think this is a point that is missed by PRT designers who assume a business model with only large, high volume stations. (If there are so many pedestrians, how did they get there?) By extension the logic of distributed, easily accessible entry points leads to those easy-to-build-on corridors that divided highways often offer. At this point the cost-effectiveness of a minimal-profile rail-type system almost begs to go farther out, since the capacity vs. cost is far superior to extra freeway lanes, and “Park&Ride” lots are already in place in most cities.

The aerodynamic drag on a given vehicle roughly cubes as the speed doubles, and so there is a natural limit on the top speed that PRT vehicles ought to go, after which they are wasting energy toting around oversized, expensive motors that may seldom be used. Actually the whole concept of individual travel becomes less and less advantageous compared to large groups as you approach these speeds, since many little vehicles catch much more air than a single, larger vehicle.  So this begs the question of what, exactly, that speed limit should be. I believe that the answer should be guided as much by psychology as by physics.

People know when they are being whisked off to their destination, and they know when they are in a sluggish vehicle. Ever get driven somewhere by a senior citizen or try hitting the highway in a weak sub-compact car?  The power needed for decent acceleration, particularly up hills, will require motors that are capable of freeway speeds, and probably a bit faster. Here I would remind the reader that a hanging system enables ground level boarding, and minimal ramps are cheaper and easier to fit into the urban landscape;  This, in turn, means that it is highly advantageous to both lift and accelerate the vehicle to merging speed in a short time…something that will require substantial power. This would indicate that there is little value in making the vehicles only capable of sub-highway speeds, just on the basis of motor size.

Way back in post 56 I outlined a method of going very fast, using “engines” that would link together PRT pods into an aerodynamic “train.” Such a method seems well suited to the “diamond track” concept that was outlined in post 141. Those large, splayed outer wheels could be on the engine/connectors, and this could free the ordinary, lower-speed guide wheels from costly wear, while enabling speeds much faster than any highway. This method of connecting vehicles addresses the both motor size and aerodynamic drag issues.

One issue that often comes up is the matter of travel between whole different cities. That is one that raises questions about tracking vehicles over long distances and imbalances in inventory between cities, as well as issues like rest stops, or the simple matter that longer journeys would be uncomfortable in the small vehicles that are best suited for economical city use. I suppose one thing that ought to be considered early on is the possibility of GRT (Group Rapid Transit) vehicles that are compatible with the track size. This is also consistent with both the need for less air drag per passenger and the lure of heavier freight loads than would be acceptable in town. In such a scenario the smaller “pods” would be able to travel the heavier GRT/freight track but not vice versa. One drawback would seem to be the possibility of speed-limited PRT getting in the way of faster long haul vehicles. But computers are amazing things, and with a couple of passing lanes and properly orchestrated timing, I would imagine that the two could coexist in a pinch. Going very fast, say 100 mph and above requires a track with enough inner area to allow air to get out of the way of the speeding bogie. This seems consistent, however, with a larger, stronger track, since the minimal-profile track used in the city is only for aesthetics. There is structural value in increasing the dimensions, even with the same amount of steel. I do not think changing vehicles on either end of a long journey represents much inconvenience. After all, the PRT vehicles should be waiting for your arrival, and even departures should involve minimum waiting with a relatively small number of passengers. (Or just pay extra for the whole darn thing… a sleeper car!) Anyway, that is a subject that is removed from the matter at hand, other than that a heavier, even faster track could (and probably should) be considered for track that parallels major highways.   

So, to sum things up, there are factors, some purely psychological, that push the optimal speed limit up to a range that is roughly comparable to what we experience in our cars. There are other factors, namely aerodynamic drag, that limit the speed to not much beyond that range. After that, vehicles should probably be coupled or GRT could be used, increasing efficiency dramatically. This probably would call for a modified track design. This is, however, generally consistent with what would be optimal for fast PRT routes anyway. 

Every endeavor has to start somewhere, and PRT surely has to start with very modest routing and so performance is not really an issue. That does not mean, though, that PRT should best be designed with those limitations baked in. The trick, however, is to take this speed requirement and integrate it into a design that does not sacrifice full 3D maneuverability or minimal track size. Otherwise the design would be quite simple!