Sunday, February 14, 2010

73> Overwhelmed by an Underpass


I couple of weeks ago I was asked to quantify how tightly I thought a PRT vehicle would need to turn. In the case of PRT, the vertical turning radius must also be considered. (going into, cresting, or coming out of a slope) I have always considered that these radii must be fairly tight, but I had not really examined how and why I drew this conclusion.

I started my inquiry with a Google Maps starting with a place I have gotten stuck in traffic in the past. 

 

It is a place where an older road leading out of town intersected with a highway “loop” around it. Then time passed, and they built a mall and widened both roads a bunch of times. Now it’s a nightmare, but with good restaurants. Sound familiar? How would the various PRT systems compare as a solution?  The bottom picture shows the view looking east from “point B” and the smaller inset (point A) shows a small open-air bus stop, typical of the southern U.S.

The bus stop is included because it is part of the problem more than a solution.
On a typical afternoon, the traffic builds sufficiently so that it takes at least three full traffic light cycles for a bus to even reach this stop. There it blocks traffic further as passengers board - unless it just happens to reach the stop in synch with the rest of traffic stopping as well. Going straight, it blocks vehicles that might be able to turn right on the red light. If buses didn’t run through this intersection it would be better for everyone - especially the bus passengers, (who have to spend nearly ten minutes on this one intersection) but the alternative routes are nearly as bad. It seems obvious that the bus routes through here are very expensive to operate, and there is no place to fit light rail. The east/west road is as wide as will fit, and the loop was just widened - again. (Note the five-lane feeder road.)

PRT could cut through this mess like a hot knife through butter. I guess my main question is whether going over forty feet (12m) high to clear a raised highway is acceptable. As a devotee of all things futuristic, I personally have no problem with it, but will it sell at city hall? If the answer is no, it exemplifies a lot of what I have been saying about sharp curves, steep slopes, variable speeds and hanging vehicles. Hanging podcars, as I envision them, would have no problem going either over or under. The concept of a uniform line speed is challenged, however, because the PRT vehicles would need to bunch up and slow down to make the abrupt elevation drop and/or turns. I guess a two second headway would be about all we could get through per track, though.  If going over is the thing to do, the stations may need to be moved back from the intersection quite a ways, depending on how steep of a slope the system is designed for.  



I Googled around a bit, looking for other examples and found this. It is Main Street, Houston, TX as it passes under I-45. Pictured are the tracks of the new light rail.  I guess the car traffic has been largely diverted to other roads. On the face of it, it looks like they are squandering enough space to move a heck of a lot of people. Again note the need for PRT to go over or under. Actually there is virtually no way out of downtown Houston that avoids this overpass dilemma. Then there is there is loop 610 farther out, with the same thing…  And then there is the outer loop. Anyway, the situation is the same in lots of cities across the globe, so systems ought to be designed to handle the situation gracefully, whether it’s going over an underpass, or under an overpass! ;o)

6 comments:

afransen said...

That last picture seems to indicate that there is nothing under the highway. Seems that a PRT line could just divert slightly to the left or right, and come close to ground level to pass under. You'd need fences/bollards to protect the vehicles as they are close to the ground, but you avoid using up any existing road lanes.

A bigger challenge might be such a highway where there is a road that runs underneath, such as the Gardiner Expressway (you've seen it in a lot of movies!) here in Toronto. On the other hand, I believe the Gardiner might be high enough (in places) to allow a PRT line to run between the levels of traffic.

cmfseattle said...

you need to change elevation by 20 feet. 10% grade = 200 feet, or about where the A is in the upper half of your first image.

the bunching on the uphill side is matched by an un-bunching on the other side. at 30mph, the whole thing happens in about 15 seconds.

Ray MacDonald proposed some figures for standardization of PRT. you can find the info on page 379
http://books.google.com/books?q=automated+people+movers+2009&btnG=Search+Books

Dan said...

Dan The Blogger finally gets back to his own blog...

A couple of points. The bottom pic, like the top, has a parallel “feeder” road. To cross that with 16-17 ft. clearance and to then drop beneath the overpass is only an elevation change, which equals the height of the PRT vehicle plus the track. A system like Skyweb Express, if it could only get to the right height, could pass underneath with 8 ft. below it for pedestrians. For a self-leveling hanging system it looks like fences would be needed, but the ease of elevation changes and tight turns would enable a greater percentage of applications where going under would work. Of course there may not be room for both PRT and pedestrians if the PRT hangs down too low to go over them. This is a one pretty compelling reason to try to keep the track/vehicle profile at 8’ or less. For designs like mine, this severely restricts bogie wheel size. Alfransen, do ordinary (non-engineer) Canadians primarily use metric? I confess to thinking in feet and inches, and I try to convert for my posts, but this is just the comments section, and I have a converter in the sidebar, so I’ll be lazy.

Going over the highway does indeed mean a 20 ft. rise from a normal elevated track height… and a 200 ft. free-span bridge, in the case of the first example, although it’s a pretty wide highway for most cities. Going over means no slowdown, no bottleneck, though.

I was wondering why it seemed like Raymond McDonald was more of a spokesman for Taxi 2000 than someone without a vested interest. I guess the fact that he helped develop Skyweb Express has a little something to do with it. I mean really… Standardize OUT any parallel or back-out boarding? Standardize IN a speed limit and maximum grade? Why standardize having a TV or which way tilt-up seats face? ‘Sounds to me like he is describing the amenities and limitations of Skyweb Express as much as following any logical standardization criteria. I just get a little perturbed when people try to label their own designs as the only possible true PRT and then extend generic logical arguments from there, such as the one where he concludes that only high-capacity PRT can be economically viable. (He didn’t quantify as far I noticed) He includes nothing on hanging systems, which could employ dirt-cheap ground-level open-air stations, nor any sort of tensioned truss, which would have the potential to cut down the number of support structures. Then there is the fact that he presumably assumes this will be done by a single “Swiss-army-knife” company instead of a consortium. None-the-less, I thank you for the interesting read, cmfseattle, and I do not mean to suggest there isn’t a lot of worthwhile stuff in there. There is, (including the other books and articles) and it will make a fine addition to my (very slowly) expanding alphabetical index of links. It’s just a little hard to read while I’m rolling my eyes.

cmfseattle said...

i don't like backing-up movements in stations, either. i prefer a parallel design like in post 47. mostly, i thought it was interesting that he had numbers for max Gs in sloping segments.

i think PRT should go over freeways if possible, so that freeway maintenance doesn't shut PRT down. this scenario is both a challenge and an opportunity: good architecture and a freeway's worth of capacity, without splitting a city.

afransen said...

Canadians are mostly comfortable with both metric and imperial. It's a bizarre hybrid really. Height of a person is always in inches, speed of a vehicle is always km/h, temperature is Celsius if its outside, or Fahrenheit in an oven. Construction is done in imperial. It's a real mess. We'd probably be further along in metric uptake if we weren't next to the last bastion of imperial.

Dan said...

Sorry for my absence. I’ve been dealing with the medical emergency stuff that comes with aging parents, and it’s not over yet.

Cmfseattle, I was a bit underwhelmed by the advantages of quick elevation changes in the overpass example, although I’m sure there are cases when only a system like I suggest will work. I was surprised to find however, that there are cases where the Anderson designs would work and my system wouldn’t. The example of parallel boarding you site in post 47, though, is one where the steep slope, tight vertical radius capabilities really shine, since they allow the vehicle to drop down for boarding, meaning pedestrians need not cross the tracks.

I’m with you in theory, as far as your reasoning for going over the highway. I would love to see an estimate for a cable stayed or suspension PRT bridge. I bet it would be surprisingly cheap. I worry, though, about getting away from the fundamental proposition of dirt-cheap track. What is the trade-off of a bridge vs. more track coverage?

Thanks for the help, Alfransen. I guess I should have known. It’s pretty hard to avoid U.S. culture being as close as you are. You know, I grew up just south of Toronto, downstream from Niagara Falls. I could see Brock’s Monument right across the gorge from my bedroom window, and even Toronto, on a clear night.

Every time I write a post I wonder how much translation is needed. Do English (as a second language) speakers have any concept of horsepower? (perhaps from an American made motorized product?) And feet… I mean… we all have two of them, even if it’s not very accurate. Does that mean they can envision 50 ft. better than I can envision 15 meters? They taught us metric in school, even “back in the day”, but guys who build stuff (especially motorheads) are late adopters. In the machine shops I use (they mostly do oilfield related parts) everything is measured in thousandths (of an inch).