Weight transfer mid-S bend

[Horse]

http://maps.google.co.uk/maps?f=q&sourc ... 04742&z=18

I travel through this 'S' bend most days, and feel uncomfortable with the sideways weight transfer during (or, I suppose, as a result of the switch from 'left' bend to right.

On many if not most 'S' bends there is a moment or three when the vehicle is travelling straight, allowing the 'outwards' pull (I'm deliberately avoiding centrifugal/centripetal type words ) to come back in line with travel.

But here - and it's narrower (do a 180 U turn on-screen and check how much of the lane width the following car takes up) and tighter than it looks - the left-right switch has no 'straight' in between, and as a result 'outwards' swaps sides after turning right.

It feels uncomfortable (to me), and I don't like it. The only option I can think of is to go slower - and I'm not exactly hurtling through them now!

Instead, what I tend to do is turn everso slightly earlier, then allow the force 'changing sides' to pull the car slightly left, back towards the left verge.

Any other suggestions/comments? [Other than 'get used to it', 'go even slower' and 'find another route' ]


As a matter of interest, coming the other way there's a lovely set of 'observation links':

Note roof or two just visible in the trees (remember the camera is well up on top of the Google car)

http://maps.google.co.uk/maps?f=q&sourc ... 01186&z=20

Hidden, now, in the undergrowth . . .

http://maps.google.co.uk/maps?f=q&sourc ... 01186&z=20

Entrance just around the blind left bend . . .

http://maps.google.co.uk/maps?f=q&sourc ... 01186&z=20

[Gareth]

What is your entrance speed for the initial bend, compared with your speeds through each of the bends?

Have you tried starting a bit slower allowing positive acceleration into the first bend, lifting slightly as the second bend becomes visible, then a second lot of positive acceleration into and through the second bend?

[Horse]

What is your entrance speed for the initial bend, compared with your speeds through each of the bends?

Have you tried starting a bit slower allowing positive acceleration into the first bend, lifting slightly as the second bend becomes visible, then a second lot of positive acceleration into and through the second bend?

I can see what you mean - and, trust me, I'm not going into the sequence particularly 'fast' - but there are a few other factors. Edit: it's a 60 limit, BTW. Not that I would try at that speed

One - which I've only realised after looking at it in the Google satellite view - is that is a 'short sharp' left into a sharp but opening long right. There is no 'slack' between seeing the left and needing to be taking action. Well, not at even 'sensible' speeds, anyway. What I [think I] do is set speed for the entry of the left, enter under power, then add drive before I start the right.

Second point is that immediately prior to the bend is a junction to the left. 'Undue' slowing their could give the wrong message to anyone waiting to emerge.

Like this: http://www.cooperbiketraining.org.uk/ne ... emloop.ppt

[Gareth]

I'm not going into the sequence particularly 'fast'

I think it was more the relative speeds I was asking about.

it's a 60 limit, BTW. Not that I would try at that speed

I should nip over and try it out - it's not far away, and anyway I remember the road from when I lived in Bracknell and worked at ICL - because it's easier to discuss something from recent knowledge.

[it] is a 'short sharp' left into a sharp but opening long right. There is no 'slack' between seeing the left and needing to be taking action.

There may be no 'slack' but it might work better if you put some in. If you're still accelerating as you pass from the left to the right, you may not be able to accelerate enough in the right to make it comfortable.

[Horse]

I'm not going into the sequence particularly 'fast'

I think it was more the relative speeds I was asking about.

I knew that

it's a 60 limit, BTW. Not that I would try at that speed

I should nip over and try it out - it's not far away, and anyway I remember the road from when I lived in Bracknell and worked at ICL - because it's easier to discuss something from recent knowledge.

Try it when it's quiet, see below.

[it] is a 'short sharp' left into a sharp but opening long right. There is no 'slack' between seeing the left and needing to be taking action.

There may be no 'slack' but it might work better if you put some in. If you're still accelerating as you pass from the left to the right, you may not be able to accelerate enough in the right to make it comfortable.

Wouldn't accelerating more increase the outward 'pull'? i.e. Same radius at higher speed.

I meant 'slack' as in a short straight section where the body roll can catch up and come back into line. The problem is where that 'lag' comes back when the car is already heading 'right'.

16:20 this evening, and as I approached there was a car waiting to turn out of the junction, another waiting to turn across in, and another car following me - so I had to choose a sensible speed which covered all options (see previous post).

Entry speed was about 36, up to about 38 as I was into the right hander.

Speed was set prior to the junction; there's no real room after to adjust and, as I said, there was a car following so I didn't want to brake where that driver wouldn't - probably - expect.

Throttle on to take it through the left, slightly more applied into the right, but as I was a. concentrating on getting around and, b. checking the speedo, I didn't note exactly where - if - I altered the throttle.

[Horse]

Pt 2

Have a look at this view:

http://maps.google.co.uk/maps?f=q&sourc ... 00493&z=18

Note that the bend starts alongside the junction, and the 'switchback' is visible through the trees.

[Astraist]

Slower entry speed and perhaps a different cornering line where possible might give better "cornering stiffness". A decisive turn-in motion when going into the S-bend can make the car "take a set" faster, making it easier to change direction quickly afterwards. And I believe that is the main problem. The input should be as quick as possible that can still generate a smooth output.

A forward weight transfer might be beneficial during turn-in, but I am uncertain as to it's benefit during the change in direction (particularly that you care for it's feel), so I would recommend maintaining throttle, but not accelerating, as this will generate understeer for you to overcome, which will give you that sensation.

Unless the steering inputs required are very large, fixed-input is very good when you need to steer and than quickly change direction. How high do you keep your eyes through the turn? A further point of focus might smoothen out your inputs.

[jcochrane]

With reference to "atraists" post above and to avoid further duplication, other views on cornering have been fully discussed in the "Cornering - Balance" thread. Contributers in that thread include respected experts such as "StressedDave" who are well worth listening to.

[7db]

There are some sharp switches on the Millbrook hill route which upsets even the 7 on the lateral transfer. Obviously all the usual thinking applies about the balance corner and the importance of optimising the second over the first, but I find a slight lift can help the weight over.

[7db]

Alternative thought -- if the junction is nice and open and clear, how about a faster approach speed and trail-braking through the left -- treating it as a curved braking zone for the following right? Vision appears not to be especially limiting -- more comfort and grip.

[StressedDave]

Ignoring the difficulty of analysing a bend based solely on Google Streetview and some dodgy writing

There are always going to be some double bends where there is no 'straight' where you can transfer weight from left back to centre before transferring it right. In general, I find there are three ways of dealing with the situation:

1. Accept that it's not going to feel nice and just drive the bends as safely and smoothly as you can. There is some truth in having 'a decisive turn-in motion', but without seeing this demonstrated by our Israeli friend, it's hard to tell whether this is a deliberate 'chuck' of the steering wheel or something more controlled and deliberate. If it's the former then adding to the destabilisation of the car (and certainly previous descriptions of the technique make it sound like the former) isn't going to improve matters.

2. Try and find a way of engineering a 'straight'

3. Continuous hand movements


In explanation: As most know I spend a lot of time on the Outer Handling Circuit at Millbrook - here, which has a couple of high speed (relatively - you're doing well to exceed 70mph anywhere around here). The first is on the back straight (the little kink to the right of the large triangular forest on the right of the satellite image). If you notice almost immediately after the chicane you've got a long sweeping right hander so being as straight as possible on the exit is important. So the first half of the chicane is taken as late as possible to give the largest possible radius and the straightest possible exit. It may well be that there isn't enough width on your particular bend sequence to play fast and loose with the gods of line.

The second chicane is at the bottom of the image, after the long sweeping right. Many of my clients, particularly those that style themselves as expert track day goers and amateur racers, have a tendency to treat this chicane as a 'left-centre-right' type manoeuvre, i.e. decisive steering inputs. It doesn't work for a number of reasons, not least of which is that it's inefficient in its use of grip (and not to mention that it's downhill with varying cambers). I know we have plenty in reserve on road, but that's no excuse for not using as little of it as possible. The key to smoothness and thus efficiency and speed is to have continually moving hands. There's a fuller description of my ethos to cornering in the 'balance' thread, but effectively it's a case of smoothly and continuously steering to the left until the car is on the desired path and then smoothly steering in one movement to generate the exit path and the entry to the right. If you recognise that the car will steer to the left as long as the wheels are pointed to the left, you can the technique to work pretty much everywhere you have a transition like this, but it does take practice as it's a significant change to the way you might 'normally' approach the problem.

[MGF]

...Contributers in that thread include respected experts such as "StressedDave" who are well worth listening to.

Respected by whom? (And I don't mean no denigrate SD by asking this question). Can we be left to use our own judgement as to who is worth listening to and who isn't? (Personally I find both, at least, worth a read).

[7db]

I think John's aim is to apply a prophylactic to some of the views expressed here which he believes might be dangerously misunderstood by someone without your judgement, rather than necessarily to blow SDs rusty trombone.

Someone with your judgement can spot this and opt to dabble without such protection.

[Astraist]

I stated what I mean by decisive: The fastest input possible without hindering the smothness of the output. It must also be accurate (thank you for reminding me) and applied late, in this case.

A slightest transition in weight can be beneficial in that phase. Likewise, it can be ineffecient, it is situation-dependant, mainly dictated by the corner and it's cornering speed. I believe it could assist in the given situation by making the initial turn-in more effective.

Other than that, I agree with Dave's post fully.

[Horse]

Ta All.

Now, possibly in order:

Slower entry speed and perhaps a different cornering line where possible might give better "cornering stiffness". A decisive turn-in motion when going into the S-bend can make the car "take a set" faster, making it easier to change direction quickly afterwards. And I believe that is the main problem. The input should be as quick as possible that can still generate a smooth output.

Unless the steering inputs required are very large, fixed-input is very good when you need to steer and than quickly change direction. How high do you keep your eyes through the turn? A further point of focus might smoothen out your inputs.

Slower entry: yes- ish, in that the would be less outward force in both direcitions (consecutively ) so less of a problem in the first place.

Line, quick turn, steering input: irrelevant; the lane is narrow, there are limited option on position and steering unless intimacy with hedge or oncoming vehicles is desired.

Eyes up: looking for the limit point - check again Google Streetview to see what I mean.

Alternative thought -- if the junction is nice and open and clear, how about a faster approach speed and trail-braking through the left -- treating it as a curved braking zone for the following right? Vision appears not to be especially limiting* -- more comfort and grip.

Tried this evening going slower (trailing/closed throttle) through the first bend (steady speed past the junction). That helped, seemed to allow (poss. because of slower overall speed) a slight straightening during the transition.

* It is, honestly. Go to the second link I posted where the bend is just visible through the trees.

With reference to "atraists" post above and to avoid further duplication, other views on cornering have been fully discussed in the "Cornering - Balance" thread. Contributers in that thread include respected experts such as "StressedDave" who are well worth listening to.

Ploughed through. See, you can play nicely sometimes, chaps

Ignoring the difficulty of analysing a bend based solely on Google Streetview and some dodgy writing

Oi! I resemble that remark!

There are always going to be some double bends where there is no 'straight' where you can transfer weight from left back to centre before transferring it right. In general, I find there are three ways of dealing with the situation:

1. Accept that it's not going to feel nice and just drive the bends as safely and smoothly as you can.
2. Try and find a way of engineering a 'straight'