Help with transportation superhighway
 

Hello, guys,

About a week ago, I was asking for help on a little project, to predict the ideal high-efficiency car. Well, I have labeled that job "complete" and I'm on to a new one. (The document from the last project was posted today in the general efficiency forum.)

New project: To define the ultimate public transportation vehicle. Something better than a train, better than a bus, faster than a plane for short distances, say up to 200 miles. The premise of the new vehicle is that it will operate in a "transportation superhighway," a network with distributed control, with traffic routed for maximum throughput and efficiency.

The vehicle is presumed to travel on a rail and be small enough to provide near-point-to-point transportation. I am imagining a body of revolution sized to carry eight passengers--

7.8 ft diameter x 35 ft long (D/L = 0.21?)
200 mph design speed
Up in the air, no ground effect. Re = 20 million? (from memory)

My question: I guessed that a drag coefficient (based on frontal area) of Cd = 0.02 should be achievable. Could somebody help with this, verifying or improving this number? I was unable to find a place to get the information.

Ernie Rogers

Ernie, you got your Cd figure wrong for sure. Maybe you meant CdA, since you mentioned frontal area, but even then 0.02 figure does not seem right.

Honestly, what's wrong with trains? I traveled to Italy last year, taking the train for more or less 200 miles everyday. There were trains every 15 minutes during peak hours and those things travel up to 100 mph with no traffic! And it was insanely cheap!

I agree, the number sounds wrong.

And I also wonder how another form a rail transport could improve on trains dramatically, since the cost will be much higher for multiple smaller vehicles and as far as CD goes trains will win over shorter vehicles because of the length.

My grandma says one of the biggest problems with America is that it did not take trains to the level of integration that Europe did.

The reason for that is that the automakers and big oil lobbied against railroads, and in favor of government building more highways.

superhighway
 

I did two or three posts on trains.The data was from Hoerner.If you can chase those down with the search engine,there might be some preliminary info to get you started.-------- Cd0.02 would be a passenger airliner wing.That would be hard to pull off.--------- The rail system,cross-ties,girders,and pylons would still create some ground-effect interference.----------- An exposed underside would aggravate lift,especially with a truly "light" rail system and the velocities you're talking about.---------- Since power varies as the cube of velocity,I believe there are already concessions made to speed,as there becomes a point where the cost of some velocity can not be justified in fares.---------- Since you are describing an unknown quantity,there won't be any real empirical data to guide you.--------- You may want to study roller-coasters.Coaster-based systems can operate mostly on gravity.Empty"trains" lifted with renewable energy.Passengers ride up to load from elevators run on renewable energy.Any 34-story building (there are over 55 cities in N.America with them) can provide the base structure which receives the add-ons.Once loaded,the cars accelerate to over 100 mph on gravitational pull alone,and can coast great distances when aero-maximized.Rolling resistance is already the lowest known,short of Mag-lev.The infrastructure requires a very small footprint across the landscape.Many possibilities.

If you have control over your lane, as with a railway, you don't need heavy crash protection. With less weight, there's less incentive for steel wheels, although they are wonderfully slender. In any event, adding wheels and a ground plane will have a lot of effect on the CD. To get those .0___ cd ratings, you need flawless shapes. Maybe our kids will think of trains as something that you can run your fingernails along without feeling anything until the little tick for the door seam, near the back, but it is a stretch for me. If fit and finish can save half our oil, it seems like a noble crusade.

Around here and in other major cities, there also used to be much more extensive public transit with street cars, but those tracks were torn up in the early '20's so more automobiles could be sold.

Hi, Tas,

You aren't facing the question. Do you have a source for drag coefficients of bodies of revolution in incompressible flow and high Reynolds numbers?

Access to a CFD program should allow calculation of a Cd, but that's more work than I would expect.

I calculated the Reynolds number:

Sea level
200 mph
35 ft long ------> Re = 65.5 million

Ernie

Okay, guys,

I think you have wandered off the topic. (It was aerodynamics of a high speed car without ground effect.)

But, let me join in. Here are CO2 emissions for different transportation choices, including the new "aerocar" high speed 8-passenger "tear drop."

Shinkansen Bullet Train....................100 gm CO2 /passenger mile
U.S. Train.......................................95
Transit Bus.....................................64
Prius (two people)..........................131
Airlines.........................................3 23
X Prize car (100 mpg, 2 people)..........59

New 8 passenger rail car...................13 gm CO2e /mile

/Ernie Rogers
Some of the data are from here:
"High Speed Rail and Greenhouse Gas Emissions in the U.S."
Center for Clean Air Policy and Center for Neighborhood Technology

winkosmosis -

I thought the highway program was also part of a Cold-War strategy to have a highway system that could move miltary equipment quickly and efficiently. This served the dual-purpose of favoring the automobile industry.

I also thought that the airline industry (Boeing, McDonnell-Douglas, etc...) was favored over the railroad industry for similar reasons.

CarloSW2

Still Waiting
 

I'M STILL WAITING--ANYBODY OUT THERE HAVE ANY INFORMATION?

(Correction on Reynolds number: 65 million)

Ernie Rogers

If you want "near point-to-point" you need either a lot of new rails, or two sets of wheels on this. Maybe you can keep the speed down on the road, to minimize the need for crash protection and weight. The optimum size will vary wildly, depending on location and time of day. If you build a rail system, please leave the vehicle options open, up to a set weight.

Good thinking
 

Ah, excellent, Bob,

Thinking is definitely appreciated.

Ernie

Hi Ernie,

Hey! If your going to fill up the train with people, then you got to fill up the Prius with people too. That makes for a 131/2 = 65.5 gmC02/passenger mile.

On your Cd question, yea, .025 might be doable. That is the Cd similar to a passenger aircraft such as a 787. The vehicle would need to have some sort of struts down to the track to support it high above the track to avoid compressibilty drag between the track and the body. The struts need to have minimun section area. Kinda like wings.

Thanks for the Cd
 

Hello, Donee

That's a good point on the occupancy. In which case, a train uses more fuel /makes more CO2 than a Prius.

On the Cd, so far it looks like I need to lower it a notch

Not sure if this is what you want but - http://www.2020engineering.com/pdf/2020eels.pdf

BTW that link mentioned the CyberTran system this you can find more about here- CyberTran International

information
 

The ideal teardrop form(about 2.7:1 fineness ratio) outside of ground effect has a Cd 0.04.I believe Morelli developed a low drag form of which Aptera borrows from.It was never tested in ground-effect.---- Any lengthening of the ideal teardrop body of revolution exhibits an increase in drag do to skin friction.I believe Cd 0.04 is the limit.Any environmental structure within close proximity will degrade the Cd.

From another angle that looks a a slightly bigger picture...

BTU per passenger mile (2006)
Cars............................ 3512
Personal Truck.............. 3944
Motorcycles................. 1855
Demand Response*........ 14301
VanPool....................... 1322
Buses (transit).............. 4235
Air (certified Route)........ 3261
Rail (all)........................ 2816

*Dispatched services

Source
Chapter 2 Energy - Transportation Energy Data Book

What's confusing me is why you want to ignore ground effect while having rail (I understand you're thinking raised rail - but losses from these stationary objects will be significant).

But that said, I concur with aerohead - I don't remember the specific source, but I have read of theoretical limits approaching .04.

Good job, thanks,

The CyberTran is just about what I had in mind.

Erine Rogers

Now we are getting somewhere
 

Good job, guys,

Now we are getting somewhere. My Mark's Handbook indicates minimum drag occurs at L/D = 5 and equals Cd = 0.04 at Re = 6 million. It further suggests that Cd rises at higher Re. But, the data are ancient and I'm not ready to trust them.

There are a number of Wortmann airfoils that reach Cd = 0.006 on chord basis, at around Re = 3 million. These published Wortmann sections are about 18% thick. But, I am pretty sure when I worked on this two years ago that I was able to reach half that value, maybe to 0.002 at higher Reynolds numbers using a commercial airfoil calculator. (Don't remember the name, I've got it here somewhere.)

I'm still hoping to find a recent paper or something that has reviewed the situation for bodies of revolution. My handbook is quoting 1930s data.

Ernie Rogers

Odd Numbers
 

Do you have a reference showing attached flow on such an abrupt shape?

According to "the Theory of Wing Sections" by Abbott & Von Doenhoff, the NACA 66021 has a cd of .0035, but the 66009, twice as fine, gets .003. Even the 63 series is only listed at up to 21% thick, although I think I did OK with a 64025 for a strut.

For a good example of balancing volume with frontal area, I'd look at the Zeppelins. It is unfortunate that so much data on shapes pertains to wings. The earlier NACA series 0010-35 shape, with 10% thickness and a continuous convex curve to the back edge, more like a zeppelin outline, got to .003 cd, but had trouble with pitch instability as a wing.

To modify a pure shape for running near the ground, the bottom is squished in proportion to how low it is. This produces a wing shape, but the lift can be cancelled by a bit of rake and the venturi effect underneath. The overall effect on drag is not as bad as the addition of wheel exposure. Successful LSR cars are not jacked up. One such HPV was made, but it embarrassed the builders.

Hello, Bob,

You seem to have more information at your fingertips than I have in my books.

You have been sucked in by a difference between cars and airfoils. Somebody correct me if I get this wrong--

First of all, airfoils are two-dimensional and cars are three dimensional. Here are the drag formulas for both:

Cars: D = Cd A 1/2 rho V^2

Airfoils: D = Cd L 1/2 rho V^2

L is the length of the chord line. So, D for an airfoil has units of force per unit length of wing.

The real problem is that the Cds above have different definitions. It may not be possible to translate from one to the other, but let's try. Suppose the car has a rectangular cross-secion of unit width. Then, you could replace the A in the car formula by H, the height of the car. Equating the two drag formulas above gives--

Cd(car) H = Cd(airf) L,

Cd (car) = L/H Cd(airf) For the airfoil, L/H would be the "fineness"

Okay, so how do I translate for a car with a circular cross-section? Let's say that the car and the airfoil have the same drag when they have the same frontal area. Then, I could write an equation--

Cd(car) A 1/2 rho V^2 = Cd(airf) L W 1/2 rho V^2

and A = HW.

W is the length of the wing, or the width of the rectangular car.

We get the same result as before,

Cd (car) = L/H Cd(airf) For the airfoil, L/H would be the "fineness"

Okay, you say there are airfoils of about Cd = 0.0035 for which L/H = 5 about.

So, Cd(car) = 5 x 0.0035 = 0.0175, close to my initial guess. Well, I have settled on the suggestion from (who said it?) and used Cd = 0.025 for analysing the super-efficient rail-car.

Thanks, Bob, your words have been very helpful.

Ernie Rogers

Instead of putting your theoretical future transportation system up in the air, run it underground, in a tube - a larger version of an oil or gas pipeline. Then you evacuate the pipeline, and air drag becomes moot.

A tube does not have to be buried to be evacuated. It could even be clear. However, even long, dedicated holes like the Chunnel have never traded the fuss for the savings.

True enough: my idea on that was to get it out of the way of other stuff, for what I assume is the same reason (but even more so) that the original concept was elevated on pylons.

As for trading fuss for savings, I don't think the original concept does that. I'm a great fan of trains in their proper place, but that place is where you have lots of people wanting to get from one place to another. Building all that dedicated infrastructure to move just a few passengers seems counter-productive, especially as you'd have to have some other way (roads?) to move heavy/bulky objects.

Ahh, if only we were starting from scratch, with lots of money, and helicopters to get started with. :-) If you connect any two points on earth of equal altitude with a straight, evacuated tube and a frictionless track, and let gravity power the cars, they go anywhere in 42 minutes!
Overall, I think that reducing the need for long commutes and freight routes is the way to go, but it might still produce branching or grid systems with high-volume corridors. To get the best features of cars and trains in one system, small units can be combined into short railway trains under automatic control. They can be private cars that also have road wheels, or dedicated public transit cars that enter from stations. These stations could also dispatch cars filled with a few pallets of freight, robotically handled for most of their journey. The pallets, in turn, might be loaded with small containers or standard, bar-coded boxes for courier delivery.
To keep the merging lanes short from the road system, the entrance ramps could have progressively flashing LEDs, run by the RR traffic sensors. Stay beside a green zone of "moving lights," and you can merge gracefully.
One of my favourite pipe-dreams has been a transit device divided into a double or triple decker, so that people can drive on and off in velomobiles, and relax in their own space while on board.

True, especially for the commuting. My 186,000 mps telecommute beats your 47 minutes :-) The problem with the rail system comes at the other end. When I leave my house (which itself is not where there's what anyone would call a lot of traffic), it's often to go to someplace like a lake or trailhead that might not get visited by more than a few people a week. Hardly economic to build rail there.

As you get less & less need for people to physically commute to job locations, though there'll be an overall decrease in traffic volume, a larger share of what remains will be trips with one or both ends at less-dense locations.

Something Bicyclebob said reminded me of this-
http://www.youtube.com/watch?v=rtrB8...eature=related
The pedal powered version-The Shweeb – World’s First Human Powered Monorail Racetrack, Rotorua

Abbott&von Doenhoff
 

What I take away from their work,is that the wing sections have really low Cds,but they've got all the induced drag associated with lift.The other thing is that with aircraft,the wetted area is equally if not more important than Cd.Lots of skin friction and much interest in laminar flow design.-------- As far as bodies of revolution,I did take another look at Mair's boattail work,and if you look at his data table,his data infers that he achieved Cd0.035 with his body of revolution,with prolate elipsoid nose,and one-diameter length boat-tail@ 22-degrees.I think this is the lowest thing I've seen for free air.Hucho has this in his book.-------- Ernie's 5-1 fineness ratio for low drag in ground effect seems to still hold true.VW has found it to be the "sweet-spot",also Mercedes-Benz,with their record-car projects.Mair's body in ground-effect according to Hucho's relationship,would come in around Cd0.70,and with ther addition of wheels,would final out in Cd0.12 territory.With extreme attention to the under-carriage,they might shave it to Cd 0.10,as Honda achieved with Dream-2.

Creating induced drag is the purpose of wings, with as little form drag and surface drag as possible. So, if you align a wing with the airflow, to eliminate induced drag (as in the Vomit Comet going over the top to produce weightlessness) you have a pretty well streamlined shape, with low enough form drag that the unavoidable surface drag is a significant part of the whole.

I goofed on my numbers
 

Bicycle Bob,when I posted last,I was shooting from the hip without my references with me,as is common for me.I got into Theory Of Wing Sections and found your airfoils.There was also a NACA 0010-35 that had Cd 0.035 @ a couple degrees each side of 0.At some time one of the pop sci magazines did an article about modern airliners and tossed around numbers as low as 0.002.-------------- In Schlichting's Boundary Layer Theory,he was big on NACA 63(4)-021,as it is a laminar flow wing,but it doesn't score numbers as low as the 66 series airfoils.--------------- Where I really missed the boat,is with Mair's numbers.I enlarged his graph and added resolution to the data points,and when extrapolating out to a full boattail,his body of revolution scores on the order of Cd 0.0225.---------- This conflicts with Hucho's portrayal at Cd 0.04.Perhaps the art reproduction in the book is off.I don't know.---------------- Going back to Aerodynamic Drag by Hoerner,the lowest drag he publishes,is for the ideal teardrop,at Cd0.04,which might take form as an airship or wingtip tank.A basic fuselage scored Cd 0.06.A 12% fuselage got Cd 0.066.15% fuselage got Cd 0.05.Gee Bee R-1 I estimate at Cd 0.044.37% teardrop Cd 0.04.Mair body of revolution with prolate ellipsoid nose an full 22-degree boattail an 39.6% dia/length = Cd 0.0225.---------- The best float I could find was Cd 0.056.Flying boat hull Cd 0.063.Torpedo Cd 0.35.50-kg bomb Cd 0.11.------------ The wing sections are slick,but its two-dimensional flow and packaging kinda weird.--------- I don't know what to make of Mair's form(secret-sauce streamlining).--------- If we take Cd0.04 as the minimum for three-dimensional flow in free-flight,at a fineness ratio of around 2.5 to 1 we get 5-to-1 due to ground-effect mirroring and Cd 0.08.Adding wheels,Cd jumps to 0.13.Fairing in the wheels and we get back to Cd 0.10.the theoretical minimum.I don't know if Ernie can break Cd 0.10.If you introduce any angle of attack,then induced drag eats into overall drag,and at 200-mph,which I think was Ernie's target,would stability not be a problem if lift is varying as the square of the velocity?

The wing sections were selected for types that had a fairly consistent center of pressure at varying angles of attack, to reduce the need for compensating forces from the tail surfaces. When designing for streamlining, the variable lift figure (side force in a crosswind) is an unwanted by-product, but it matters less just where it is centered; it is almost inevitably too far forward. We are now seeing blowover accidents moving from hydroplanes, to Funny Cars, to Le Mans, and side winds have always been a problem for streamlined HPVs. Since better streamlining almost always produces more potential for lift, I think it behooves us to move the masses forward as much as possible, and consider tail fins to restore directional stability. If you hit a patch of wet glare ice in a strong crosswind, you should get shoved over, not spun around.
The silver lining is that a crosswind can actually reduce drag. Someday, we might see rear-wheel steering to let cars angle into the wind like sailboats.

crosswinds
 

Yes,everyone seems to emphasize keeping the CG ahead of CP,and if you have to add some rear fin,to do it.Bonneville closes the course at around 7-mph crosswind.I think it's the same at El Mirage.Battle Montain in '05 had some wind we had to wait on.Varna is better up than down!

"Ecotopia" by Ernest Callenbach is worth your time. Bicycle Bob has similar ideas. See also Curitiba, Brazil.

Bob, you are a prophet!

Isn't this the thread I started about a super-efficient car? It was postulated to be attached to a rail. (Or rails.) Cross-wind lift is probably okay and should augment propulsion. (The lift vector has a forward component.)

This minimum drag issue is of such fundamental importance that I think somebody has to help us get it settled. It seems to me that a CFD program is the best way to find the answer. And, based on what we have seen here so far, a good study (though brief) should be suitable for publication in a journal. Any takers? Should be good for a master's thesis, don't you think?

Ernie Rogers

cfd
 

Ernie,a number of months ago,a member posted that MIT had solved the Navier Stokes equations for
the 3-dimensional coordinate matrix.If true,CFD would be able to predict actual 3-dimensional flow for any real form.I don't know if anything like this would exist outside the academic arena,however,it sounds like the kind of tool you might use to model your architecture.------ Wind spectra within an urban environment might play havoc with the dynamic range of results but short of building a scale model and subjecting to empirical testing,I can't imagine an alternative to the 3-d CFD.Perhaps you could apply for a grant to fund a graduate study.Let us know how things shake out and good luck!