unwinding steering lock
Discussion
Guys,
Back in Feb, I did my first trackday at Oulton Park which I thoroughly enjoyed and learnt a lot from.
The best £25 I have ever spent was to invest in some tuition, which helped immensley in going faster, safer and more in control of my car.
I have just started to read through the guide to tackling the 'ring, and one of the first things that jumped out was the mention of "unwinding steering lock after the apex".
It brought back memories of coming out of lodge corner and my instructor telling me to unwind the lock, get the wheels pointing straight, and getting on the power.
In the aftermath of adrenalin following my tuition, I forgot to enquire more into how this is actually achieved.
I think I understand the principle- no lock means straighter wheels thus allowing you to put the power down earlier. However, I always felt that the point in which I was being told to do this would mean that my car was pointing "off" the track and that putting the power down would have meant driving in this direction.
So my question really is am I right in my assumption on why it is neccesary, and also, how exactly is it achieved successfully and are there any deviations neccesary to enable it to be done in different cars (i.e FWD/RWD)?
Back in Feb, I did my first trackday at Oulton Park which I thoroughly enjoyed and learnt a lot from.
The best £25 I have ever spent was to invest in some tuition, which helped immensley in going faster, safer and more in control of my car.
I have just started to read through the guide to tackling the 'ring, and one of the first things that jumped out was the mention of "unwinding steering lock after the apex".
It brought back memories of coming out of lodge corner and my instructor telling me to unwind the lock, get the wheels pointing straight, and getting on the power.
In the aftermath of adrenalin following my tuition, I forgot to enquire more into how this is actually achieved.
I think I understand the principle- no lock means straighter wheels thus allowing you to put the power down earlier. However, I always felt that the point in which I was being told to do this would mean that my car was pointing "off" the track and that putting the power down would have meant driving in this direction.
So my question really is am I right in my assumption on why it is neccesary, and also, how exactly is it achieved successfully and are there any deviations neccesary to enable it to be done in different cars (i.e FWD/RWD)?
FWD experience here.
I had an instructor tell me the same thing. I ended up over the kerb and on the grass protection grid stuff.
I took it to be the instructors way of saying 'use ALL the track'.
I didn't need to be told that, so the instruction just led to a partial off.
Needless to say I have ignored the advice on that particular 'technique' since, and have progressed nicely.
3 years and 10-15 trackdays later I went out with another instructor, at the same circuit, and he did not suggest I needed to straighten the wheel earlier.
RWD may be different. The looser your rear end, the earlier you are likely to be winding lock off I guess.
I had an instructor tell me the same thing. I ended up over the kerb and on the grass protection grid stuff.
I took it to be the instructors way of saying 'use ALL the track'.
I didn't need to be told that, so the instruction just led to a partial off.
Needless to say I have ignored the advice on that particular 'technique' since, and have progressed nicely.
3 years and 10-15 trackdays later I went out with another instructor, at the same circuit, and he did not suggest I needed to straighten the wheel earlier.
RWD may be different. The looser your rear end, the earlier you are likely to be winding lock off I guess.
I don't know what line you were taking or what the instructor was trying to change, but perhaps he was trying to get you to smoothly blend the transition from cornering to straight line acceleration. What you'd typically aim to do is smoothly change from straight line deceleration to cornering to acceleration as you pass through the corner. Since the total grip is fixed this means that as the deceleration eases off you can corner harder, as you start to accelerate you have less grip for cornering and have to reduce the lateral acceleration. Since you'd come off the brakes and onto the throttle progressively, this naturally leads to a line through the corner which is parabolic rather than a constant radius.
You can get round the same corner by mashing the brakes in a straight line, taking the corner at a constant radius and then switching to acceleration once you have left the corner, but it's wasting a lot of grip. So maybe the instructor was just trying to encourage you to make the transition sooner and/or more progressively.
You can get round the same corner by mashing the brakes in a straight line, taking the corner at a constant radius and then switching to acceleration once you have left the corner, but it's wasting a lot of grip. So maybe the instructor was just trying to encourage you to make the transition sooner and/or more progressively.
It's not something you would be doing if you were subscribing to the ethos of driving on a constant radius line. Winding off lock in such circumstances would normally then involve a trip to the outside of the bend, unless one had a fair chunk of grip left to sort out the impending car/armco interaction.
I tend to teach a slightly different system. Getting someone to go slower into a corner is the first stage, so that they can balance the car properly on the power before they start to steer. They then keep on adding steering until they are on the line that they want to take through the corner, while maintaining the power. Then, once at the apex (which is later than the so-called classic racing line), you would then use additional power to drive the car out on the exit line rather than winding steering off to take you out on the correct line. As Green says, it tends to give a more parabolic line, although a mathematician (7db ?) would argue that it's absolutely not parabolic.
Why do I teach this? Firstly, from a vehicle dynamics point of view the steering is a blunt object and not that easily wielded. Twiddling the wheel makes direct changes to the slip angle at the front wheels only and the rear wheels then have to play catch up which can result in 'squeaky bum' time if you're in any way ham fisted. Playing with the throttle means moving a few bags of sand around the car and from a kinaesthetic point of view the visual feedback is much easier to learn. Secondly, it's quicker - a lot quicker - even for those expert track day nutters because you're using the grip in the right direction all the time. Thirdly, it works on every type of drivetrain configuration, be it FWD, RWD or 4WD; FF, FR, RR or MR.
The hard part of this style is normally:
a) convincing someone to slow down enough to be able to turn tighter at the start
b) convincing someone that you need more power (a lot more power) than you think you really ought to be using and getting it on vefore you start steering really makes a difference
c) convincing someone that because everything looks all relaxed and easier you really are travelling quicker.
I tend to teach a slightly different system. Getting someone to go slower into a corner is the first stage, so that they can balance the car properly on the power before they start to steer. They then keep on adding steering until they are on the line that they want to take through the corner, while maintaining the power. Then, once at the apex (which is later than the so-called classic racing line), you would then use additional power to drive the car out on the exit line rather than winding steering off to take you out on the correct line. As Green says, it tends to give a more parabolic line, although a mathematician (7db ?) would argue that it's absolutely not parabolic.
Why do I teach this? Firstly, from a vehicle dynamics point of view the steering is a blunt object and not that easily wielded. Twiddling the wheel makes direct changes to the slip angle at the front wheels only and the rear wheels then have to play catch up which can result in 'squeaky bum' time if you're in any way ham fisted. Playing with the throttle means moving a few bags of sand around the car and from a kinaesthetic point of view the visual feedback is much easier to learn. Secondly, it's quicker - a lot quicker - even for those expert track day nutters because you're using the grip in the right direction all the time. Thirdly, it works on every type of drivetrain configuration, be it FWD, RWD or 4WD; FF, FR, RR or MR.
The hard part of this style is normally:
a) convincing someone to slow down enough to be able to turn tighter at the start
b) convincing someone that you need more power (a lot more power) than you think you really ought to be using and getting it on vefore you start steering really makes a difference
c) convincing someone that because everything looks all relaxed and easier you really are travelling quicker.
StressedDave said:
As Green says, it tends to give a more parabolic line, although a mathematician (7db ?) would argue that it's absolutely not parabolic.
I thought about it, but then I didn't think anyone else would notice. It's one hell of a tight constraint on a curve to accuse it of being parabolic - not least on one that you'd hope was asymptotically straight... I think what we mean is a curve of decreasing radius.I ran the numbers on SDs assertions (when, young and foolish, I didn't believe him) and it's clear that - devoid of the human element - it is as worst not much slower than the fastest particle route through the corner. The rubber really hits the road when you have a human driver - the late apex allows you to get it less than perfect more often and that to have less effect on the outcome -- it's all very well trail-braking to the apex and accelerating out, but it's rare that a driver can get that right repeatedly and have room to deal with the chap coming the other way who can't do it either.
PS - on the OP's issue - the "unwinding the lock" comment is simply about making sure that as the power goes down the traction required reduces.
As you get to full power, the car starts to go faster, and so - on a constant radius corner - the centripetal (lateral) acceleration would be *increasing* towards the exit. This is uncomfortable, and feels to a passenger like you've got it all wrong and just put an extra armful of lock on to make it round the corner.
Look out for it next time you are passenger - a definite shove at the end of the corner. Nasty.
What should really be happening is that you should be gently decreasing the lateral grip required whilst increasing the longitudinal grip, and that means unwinding lock as the power increases.
As you get to full power, the car starts to go faster, and so - on a constant radius corner - the centripetal (lateral) acceleration would be *increasing* towards the exit. This is uncomfortable, and feels to a passenger like you've got it all wrong and just put an extra armful of lock on to make it round the corner.
Look out for it next time you are passenger - a definite shove at the end of the corner. Nasty.
What should really be happening is that you should be gently decreasing the lateral grip required whilst increasing the longitudinal grip, and that means unwinding lock as the power increases.
7db said:
StressedDave said:
As Green says, it tends to give a more parabolic line, although a mathematician (7db ?) would argue that it's absolutely not parabolic.
I thought about it, but then I didn't think anyone else would notice. It's one hell of a tight constraint on a curve to accuse it of being parabolic - not least on one that you'd hope was asymptotically straight... I think what we mean is a curve of decreasing radius.StressedDave said:
Indeed, and one where the decrease in radius (d²y/dx² for those who like their calculus unadorned) is not constant - I'm pretty sure it's inversely proportional to the remaining traction budget...
I think the actual shape of the optimum path is very complex to describe and I wouldn't like to say more than that typically it looks more like a parabola than a circle.It's actually reasonably easy to describe: http://mathworld.wolfram.com/PursuitCurve.html gives most of the maths behind the problem. Path optimisation has been quite a big field for vehicle dynamicists and race car engineers for quite a few years since processing power meant that models could actually be simulated in a reasonable amount of time and you could iterate. A bit of a far cry from 1992 when I first worked in the field and you could happily wait an hour for a 10 second data run.
StressedDave said:
It's actually reasonably easy to describe
It's been far too many decades since I did any serious maths but I don't think I would ever have described that as 'reasonably easy' and in any case it doesn't look to me as if that is actually describing the optimum path through a corner - I think that finding the optimum path will involve far more than is implied by that pursuit algorithm. I wouldn't even attempt trying to derive the optimum path mathematically. It would probably be possible to brute force it using a finite element simulation if you were willing to spend enough effort on it.
StressedDave said:
I'm pretty sure it's inversely proportional to the remaining traction budget...
Strewth. That's also an incredibly tight constraint to put on a curve - making it hyperbolic instead of parabolic. Swapping one conic for another doesn't seem to make the situation any better I'd settle for "decreasing".Continuously differentiable would be nice too, as we all like a bit of smoothness...
StressedDave said:
I'm an engineer though and brute force is what I understand, even though occasionally I miss the purity of mathematics.
As posted above - I think the interesting thing is not the optimal particle path (as given by the pursuit curves - and incidentally the target particle in the pursuit curve is the limit point for interested parties), but stability under small perturbations from the optimal path.
For the engineers in the thread, that's about how close the brute force is to the bodywork when you get it wrong.
I can see a bit of a struggle here, and good folk trying to understand the full contexts of the question.
An instructor maybe telling you to unwind lock, because he’s nervous that as you apply full power, you won’t cope with any oversteer that may raise it’s head!
But in a wider context, you would unwind the lock to enable a higher exit speed (using more power) by decreasing the turning radius. Shorthand for this maybe for a beginner, ‘use all the circuit on the exit’.
This would not include pointing off the circuit Red Zed as you point out, that would be ‘oversteer’ and would give rise to a whole fresh chapter.
In an effort to clarify ‘unwinding steering lock’ or why, I think we need to go back to basics. The first assumption here is that we are driving on a circuit, to go as fast as possible (within whatever criteria you’re applying).
Right. The fastest route between two points is a straight line. So if your start and finish line are A & B we are trying to ‘straight line’ any corners or corner combinations to enable us to maintain the highest possible speed, by keeping the throttle open the longest, least on brakes.
Dependant upon the power of the car, the tightness of bends and grip available (contours and cambers) means that ‘finesse’ is used. You would for example at a hairpin where you usually have a surfeit of power over grip, ‘give away’ a bit of entry speed in order to get the car turned early, use a ‘late apex’ allowing a wider shallower radius leaving the corner which in turn allows you to open the throttle and apply more power earlier, ‘unwinding steering lock’ as we go. And an overall faster lap time.
How much we can unwind steering lock will be dependant upon how tight the corner exit is and how much power you can use within the grip you have. Again, as the easiest analogy that we can visualise, maybe I could suggest the following.
Clean sheet of paper, describe a circle. Fully within the circle now draw a cross, a NSEWest cross. Now visualise this circle as being the tyre footprint (similar to the print of your shoe sole on the ground) on the ground. The entire weight of the car sits on these four footprints.
Go back to your circle and cross within it, the North/top end of the cross is your maximum braking grip available and the south/bottom end of the line is the maximum traction grip you have available. Likewise the extreme ‘west’/left end of the cross is the maximum grip you have available to the right; the extreme ‘east'/right end would be the limit of grip going left.
So now you can now see by following the route of the circle around the four points that when you are braking at the maximum of the tyre’s grip, it cannot give any grip left or right. Likewise, if your are cornering at the limit of grip available in either direction, you cannot then introduce any accelerative or brake forces, WITHOUT reducing the amount of lateral grip you have.
So demonstration over; if you are ‘unwinding steering lock’ you can apply more power, sooner, and be quicker.
Yes, I fully appreciate the variations and alternatives that all you clever people can now introduce. But this basic essential is what makes our high speed motoring almost impossible to get right and the challenge it is.
And there are a few further logical observations you can make from this simple idea. Like ‘managing’ quick laps by being ‘consistently’ quick maybe not necessarily the ‘fastest’ at any one point, rather being ‘closest’ to the extremes of that circle, for ‘longer’ may produce the required result.
I know the above is gormless, but so often, answers lies in simplifying the question and logic. For example, the road is only wet when it remains raining, …. !
I go lie down now. What was the question?
An instructor maybe telling you to unwind lock, because he’s nervous that as you apply full power, you won’t cope with any oversteer that may raise it’s head!
But in a wider context, you would unwind the lock to enable a higher exit speed (using more power) by decreasing the turning radius. Shorthand for this maybe for a beginner, ‘use all the circuit on the exit’.
This would not include pointing off the circuit Red Zed as you point out, that would be ‘oversteer’ and would give rise to a whole fresh chapter.
In an effort to clarify ‘unwinding steering lock’ or why, I think we need to go back to basics. The first assumption here is that we are driving on a circuit, to go as fast as possible (within whatever criteria you’re applying).
Right. The fastest route between two points is a straight line. So if your start and finish line are A & B we are trying to ‘straight line’ any corners or corner combinations to enable us to maintain the highest possible speed, by keeping the throttle open the longest, least on brakes.
Dependant upon the power of the car, the tightness of bends and grip available (contours and cambers) means that ‘finesse’ is used. You would for example at a hairpin where you usually have a surfeit of power over grip, ‘give away’ a bit of entry speed in order to get the car turned early, use a ‘late apex’ allowing a wider shallower radius leaving the corner which in turn allows you to open the throttle and apply more power earlier, ‘unwinding steering lock’ as we go. And an overall faster lap time.
How much we can unwind steering lock will be dependant upon how tight the corner exit is and how much power you can use within the grip you have. Again, as the easiest analogy that we can visualise, maybe I could suggest the following.
Clean sheet of paper, describe a circle. Fully within the circle now draw a cross, a NSEWest cross. Now visualise this circle as being the tyre footprint (similar to the print of your shoe sole on the ground) on the ground. The entire weight of the car sits on these four footprints.
Go back to your circle and cross within it, the North/top end of the cross is your maximum braking grip available and the south/bottom end of the line is the maximum traction grip you have available. Likewise the extreme ‘west’/left end of the cross is the maximum grip you have available to the right; the extreme ‘east'/right end would be the limit of grip going left.
So now you can now see by following the route of the circle around the four points that when you are braking at the maximum of the tyre’s grip, it cannot give any grip left or right. Likewise, if your are cornering at the limit of grip available in either direction, you cannot then introduce any accelerative or brake forces, WITHOUT reducing the amount of lateral grip you have.
So demonstration over; if you are ‘unwinding steering lock’ you can apply more power, sooner, and be quicker.
Yes, I fully appreciate the variations and alternatives that all you clever people can now introduce. But this basic essential is what makes our high speed motoring almost impossible to get right and the challenge it is.
And there are a few further logical observations you can make from this simple idea. Like ‘managing’ quick laps by being ‘consistently’ quick maybe not necessarily the ‘fastest’ at any one point, rather being ‘closest’ to the extremes of that circle, for ‘longer’ may produce the required result.
I know the above is gormless, but so often, answers lies in simplifying the question and logic. For example, the road is only wet when it remains raining, …. !
I go lie down now. What was the question?
ph123 said:
But in a wider context, you would unwind the lock to enable a higher exit speed (using more power) by decreasing the turning radius. Shorthand for this maybe for a beginner, ‘use all the circuit on the exit’.
<pedant>Increasing the cornering radius</pedant> 
I'm not a big fan of the friction circle either because of a couple of fundamental things:
1. The top of the circle is unobtainable - if you draw a line across the circle at peak acceleration that gives you the envelope you can really play with and the ability to use far more power than you would otherwise use.
2. There are some really useful interactions between power and cornering grip that skew the importance even further towards using less of the cornering grip and using more power to fill up the budget. But that's straying way, way beyond this discussion.
It's all so much easier with data logging, cause you can't hide behind the 'that's what I was doing' argument...
My take on the 'unwinding lock' comment is that when starting out on a track most people tend to 'choke' the exit because they keep the lock on. It could just be a remnant from driving on the road, where one is not used to using all of the available space and usually just steer through a corner to stay on your side of the road safely. Choking the exit is also a symptom of early apexing, which is also easy to do when starting out.
Hi All
I've been looking through some old threads and about 9 months ago
know a bit more about those interactions.
Over to you Dave on a new thread
Best regards
Martin
I've been looking through some old threads and about 9 months ago
StressedDave said:
There are some really useful interactions between power and cornering grip that skew the importance even further towards using less of the cornering grip and using more power to fill up the budget. But that's straying way, way beyond this discussion.
Maybe Stressed Dave could stray way ,way beyond that discussion now and let those of us who are interested in "using more power to fill up the budget" know a bit more about those interactions.Over to you Dave on a new thread
Best regards
Martin
Edited by Martin A on Monday 2nd February 09:11
Simply speaking, when negotiating a bend in a car you have a budget of grip to use which is shared out amongst cornering and power application (and braking as well if you start trailbraking...). What your instructor was referring to was that the exit phase of a corner is all about acceleration, and therefore from the apex onwards you should be geared up towards doing that, and not tightening or even holding a line. If you're running out of road, I would suggest that your entry line may be incorrect and the instructor should have pointed that out. Try and think of the whole corner in terms of the exit, because for 90% of bends that's where the lap times lie (oh, and its probably the safest way to corner to!!).
From the above, it is reasonably easy to work out that the standard track technique is to turn in as you release the brakes (using weight transfer to give the car the balance you want - note that this is very subtle in many cars so don't get too hung up about it), and then progress from turn-in to apex on a steady throttle (whatever throttle is necessary to maintain your speed). Once you reach the apex you can start opening up the power and unwinding the steering. If you've done everything correctly it should almost look like the throttle is connected to the steering by a piece of string - one comes in, one goes out. If you're running out of road, then change your entry to make the car point more down the straight at the apex.
Put simply: steady in, fast out.
Once you understand and can do all of the above, then you can start looking into more advanced theory etc, but to be honest, mastering and understanding the standard technqiue will get you 90% of the way there.
Lastly, watch someone like Jackie Stewart or Jenson Button driving on an in-car view and just copy that There's loads of videos out there on You Tube.
Try this one:
http://ca.youtube.com/watch?v=tCVSc1L6EUs
From the above, it is reasonably easy to work out that the standard track technique is to turn in as you release the brakes (using weight transfer to give the car the balance you want - note that this is very subtle in many cars so don't get too hung up about it), and then progress from turn-in to apex on a steady throttle (whatever throttle is necessary to maintain your speed). Once you reach the apex you can start opening up the power and unwinding the steering. If you've done everything correctly it should almost look like the throttle is connected to the steering by a piece of string - one comes in, one goes out. If you're running out of road, then change your entry to make the car point more down the straight at the apex.
Put simply: steady in, fast out.
Once you understand and can do all of the above, then you can start looking into more advanced theory etc, but to be honest, mastering and understanding the standard technqiue will get you 90% of the way there.
Lastly, watch someone like Jackie Stewart or Jenson Button driving on an in-car view and just copy that
There's loads of videos out there on You Tube.Try this one:
http://ca.youtube.com/watch?v=tCVSc1L6EUs
Gassing Station | Advanced Driving | Top of Page | What's New | My Stuff