Friday, August 12, 2011

127> Really, Really Fast

As anybody does much design work knows, you can always do better. Second guessing one’s own designs is something that is best not rushed, however. So here I am starting from scratch once again, with fresh eyes and a few different conclusions.

There are several considerations that motivated me to rework the bogie design. First, I think I put too much emphasis on a system that could use off-the-shelf tires, even at high speeds. This led me to motorcycle tires. Actually, though, what is the function of a tire? It is for vibration dampening, shock absorption, and traction. Since we are talking about running on smooth (finger-jointed?) steel, it is mostly just traction that we’re worried about. The problem here is that to achieve it, tires create a flat spot where the tire meets the road. Taking the wheel “out of round” in this way increases rolling resistance. In other words, it wastes energy. Any emergency stopping should undoubtedly be done by clamping the track, and no standard tread design is going to climb very steep slopes anyway.  Finally, I suspect that it would be easy and cheap to outsource, even in small quantities, a solid rubber tire designed specifically for PRT. 

Secondly, there is the matter of flanged wheels. I don’t like flanges for hard and fast use because I if they are of a hard material, they will make noise and vibration. If they are of a soft material, the area making angular contact will wear quickly. This is because a flange is in effect, a wheel with more than one diameter. Since any given diameter will make a wheel travel just so far per revolution, if wheel portions with more than one diameter make contact at the same time, one or the other must skid to compensate. Thus you have designed-in a wearing surface. Position-locking angular contact can be made with equal diameters however. Consider the example of a rounded pulley wheel on a square bar. There, two point contact can be made and, if the materials are hard, there is little frictional tradeoff. Anyway, I have softened my position to consider using flanges because they so simplify the mechanics involved. There are better plastics these days, (such as Dupont’s Hylene) and with large diameters and geometries that minimize load, it’s worth a look, even for continuous high speed applications.  
Another matter that I have been recently considering more is the matter of aerodynamics of the bogie itself. If the bogie takes up all of the room inside of a box beam track, then it must push all of that captive air in front of it. Any bogie design must take this into account, and obviously smaller is better.
The design shown fits into a track with and internal height of about 20” (500mm) high. This is where I may have gone a bit overboard. You see, I wanted to fit the wheels with a commercially available hub motors and although I have seen many Chinese offerings from companies I have never heard of, these don’t even come with technical data sheets and are hard to design around. Unfortunately, western motor manufacturers seem to only want to design for a very large customer base, and really haven’t tried to get into the direct-drive vehicle business, so I was left with a somewhat oversized British offering. 

Protean motors are very powerful wheel motors which are designed to fit on ordinary cars with minimal modification or loss of power. Since I don’t want to design track that is too small to transport people at speeds they have already become accustomed to, and I don’t want to design a system that will constantly require wheel changing, these 16” offerings seem like a reasonable top end, as far as rim diameter goes. This does, however, make it into one heck of a hotrod. 
The Protean wheel motors, you see, produce (together) up to 320 HP continuously. (240 KW) These motors ARE the wheels, of course, so there is zero drive-train loss. So the thing can pretty much go as fast as we want. (For comparison a Tesla Roadster goes 125 mph (0-60 mph in 3.9 seconds) pulling  a roadworthy steering and suspension system, a transmission, and a 450kg battery pack at “only” 288 HP. (185 KW) So we are talking fast. Very fast.  Note that a motor’s power draw is proportional to the work it does, not its potential, so you still use very little power while cruising if the vehicle and bogie are well designed aerodynamically. In the pictures these motors are seen in green. 

The geometry that I am exploring in this design centers around eliminating upper guide wheels by having the vehicle press against a “ceiling” within the track to eliminate tipping or derailing when turning off of the main track onto a fork. The steering guide wheels are angled and flanged to fit more compactly. These guide wheels could also be external to the track, something that I have avoided for noise reasons, but my fears may well be overblown on that issue. Anyway, I have shaved a few inches from the track girth and, well, made a rocket.


Andrew F said...

Hi Dan,

I read this a few times and thought of some comments.

I think you made the right choice going with the solid rubber tire. I was a little puzzled when you initially made that design decision. Another advantage of solid rubber tires is that it eliminates the risk of blowouts at high speed. It should be straightforward to have the vehicle detect skidding on the rubber tires and engage the emergency brake.

On the topic of the aerodynamics of the bogie, you could also consider venting the track to relieve the high and low pressure before and after the bogie, respectively.

It's good to know that there won't need to be a tradeoff in speed of the vehicles and keeping the guideway height to a reasonable level. Do you think solid rubber tires would be able to handle higher rpms better than pneumatic tires?

I also liked your idea of electromagnetic steering, which should help with safety and reliability.

My only concern with your change of the running surface for the wheels is that it now forms a channel. Could it gather frost in the winter and puddle as the frost melts daily, forming ice the next night? Also, how does going to a plate between two tubes affect manufacturability?

Dan said...

Dan the Blogger – Still responding when possible! (from the great north woods)
Thanks for your comments Andrew… I have indeed considered venting the track to relieve the air pressure… I’m still playing with dimensions, though, so no decision yet. As for the solid rubber, I can’t see that it would make a lot of difference if the tire is airless. The smaller the wheel, the faster the wear, period. One can argue the logistics of changing out the rubber more often against the benefit of a smaller track profile either way. It’s a judgment call. I think the real improvement is spreading out the “hold-down” wheels to prevent tipping. This eliminates the upper guide wheels, and gives space where it is needed structurally, where the top joins the sides.
I don’t know what you mean by a channel. Are you referring to the joints between the plate and the pipes that support it? In that case it is just a drawn that way because there is little sense in trying to draw a weld, which would fill it. (don’t want to confuse the reader into thinking I’m using complex steel profiles that don’t exist, when I’m just using flat stock and pipe.) If it was “stitch” welded, or recessed, I would fill the gaps with an epoxy filler. Then the joints would be ground flat.
My thoughts on condensation are that the inner wall would be designed to drip without it landing on the running surfaces. If the system is designed to go reasonably fast, there is a fair amount of heat generated by the motors. This can be directed at the running surfaces. There is also the possibility of a heating element, such as they use for melting sidewalks of keeping pipes from freezing, as it would be almost never used, except for very humid weather and rapid temperature rise, such as a winter morning fog burn-off with no PRT traffic.
One concern with the flanged wheels is the fact that the flange contact area is a slightly larger diameter than the rubber wheel. (creating constant slippage) This can be dealt with by allowing the flange to rotate independently, have BOTH guide wheels engage for general travel, (to center the vehicle to minimize flange contact) or just figure that it’s so close it probably won’t hurt anything anyway.

Aurora Lindstrom said...

Hi Dan,

my name is Aurora Lindstrom and I work for a NGO called the Institute of Sustainable Transportation (INIST). We are currently designing a multidisciplinary program for university students where they are challenged to work on designing and building prototypes of solar-driven PRT. I came across your blog and got very excited to see well-made images portraying suspended PRT vehicles. The reason why is that I am searching for images to use in our student brochure that show exactly that. Yours are the best I have seen and I would like to ask for your permission to use them. I can give you more information if you email me at Thank you!

Dan said...

Sure, Aurora, use them as you please!