Toe out for turn in

[raymondu999]

Hey all. A friend of mine told me recently that if I wanted a more responsive turn in, I should be setting up some toe out. When I think this through roughly, in rough geometry, it doesn't make sense to me. When turning, if you turn only so little, then one wheel will still be facing outwards, no?

If you think of it in terms of conventional geometry, when you turn more, then it as it's toed-out, the inside wheel will be turned more, and your two front tyre slip angles will be more akin to ackerman angles.

Isn't conventional wisdom more towards anti-ackerman angles though? I suspect setting up with a (static) toe out would wreak havoc with that, no? And why, in the name of the Lord, would toe out in fact help turn in? I've never actually experimented with that yet, so I don't have any experience with how it feels. But anyone know the theory behind it? JT perhaps?

[machin]

Mark Ortiz answered this question in the most recent Racecar Engineering magazine:

Essentially When the inside wheel turns in it creates a lateral force and drag force -both of which create positive yaw moment into the corner...

When the outside wheel turns in it creates a lateral force (which adds a yaw moment into the corner), but also a drag force which has yaw moment out of the corner...

So the toe out case removes the yaw moment out of the corner caused by the drag of the outside front wheel.. hence aids turn in...

[Jersey Tom]

I call BS on the whole notion of trying to use drag to 'yaw' into a corner. Maybe it's just my opinion, but (a) the longitudinal 'drag' forces are tiny in comparison to the lateral forces... sine vs cosine functions.. (b) the moment arm of drag forces (half the track width) is likely going to be significantly smaller than the moment arm of lateral forces (roughly half the wheel base). Point being, you will get the biggest effect by working on the true lateral component rather than longitudinal.

Will front (and rear) end toe in/out have an effect on "turn-in" ? Yes. Can run it in a simple dynamic sim pretty easily and see what it does. Whether or not you want front toe in or out depends on the rest of your setup and what exactly you're after.

[machin]

I call BS on the whole notion of trying to use drag to 'yaw' into a corner. Maybe it's just my opinion, but (a) the longitudinal 'drag' forces are tiny in comparison to the lateral forces... sine vs cosine functions..

Good point... and I also think "general rules" are dangerous especially with suspension where there are so many factors at play.... and yet "generally" toe-out causes better turn in... so something's at play here isn't it....?

Sines and cosines being what they are means that as slip angle increases the drag % gets bigger and the lateral % gets smaller.... so taking Mark Ortiz's yaw explanation you can see that you want higher slip angle on the inside wheel (where the side slip and drag forces aid turn in) than on the outside tyre where you want to minimise the drag which takes a little away from the turn-in yaw moment... I agree it'll be a small effect... and other factors may have a bigger influence on how you want to set the toe angle... but it might explain what people are experiencing?

[HampusA]

I added some toe out on my BMW to make it abit more "agile" or alert in the initial turn-in.
Works great and doesn´t affect the tire wear much, if anything.

[Jersey Tom]

and yet "generally" toe-out causes better turn in...

Does it? Says who? For that matter, define "better."

Honestly, a lot of this stuff just gets passed around and spread when one man's word or hand-wavy explanation is taken as gospel with no attempt to verify it. I will admit I was bad with this behavior myself some years ago. Regardless of what M. Ortiz, or C. Rouelle, or C. Smith or anyone says - the proof is in the data. Show me the money!

To get a real answer to this, all it takes is some concept of dynamic responses and maybe an hour or two in MATLAB or Excel or whatever. Make a simple vehicle model, simple (even linearized) tire model... run a few sims... it will become fairly apparent what the dominant effects are. Fairly straight forward in the same experiment to disprove that the slip angle drag thing has much to do with it.

[machin]

and yet "generally" toe-out causes better turn in...

Does it? Says who?

As I said in my post; I don't like general rules, however it is interesting to note that a lot of people agree with it: HampusA for one. Raymondu99's friend for another, Mark Ortiz for a third, the poster of the original question in Racecar Engineering for another.....

For that matter, define "better."

By "better" in this case most people mean "more responsive", i.e. more yaw moment for a given amount of steering angle. On a "twitchy" car "better" might be considered less responsive, but that's by the by.

So I made a very simple little excel spreadsheet calculating yaw moments against slip angle, and low and behold the answer came back that adding toe-out does improve the yaw moment for a given steering input... however I will caveat that slightly by saying it is heavily influenced by the slip angle vs. side thrust curve for the tyre.

Explanation below:-

For a simple model lets assume that the thrust generated by the tyre is proportional to the slip angle, so 1 degree equals say 10 units of force thrust, and 2 degrees equals 20 units of force thrust. Looking at some tyre curves this is approximaltey true at low slip angles, so is a fairly good assumption here. Also I've assumed that the vertical load on the tyres is equal at all times (not that accurate, but lets assume we have a really low c.g)

case 1: If both front tyres point straight ahead with zero steering angle applied (i.e. zero-toe) and that adding 1 degree of steering angle causes both the front tyres to have 1 degree of slip angle then they both generate the same thrust, but due to the effects described in earlier posts, the outside tyre contributes a little less to the yaw moment due to its drag moment being out of the turn, agree?

case 2: Now if we add 1 degree of toe out to both wheels and then add one degree of steering angle (causing 2 degrees of slip angle on the inside tyre and the outside tyre to be straight ahead) then the thrust from the outside tyre will be zero, and the thrust from the inside tyre is twice that of the inside tyre in the first case. Since the lateral force and the drag force on an inside tyre both cause a positive yaw moment into the corner then in this case the total yaw moment is higher than the first case, and the car turns in faster...

Agree? I know, its a bit simple, but it would explain what a lot of people are experiencing.... But I will add another caveat to say that there are of course other factors that will influence what the "ideal" toe angle is for a particular car....

[raymondu999]

Yes, but a part of the load would be transferred outwards at turn in, no? Adding vertical load to the outside tyre and decreasing tyre load from the inside tyre, and with it the friction forces & grip

[Jersey Tom]

Also I've assumed that the vertical load on the tyres is equal at all times (not that accurate, but lets assume we have a really low c.g)

I'd argue you've eliminated the most important part of this then.

Also, should be dynamic rather than steady state (which can still be done in Excel I think). Put those together and the major effects will drown out the drag bit by a wide margin.

Also, consider this: For "good" turn in (agile, precise, whatever adjective), in engineering terms do you want a big response or a fast response? If this were only about getting a bigger control moment gain, you could just throw in a faster steering rack. There's more to it.

[machin]

Also I've assumed that the vertical load on the tyres is equal at all times (not that accurate, but lets assume we have a really low c.g)

I'd argue you've eliminated the most important part of this then.

Its easy to criticise of course... providing an explanation as to why a lot of people claim toe out improves turn in is what we're here for and I'm more than willing to listen to an inciteful answer.... "because it does" or "people are wrong" doesn't really cut it with my mind... its too prevalent a belief just to dismiss it...

[Jersey Tom]

There are plenty of prevalent beliefs in racing which are BS. Someone has something they observe or rationalize, pass it along, and then enough people pass it along that it becomes gospel. Taking air out of the tires = bigger footprint = more grip... adding air = stiffer sidewalls = more response... jacking forces are bad... all BS. The more I do this the more I tend to dismiss things by default until it's proven out.

But I digress.

I was hoping that by trying to point the discussion in certain directions it would give enough of a hint for someone here to investigate the dynamics of it and arrive at an answer. I was very curious about this myself a while back and did just that, wrote a dynamic sim and took a look. Suffice to say, the load transfer aspect has a major impact on the yaw response of the car.

Not going to give away the specifics, but in my experience front toe out could be said to make turn-in "better" or more "precise" or "crisp" because you reduce response lag. As I said earlier, more an impact of speed of response than size. If big yaw responses were what you wanted for "good" turn in, then people would probably run heaps of toe out on the rear end. Anyone who has tried a setup to this end knows that depending on the platform, that can just make the car wild, uncontrollable, lazy, imprecise, etc.

[machin]

Not going to give away the specifics....

Why not? Isn't the reason you're posting on here so that you can help educate other people, help them with their questions etc? Rather than give cryptic answers and then pooh pooh people's attempts to work out what you're getting at? People will respect you more for that.

But I digress....

in my experience front toe out could be said to make turn-in "better" or more "precise" or "crisp" because you reduce response lag. As I said earlier, more an impact of speed of response than size.

Isn't that where we got to before; 1 degree of steering angle with toe out gives a bigger response than 1 degree without toe... if it takes you 0.1second to add 1 degree of steering then you've made the turn-in quicker (i.e more responsive to steering input)... it doesn't increase the maximum amount cornering force possible, it just gets you there quicker...?????

come on man, just help us out here and we'll love you for it!

[hardingfv32]

1) "in my experience front toe out could be said to make turn-in "better" or more "precise" or "crisp" because you reduce response lag"

Can you expand or provide any details?

2) "Not going to give away the specifics" Is this to protect a competitive advantage?

Brian

[Jersey Tom]

come on man, just help us out here and we'll love you for it!

I'm trying to.

With regard to respect on an internet forum I'm not too concerned. My main objective is to win races every weekend, this is a bit secondary As far as competitive advantage is concerned, this is just fundamental vehicle dynamics.. but I do get a little uncomfortable giving things away.

As far as dynamics are concerned, think about the dynamics of the whole car, the system, not just putting some steer angle into the front tires. The response of the CAR is what we're interested in.

My suggestion would be to make a car-level dynamic simulation.. ie with yaw and lateral dynamics. Can start with a 2DOF bicycle model and then pretty easily expand it to 4-corners with load transfer effects.

[machin]

I do get a little uncomfortable giving things away

I used to feel that way.... then I thought "what if I get knocked down by a bus? Who cares that I speant all that time learning all that stuff?"

However, if its company IP then I respect that. If its simple dynamics, why not let us in on the secret and we can all move on to try and work out solutions to some of the world's much bigger problems....?