Ackerman

[Macs Stang]

Hi guys, I am new to this site. I am working on Minnesota State University-Mankato's '07 FSAE car.

My first question is about ackerman. I am using susprog3d to design the steering system. I understand ackerman, but from some of the information I have gathered it is confusing me as to the amount of ackerman that is good and bad.

When it is said to run 100% that means parallel steering. I have heard that between 70-80% ackerman is the goal. Now is that in front of the rear axle or behind it?

Thanks,
Tim

[Macs Stang]

Hi guys, I am new to this site. I am working on Minnesota State University-Mankato's '07 FSAE car.

My first question is about ackerman. I am using susprog3d to design the steering system. I understand ackerman, but from some of the information I have gathered it is confusing me as to the amount of ackerman that is good and bad.

When it is said to run 100% that means parallel steering. I have heard that between 70-80% ackerman is the goal. Now is that in front of the rear axle or behind it?

Thanks,
Tim

[Jersey Tom]

This has been discussed at length in earlier threads. Search em up.

You don't know what your car's optimum Ackermann setup is. If its pro-Ackermann, parallel steer, or reverse Ackermann. Build it to be adjustable.

[Macs Stang]

Right now I have the ackerman setup so that it is 70% of the wheelbase after the rear axle. I have a feeling that is wrong.

What are your thoughts?

[Jersey Tom]

As a baseline setup I'd reccomend a Pro-Ackermann setup. Inside wheel steers more than outside.

The axle thing only works if your tie rods are perpendicular with the centerline of the car. Its really about the angle made between the tierod and the line made by the kingpin point and the steering pickup. For steering pickup points forward of the kingpin point, acute = pro, right = parallel, obtuse = reverse.

You have no idea though what the best Ackermann setup is for your car and tire. Nor do I. Nor does anyone. You have to test it. Make it adjustable.

[jack]

Macs-

what you described sounds like parallel steer which would be 0% akerman.

tom-

know one knows what optimal ackerman is for there car without testing or really good lapsim. but on the other hand it should be quite strait forward to firgure out where you want to be within 30% or so. it would be retarded to make it adjustable from -100% to 200% or something. this shouldn't be too hard to firgure out if you have tire data too. and i you dont have tire data....what the hell are you building a raccar for...

[Jersey Tom]

Quite straight forward hmm? Well without tire data or cG height or logged lateral accelerations I'd say you have nothing to work with towards designing your Ackermann correctly.

This was discussed at length before, and given the force vectors and slip angles involved it is damn hard to analytically come close to an ideal Ackermann setup, if you ask me. I'm designing our car to be able to go fairly easily between reverse, parallel, and pro ackermann.

[kwancho]

jack looked like he was a little tired writing that post. We've had a conversation before that it's hard to draw conclusions from the tire data because the peak slip angle of our lateral force graphs is greater than the sweep TIRF ran through.

[jack]

i guess i wasnt clear about what i meant.

what i mean is it would be stupid to design a car where you could adjust your upper a-arm inboard points verticaly six inches to change your IC length from 0 to 200 feet. no one does that.

the same can be said for ackerman. just like you do with your IC lengths, you say "ok the goal of our vehicle is ___ so the overall behavior should be like ___ so our ackerman probably needs to be in the range of ____ to keep our slipp angles at _____ during a condition like _____"

to do that you dont need ackerman that is adjustable from negative something to crazy amounts of positive. if you are doing that, you have no overall goal for the suspension of your car. other than maybe geting lost in different setups.

it should be pretty easy to aproximate cg height and lateral forces too.

to paraphrase, at FSAE west i had to talk to Rouelle during our design presentation. he asked what our ackerman was. i said; "our ackerman is ___ because we were going for a behavior of the vehicle like_____ and we were constained by ____ so we ended up with ___" later i talked to Milliken and he also agreed that is the correct answer.

i dont think there is much engineering involved when you throw up your hands and say "we cant make an educated guess or better, so we'll just make is super adjustable and forget about it"

i think making it asjustable is a good idea, but adjustable from -XX% to XXX% is silly. 30% or so of adjustablility should be within the goals of your design.

[Buckingham]

Also, don't limit yourself to thinking that your Ackerman has to be constant. There is nothing wrong with having it increase or decrease as a function of steering angle if that is what you design for.

[Frank]

this is a rant

100% is at the rear axle
70% is behind the rear axle

ackerman will never be constant..

the position of your rack (for-aft) will dictate the rate of progression AND the magnitude of ackerman..

the steering arm angle (viewed from above) dictates magnitude of ackerman only...

most SAE cars use heaps of ackerman, I'd say +100% at full steer is good....

BUT, you need to include pretty serious steering angles on SAE cars, in case they throw a 180deg minimum radius (according to the rules) hairpin corner into the comp track...

+100% ackerman at high steering angles means a really short "effective steering-arm lever" (i made this term up, there might be a correct term for this) for the inside wheel, and may cause near zero scrub radius, which together do bad things if someone hits the brakes mid hairpin..

NEWBIES drivers do this... and i hit them when they do...

thats what ackerman does, by the way, is change the "effective steering-arm lever" WRT steering angle...

net result... unless you limit the steering angle somewhat, you can find it hard to get loads of ackerman....

and you need long steering arms to achieve both high steering angle and heaps of ackerman... or else complience bites you...

Frank's bush mechanic observations say....

"you know you've got enough ackerman when the inside wheel makes noise before the outside wheel" AND "and you can feel the loss of steering effort due to ecessive steering angle at the inside wheel first"

[Bill Kunst]

I think what donovan was saying is that you can either think of it as : "What amount of ackerman do I have at full lock?"(totally disregarding the progression) or "What is the ackerman curve that I have leading up to full lock?"

Ideally, one should be looking at the second if they want to do anything for design, as full lock is a scary place to be. If you disregard your ackerman curve, you may be missing some of the ackerman that you desperately need, or could do without.

Also, if you tell the design judge that you have 100% ackerman, they should ask you at what radius, and if you could procure any type of curve, as this could show your attention or neglect of the finer details in design.

Bill

[Jersey Tom]

Still dont see why you need 'heaps' of Ackermann, nor have I seen an explanation of why GP cars at Monaco use heaps of reverse ackermann.

[Bill Kunst]

Tom,
I think that if the GP cars were only doing a max of 75 mph, and had to turn around the radii that we have, that there ackerman philosophy would most likely change. I am not saying that they would go to a 100% ackerman setting at the tighest turn radius, but that you would probably see a ackerman setting similar to what a lot of teams are doing. I just think that if you are comparing our cars to a 200mph car and our turns of 15mph to the 120+mph sweepers that they see, you could be missing the boat.

Is there a team that has done significant ackerman testing, realtime or sim, that has any type of data regarding the effects of ackerman on steady state turning and the related cornering ability of that vehicle (g's of force or understeer produced)???
You don't have to give your results, I just want to know what teams have tested this on there car.

Bill

[Jersey Tom]

That's what I'm going to be doing this year (and probably this winter). And is why I want a WIDE range of Ackermann adjustment.

[Mike Cook]

Bill, our team has used our tire data to solve for ackerman. This has lead to much improvement in steady state handling. Steady state is one aspect to ackerman though. I suspect that ackerman will also play a large role in transient manuevers, much like toe does. This is often neglected and much harder to sim. We plan to test using a rack limiter and just doing a step input (steering) and measuring yaw response.

[Macs Stang]

Well I really have no tire data. I am using front steering. According to my suspension program I have the ackerman setup for 70% of the wheelbase after the rear axle center line. It is pretty much impossible to have ackerman infront of the rear axle with front steer due to the tie rods hitting the disk brakes.

How would the car respond with negative ackerman (crossing infront of the front axle centerline)?

[jack]

i couldnt get more than 84% ackerman because of the same problem (we run 10's too) if i could of, i think i would of gone for 120% plus...

what does reverse ackerman do? well, think about slip angle, steering angle, and weight transfer all at the same time. at one point i thought it would be a good idea, but for other reasons, i found what i ran to be a better choice to fulfill our vehicles goals.

[Hammock]

In our steering design, we use the tire data to analyze the forces at each wheel across the usable range of the steering. That is the kicker, what is the usable range?? We have been using a range of 17 to 125 foot radius thus far.

Similarly, it is the analysis of these same forces as they transfer from straight ahead under braking to the cornering radius at steady state that plays such an interesting roll in the transient. All I can figure is that it is best to create as smooth a curve as possible for this transition. I would much appreciate any ideas on this as well!!

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> I understand ackerman </div></BLOCKQUOTE>

lol

[Korhan]

Hey, we've got our steering set up and everything, but now I'm trying to fill out the spec sheet. How do you calculate Ackerman percentage? I've searched all over the site and couldn't find anything. Thanks for any help.

Korhan
Navy 07

[flavorPacket]

mitchell's new paper is what you want

[Korhan]

Thank you, where can I find it? I appreciate the help.

Korhan
Navy 07

[GSpeedR]

We had a good discussion of Ackermann and why you would want a bunch or none at all (including reasons why F1 cars aren't good examples), though that thread is somehow gone. Sucks to the person that deleted it.

[terra_dactile]

On the racecar enginnering website, their are some article (old files), three parts on steering geometry, written by eric zapatel formely know as Z on this forum back in the day, these articels bring up many interesting points about the need for ackermann or antin ackermann, not sur eif all the information is good as i am in no way an expert but its always good to read many different views on a subject to check for dicrepencies and keep an open mind,


Jude Berthault
ETS FSAE 2003-Current
Vehicle Dynamics Leader