Sunday, January 31, 2010

71> Switching Tracks


Here is a little project that is still under development. I would have preferred to have solved all of the problems with it before posting, but I have had precious little time to devote to my PRT “hobby” lately, so I’ll go with what I’ve got.

The design is inspired by a very innovative aspect of the MISTER system as shown most clearly near the end of this video, Mister not only doesn’t require switches in the usual railroad sense, it doesn’t even need the connecting track to touch the main line.  One great aspect of this scheme is that it gives remarkable flexibility in routing, even after a line is finished, because a branch can be added with little or no disruption to the existing track.  

While I have taken a general track design approach that is much more similar to the designs worked out by PRT pioneer Dr. J. E. Anderson, (but turned upside down) I still wanted to maintain the possibility of track that could be added to with minimal disruption. The bogie designs I have been working on all share the quality of being able to completely support a PRT vehicle from either their left or right sides, like the MISTER system. I cannot claim the ability to add a switching point without any service interruption whatsoever, but the track shown may have either side removed and still be functional, if supported properly. Sections of such track could be used at points where future expansion plans might require branching. This would include entrance and exit points for sites of possible future stations. The trick, design-wise, is to build a box beam with the bottom element removed, without greatly weakening the structure. The main trusses must remain parallel to each other and square to the overall structure, yet removable. 


The use of turnbuckles (shown below) evolved from a realization that the chances of getting this thing to fit together while hanging it from a crane were pretty close to zero, yet I had wanted cross members to stiffen those areas of the structure. In theory the turnbuckles could bend the plates enough to get the assembly started, exerting forces shown by the arrows. They then could be loosened, the plates bolted, and then retightened.  


After assembly the cross bracing is continuous, as shown below. 

I now believe the better course would be to have rigid bracing with permanently angled plates, but that involves a total redraw, and this series of pictures illustrates the general idea well enough, I think. I would also note that most designs can be greatly simplified after some careful study, but this project has not reached that stage.


cmfseattle said...

your trusses look really good; i know that drawing them takes some time, even in sketchup.

something i've been thinking about is platooning or coupling vehicles into groups of 10 or 12 during peak demand. it might mitigate risk by being able to operate the system like an APM.

so, that would call for a guideway design that could handle "PRT train" dynamics, at least for the initial segment(s).

aussiejoshm said...

Interesting concept. Its a good idea to be able to easily add connections to the track if required. However, there are two significant problems with trying to along the connection to be added while track is in operation:
1. Its really not worth the risk and to passengers and authorities that risk will not be acceptable.
2. The entire network will need to be designed such that half the track will hold any dynamic loads from the vehicles.

Dan said...

The idea of platooning for safety, Cmfseattle, seems rooted in the concept of eliminating the speed differences between vehicles. In Irving’s and Anderson’s works, both allude to fact that a first vehicle could not stop instantly from most causes, meaning that a following vehicle needs only to brake equally or better than the first to avoid collision. With either LIMs or direct drive wheels, the braking is accomplished by decreasing a frequency in the controllers, which can be set identically for a decelerating group. This means that the group will behave as a platoon save for possible traction issues. (Frictional and/or magnetic) On a dry surface this would presumably only happen in an emergency, “jamb on the breaks” situation. To address this, I personally favor safety devices on the bogies that can physically establish a healthy spacing between vehicles. Think airbags for bogies. Thus vehicles, in a crash situation, would behave as though part of a spongy platoon, where the closer they got to each other, the stronger the resistance holding them apart. This would be especially beneficial for the frontrunner, which would not have to withstand the simultaneous impact of the entire cumulative mass of a train of solidly linked cars hitting it from behind. Anyway, one of the last things I want to explore in establishing a standardized track is braking, and much of this was to come up in the next post anyway, so I’ll leave it at that.

Hello, Aussiejoshm – Welcome to the blog.

I did not mean to suggest that switching points could be added without ever closing the track to traffic. I was just suggesting a method by which such changes could be better facilitated. I imagine that by working at night the project could be accomplished in stages where the track could reopen each day. Special scaffolding could be engineered to help support and stabilize the track for day traffic. I acknowledge that such a method would require the same rigorous testing and approval process as any other aspect of the system. I would point out, though, that in a PRT system the traffic count, speed and spacing may all be precisely controlled. This could theoretically minimize the amount of special bracing required, if this were to be a problem for some reason.

As per your second point, the track is already capable of handling the loads on either side. The load is distributed to even out the wear. If it seems redundant to use matching half-tracks when only one is theoretically required I would point out that the truss requires the full width anyway or it could “fold” from sideways forces. It also has the function of keeping weather away either half of the bogie and keeping sound in. Anyway, as I stated, this splitable truss is designed for areas where branching is anticipated. A somewhat less complicated truss structure would be used elsewhere.

cmfseattle said...

i should have written "investment risk" (of a small-vehicle system). most proponents of current systems cite the uncertainty (except in simulations) that small-vehicle systems will have adequate capacity as one of the reasons not to build them.

so platooning would be a brick wall stop-compliant method of meeting perceived capacity needs (although with the likely trade-off of higher-cost guideway).
grt vs prt with train formation

PRTcap.xls Last updated by Mike C - Jun 16 2009

keeping all the vehicles near the same velocity reduces control-system complexity, thereby raising its reliability (it also reduces the potential for congestion).

setting the same motor frequency for several vehicles may not be sufficient:

in an emergency, each vehicle's deceleration rate would be monitored by the ZC. it's possible that deceleration(s) would be faster or slower than predicted by the control system; the system should be designed to handle this safely.

Ben said...

Each vehicle should travel at minimum distance of one and a half car spaces so that cars that merge onto the main line are always able to easily negotiate entry before the track runs out. Linked cars would make the controls more complex and incorporate increased risk.

I would suggest a fabric "airbag system" much like those used in cars. I would contract vertigo air-beams technology that can be rapidly deployed at lengths of over 10 feet easily. You could use cheap material to create such systems because they would be designed to fail. These systems would be deployed as the result of a laser range-finding instrument.

Dan said...

Your first paragraph, Ben, is bit of a simplification. Headway is a function of speed, not fixed distance, at least that is the way safety regulations are written. Of course with the safety features mentioned in the next post the current regulations for rail-based transport obviously shouldn’t apply. Assumedly, rewritten headway regulations will be based on what can be proven to be safe, so platooning will have to pass that bar. I want to note that with controller based motors the RPMs are controlled by a precise frequency. Any speed difference between two vehicles running on the same frequency is negligible, as are their electronic braking characteristics, as long as the tire wear and pressure are close and there is no magnetic slipping. Platooning maneuvers should not be very complex for digitally controlled vehicles.

No vehicle should start acceleration towards a merge without a slot being available. Vehicles can be preparing for this way ahead of time, based on track sensors or markers well in advance of the merge point. Merging might even be a reason to create a platoon, so as to make space.

Those Vertico “air beams” are an interesting concept. They use the weave of a Chinese finger trap to reinforce inflatable “beams” to great rigidity. High-pressure inflatable shapes are a technology that someone will make millions on someday. It’s not really a question of what you can make, but what can’t you make. The idea of long airbag is intriguing. I suppose it could even expand to fill the whole inside of the track. In the next post I have a cylinder style instant shock absorber, which would be simpler to engineer initially because its dimensions are so controllable. An airbag would be lighter though, if the exact cushioning profile can be quantified and controlled, so as to make the stop with the least G forces possible.

Dan said...

Oops! Sorry, cmf...forgot your comment was up there... I don't have Excel on my computer and thought I downloaded a reader for xls files but I can't find it now.

That was an interesting discussion though.. I missed it when it happened. I was probably at my cabin.

Later today I'm doing a post called "Autonomy" where I spell out what my veiws on the ZC thing