Morelli Shape

[RobertSmalls]

Jim Bullis,

What's the full title and author of [Morelli 1983]? I'd like to see if my local university has a copy.

While searching for it, I did come across the title of this paper: Automobile aerodynamic drag on the road compared with wind tunnel tests (Book, 1981) [WorldCat.org]

Which is a reminder that unless your wind tunnel / CFD simulation is incredibly complex with rolling road, rotating wheels, air passing through the vehicle, flexible vehicle bodies, crosswinds, and weather, there will be differences between simulations and real world results. How close your simulation can get is a function of your budget. In any case, I'd expect a dozen good simulations to cost FAR less than one failed prototype.

My own efforts at low-budget CFD have been unfruitful. I've used COSMOS FloWorks to generate airflow visualization, but it's really not up to the job of estimating Cd. I've also tried ANSYS FLUENT, which is said to be one of the easiest CFD packages around. Maybe, but I couldn't get the mesher to import my models correctly. OpenFOAM is free, but you'd better have a little programming and engineering background before you use it.

Have you talked to any engineering consultants who specialize in CFD?

I think the concept of operating outside of ground effects is interesting, and a tandem Aptera would be really cool. I wouldn't dismiss solar car style vehicles out of hand. There might be some way to package a human in comfort and safety in a very thin body.

[RobertSmalls]

Neil,

I'll grant you that the Boxfish model is aerodynamically very good, but I wouldn't use it as a starting point. You should consider it to be a demonstration of the fact that tapering a vehicle in from both sides and the top gives excellent aerodynamic drag. Start with as little frontal area as you can get away with, and work your way back from there, ending with as small of a trailing wake as you can.

You also have to realize that a clay model with a Cd of 0.095 will produce a car with a much higher drag coefficient.

The blue model would not be roomy inside. I'd say it's the basis for a tandem two or inline three seater.

Originally Posted by NeilBlanchard

Can you point to *any* four wheeled street car with a Cd within even 50% of this?

Ford Probe V.

[aerohead]

Originally Posted by Jim Bullis

aerohead,

Hoerner's book is also informative but not all that free of the conventional automobile straight-jacket.

I think the Los Angeles is another airship along the lines of the 1906 Fuhrman (Prandtl's student) airship. They really seem about the same, and the only reason I specifically use the USS Akron zrs4 is that the testing reported by Freeman 1934 is by far the most complete and the best explained. It also was done at a scale where the actual test body was about the size of a car, so no real scaling is even needed. That report is free and very well detailed.

Jim,I think the Los Angeles is patterned after Jaray's LZ-126 Graf Zeppelin of which all post LZ-126 Zeppelins owe their design,including Hindenburg.
I looked up the Akron and for its Length/Diameter came up with a free-air,frontal area Cd of 0.058,with Cd 0.05 skin friction and 0.008 pressure drag.
If you did your body as small as my 1984 CRX ( around 42-inches ) the body would be 19-feet in length and would require 8-feet of ground clearance to nullify the drag increase due to ground proximity.
For comparison,a 1/2-"Akron" pumpkin seed form with skinny tires and 'normal' ground clearance would have Cd 0.15.
I don't know how to figure the lower portion of your car.The 1947 Taruffi twin=boom "Tarf" tunneled out at Cd0.10,so it's possible to make the lower section really clean.
I don't remember strut drag minimum ratios but they're tabulated.
The only cautionary note I'd like to share is on safety.If you'll remember,Bucky Fuller had a German industrialist aboard his Dymaxion Car who was killed when an over zealous rubber-necker collided with them,causing the roll-over crash.

[Grant-53]

There's lots of history to look at. There were two SAE papers by Dr. Alberto Morelli that were published and led to concept vehicles. The first paper led to the Pinifarina body for the Fiat 124 with a road going 0.17 Cd. Popular Science did a cover story on it. The second is the Aptera shape shown earlier in the thread and discussed breifly by Dr. Joseph Katz in his book on racecar aerodynamics.
Seating can be side by side, offset with passenger seat closer and about a foot behind, or in-line as the GM Infinity. In actual driving, practical ground clearance is at least 4" and seat height is 20". The newer Aptera shape will likely see use over recumbent bikes and scooters.

[Jim Bullis]

RobertSmalls,

I had this in an email from Prof. Eric Thacher (who is author of a book on solar cars that is also helpful):

"The Morelli paper is in the book Impact of Aerodynamics on Vehicle Design. It was a paper given at an International Conference held at the Café’ Royale, London , 16-18 June, 1982. The book is the proceedings of that conference and was published by Inderscience Enterprises Ltd., La Motte Chambers, St. Helier, Jersey, Channel Islands, UK , in 1983. It is a special publication (SP3) published “in parallel” with the Int’l. J. of Vehicle Design. The ISBN is 0 907776 01 9. The sponsoring technical society was the International Association for Vehicle Design. The Inderscience site is: Inderscience Publishers Ltd. "

[Jim Bullis]

RobertSmalls,

Your CFD experience is similarly frustrating as mine. I see no use in CFD unless it can match at least somewhere close to wind tunnel results, so all I even tried to do was to compute drag for the USS Akron shape using the CFD. Since results were not at all close to the goal, it had no chance of functioning in an optimization process.

It did however show general flow effects that helped with perception, maybe. I think I got a lot more insight from the old Freeman 1933 paper.

[Jim Bullis]

aerohead,

Zepellin and Jaray did not lead the airship shaping part of the game. Fuhrman in 1906 published a shape and drag calculation results that clearly set the standard, which did not get picked up until 1913 on the LZ17, according to my looking at the shapes of the early airships; where Zepellin made long cylinders fitted with nose and tail shapes that helped a lot, but the ideal drag form. Jaray might have 'designed' the LZ120 and even measured it in a wind tunnel, but this came some years after 1914 when he first started to work there.

In parallel with Zepellin manufacturing oriented work, activities by Prandtl and Fuhrman (student of Prandtl) took place at a more theoretical laboratory level. Prandtl is referred to in some of my books as the father of modern aerodynamics, whatever that means. He is also credited with the Prandtl wing used by British and German WWII aircraft. My old fluid dynamics book by Rouse and Howe 1960 shows Fuhrman reported data and includes that airship on the comprehensive plot of drag coefficients for various shapes. Data there shows Cd for that Fuhrman airship for various Reynolds number values, but generally these range around .05.

The NACA wind tunnel tests offer the best info I can find on the airship, and the USS Akron is the most thoroughly tested model, particularly the ZRS-4 model. Getting the exact Cd in our modern terms is not straightforward, since most of the data reported was in a different form of Cd where the coefficient is with respect to, not frontal area, but the cube root of volume squared. This gives numbers like .02 which can not be directly compared with the kind of numbers we talk about here.

I do not think there is any difference between the USS Akron and USS Los Angeles as far as basic shape goes. I especially liked the Akron data because it was done on a scale model of the airship, but the model is about the size of the car I am interested in, so there is much less concern about accuracy of the scaling process.

[Jim Bullis]

aerohead,

Sorry but I forgot to say that the Fuhrman shape was published in 1906.

[Jim Bullis]

aerohead,

Sorry about the previous un-necessary correction.

[Jim Bullis]

aerohead,

You said, "If you did your body as small as my 1984 CRX ( around 42-inches ) the body would be 19-feet in length and would require 8-feet of ground clearance to nullify the drag increase due to ground proximity."

I say,

Where the heck does that '8-feet' number come from? I might guess that this is a standard for wind tunnel size such that very accurate measurements can be made. I see that as a very different standard than the needed road clearance to minimize ground effect to a level that would not increase drag by more than about 20%. I argue for two feet clearance based on Morelli data in the book I gave reference for a couple comments back, as well as a variety of less direct indications from other publications.