Sunday, October 30, 2011

130> Progress Report

As many of you know, I have been advocating the standardization of key PRT technologies in order to allow PRT to be developed, produced and deployed by a consortium, rather than a single company. This, in turn, requires that this development be started on a basis of the most promising design approaches. I have concentrated my efforts on a suspended system, rather than the bottom supported approach, even though I suspect that the majority of readers prefer the latter. One key to why I think suspended systems represent the best way forward is referred to in my latest iteration of the acronym “SMART,” which is what I call this effort.  (“Standardized Multi-axis Automated Rail Transport”)  
As I have pointed out numerous times before, all “ground” transportation suffers from the same problem: Vehicles or people going in different directions will run into each other unless they stop and wait their turn. Going over or under solves this problem, but that solution is too expensive to deploy universally with current modes of transportation. That indispensable artery of modern commerce, the freeway, clearly shows how effective high-speed non-stop transportation can be. This is simply the result of what happens when a transportation system is modified to be multi-axis instead of existing on one plane - the ground. Unfortunately making multi-level (multi-axis) routing for large vehicles such as trucks and trains takes huge amounts of money and space. When it comes to multi-axis transport, smaller is better. Luckily, we mortals are pretty small.
A multi-axis automated rail transportation system is essentially a new infrastructure designed to do what the freeway can’t. Go to any street, to any bus stop, to any building. It would be designed to be faster and safer than driving, more energy efficient than the most advanced electric car, and expandable for a fraction of the cost of roads. Being natively multi-axis, a suspended system can be employed in areas where long ramps are undesirable (that’s basically everywhere) and the system can be elevated higher than would be practical for supported systems. This can minimize visual impact. While it is true that a supported vehicle can be made to self-bank and keep its cabin level on slopes, it is much more cumbersome to engineer. Vehicles with wheels on the bottom are just ill-suited for extremely steep travel, while hanging vehicles have no such problem. Traditional PRT designs require raised, elevator equipped stations because otherwise the entrance and exit ramps into the station would block driveways, be subject to climbing, and be visually intrusive. A native 3D system has no such restrictions. A suspended vehicle can either taxi in like an airplane or come down like a helicopter. This means that stations can be put nearly anywhere, and they can be very minimal and inexpensive. They do not require high traffic volume to pay for themselves, so they may be placed with high frequency, like bus stops rather than actual stations.  This will increase ridership. The question is this: If we are going to build a whole new infrastructure, do we want it to be raised, single-level, multilevel with ramps, or natively  multi-axis? A consideration of the various routing situations likely to be encountered in a widely deployed system leads me to believe that it would be better to have true multi-axis capabilities from the start.  Anyway, here is the latest iteration of the SMART PRT vehicle concept. Hmm… How about “SMARTPOD?”….Sorta has a ring to it…

Unlike previous versions, the steering guide wheels have been moved outside and under the track. This shaves off about five inches from the track height, bringing it down to about  30 inches/76 cm. (less if it the track is hung from a ceiling.) What is shown here is a high-speed vehicle, (highway speeds and higher) designed for many tens of thousands of miles between tire changes. (hence the large wheels)
The wheel flanges are designed to outlast the tires in two ways. They turn independently of the wheels, so if they contact the track at a different diameter than the tires there will be no conflict. Secondly, they are only deployed during actual turns. Otherwise the bogey is centered by leaving both left and right steering guide wheels in the upright position. The upper “hold-down” wheels replace the upper steering guide wheels of previous designs, prohibiting any rotation of the bogey within the track. This design is extremely maneuverable with a turning radius of a mere 8 ft., including vertical turns. (The spacing between various track surfaces must vary, however.) The pictured design is missing most of the components of the bogie at this stage of development. The sprockets pictured are for vertical climbing, although I plan to adjust the sprocket size somewhat.  

The track has been designed to be extremely easy to fabricate into sections that are straight or curved. There would be no problem finding shops willing to bid this work, even in small towns if the pipe bending is outsourced.  Removal of a left or right truss section will not mean that vehicles cannot pass, although there is a small temporary rail that needs to be placed as insurance against any freak events that would make the whole vehicle sway with great force. I am still working the best way to attach sheathing, although the reader will note that there is a slight arc to the outer edges of the truss. This is to make light-weight metal or plastic sheathing more rigid.

Alert readers will notice an air scoop. At this point I am leaning toward liquid cooled motors. This greatly increases the performance-to-weight characteristics of the motors, and hub motors are ideally suited for this, as the copper coils that need cooling are stationary and accessible radially from where the wheel attaches to the frame. A simple little electric pump that is remote from the motor itself is all that is needed. No moving parts are added to the motor. The scoop is for a radiator/heat exchanger.
Finally, I want to emphasize that this vehicle has capabilities that go well beyond what is likely to be deployed early on. Nonetheless, I am designing with the future in mind so aspects that are practical today but foreclose later improvement can be avoided. Water cooling, high speeds and vertical climbing are features that might be expensive complications to first deployments. However I see little point in building an infrastructure project whose inherent design limitations will become apparent as soon as it is deemed a success.  This is, in part, why I favor a full multi-axis approach. Future cities are only going to get more crowded and time is only going to get more precious.   


Juho said...

Good progress. I like the idea of truss based suspended system. Truss might mean tht the system is cheap and fast (free flow of air), and it can still be covered. One key reason why I sympathize suspended systems is that they can be covered (against snow, ice, water, dust, trash and birds).

Higher poles are also a negative point in suspended systems (price, visibility, need to bend the pole to leave space for the vehicle).

These design principles could really work for both first cheap implementations and long term fast systems.

Some more comments / questions:

- Is the piece of metal where the main wheels are running strong (or light) enough? (I-shape vs. L-shape vs. welded to the tubes)

- No crossings shown this time. Can this track switch in high speed? Is one half of the track enough to support the vehicle while switching? (or does the bogie lean on the other side (broken tracks) too)

- Climbing is interesting too but wouldn't low tracks be sufficient in most cases? (for example in my home yard)

- Electricity from top middle?

- Do high speed tracks consist of two standard (curved and straight) sections or are the transitions from straight to curved track smoother?

- Is this wheelbase sufficient for high speeds? (could be extended for high speec bogies, but limits turning radius)

Anyway, for suspended systems with C-shaped track, I think this design could already be proposed as a starting point for standardization.

Asko K. said...

You need an implementation for successful standardization. Only after the feedback coming from such actual work and operation good practical standards arise.

Having said that, I am impressed (again) by Dan's pictures. This certainly can be used as a basis of implementation, but just forget the standards for now. Just make it. Keep it open. If it works it is a de facto open PRT standard. As easy as that. :)

Dan said...

Dan the Blogger responds ...
Hello Gentlemen! I'll have to do these separately. Juho first.

Strong enough? Toward the middle, certainly. That’s 3” pipe with supports every 21 inches. Toward the outside, it would be fine with thick flat stock but you are right, better to use a vertical stiffener, as that would allow for the least weight.(thinnest stock) I had that included but removed it as I was studying the space needed for a chain “rack” for climbing.

Switching at high speeds- yes. It needs both top wheels though. (And a cross-over on the “ceiling” for the wheel to remain engaged. Low speed can be done on a half-track, as the bogey is captive unless some force twists it in a way that defies gravity. A small piece of angle steel could be used to prevent this if the track is being opened to construct an interchange.

I think climbing is the key, actually. I would warn against assuming that people will accept a new infrastructure. The ultimate way to “step lightly” around potential adversaries is to be able to put the very minimum of track in their faces. Also, it is not unusual to have a driveway every 100 ft. or so. To get clearance for trucks that is already pretty steep, if you’re going from ground level.

I am thinking communications from top and electricity from lower sides (another reason that stiffener is missing) This is to prevent it from being touchable with a straight piece of metal or an arm. I believe I can add angle steel guards (to the pipes that form the opening) in areas where the track gets within reach. (The bogie would only go very slow in this case, so the margins can all be tightened up) At present I think I can close the gap to a distance too small to get a hand into. (With those guards)

Transitions – That is all about dealing with the expansion/contraction of those sections, and is pretty tricky. I really haven’t got all of that figured out yet. I imagine, off hand, that airs of truckable sections would be site connected, and these combined would be roughly the distance between supports. Curves present a challenge because they hang their weight out in a precarious way. I would just pass this all off to the experts but for the fact that I would love to have a better grasp of how much this track will cost and how much it will weigh early on, just to see where I stand. It’s going to require a certain vehicle density on the track to make a compelling business model, after all.

Wheelbase – I must admit to a certain instinctual discomfort with the short wheelbase myself, but I think this is just because of being used to vehicles that rely on gravity for their stability. If it can’t tilt, it can’t tilt. Keep in mind that the rubber can be of a hard type with little give, so long as it isn’t so hard as to not stifle vibration/sound and it can still absorb some minor dimensional irregularities.

I almost included the whole fabrication procedure for the track in this post but it was getting too long so I cut it out. I probably will combine that with the climbing aspect in a future post. There is a little trick to getting the teeth to mesh when the bogey first starts “biting”.

Dan said...

I tend to disagree, Asko. I think a rush to implementation will only produce half-baked standards. Behind every standard there goes a lot of engineering effort. So much so that the next guy doesn’t really care to do it all over again and so throws his hat in with yours.

A lot, at this point, hasn’t even been decided in terms of division of labor. For example, what if there were no motor makers, but only makers of complete motorized things. Then you clearly wouldn’t need a NEMA standard, so that you know the motor will fit and the bolt holes will match. Or who needs standardized threads if everybody makes their own on the spot? That’s where we are with PRT. It hasn’t been engineered into the simple blocks for outsourcing. I know from experience that without this step, almost no amount of money will be enough to get the job done. In many cases there is only one good way to engineer these aspects, and by identifying the best way and publishing it (which prevents patents) the seeds of standardization are planted. Each trade has a different language, but nobody speaks PRT… yet. Box beam, suspended, supported, two rail… These are all quasi-standards already, as they are all descriptive classifications. What is a class 4 SV boxbeam? Nothing,yet. But it could mean a box beam for suspended vehicles of a certain weight…Someone has to propose it before someone else can find it useful. Standards, it is true, are formed by a consensus of practitioners, and I take this to be central to your point. But I think it is worth putting this stuff out there anyway, in that I frame the arguments that this consensus can be forged from. (not that I’ve done a particularly good job at this so far)

Anyway, I have no interest in commercializing my designs. Raising capital,budgeting and promotion, etc… These are huge commitments and a major lifestyle choice…one that takes a certain type of personality. Been there, done that. Didn’t get a good night sleep for ten years!

Andrew F said...

Dan, the idea of a standard for PRT is apparently somewhat controversial. I noticed some people over at transport-innovators think that a standard for PRT would be disastrous, and lock the technology into some inferior form. Personally, I think it would be a disaster to see any significant PRT development without something of a robust standard in place. The network effects are so powerful that once there is a critical mass of any system, it will be hard to uproot, and we might end up with a situation like they have in Europe, where the rail gauges are incompatible, or the world-over with left- and right-hand drive cars. These are a real shame, as the choices were essentially arbitrary, and agreeing on a compatibility standard early on could have saved many billions over the decades.

Leaving that aside aside, I just wanted to say that I liked the latest iteration of the track, particularly the reduction in track height. I'm curious about the climbing mechanism, but since you mentioned you would post on it later, I'll wait until then.


Dan said...

I can most certainly see how they could feel that way, Andrew. After all, the very examples you give to support the idea of standards are examples of standards themselves. You are quite right to point out that some of these standards were indeed based on decisions that were essentially arbitrary, and that this is to be avoided. All I can say is that this subject calls for its own post. It’s far too important of a subject to delve into exclusively in the comments section. I will say this much, as food for thought.

First, there needs to be clarification between ISO style standards and general classification and definition. What are the relative benefits and dangers when we dissect the problem that way?
Secondly, is there a general way to avoid the pitfalls? Is it just to be first, so a parallel technology (i.e. right hand drive) hasn’t already taken hold somewhere else?
And what about the situation where immutable factors make certain structural design decisions unavoidable? Shouldn’t the resultant solution at least be named?
Finally, can’t some of the danger be mitigated by structuring a standard properly? You used the example of incompatible railroads. Yet there is the commonality of a pair of rails supported by wooden ties, which are in turn supported by gravel. Couldn’t a set of the most promising dimensions have been introduced with version numbers, so that the “arbitrary” decisions would tend to fall into round numbers which have been vetted? And by the way – there is was no need for a standard for the gravel size. Only one general size will work. It’s an example of something that could have been standardized early on without the possibility of creating conflict.

Anyway, those are the main ideas I’m wrestling with to address the issue. If I left something out, let me know. I could use some help here to digest some of this stuff before I post. I’ll be posting on that climbing mechanism next, though.

Juho Laatu said...

I'd like to see some standards. If we will see just various very local proprietary PRT systems, I'm slightly interested but not very much. What is interesting to me is the potential of building a larger PRT system that might one day be good and strong enough to take care of large part of all traffic.

All we need is some minimal set of standards. In practice that would mean a standard track that can be manufactured by many providers, and that can be used by vehicles of many providers. And of course that standard allows different initial and separately financed and planned local systems to be connected to a larger system. A standard is badly needed to make other than local airport style investments sensible.

It is not a problem to have even multiple standards. Markets will hopefully pick one of them as their favourite. One problem with standardization is that typically the leading companies are less interested of the standards since from their point of view standards would only level the playing field for their competitors.

If we want PRT to ever grow to a major approach in transport, we need some solid standards. They can be very minimal and cover only the interoperability related aspects of the track design. This is not an easy task, but a good standard might be also successful.

Dan said...

Sorry, Juho. I wrote a response and I guess I forgot to post it. It basically said that I agree with your position... I am currently trying to quantify and qualify this whole thing with some solid logic. Are there guidelines for judging what is good material to standardize vs. what is personal design preference? How fine-grained vs. general? I know first hand that it is very easy to conclude that a certain design or method is the only way to do something when, in fact, there are options. Anyway, this will probably be the subject of my next post.

Juho Laatu said...

I think the simplest specs specify only what is absolutely needed to make sure that my bogie will run on your track. The specs could be more comprehensive too, but those parts would be just guidelines and suggestions for implmementing the system, not normative specs.

In practice all I need is some numbers on where the surfaces for the wheels and eletric contacs are, how wide they are, how much free space I can assume around them for the wheels and other structures (including the gondola), and maybe something about the curvature and tolerances too. Some things like allowed maximal weight and speed need not be standardized since we may have different kind of tracks, but of course we need to have some understanding on which categories we should support and what kind of track you plan to build. I also need to know the voltage (but I could accept different solutions and variation too).

There may be multiple interfaces to standardize. I mentioned only the bogie-track interface above. A standard for the platform location at the stations (for human traffic) is closely related. We could also specify an interface between bogie and gondola to allow these two businesses to be separated. Then we could specify some communication systems, although here we can again be flexible and allow even different solutions since it is not a major problem e.g. to support two communications interfaces in the vehicle (and allow them to evolve independently in time).

Before the final specs we may need also some discussion papers to agree on the concepts and what the requirements and standardized interfaces of the system are.

So, at minimum I need to know where the surfaces for the n wheels (and electric contact) are and how much free space there is for my bogie and gondola design. That picture would be much simpler than the pictures that you already have - just surfaces and spaces. I want to know the platform location too (maybe later). Curvature and switches need some special attention. In the switches I need to know when each supporting surface is available for me and when not. One important detail is to know when the switching section starts and ends (and I can not change my choice of direction). Then we could agree to arrange some interoperability tests after one year when our products are ready for that.

Dan said...

Juho, I wrote a lengthy response but never posted it because I keep second-guessing different aspects of the design. What I show in this post is absolutely chocked full of trade-offs and judgment calls. I’m really not comfortable with committing to exact dimensions yet. Bad standards are worse than no standards. At the moment I am leaning toward dimension ranges rather than absolute values.

If you are designing a system, I would strongly suggest that you download Google’s free Sketchup (3D modeling) program, so I can email you the files. That way you can dissect the designs for yourself. There are only about a dozen tools, and one is a tape measure. (You said, “The picture would be much simpler…”. These illustrations are not drawings. They are screenshots. Making any drawing is a lot more time consuming. I will say, though, that I am considering investing in a white board so I don’t have to spend a half hour in a paint program to express something that I could draw with a marker in 2 minutes.

Anyway, here are some guidelines. For bottom of track to ground - about 8 ft. (Out of reach for kids and most adults, but can still fit between floors of an average building.) For the “slot” at the bottom of the track- this could be as small as 1.5 inches. (good to keep birds and hands/arms out) The wheel width – This is a tradeoff. More width equals longer rubber life, but pushes more air in the enclosed space. I think that 4.5” of flat space for each wheel is plenty, though. The electric – I don’t like the top center because someone might stick a metal broom handle up there or something. If the middle slot can be made to prohibit a straight object from reaching there, I would prefer the top, just inside of the hold down wheels. That or the sides, down low. (needs both sides for splitting track) The track inside height is all about speed and tire width. I wouldn’t go less than 20” though, if you want to go highway speeds. As for the switching, since it is done by the vehicle and not the track, you can go right up to the point of physical divergence, although you would presumably set your steering guide wheels and do all checks way in advance.

Juho Laatu said...

Some technical comments / thoughts in my mind.

- Ability to run to/through buildings is a nice property. How about the height of the top of the track? Should the whole track fit within one floor?

- High speeds are as well a good target. Some bogies might have also narrower or lower wheels (maybe allowed unless narrow wheels wear the track too much). Some light and fast vehicles might try to save energy with narrow (and hard) wheels.

- Also tracks could vary. High speed tracks could have more air space inside.

- So far I guess the wheels are not designed to turn in curves. I guess in the bogies of this article the left and right wheels are independent and can have slightly different speeed in the curves. If the wheels of some bogies can turn (to avoid friction in the curves), they could take the required "direction guidance" from the vertical surfaces of the track. Maybe not needed at all?

- In a switch the bogie could lift the left side wheels a bit when it turns right. That could make the ride smoother. Maybe not a requirement since also fixed wheels work quite fine.

- What is the mechanism that tells the bogie that a switch starts and the bogie can no more change direction. Is it mechanical or maybe electric. Maybe timing is designed for the maximum speed of the track and slowest allowed direction switching time of the bogies.

- Do you have a policy for long term standardization (after the project becomes popular)? Would the process be open to all, to some core team of neutral experts or mainly intended for commercial participants? Do you encourage or discourage creation of IPR? What kind of technical evolution of the specs (from version 1.0) do you expect? What interfecaes to specify? Will the working style be Internet based like now?

- I downloaded the program but I didn't generate anything useful yet. For most part of the normative specs basic 2D drawings may be enough. For illustration 3D works better.