steering

[mm davari]

can somebody tell me what does 100% or 200% akerman means? what about 50% , do we have?
i didnt face with it in books like RCVD

[mm davari]

can somebody tell me what does 100% or 200% akerman means? what about 50% , do we have?
i didnt face with it in books like RCVD

[RacingManiac]

It has to do with ackermann point and its relative ratio to wheelbase. 100% I believe has it located at the real axle(100% wheel base), 200% is at 1/2 wheel base, and 50% is at twice wheelbase, all positive, as in behind the front axle. This is assuming your tierod is straight I think.

[mm davari]

so, according to your post , i visited a website and that mentioned more akerman insteade of 50% and less akerman for 200% and true akerman for 100%.
thanks "racingmaniac" that was so helpful

[mm davari]

i have read RCVD steering section and there was mentioned for steering ball joints placment (front down and rear up of the wheel).other than understeer effect and reducing wheel kick back ,what is the effect of putting steerig BJ in this area (especially front down)

[^_^]

The ackerman percentage us basically the amount of toe difference between wheels at your steering geometry divided by the toe difference at geometric ackerman.

The projection of the ball joint and outboard joints is not correct. Do not use this method. When you project the points to the center of the rear axle, or wherever you want them to be, you are only finding the instant center of the linkage. This will change a great deal over the range. You need to do a kinematic analysis of the points over the whole range. You will be surprised how far off it is from the projection approximation.

[mm davari]

any team use front down steering lever?
it would be appreciated to explain why?

[fade]

packaging

[PatClarke]

MMD,
Ackermann is a steering geometry designed to get a specific steering response. It is the effect of ackerman that is important rather than the ackermann itself, that is, that the toe-out on the front wheels increases as the steering is turned. This can be achieved without using 'classical' ackermann geometry.
The position of your steering arms is immaterial as long as you do not get excessive undesired steering angles or bump steer (where the steering alters as the wheel moves vertically.
You need to design the whole thing in a suitable software program (Google for Bill Mitchell Suspension Software) and iterate it until you have the amount of toe out increase you desire with minimal undesired effects. Its not easy but worth while.
Tip, you will need plenty of toe-out when you turn, look at pictures of FSAE cars turning. You need the wheels close to parallel (a little toe out is okay) when travelling in a straight line.
I hope this helps a bit
Regards
Pat Clarke

[mm davari]

any suggestion for using tire data and make that informations usable for steering design.
any help would be appriciated.

[Hoosier Daddy]

Send $500 or 4,572,500 ریال to MRA and they'll hook you up!

Thats alot of cash to come by for a college team. Perhaps liquidating some of that surplus depleted uranium could help.

I keed I keed. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Garlic]

If you are building a car, make the ackerman adjustable.

Even in major racing it is one of the most changed variables... besides cambers and toes.

[mm davari]

is there any relation between steering and front wings?
how does that effect handling?

[PatClarke]

What do YOU think MM Davari?? Is there a relationship between steering and front wings?
Pat

[Jersey Tom]

Reverse ackermann is sexy.

[Garlic]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mm davari:
is there any relationship between steering and front wings? </div></BLOCKQUOTE>

http://www.f1tip.ch/f1/grafik/1997/9.jpg

Last I heard, they broke up.

[mm davari]

hey guys why you act so unfriendly?
i am not expert in aerodynamics. i just want know that is there any dynamic relation between front wing and steering or may be handling.
thanks for your help

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mm davari:
hey guys why you act so unfriendly?
i am not expert in aerodynamics. i just want know that is there any dynamic relation between front wing and steering or may be handling.
thanks for your help </div></BLOCKQUOTE>
MM,
There is a dynamic relationship between EVERY part of a racecar and every other part. If you fit a front wing you will need to make compensatory adjustments to maintain a car balance (maybe fit a rear wing!).
All racecar dynamics are reacted through the tyres, so that is where you should start your research. Look for tyre data such as the relationship between slip angle and vertical load, and you will be able to analyse such problems for yourself.
Regards, Ian

[mm davari]

is there any relaiton between understeer or oversteer and the location of steering lever ?
why ?

[Patrick W. Crane]

This is the way i have been looking at ackermann. (somebody correct me if i am totally of base so i don't screw up our car...)

design 1.
lots of extra ackermann. - your inside wheel turns "significantly" more than your outside. This does little to help latteral force but the drag created by that inside tire causes a moment about the rest of the car, helping it rotate.

Design 2.
Use reverse ackermann. Your inside wheel turns less than your outside and - based on the tire data that you have kicking around - you know the precise ideal slip angle for the given loads. So you design your inside tire to be at max lateral force along same as you would the outside tire. this would provide a higher maximum lateral force.

Design 1 seems to be much simpler and less sensitive to error, but seems a bit of a comprimise. If the data is available, designing a car to make the best possible use of the max available lateral force, would seem, to me anyway, that this would be the way to go. the only serious counter argument i can see is that SAE cars need to change directions so frequently and quickly that the extra help with pivoting would possibly out weight the lateral force gains.

For our car, as simple bracket allows a quick swap between the two setups. Testing will show which is better.

[Patrick W. Crane]

mm,

i would have to say yes and no. if you change the location of your rack, while keeping everything else constant, your understeer/oversteer charactersitics would probably change. chances are you wouldn't do that though...

[Tommo]

mm davari,

if there was no relation between understeer/oversteer and the location of the steering lever/rack/7th extra ball joint, and it fell in the woods would murpia hear it

the purpose of this ackermann business is to increase the grip/goodness and overall optimise the affect of the front end (assuming no rear steer)

if the grip at the front goes up and the grip at the back stays same...................

[awhittle]

Simple test requireing no calculators.

1) Put the car on the floor with driver installed
2) Turn the wheel to half lock
3) Get some salt and sprinkle it on the floor where the inside front tire will run over the salt
4) set a 12" square if flat metal on the salt so that it slides on the floor very little force
5) Slowly push the car over the plate

If the plate under the inside tire gets pulled in a lot, you may have too much Ackermann

If the plate stays in the same spot you are close to 100% Ackermann

If the plate gets pushed to the inside of the turn and away from the centerline of the car then the car will push any time the inside tire is on the ground.

Remember if you have a Torsen then you should carry the inside front tire at any time the car is coming off the apex so all this should not be an issue after the apex. It is a big deal on corner entry on the brakes when there is plenty of load on the inside front tire. At this point the inside rear tire should be unloaded an if you are running a Torsen and a single rear caliper then the rear brakes will not be working. 0*3.5=0

Hope this helps

AW

[Jersey Tom]

I disagree on your test conclusions, AW.

What kind of Ackermann you should be running is all about load transfer and the tires you have. Pushing a car around inside will not induce load transfer like a 1.5G turn will.

[awhittle]

Jersy

Well... Are we talking about weight transfer as the karters think about it. We realy need to start breaking this down. Ackermann is the differance in the angle of the outside front tire compared to the angle of the inside front tire. I would guess it is possible to have 0 Ackermann and still have tons of caster induced weight transfer. Or the other way around. I have a 125 shifter kart and it has tons of both. It is all a little clouded when you ad in slip angle (front and rear) and power induced oversteer and opposite lock.

See http://www.ncs-stl.com/Files/kart%20handling.doc

I wrote that a few years ago.

Hope this helps

AW

[Sathersc]

Well as near as I've figured it out, moving the rack will change the amount of ackerman your steering observes. I forget how and by what amount though.

Moving the steering joint connected to the upright in or out will change you from positive to anti ackerman, but will also change the ratio between steering wheel input and wheel turn, thus determining how quickly the car can change direction. I've only thought about it in positive ackerman, but I imagine that the absolute distance in/out from the UBJ changes your "steering ratio" works the same for anti ackerman.

For our car, I'm leaning twords regular ackerman for the simplicity, and for the idea of changing directions quickly at lower speeds/lateral forces. I'm either going to prototype with a steering joint mount for the upright that's adjustable, or make multiples, to test varying degrees of ackerman with our setup.

[JP Venturi]

This is a kind of esoteric topic to say the least. Carrol Smith said it the best that at the end of the day, it doesnt really matter anyways, lol.

However, my own personal interests will not jsut allow me to place a rack, put some links in, and settle on that.

From my experience on Pro/E using the kinematic analysis package, chaging just the longitudinal rack position will affect the progression of the ackermann effect (how fast the toe out happens).

Playing around with the out board points will tool around with the actual angle differences.

Now, in terms of waht all this means in performance..i have no idea. I have a few diff top uprights designed and ready to go on, plus an idea to move the rack around. General rule of thumb from what i've read is to run a little bit on the reverse side to compensate for g-accelerations

[mtg]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by awhittle:
Ackermann is the differance in the angle of the outside front tire compared to the angle of the inside front tire. I would guess it is possible to have 0 Ackermann and still have tons of caster induced weight transfer. See http://www.ncs-stl.com/Files/kart%20handling.doc

</div></BLOCKQUOTE>

Andy, sorry if this comes off as harsh, but some of the things you posted are incorrect.

Ackerman is not the difference in steer angle of the front wheels. 100% Ackerman is when at zero slip angle the yaw axis of the car is on a line collinear with the rear axle.

Less than 100% Ackerman is when the inside front wheel steer angle is less than the above angle. And vice versa. For those that went to an OptimumG seminar recently, if Claude included the K&C section, I put in the mathematical definition of % Ackerman at the end.

And it is very possible to have caster induced weight transfer regardless of the Ackerman on the car. With caster, one wheel goes up and the other goes down as they are steered- causing weight jacking (unless you have zero K-value springs on the car, which produces other sets of problems). Changing the Ackerman slightly changes the steer angle of each wheel relative to each other, adjusting the weight jacking, but it's always going to be there if you have caster and springs/chassis with a non-zero spring rate.

With parallel steering, you can set the car up with no steering induced weight jacking with 0 caster but with non-zero KPI.

As far as the kart handling article, the coefficient of friction of a tire is not constant! It is most definitely a function of the vertical load!

[BryanH]

Just my 2c worth, As some of you know Rotor is a designer/data engineer at ProdriveFPR V8 Supercar team. He told me that Ackermann is the least changed/smallest effect in front end setup. I think they currently run xxxx, The FPR Fords have run fastest lap in the last 6 races.
Cheers Bryan H
RMIT fan club

[awhittle]

MTG

I admit Ackerman is not simply the angle A - angle B I guess I should have made the sentence longer and made it include ...and the resulting instantianous center of rotation of the chassis when the slip angles are near 0. But.... I was trying to make the point that my simple plate on the ground test is close enough for most all of us. When the constently changing slipangles and the weight transfer generated by 2 geez of weight transfer and the operation of the rear differential all come into the equation, as Carrol Smith comments, this all gets a little overrated.

Re:CoF You ought to see some of the explanations of oversteer and weight transfer in the kart circles. It makes no sence at all. The few percent of CoF change that normal changes in sway bars will make pails in comparison to a chassis of a kart picking up the inside rear tire 3" and the resulting changes to the camber of the outside rear tire. Or the result of not unloading that inside tire due to not enough weight transfer due to the driver not getting enough load transfer by not driving fast enough.

The drawing below is exagerated but puts this into perspective how the Ackerman center can be a big distance from the center of path. Ackerman can be a big deal in corner entry and the car is on the brakes. Bump steer may be huge issue at this point. Most of my tuning is done with brake ballance for the corner entry stage of a corner. Once the car takes a set and begins the corner exit stage....
http://www.ncs-stl.com/images/ackerman2.JPG
AW

[Jersey Tom]

Importance of Ackermann steering as I see it -

To get the most cornering capability out of your car, you want each tire to generate the maximum amount of cornering force.

Cornering force is generated by slip angle. For a given tire load, if you plot cornering force versus slip angle it will start at zero, climb to some peak, and then fall off slightly before breaking traction completely.

Depending on what racing tire you have, that 'peak' cornering force will either be at a lower slip angle with a lower load.. or a higher slip angle with a lower load. FSAE cars pulling 1.5+ G in a cornering experience significant lateral weight transfer, and thus your outside wheel becomes very heavily loaded and the inside wheel very lightly loaded.

For tires that experience peaks at high slip angle under low load, you want your inside tire to steer more (to a higher SA) than the outside tire. Conventional Ackermann steering. For tires that experience peaks at low slip angle under low load, you want your inside tire to steer LESS (staying at a low SA) than the outside tire. Reverse Ackermann steering. You see this a LOT on F1 cars. If you push that inside wheel to a high slip angle it will just slip, push rubber, and generate drag rather than cornering ability (though some people say this is a good thing as the drag induces a yaw moment into the turn).

For high speed turns since the steering angle is low anyway, Ackermann is not a big concern. For low speed turns with high weight transfer, it definately is.

Since its a pain to push/drive a reverse ackermann car around at low speed with low weight transfer, the compromise often is something close to parallel steer.

It all depends on your car's geometry, how much weight transfer you expect to see, and most importantly your tire's slip angle curves!!

As far as I can determine from making simple motion models in Solidworks, there are two good ways of changing your Ackermann progression. Its all about the angle made between your tie rod and the line between your upright's lower ball joint and the steering joint. A right angle yields parallel steer. For cars with the steering pickup ahead of the lower balljoint, an acute angle yields progressively positive Ackermann, and an obtuse angle yields progressively reverse Ackermann.

You can change your Ackermann by moving your rack around, but it takes a LOT of movement to have a big impact on Ackermann. Easier to do at the upright.

[mm davari]

Anybody know about the reasons for choosing the lever arm position (up or down on the knuckle)?
is there any dynamic reason for these?

[LU-Bolton]

That would be one of the side effects after you decide how much ackermann you want. You actually have to find a height of your joint so that there is as close to zero bump steer as you can get. Very easy to do if you have a 3-d simulation of your suspension.

[mm davari]

thaks bolton
but i think akermman geometry is not depend on height of my joint. if it is,how?
where i can find a good resolutoin for height of steering arm on the nuckle?

[Ashley Denmead]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Halfast:
Just my 2c worth, As some of you know Rotor is a designer/data engineer at ProdriveFPR V8 Supercar team. He told me that Ackermann is the least changed/smallest effect in front end setup. I think they currently run xxxx, The FPR Fords have run fastest lap in the last 6 races.
Cheers Bryan H
RMIT fan club </div></BLOCKQUOTE>

I think jersey tom summed it up well regarding the effect of ackermann for different applications. I'de also like to add that if you drive a V8 supercar around bouncing off ripple strips because its the quickest way to get the thing to change direction and as a result you only have 2 wheels on the ground then ackermann has no effect at all and there is no point in changing it http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Jersey Tom]

With pole positions sometimes decided by .002s I'll take every pound of cornering force I can squeeze out of a system.

Highly dependent on track type though as I said. Fast tracks.. low steered angle, no big deal. Twisty, tight tracks.. much bigger impact. Take a look at the reverse Ackermann they were running at Monaco!

[kwancho]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:
As far as I can determine from making simple motion models in Solidworks, there are two good ways of changing your Ackermann progression. Its all about the angle made between your tie rod and the line between your upright's lower ball joint and the steering joint. A right angle yields parallel steer. For cars with the steering pickup ahead of the lower balljoint, an acute angle yields progressively positive Ackermann, and an obtuse angle yields progressively reverse Ackermann.
</div></BLOCKQUOTE>
I'm trying to make a Solidworks construction that'll help me look at Ackermann, and either I'm doing something extremely wrong, or I'm going to have to do some retarded things with the outer steering tierod point to get the Ackermann I want. Here are two pictures of it:

http://img237.imageshack.us/img237/4016/ackermann1nu6.th.png (http://img237.imageshack.us/my.php?image=ackermann1nu6.png)
http://img201.imageshack.us/img201/5139/ackermann2pv1.th.png (http://img201.imageshack.us/my.php?image=ackermann2pv1.png)

Sorry if it's hard to decipher. First, the UBJ is 23.75" out, 1.5" backward of the origin. The steering tierod is in plane with the upper a-amrs, and the outer steering point is 3" forward, and a ridiculous 5" out, just to prove my point. The angle between the steering tierod and steering arm (line between UBJ and outer steering point) is 39.71deg, very very acute. The steering rack is 4" forward, and the inboard steering points are 12.5" out from the center.
The second picture shows it steered. For the record, from my calculations, perfect ackermann for a 50" track, 68" wheel base around a 177" radius turn (hairpin in the autox) is 15.774 outside, 24.617 inside. Is that about right? Anyway, even with the obscenely acute angle, I'm barely at 50% ackermann. What gives?

Also, mtg, I've got the binder in front of me, and I can't find the mathematical definition of Ackermann. Is it just the delta = atan(wheelbase/radius turn) ~= wheelbase/radius turn? What about various percents? I would think that 50% would be halfway inbetween parallel steer and 100%, and then 150% would be parallel + 1.5* the difference between parallel and ackermann.

[Jersey Tom]

For perfect Ackermann, assuming your tie rods are parallel with the SAE y-axis (which is nice because it doesnt put the rack in bending and more importantly if you use bushings for guiding your rack extenders they wont want to bind!), drawing a line thru the kingpin axis and steering point should intersect with the center of the rear axle.

Check p713 in RCVD. Note Ackerman steering is pretty much good for low lateral acceleration only.

Not sure but I want to say 50% Ackermann puts the intersection of those kingpin - steering lines at 50% wheelbase. Parallel steer would be infinite % Ackermann, and reverse is negative. Actually I have a feeling that's wrong haha.

[Matt Gignac]

My view is somewhat like Tom's regarding ackermann, except 50% would be if the itersection would be at 200% of wheelbase in the rear, 25% would be 400% of wheelbase, and so on and so forth. I would define reverse ackermann as the case where the two axes in question intersect on wheelbase length in front of the front axle.

Matt Gignac
McGill Racing Team

[kwancho]

Alright, well, first of all, the rack is in the location that it is because the wheel is "swept back", which serves to shorten the wheelbase. There is the possibility of sweeping the wheel back forward without changing the hardpoints, but it seems that lenghtening the wheelbase just to get the ackermann that I want has more cons than pros

I think Matt's definition sounds good, but that holds true only if the steering tierod is approximately horizontal.
I've looked at RCVD, and it suggests reverse ackermann for high lateral accelerations because the peak force slip angle increases with lateral load, thus an inside, lightly loaded wheel needs a smaller slip angle. (By the way, figure 19.4 on pg. 715 shows that perpendicular tierod and steering arm = parallel steer) However, 1: our tire data not show a peak force slip angle, 2: some other teams have used very high amounts of ackermann (&gt;100%) to induce a drag on the inside wheel to help yaw the car.

Either way, I'm not sure what kind of Ackermann I want, so I'd like it to be adjustable. And I'm not sure extending the line of the steering arm to the center of the chassis holds when the steering tierod is angled so far back. Anyway, my question is mainly about why my geometry construction is so messed up. Even with something like 300% theoretical ackermann (by Matt's definition), I'm not getting the difference in the two steered angles that I expect. double you tee eff.

[Jersey Tom]

Yea Matt's definition is right, my bad.

Your tire data doesn't show peaks.. as I imagine it only goes to +/- 7deg SA. But from the concavity of each plot (varying normal load) you can tell where the peaks will be, roughly.

[kwancho]

Nope, goes up to 15. And still no peaks. This is the TTC data we're talking about. Very frustrating. It gets very flat. Either way, like I said before, I'm not worried right now about what's the theoretical best amount of ackermann.

edit: Actually, I'm retarded. It goes up to 10. Nevertheless, there's still no peak, and I'm not too convinced about the concavity argument..

[Paul Garcia]

So...

If I do consider the slip differences inside and out induced by weight transfer (I don't see how you could ignore this) where does that leave us? Reverse-ackermann compesates for this load transfer by steering the outside tire more to make up for its greater slip under greater load. Do we want ackermann regardless of this benefit? According to Milliken this is the solution for "high lateral g's" which is what I would consider our driving conditions.

Another clarification on tire slip angle. When I load my outside tire it is CAPABLE of generating greater lateral force for a given slip angle. But with two equally steered wheels (parallel steer) what happens? Are the inside and outside tires always going to slip at the same angle??

Milliken and Steve Smith talk about the outside tire wanting to slip more due to its lateral load transfer but it seems to make more sense to just say...

the outside tire develops more lateral force for a given slip angle, both tires always at the same slip angle...

I'm confusing myself at this point, can anyone shed some light...

[kwancho]

Paul, I'm not ignoring it. I just A: can't see it in the tire data, and B: am not convinced it's that big of a deal in our case, where an inside wheel has maybe 50lbs on it in a high G turn.

For parallel steer, the two tires will have different slip angles because they're traveling on different radii turns (significantly in the case of tight turns). The two tires do have different slip angles for sure. Having something like perfect ackermann will give you equal slip angles on both wheels, but like you've said, from a force vs. slip angle perspective, that might not be what you want.

By the way, weight transfer doesn't induce slip differences, really. Correct me if I'm wrong, but from a quasi-steady state perspective, slip angles are really a function of the body side slip angle (beta, for those of you familiar) and yaw rate. Once you know those two numbers, you can translate that to each wheel's slip angle (that also means you really only need one slip angle sensor).

[Alan]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Alex Kwan:
By the way, weight transfer doesn't induce slip differences, really. Correct me if I'm wrong, but from a quasi-steady state perspective, slip angles are really a function of the body side slip angle (beta, for those of you familiar) and yaw rate. Once you know those two numbers, you can translate that to each wheel's slip angle (that also means you really only need one slip angle sensor). </div></BLOCKQUOTE>

Maybe I misread what you just wrote and you will probably misread what I'm about to write since I'm not really good at explaining things. But slip angle is based on steer angle, which induces a yaw rate, from which you can caluculate a slip angle (and resultant beta) based on the wheelbase. What you have (beta resulting in alpha) doesn't make sense to me. But like I said, maybe I'm just not reading it right.

[Paul Garcia]

I don't know it would just be arguing which comes first, the chicken or the egg, but it does make more sense to say that vehicle slip angle and yaw rate are both products of how your tire slip angle and not the other way around. But it also seems reasonable to measure one vehicle slip angle (beta) along with the yaw rate and then back out the tire slip.

But I'm still confused. One thing Alex said made perfect sense, the two tires (in/out) slip differently based on the radius of turn. But is this slip angle difference merely a function of geometry and owes nothing to lateral weight transfer?

This is my understanding, maybe someone can shed some light on whether this is right:

Lateral weight transfer into a corner increases the outside tires lateral force "potential" or a higher lateral force for a given slip angle. As the slip angles in the tires build, the outside tire makes up a greater component of the lateral force generated by that track. They both slip near the same angle throughout the turn (which is determined geometrically by the radius of turn).

As the inside tire saturates the outside tire (may) still have potential. As the outside tire continues to slip it would leave the inside tire no where to go but to be "drug" to a higher slip angle. This creates drag and maybe that "ackermann yaw" people have discussed. Although would this not be a stabilizing yaw?? Is that good??

[Kerry]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Alex Kwan:
It goes up to 10. Nevertheless, there's still no peak, and I'm not too convinced about the concavity argument.. </div></BLOCKQUOTE>

No peaks? Maybe not in the tire coordinate system, but what about relative to the chassis? You could take a tire and turn it perpendicular to the direction of travel and drag it along, and it would still produce similar lateral grip to what it produces at normal slip angles (maybe 4-8 deg). The difference is the force is lateral to the tire, but in the longitudinal direction of the car. No help cornering there. Even after resolving the forces to the chassis coord. system they peaks are very small (out to 10 deg) but they are more pronounced (or appear vs. not being there before?).

[Jersey Tom]

Calculate how much load is on your inside front tire at 1.7G. That drag force wont be hardly anything. Get more out of the outside wheel!

The yaw moment generated by the increase of outside wheel force multiplied by half the wheel base will be more than the 'drag' force times half the track.

And hell, that drag force is not making you corner any harder. It is only (a) slowing you down, and (b) trying to spin the car.

Booo pro-ackermann

[kwancho]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Kerry:
No peaks? Maybe not in the tire coordinate system, but what about relative to the chassis? You could take a tire and turn it perpendicular to the direction of travel and drag it along, and it would still produce similar lateral grip to what it produces at normal slip angles (maybe 4-8 deg). The difference is the force is lateral to the tire, but in the longitudinal direction of the car. No help cornering there. Even after resolving the forces to the chassis coord. system they peaks are very small (out to 10 deg) but they are more pronounced (or appear vs. not being there before?). </div></BLOCKQUOTE>
Correct me if I'm wrong, but isn't the TTC data also measured relative to the "chassis" (the machine that rotates the tire)? If not, are you suggesting adding a cos(SA) term to all the lateral force data?

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> Originally posted by Jersey Tom:
Calculate how much load is on your inside front tire at 1.7G. That drag force wont be hardly anything. Get more out of the outside wheel!

The yaw moment generated by the increase of outside wheel force multiplied by half the wheel base will be more than the 'drag' force times half the track.
</div></BLOCKQUOTE>
Yeah, this is the argument that made me decide to put off ackermann for now. How much does 50lbs of normal force give you laterally? How about with plus/minus a few degrees of slip angle? Not a big difference.

[Marshall Grice]

Lots of ackerman does one thing for sure. It really screws up your A-arm packaging for a given minimum radius turn.

[Kerry]

I don't have the TTC stuff with me right now, so maybe I should be waiting to say this, but I think the tire data is given relative to the tire, not the machine. Somewhere in the documentation it does say which coordinate system is used, and I think RCVD may say something about converting tire data into the vehicle coord. system (Ch. 14?). If I'm wrong, then ignroe my previous post, but if the data is collected relative to the tire, then I would multiply by cos(SA).

[Paul Garcia]

Jersey,

Could you be a little clearer about how the yaw moment generated by the outside tire is effected by ackermann steering? The drag I was referring to would be caused by ackermann dragging the inside tire, which I agree seems like it might not be very much. But then you say boo pro-ackermann. I'm confused, maybe you can shed a little more light.....?

[Jersey Tom]

Gladly.

Starting from some baseline setup..

Going with lots of pro-Ackermann can drive the inside tire to a high SA such that it starts to drag/scrub accross the track, creating a force in the SAE -x direction. Does two things, slows you down and generates a more yaw moment equal to the force times half the front track.

Going with reverse Ackermann generates more cornering force at the outside tire in the SAE y axis. This does two things. Generates more cornering force allowing you to take a faster or tighter line, and generates more yaw moment equal to the additional force times half the wheelbase.

You get that additional yaw either way. But one way slows you down, the other lets you corner harder.

[Paul Garcia]

Jersey,

That makes perfect sense and is just the way I am interpreting it! But I think I/We may be missing something (??). All our teams have run Pro-Ackermann in the past with good success as have many at competition which is what makes me skeptical.

What are the disadvantages to Anti-Ackermann? I understand that with that setup the additional steer of that outside tire counteracts that outside tires greater slip essentially driving the car towards "parallel steer."

But other people are saying that for slow transients with low weight transfer (or even high weight transfer) that anti will tend to plow. For example, in a slalom it would seem your outside tire would have to push "through" your inside tire to corner hard through the manuever. Another thought was that Anti-ackermann was seen run primarily in high speed applications where stability was a problem...I'm not sure I understand this line of thinking.

Any thoughts on this?

[Bill Kunst]

Didn't Mr. Smith have something to say about this. Something about it didn't take much to figure this out when pushing a car in the pits. And then about the eventual conclusion being that ackerman might be good? I can't remember, but I think he wrote a few books, all that I do not have. Can anyone score some quotes on the subject?
Bill

[Jersey Tom]

Here's a good quote (well, paraphrase) from Carroll Smith -

Don't be afraid to challenge racing dogma!

Our team has also always run pro ackermann. There was no engineering involved. Check your design reports. Why did you run pro?

Just because it worked doesn't mean it cant work better. Trust no one, challenge and question everything. Theres another phrase to remember.

Carroll Smith has some words on the subject, but nothing really substantial. On the teams he worked with, running more ackremann helped. He did not use FSAE tires.

[Bill Kunst]

Like I said, I knew that he said something about it, and that it was positive. But, I also did not have the books here to look at.

Thanks, Tom, for being on top of the quotes.
Bill

[Paul Garcia]

How about Steve Smith who said "The modern race car suspension, no matter what type of vehicle, should not have any amount of Ackerman built into the steering geometry." That comes from Advanced Race Car Suspension Development which WAS published in 1976 but seems pretty clearly stated. So my question is, what's different now?

His arguement is the same as Jerry's, higher outside loads leads to greater slip from that tire.

[GSpeedR]

Running anti-ackermann on a FSAE car means you are throwing away performance in low speed turns. A vehicle must have a very high yaw rate in order to change direction fast enough to avoid understeer in a low speed turn, which I believe makes a larger impact on lap time than going slower through high speed turns. Look at the turn center for a low speed corner. The turn center is located further behind the front axle line relative to a high speed turn (because it's laterally closer to the vehicle). Any force vector projected in front of the turn center serves to yaw the vehicle OUT of the turn; likewise a force projected behind the turn center is yawing the car INTO the turn. You guys are looking at one snapshot of the vehicle in time, the vehicle's heading must change throughout the turn.

Also, the yaw moment of the Ackermann vehicle is significantly larger than that of the anti-ackermann vehicle (looking at front axle forces only), particularly if the track is nearly as large as the wheelbase. With the inside wheel steered nearly 45 degrees, the slip angle drag makes a significant contribution compared to the overall force. This drag is bad in high speed corners, but in low speed turns, FSAE cars have plenty of power to overcome the drag. Refer to the Ackermann series in Racecar Engineering.

The steer angles of a low and high speed corner are very different and highly nonlinear ackermann (static toe-in) can give the best of both worlds.

[GSpeedR]

Here's another quote since you are all acting so damn quotatsic: "...Speed and noise will make this fight less obvious, but while both front-wheels are on the ground the anti-ackermann car won't like sharp corners." -Erik Zapletal (Racecar engineering writer).

Milliken's book is great for understanding vehicle dynamics and kinematics but I'm pretty sure he's never made an autocross car. He worked with high speed racing cars and if you are designing your car completely around his book then you better start petitioning for a Formula CART series ASAP.

[Paul Garcia]

Doherty I think what your saying makes sense and I agree RCVD seems to be discussing higher speed applications than apply to us. I read the article your referring to and what I took away was the "run Ackermann" but I was unsatisfied with his justification.

Let me see if I understand. Are you saying that our tracks have turns that are tighter and so the necessary steer angles are greater. That would seem to reduce the slipping effect, say a 20 deg turn vs a 10 deg turn with 5 deg of slip. I say 5 deg (or so) because we should still be considering our PEAK tire cornering forces shouldn't we?

Also, bat this thought down if it doesn't make sense. Would ackerman be better in, say, a slalom? I would think that the outside tire turning less means it can transition to become the inside tire more quickly and easily when jumping back and forth...But even as I write this I doubt if that makes sense...

[Jersey Tom]

I still dont see any good explanation of why reverse ackermann is bad in a low speed turn or why the yaw moment is greater on an ackermann car.

[GSpeedR]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:
I still dont see any good explanation of why reverse ackermann is bad in a low speed turn or why the yaw moment is greater on an ackermann car. </div></BLOCKQUOTE>

1.) Look at a vehicle making a tight turn. The track width is now of comparable size to the turn radius which means that the front wheels follow significantly different paths (inside tire follows smaller radius). You have to remember that on most turns the turn radius is much larger than the track width and thus it's assumed that both front wheels follow a similar path (direction vector is the same). This means that when we look at slip angles of the front wheels for a sharp turn, the inside tire must be steered a lot more than the outside tire (as much as 10 degrees I've read) or the inside slip angle will be much smaller. The tire data we've seen does not show a significant variation in peak slip angle with varying load (definitely not close to 10 degrees) and thus it seems to be best for grip to have positive ackermann. So when we are trying to be clever by using anti-ackermann to account for weight transfer, for a low speed corner we acheive the same result by using LESS positive ackermann. Anti-ackermann geometry in tight turns produces a huge slip angle differential (too much) between the front tires: worse grip.

2.) The force vector for the inside tire of the anti-ackermann racecar is pointing ahead of the turn center. This forward component is similar to slip angle drag detracting from cornering force (it does not make you corner harder) but also yaws the vehicle OUT of the turn instead of IN. With high steer angle (low speed corner) this forward component becomes very large especially since the 1/2 track distance isn't negligible compared to the 1/2 wheelbase. We're not going faster we're just causing understeer making a slow corner even slower.

3.) Yaw moment on ackermann car...it would be easier to draw a picture but basically it comes down to force vectors. The inside tire is steered a lot the outside is steered less. Force vectors all point toward the turn instant centers which are behind the front axle line. The perpendicular distance from the CG to the inside wheel axis (when turned) is larger than the perpendicular distance to the outisde wheel axis. For our anti-ackermann friend the outside tire is steered more...the more the outside guy steers, the closer its force vector gets to the CG. I didn't realize this until I read Zapletal's article: Ackermann (from Denny Trimble) (http://students.washington.edu/dennyt/fsae/zapletal/Toe/TOEBYZAP.TXT)

IMO anti-ackermann means that on low speed turns we are producing less grip, giving conflicting inputs to the front end, and yawing the car out of the turn...all leading to more plowing than Farmer Jacob's mule.

[Marc Jaxa-Rozen]

A decently detailed cornering simulation with the tire data can be useful to highlight the effects mentioned by DohertyWins! when you take a look at the slip angles and corresponding vectors through a few different corner radii.

Honestly though, just looking at your tires after a pretty intense test session with anti-Ackerman will probably help the decision process more than calculations.

Marc Jaxa-Rozen
École Polytechnique de Montréal
http://www.fsae.polymtl.ca

[Jersey Tom]

Good points.

Why then I ask is reverse Ackermann used so aggressive on F1 circuits with very sharp, relatively low speed corners? (A la Monaco). They have a much higer ratio of track to cg height I'm sure.. but at 5G, theres going to be some serious weight x-fer regardless!

Being able to correlate steered to slip angle would be awfully nice. I would say without it you arent really able to do the force vector analysis to any degree of accuracy, and the best choice of action is to just be aware of the effects of Ackermann and tune it on the skidpad.

[Jersey Tom]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">For our anti-ackermann friend the outside tire is steered more...the more the outside guy steers, the closer its force vector gets to the CG. </div></BLOCKQUOTE>

I'll also bring up.. for a steady state turn, do you not want the overall yaw moment to be zero? A non zero yaw moment will cause your car to plow or spin. Of course thats all dependent on your cornering stiffness front to rear, as well as the weight distribution and overall lateral grip produced at the axle.

All a series of unknowns leading to more reason to not discard anti-Ackermann and just do some solid skidpad comparisons.

[GSpeedR]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:

Why then I ask is reverse Ackermann used so aggressive on F1 circuits with very sharp, relatively low speed corners? (A la Monaco). They have a much higer ratio of track to cg height I'm sure.. but at 5G, theres going to be some serious weight x-fer regardless!

Being able to correlate steered to slip angle would be awfully nice. I would say without it you arent really able to do the force vector analysis to any degree of accuracy, and the best choice of action is to just be aware of the effects of Ackermann and tune it on the skidpad. </div></BLOCKQUOTE>

I can't say for sure why F1 runs certain setups on their vehicles, but I can say that I would probably have a different suspension design if I had to deal with 3000lb of downforce. They have 200mph sweepers and a "conventional" suspension so I imagine they've struck a compromise. I don't know enough about F1 to give a reason.

I'm not saying you don't tune Ackermann, but you can get a better understanding of what you want to test or why it may be better just using theory.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">I'll also bring up.. for a steady state turn, do you not want the overall yaw moment to be zero? A non zero yaw moment will cause your car to plow or spin. Of course thats all dependent on your cornering stiffness front to rear, as well as the weight distribution and overall lateral grip produced at the axle. </div></BLOCKQUOTE>

We are talking about additional yaw moment from the front axle, the rear axle will react by changing to a larger slip angle as the vehicle increases its yaw rate. We are creating a oversteer condition but the car will not violently spin unless are rear tires are way past their SA limit (so the fronts should be near their limit). Otherwise there is no way that the vehicle can rotate about a turn as short as some of the low speed turns at comp. So your overall steady-state yaw moment depends on the corner radius (plus the assumptions of steady-state will start to confuse things).

[mm davari]

We always read and hear that race cars are so dependent on tire diagrams.
As you know it is so expensive to have our tire data but I found some that Avon distributes them for free. Now, any body knows how we can use these diagrams to adjust our Ackerman? (Of course if we want use these tires)