Lift-off understeer

[5star]

Just found this discussion on another forum. StressedDave and Don Palmer contribute at the end.

http://www.porkers.co.uk/topic.asp?p=4& ... 72675&h=11

Page 3 describes exactly what I'm feeling:

A vehicle must have yaw angle in order to generate cornering force from the rear tyres, so when cornering the vehicle's longitudinal axis will point inwards from the tangent of heading (ie from actual direction of travel). This cornering attitude is often confused as oversteer. Sensitive drivers can detect this accumulating yaw angle within the linear range of the rear tyre slip curve, not only beyond it.

If I get on the power before I sense enough of this yaw angle, the power makes things worse and the car pushes forward.

However, once I've sensed the yaw angle is large enough, power makes things better and turns the car, all at legal speeds and well within limits of grip and vision.

When I drive, it is as if I initially need to prioritise building the yaw feeling by turning the steering wheel and only when it has built to a sufficient level can I then switch to prioritising power for the remainder of the turn.

I have no idea on how the physics are working. But I know what I'm feeling and the result it has from the driver's seat.

I understand why adding power turns the car - the drag stuff and direction of the forces. What I don't understand is why enough yaw needs to build before this happens, and why power is counterproductive before that point.

Feel free to ignore. I will ask Don Palmer about this when I book.

[5star]

Rather than worrying about the science try to feel what creates different states and how you created them. You don't really need to know the science.

This is how I've driven for 24 years, on and off the track. And I've always been satisfied with my level and the response of my instructors and examiners that have seen me drive. I did it all by "feel" without knowing the science.

I think it has come to a point where I need (and want) to understand the science.

[5star]

Trouble is, that quote is wrong. Just because someone has written on a forum (particularly PH) doesn't make it truth.

Yes, I don't doubt it, but to me, as a driver, with no Physics background, it resonates. I 'get' it and it describes a sensation that I can measure internally. And whatever combination of things is causing that sensation, it allows me to be pretty accurate.

I was really hoping I could describe what I'm experiencing, and somebody could say "Ah yes, I know what you mean, this is what you're sensing' and then give it a proper physics name or term.

You might do well to pay attention to this post in that thread.

To that end I really don't care whether the client is feeling the self-aligning torque, their guts walking about inside their bodies or the slip angle. The crucial point is that the client uses their senses to make meaning out of what’s going on.
When people focus on what their sensory system is telling them and they make meaning from this, the labels are as irrelevant as the graphs.

"the labels are as irrelevant as the graphs". Except when talking to an engineer, on an Internet Forum.

This post all started because I had trouble developing my 'sensation' in certain cars, and as a result I was needing to reduce my entry speed. Since posting that a few days back, I've figured out a way to "do stuff" in the car to maximise my 'sensation' and now the problem has gone away.

I really wanted to discuss it with you all, get some feedback, see who else had the issue and if there were other ways to fix it. But when i tried to explain what was happening (as I sensed it), all I got was dismissal because I wasn't using the right Engineering terms and my physics didn't make sense, or I was driving wrong, or I shouldn't be doing that on the road, or whatever. So I attempted to understand the physics so I could talk the language StressedDave seemed to want to, but all that seemed to do was annoy him.

So we went down that path, and StressedDave says "Road cars don't oversteer". But Chris Harris makes them. And the IAM "Better Driving" book says they do. And Mercedes Benz says their cars do. And when I drive, it feels like they do. Can you see the difficulty I've had with this?

If StressedDave would have said, "I think I know what you mean". Oh, and by the way, the strict SAE definition of oversteer means this, so in strict engineering terms it would be proper to say xxx instead. And that would have been a lot easier to understand.

Until that can happen, the labels ARE important.

[5star]

If StressedDave could answer one more question, I think I may have my answer.

1. Steering wheel is straight on approach to the turn. Call this 12 O'Clock.
2. At the right point, you begin to accelerate the steering wheel.
3. You may continue to accelerate the steering wheel, or it may reach a fixed speed, but it will remain in motion for a time.
4. The steering wheel begins to decelerate until it reaches X O'Clock.
5. You keep the steering wheel at X O'Clock for the constant radius part of the turn.

I need to relate that above, with this below:

To generate a tyre force at the rear you need either a sideslip angle or a yaw rate or some combination of both.

Something peaks at near the end of action 4 or perhaps a little way into action 5. Whatever is peaking is generated by actions 2 and 3. Also, there needs to be enough time for the something to build. If I wait in action 5 without adding power, the something falls off.

What is this something? Whatever it is, this is what I am consciously managing by the sensation I refer to.

[7db]

Sounds to me like you need to connect your mind to your grip a little more closely.

To use Don-thinking focus on what you are feeling.

You are pushing the car round the corner with your fingers, just as surely as if you were outside a tiny model with giant finger pressing on line of the front axle. The only lateral force comes from your fingers. Feel what that feels like. Make your grip on the wheel as light as the grip of the tyres on the road. Feel your fingers slipping on the wheel as the tyres slip on the road.

You then use your eyes to see if you're going where you want to and push the car onto the right line.

[5star]

OK, I'll try and give this a punt. The problem is that I'm not a psychologist and without sitting in a car with you, I've only got what you think you are feeling as mistranslated by the power of a keyboard.

Brill. I'm going to write some scenarios and describe in as much detail what I'm thinking, sensing and managing at each point.

Now the bit you're talking about (and it is as far as I can tell a slightly protracted description of turning the steering wheel) is the transient before the steady state bit of the corner.

Yes - changing from a straight line to a turn, changing from a turn to a straight line, increasing a steady turn to a tighter turn, decreasing a steady turn to a less tight turn etc. The transients.

All I can do is offer how I perceive how I steer...

This is really interesting. If we took 5 competent drivers, and analysed exactly what was going on in their head - what they are thinking, sensing, concerned about at any time, managing - I think the results would be quite different even though the end result would be the same car going through the corner at the same speed on the same line.

I'm going to write some scenarios - first by stating my objective in practical terms - and then go into detail about what is going on in my head for me to achieve the objective, using the sensations I feel along with my understanding of what happens when I make the control movements, to ensure maximum safety in the process.

[jont]

Brill. I'm going to write some scenarios and describe in as much detail what I'm thinking, sensing and managing at each point.
.....

This is really interesting. If we took 5 competent drivers, and analysed exactly what was going on in their head - what they are thinking, sensing, concerned about at any time, managing - I think the results would be quite different even though the end result would be the same car going through the corner at the same speed on the same line.

Doesn't your second statement tell you the first isn't going to be very helpful? I think you've reached the limits <rimshot> of where you can get to online. Why are you so reluctant to go and drive with Dave/Don?

And if you can get good results with some coaching - does understanding the nth degree of the underlying physics really matter?

[jcochrane]

Dare I introduce for analysis at this point the question of the science as to why, when driving a Nascar at Charlotte Speedway in America, it felt like it wanted to turn in when driving on the straight bits?

I think not as what matters is what I felt and what I did about it. I'm sure the science behind the suspension setup to induce this would have only confused me.

I know the science can be of interest but I'm not sure it can replace or is as helpful as the feel of doing it and what can be learnt from that practical experience.

[jcochrane]

Dare I introduce at this point the question of the science as to why, when driving a Nascar at Charlotte Speedway in America, it felt like it wanted to turn in when driving on the straight bits?

I think not as what matters is what I felt and what I did about it. I'm sure the science behind the suspension setup to induce this would have only confused me.

I know the science can be of interest but I'm not sure it can replace or is as helpful as the feel of doing it and what can be learnt from that practical experience.

Tyre stagger (differnt rolling radii left to right - think how a conical glass rolls) and assymetric loading. If you let go of the wheel it steers left.

Actually Dave that succinctly describes how the Nascar mechanics set up the suspension to create the left turn in. Fortunately I left it to them and just drove the thing but I have to say they seemed to be good at their job and what they did certainly helped to get round the 4 corners at speed. The American mechanics kept going on about "bump steering" like it was the holy grail of racing.

[TripleS]

Rather than worrying about the science try to feel what creates different states and how you created them. You don't really need to know the science.
If you are having difficulties with working this out for yourself there is only one solution and that is to raid the piggy bank and get some "on the road" professional coaching. If StressedD has any energy left, after this thread, he might be the one to see.

I see two alternatives here:

1. If you want to learn about all the technical stuff, and (perhaps to a lesser extent) learn to feel what your car is doing, and how you can get the best out of it, you could go to Stressed Dave.

2. If, primarily, you want to learn how to feel what the car is doing, and how you can get the best out of it, with an undertanding of the technical stuff being of secondary importance, you could go to Don Palmer.

Is that anywhere near right?

Personally, if I had plenty of money and was about 30 years younger, I'd be quite interested in either, or both, of those options, though I doubt if I'd make much sense of the intricate technical stuff.

[5star]

I'm going to write some scenarios and describe in as much detail what I'm thinking, sensing and managing at each point.

Before I do this, I'm going to describe one of the basic 'rules' I have in my head that determines whether adding throttle is going to make things better or worse.

When I'm driving, I have a picture in my head that the front wheels are taking a path, and the rear wheels are taking a path. In a turn, I imagine that the rear wheels are either taking a smaller radius, the same radius, or a larger radius than the front wheels.

If (in my mind's eye), the rear wheels are taking a larger radius, I know that adding throttle helps. The bigger the radius of the rear compared to the front, the bigger the effect of throttle. So, if the rear wheels are taking a tiny bit larger radius, I can safely use a lot of throttle to help the turn. If the rear wheels are taking a much wider radius, I need to be careful with how much throttle I use. In my head, I call this a state of oversteer.

If the rear wheels are on a smaller radius path, then using the throttle doesn't help, but it hinders the turn. I call this a state of understeer.

I'll come onto how I'm using my senses to judge how much understeer or oversteer I have in a bit, but first let me give a practical example:

I'm driving in a moderate circle at about 30 mph, and I'm on the limit of front end grip because I've accelerated very very slowly to this speed and used lots of steering. I can't accelerate (the front will lose grip), I can't turn more (the front will lose grip). But I now want to increase speed to 60mph on the same radius.

In my head, I'm picturing that if I can somehow get the rear wheels onto just the tiniest amount of wider radius than the front, that will be just enough to start applying throttle so I can start to use rear grip to turn and recover the situation.

So what I do is reduce the steering slightly (taking a wider radius) - in my head I'm picturing this is initially making things worse. As I steer the front of the car onto a wider radius, I've actually increased the state of understeer for a short time until the rear has caught up. So after relaxing the steering, I wait until the back has caught up and fallen into it's natural path behind the front. Now that I'm taking a slightly wider line, I've got a little bit of grip available at the front. Now I'll use this new grip to tighten the line again, and (if you remember my very pedantic and detailed description of how I turn the steering wheel), I've worked out for my car the point in time where, due to me turning the steering wheel, I've pushed the front onto a smaller radius compared to the rear. As long as I can get on the throttle at that point, I can ease the car from understeer back into oversteer and accelerate up to 60mph. If I get on the throttle too soon, the radius difference isn't large enough. If I get on the throttle too late, the radius difference again isn't large enough as the rear has had time to fall back in line. So unless I get exactly the right moment, I've missed my chance and throttle hinders rather than helps.

If you didn't follow that, I'll try another example.

Imagine a car simply going around in a circle. The rear wheels are following the front wheels, but depending on my use of throttle and the type of car, they will be taking a different radius (in my head). When I turn the steering wheel into the turn, there is a point in time where the front wheels are on a smaller radius than the rear wheels that is larger than the natural steady state radius difference would be. But if I stop turning the steering wheel and wait, the rear returns to the steady state original radius quite quickly. What I'm trying to do is apply my throttle at exactly the point where the rear wheels are on the widest radius compared to the front wheels, and through experimentation and practise I've worked out where that point is in relation to how I turn the steering wheel.

I'll come onto what I'm sensing to determine how much of a state of oversteer or understeer I have, and how I'm using those senses to determine available grip in each state. For now, here are some real life examples of simply managing whether I'm in a state of understeer or oversteer, but not going into detail about how much.

Imagine I'm approaching an 'S' bend on maintenance throttle. There are no other cars on the road, NSL, and good visibility through the bends. So my only limiting factors here are grip and the speed limit. I want to maximise the amount of fun and safety I have in accordance with these external requirements.

Now let's look at some of the basic rules I have depending on whether I'm in a state of understeer of oversteer (and just to remind, these are the rules in my head, rightly or wrongly), and the order of preference that I try and drive to when in a turn:

Oversteer while adding throttle: Helps turn. Safe up to the limit of rear grip. I can judge how much grip I have at the rear by feeling what the steering wheel is doing. DESIRABLE.

Understeer while removing throttle: Helps turn due to weight transfer, and slowing vehicle, but otherwise hinders tun due to increase drag. Improves grip at the front. No change at back (unless the car transitions to a state of oversteer, in which case see below.) ACCEPTABLE, proving the car does not transition to oversteer.

Understeer while adding throttle: Hinders turn. Reduces grip at front. Grip at rear cannot be utilised. I cannot accurately judge (via feedback) how much grip I have at the front - I'm just guessing. UNDESIRABLE.

Oversteer while removing throttle - can very quickly cause loss of rear grip and a spin. VERY UNDESIRABLE and all plans should be made to minimise this condition, or at least manage it very carefully.

So I want to build a plan to get through the S bend, using my priority system of rules above.

The plan will change depending on circumstances, but it's a Sunday, the sun is shining, I'm in a sports car and I'm in 'sporty driving mode'. So the circumstances right now are to maximise my 'g' sensation through the corners (for fun), and exit at 60mph.

So I build a plan based on my preference of the above rules.

1. The approach straight
I'm in a state of understeer here as I'm in a straight line and there is no radial difference in path between front or rear. At least, I hope not! If I don't brake, I have no way of actually knowing how much grip I have as there is no feedback, so I'm picking an entry speed on experience. I may decide to enter on trailing throttle, or maybe or maintenance throttle, it depends on what happened previously and how it affects my rhythm.

2. The first bend up to the apex
The beginning of this point is the transition from understeer to oversteer. I do this by turning the wheel, this moves the front onto a different radial path to the rear, putting me into an oversteer state. I then add throttle to catch and keep the oversteer state for as long as I want. If I remove throttle, then the car naturally returns to understeer.

I'm aiming to use as little steering as possible, steer at the right point which is dictated by the line. Once in oversteer state I use throttle to remain on line towards the apex. I have good feedback on grip now and I'm no longer guessing, but using real data.

3. The first bend from the apex to the connecting straight
At the apex I'm in an oversteer state with quite a large amount of throttle, but I have real data on grip so this is safe. Now, I need to think about the bend in the opposite direction, which requires me to transition from my current oversteer state to an oversteer state in the opposite direction. This requires me to reduce oversteer, transition through understeer, and then build oversteer again (the opposite way). So, what I'm doing in this part of the bend is reducing oversteer to a point of understeer for the transition to the next bend. Depending on how much grip I have at the rear, I can either start to relax throttle and use the car's natural tendency to return to an understeer state (I only remain in oversteer only as long as I'm applying throttle past a threshold point), or I can unwind steering to relax the rear and reduce throttle when I sense there is sufficient rear grip to do so. Depending on which method I choose affects the speed I'll be travelling on the straight.

4. The connecting straight
My oversteer should be almost gone by this stage, so I'm waiting for the mental picture in my head for the point when the rear wheels are exactly behind the front wheels. At this point, I'll add a little throttle to ensure the understeer state stays.

5. The second bend up to the apex
I repeat point 2, but this time in the opposite direction.

6. The second bend from the apex to the exit straight
As with point 3, I need to transition from an oversteer state to an understeer state, and how I do this depends on one thing: Do I want to be going faster out of the corner, or slower. Now on a track, the answer is always faster, but on the road, I may want to remain at the speed limit, or slowing down for the next hazard. So I will either unwind steering to return to an understeer state (while adding throttle as the oversteer reduces to increase speed), or I'll unwind steering, or I'll unwind steering followed by reduce throttle as we did in 3.

7. The exit straight
We should be back to an understeer state here.

I've probably missed a lot of stuff out, and this is a very long post. But to an observer, it should look somewhat like what StressedDave teaches - early, slow and smooth turning of the steering wheel along with correct amounts of throttle. Because everything is blended and merges into each other.

Now, the question I've got, that I want to discuss and focus on, is the transition from my understeer state to my oversteer state - I'm turning the steering wheel to alter the path of the front wheels and put them on a slightly different radius (in my head), and at the appropriate time adding appropriate throttle to catch and keep or amplify the oversteer state.

Please understand these are not jerky, on-off movements - it is all blended. But I'm aiming to get the peaks of the control movements in the right places at the right times.

I think what is happening, from what StressedDave has said, is that as I turn the steering wheel, the rear tyres are generating a slip angle, and it is only when the rear tyres are generating a slip angle does adding throttle help turn rather than hinder turn. (But I may be wrong). So in essence, what I'm doing is being consciously aware of, and managing, the rear slip angle through all phases of my driving, and choosing whether to accelerate, decelerate, move weight around, choose a speed - all based on that. But the picture I'm seeing in my head is the difference in path between front and rear wheels.

Now to be consistent, throughout this post I've only referred to the front and rear wheels being on different radius paths. But sometimes I'm picturing it as body slip angle, and using that instead of the radius paths. (But it is the same thing, because a body slip angle inwards is simply the rear wheels taking a larger radius path). When initiating a change from understeer to oversteer, I tend to picture the radius paths, but at all other times I'm picturing the body slip angle. Also, the faster I drive, the more I'm imagining a body slip angle rather than a different radius path.

So, what is the point of all this? At the end of the day, I want to be a better driver. I think that if I can understand how the pictures in my head, the senses I feel, and the physics all interact, I can optimise my system.

[5star]

Look, I said it a couple of posts before, I cannot make sense of anything unless either:

a) I'm in the car with you
b) You provide me with detailed recorded data of the car behaviour.

I'm good but I'm not a fucking wizard.

OK, so what if I said, don't read it with the intention of giving any advice or improvements, but simply read it to see if you can relate. How does it compare to what you're thinking as you're driving? How does what I've written make you feel?

Of course I understand that probably the vast majority of people just get in the car and drive without thinking about it. And for those people, they will likely read what I've written, and think "What is he on?". And for a lot of people, you'll think I'm wrong anyway, And for others, you'll think that such an approach is ridiculous and a waste of time.

But, if anybody gets it, thinks similar things, or thinks completely different things (but can see a similarity), well you sound like you're the sort of person I could have a drink with and talk about driving and stuff. If you don't want to post on the Public Forum, just send me a PM to say hello.

[5star]

I see two alternatives here:

1. If you want to learn about all the technical stuff, and (perhaps to a lesser extent) learn to feel what your car is doing, and how you can get the best out of it, you could go to Stressed Dave.

2. If, primarily, you want to learn how to feel what the car is doing, and how you can get the best out of it, with an undertanding of the technical stuff being of secondary importance, you could go to Don Palmer.

Is that anywhere near right?

I think what i want is for an engineer to 'get' what I'm thinking, and add the engineering knowledge to refine my model. So it is more a case of marrying the two rather than focusing on one or the other. Or, to put it another way, I want to know 'why' my model works, and how close to optimal it is.

[jcochrane]

Sorry 5star but I struggle with your interpretations. To mention just one starting point of your thinking I have not yet driven a production set up road car do anything but understeer at the front. So I have not experienced what you say about front end oversteer. As I said in a previous post the nearest I've come to that is driving a Nascar but that is bump steer achieved through varying suspension rod lengths laterally plus other things to create that effect. Not something you would want on a road car or racing on a road circuit as we have in Britain.

[trashbat]

Of course I understand that probably the vast majority of people just get in the car and drive without thinking about it. And for those people, they will likely read what I've written, and think "What is he on?". And for a lot of people, you'll think I'm wrong anyway, And for others, you'll think that such an approach is ridiculous and a waste of time.

I think I kind of understand what you're trying to do, and I admire anyone's appetite for finding things out, but from the start I've been trying to express that I think you're going about it in the wrong way.

This happens all the time, within driving and without. People are generally bad at determining their own training/learning needs - and this includes that self-awareness! - at least until they've been successfully around the loop, and even then the suitability of a successful method can be quite tightly bound to the subject.

This all gets a bit meta, but you have a clear top level objective of, 'sort out my ability to handle this scenario in a car', which is great, but then you seem to think your primary training need is to understand the technical detail and complex physics, possibly even without the prerequisites that underpin that (a physics degree?), and in textual form to boot?

Well, why do you think you're adequately equipped to self-determine your own detailed training needs here? Are you an experienced driving coach?

Now I'm not that person either, but even a casual observer can see this thread hasn't gone too well, so would you think about wiping the slate clean aside from that highest level objective, and delegating (at least a hefty chunk of) the training plan to someone who's well versed in it?

TL;DR: I saw a quote the other day that said, 'writing about music is like dancing about architecture', and I thought of you.