Does anyone here know about ackerman calculations?

[DMXF]

As part of a complete optimization of my suspension/steering alignment, I'm trying to determine % ackerman by toe change measurements, not geometric 2D drawings. I've created a curve of increased inside wheel angle vs steering input off my car and applied the equations shown near the beginning of this site: http://www.smithees-racetech.com.au/ackerman.html

I can calculate the R radius dimension from the outside wheel and then see whether the angle of the inside wheel is greater or less than what the equations show (equal to would be 100%?), essentially generating something like the curves shown on this site where I'm using a 0 deg Rear Slip line: http://students.washington.edu/denny...e/TOEFIG03.TIF The problem is this % ackerman varies with outer wheel angle, like shown in this last site where the R dimension changes with steering input. Does anyone know how to determine what the % ackerman is by this steering measurement approach? Is there a specific value of R at which 100% ackerman is defined?

[strictlypersonl]

You're looking for "a complete optimization" that does not exist. All configurations are geometric and mathematical approximations. In addition, a geometry that is optimized for low tire slip angles isn't correct for high slip angles. Ultimately you will have to make a choice/compromise between the two extremes.

[DMXF]

Bob - There are always compromises, so the term "complete optimization" may not have been the best descriptive choice. I'm just trying to come up with the best settings/adjustments within the constraints of what was available to SA back in the day with the original components. There are a number of key "knobs" to turn in the suspension, while most will effect other parameters, a couple of which will also effect ackerman. I know there are dynamic and tire dependent issues regarding ackerman, some of the directions gone will have to be best educated decisions based on various inputs. Determining what % ackerman the car has is just a first pass very basic understanding. So back to the question, how do I come up with the % ackerman from the measured data?

[Clois Harlan]

Morris this is right down your alley! Last time I discussed this was with Morris Clement and he was talking about sins and co sins, coeffecients, slip angles, and more math than I can remember.

Clois

[Tom Kirkham]

Dcmgt,

The original suspension bump steers enough that your Ackerman will change with ride height and roll. You need to get a handle on the bump steer front AND rear before you can do any meaningful calculations on Ackerman. Another problem you have is the rubber bushings in the rear trailing arm flexing under braking and acceleration. A good program for doing the calculations you need for the front is WinGeo3 by Bill Mitchell. His web site is www.mitchellsofware.com Note his software (nor anyone else for that matter) will not do the rear suspension for an original Cobra.

[Tom Kirkham]

Actually there are some programs that can model an original Cobra rear suspension but they are either very expensive or somewhat clumsy (Solid Works for example) in getting you the numbers you want.

[DMXF]

I understand those issues Tom. Being on the steep part of the ackerman learning curve is in large part a result of trying to correct the bump steer. SA bent the steering arms in to clear the Halibrand wheels and in the process they made bump steer worse and reduced ackerman. I have a set of original unbent arms I'm investigating using, in part because of a recommendation from the owner who races his original car in slaloms. Going to the straight arm will improve bump steer, along with other adjustments, but it will also increase ackerman, which so far looks to be in the ballpark of pretty good as is for general all around and higher speed use. It's sounding like the slalom car owner may like the straight arms in large part because he's operating more in very tight low speed turn environment where a higher degree of ackerman may be more beneficial. Another guy who races his car on the vintage circuit said he tried playing with steering rack movement a little and didn't like the on-track result of one move he made, so it's obvious one needs to have a good understanding of all the nuances of steering and suspension optimization and the ackerman is just part of the picture. Thanks for the computer program recommendation.

[Tom Kirkham]

Dcmgt,

Is the auto-x guy's initials BC?

You might want to look at modifying the rack...

[DMXF]

Yes to your first question.

Modifying the rack is probably outside of how far I would want to go, although there is an interesting aspect of that. I have copies of all the original computer analysis Bob Negstadt and the other guy at Ford did to design the Cobra suspension and they did a follow-on analysis after the car was built to determine how the suspension/steering should be optimized for racing. The main finding.......move the inner tie rod ends further inboard.

[Richard Hudgins]

DCMGT,

The Ackerman angles should be set to the course one is running.

Think about it. If High speed big radius corners are the norm then one uses low angles with the instant center very far aft.

Low speed really tight stuff moves things forward.

If the course is predominantly left hand then you move the instant center off center to optimize.

Actually quite simple in theory but a real ***** to make work on a circuit by circuit basis unless you have F1 type budgets.

My best advice to you is to just get rid of as much of the toe-steer as you can. (In particular the rear)

That will get you closer to a stable platform then anything else you can do. (Except for dampers)

And that is what you are after with this type of chassis.

Remember, a design dinosaur is after all just that.

A dinosaur.

Fun, mind you.

But there you are.

[Tom Kirkham]

Originally Posted by Dcmgt

I have copies of all the original computer analysis Bob Negstadt and the other guy at Ford did to design the Cobra suspension and they did a follow-on analysis after the car was built to determine how the suspension/steering should be optimized for racing. The main finding.......move the inner tie rod ends further inboard.

Wow. I am amazed that the reports are still around. Did you get them from BC? Do your reports have the first chassis that Negstadt designed or the latter chassis that Turner and Remington designed?

I came to the same conclusion. If you look on the KMP259 thread you will see that Morris changed his rack, for this very reason. Shortening a stock rack is not too bad on a lathe. It does however increase the minimum turning radius.

Also I believe that some of BC success in tuning his suspension came from following Colin Chapman's theory of "Any suspension will work, so long as you don't let it."

From what Peter Brock has told me I am sure that AC had a ton of steering racks left over from the 289 Cobras, and so the 427 Cobra was stuck with a non optimized rack. Actually it is amazing that Turner and Remington were able to make it work as well as they did...

[mickmate]

The "other guy" was Klaus Arning, it was supposedly the first suspension design done on Fords super computer of the day. I would concur that the whole thing is a conglomeration of available and not necessarily optimum parts (the whole car is) and the design is always going to be a compromise as to how it is set up.Some good discussion on the topic though.

[DMXF]

Tom, no the reports didn't come from BC. I've spoken with Remington about the big block car development and he said he had very little involvement.

Richard, thanks for the info. The 427 car design is actually alot better than most people give it credit. Just look at Bill Murray's results with CSX3032: The fastest replica Cobra (Tom Barnard) with all the new spaceship technology that recently raced at Road America in the '08 Kohler Intl Challenge under the Shelby banner only just matched Bill's lap times there in 3032, with the other ones (like Vincent Dean) around 6 seconds slower.

Even more telling, the top Corvettes running in SVRA (like by Peter Klutte) are tube chassis sillouette cars like modern Sprint cup stock cars, fixing all the shortcomings of the original vette, with different suspension attachment points, widened track, aftermarket brakes, modern top shelf aftermarket aluminum heads, more displacement, wider wheels, bigger front tires, etc than the Cobra and the best vette still only just beat Bill's best lap time at Watkins Glen by about a second......and 3032 is not optimized in a number of ways and is even handicapped by the SVRA forcing Bill to add 200 lb of lead ballast to the car, not letting him run webers or factory dual quads (without a venturi restrictor) and not allowing him a chin spoiler when the vettes are allowed.

Also, from a new sports car comparison perspective, Bill's pole lap time at the Glen in '07 was about 2:04 (old treaded bias ply Sports Car Special tires) and I understand the times for a new Porsche 911 GT2 is around 2:11 and approx 2:14 for a Z06. You can look at most aspects of the Cobra design and see how amazingly good it is, especially if you include the original design package of the aluminum FE engine, with only a few aspects need to be addressed - and even many of them were to be corrected in SA's race cars, like they released a memo specifying that owners move the steering rack around to optimize bump steer, etc.

[427 S/O]

A fella from spice engineering assisted with my scratch built suspension, I mentioned ackerman and he said forget it, it's not used anymore. He measured my frame and drew on my garage wall (to scale) how it should be built. All I can say is, it's like riding on rails.