The Physics of Suspension Settings

[robinm]

Hi,
I'm interested in hearing views on the effects of changes to suspension settings on trackday-related handling. That is to include settings for camber and toe and ride height, and the relevance of different spring ratings, shock absorbers and anti-roll bars. I would ask that tyres and driver technique, though both make a signifant difference to the way a car goes round a corner, should not be included in this discussion.

As a starting point I introduce my own Quantum Xtreme with supercharged Honda S2000 engine. Its current settings are 0.5 deg negative front camber, 10 mins of front toe-in and medium to soft shocks settings (the same front and rear). Do these settings seem OK? The rear settings are neutral. The overall handling is good but it is very prone to power-on understeer and will turn in quite sharply on a trailing throttle. How does rear ride height contribute towards this, if at all? The weight bias is slightly forward and front ride height as low as I can go.

I have read that stiffening suspension at one end of the car will result in improved grip at the other. Sounds strange, is it true?

I welcome any comments about my own car and also any discussion on the general dynamics applicable to this subject.

Edited by robinm on Friday 12th March 14:49

[Corpulent Tosser]

robinm said:

I have read that stiffening suspension at one end of the car will result in improved grip at the other. Sounds strange, is it true?

I am no expert but my Striker had bad understeer for a while and I messed with the setting to try to resolve this, I did find that stiffening the back gave a bit less grip there, not sure that it actually gave more at the front though.
Of course it would depend on what the setting are before you start making changes.
I resolved the problem with a slightly widening the track at the front and stickier tyres.

With a supercharged S2000 in it the oversteer is probably just power induced, may I suggest a slightly lighter right foot exiting corners ?

[Sam_68]

This is a hugely complex subject, but to make a start on the basics...

robinm said:

Its current settings are 0.5 deg negative front camber

Sounds about right. The optimum setting will vary from one car to another, depending on camber gain (the amount the camber changes as suspension moves) and a host of other factors. Basically, you're trying to achieve a good compromise between the angles the tyre likes best under pitch (braking/acceleration), roll and single wheel bump.

It's impossible to know for sure without knowing every detail of your suspension geometry and stiffness (and then doing some really, really complicated sums), but 0.5 degrees negative sounds like a reasonable starting point.

Your best bet is to invest in an infra red temperature gauge... after a few fast laps of a circuit, the tyre temperatures should be even across the tread (because this will indicate that the tyres are being worked evenly across the full contact patch). Inner shoulder too hot = too much negative camber, outer shoulder too hot = not enough negative camber.

robinm said:

10 mins of front toe-in

Again, this sounds like a reasonable starting point, but you need to experiment. Basically, the toe setting is there to take the 'slack' out of the suspension joints, so that they are under a slight but definite directional load at all times. No toe will usually feel a bit vague and indecisive; toe out will feel 'pointy' and directionally unstable (but keen to react), toe in will feel stable (but slightly less keen to react). How much toe is 'right' will depend on how you want the car to feel, but too much in either direction and you're just causing unneccessary rolling resistance and tyre wear.

robinm said:

medium to soft shocks settings (the same front and rear)

I'd be inclined to try stiffening the front; you'll almost certainly find that it helps with your power oversteer and trailing throttle oversteer, but you may find that it takes more than damper settings and that you need to stiffen the front springs a little (or fit/stiffen a front ARB). See below...

robinm said:

I have read that stiffening suspension at one end of the car will result in improved grip at the other. Sounds strange, is it true?

Yes, it is. Well... sort of. It doesn't result in improved grip at the other end, it results in the tyre being worked less hard (ie. less of the available grip is being used). The corollary being - as CT says - that you are working the tyres more at the end you are stiffening (ie. using more of their available grip).

You need to understand the concept of tyre slip angles, first, so if you don't, go read up on that before you go any further.

The easiest way I have found of explaining how to visualise what is happening with the weight transfer that happens under braking, cornering or accelerating is is to imagine the car sitting stationary, with a piece of very strong string (which can magically pass through any structure that happens to be in the way) tied to the cars centre of gravity. Pull directly sideways on this bit of string and you are simulating steady state cornering; straight forward or back is straight line braking or acceleration, respectively, and diagonally forward or back is trail braking into a corner or accelerarating out of one - you get the idea?

Now if you visualise pulling the string directly sideways with (say) a car that is much stiffer at the front, you will find that it is the front corner that is doing most of the work to resist it rolling. The tyre at this corner, being worked harder, will run at an increased slip angle.

Stiffer at the front (work the front harder) = increased slip angle at the front & reduced slip angle at the rear = understeer.

Stiffer at the back (work the back harder) = increased slip angle at the rear & reduced slip angle at the front = oversteer.

There are a whole lot of complications to the above - not least that not all of the loads transfer through the springs, which is where roll centre heights become critical, but since you can't easily change the roll centres on your Quantum, let's not worry about that just yet - and this being PistonHeads, someone will be along soon to bicker about the fine detail and semantics ( Are you listening, Quinny?), but that'll give you enough to think about for now.

If you're seriously interested, I'd suggest reading 'Competition Car Suspension' by Allan Staniforth (particularly the chapter on weight transfer) and Carrol Smith's 'Tune to Win' as a next step, although neither is perfect (and if anyone mentions the term 'mass centroid axis', just laugh...).

[singlecoil]

Excellent stuff, Sam. Normally I would suggest that you write a book, something along the lines of the 'for dummies' series, (not that I am suggesting anyone here is one ) but I think like all good explainers you would maybe have a job to fill the whole book, as you seem to have the knack of getting the main ideas across without any excess wordage.

[Sam_68]

singlecoil said:

Normally I would suggest that you write a book, something along the lines of the 'for dummies' series

I have, in fact, started to do so.

I'm about 5 years into it, now, and have almost finished the contents page.

One for my retirement, methinks...

[robinm]

Thanks for those replies.

Excellent stuff Sam. The string analagy is very effective in helping to visualise what's happening.

I have an infra red sensor and will put it to good use on the next trackday. Up until now I've been using palm-of-the-hand sensing after spirited road use - the outside of the front tyres are always noticeably cooler than the rest, but then the cornering forces are much less so I havn't made any camber changes.

If the rear ride height was lower than the front would that not have a tendency to cause the nose to pitch up even more under acceleration and contribute towards the power-on understeer in a corner? I can feel the progression towards understeer as soon as power is applied and it increases as power increases.

I will increase the front shocks stiffness and see what happens. I have tried running stiffer settings at the rear but after loosing most of my fillings on the concrete section of the M42 have reverted back to a less firm setting.

I know nothing of slip angles but will do my homework.

Many thanks again.

[66Quinny66]

Sam_68 said:

( Are you listening, Quinny?).

No... I'm reading

[66Quinny66]

Sam_68 - it is power-on understeer the OP mentions, not power oversteer.

(ETA to add the word "power" in front of oversteer just to clarify the point as oversteer is mentioned)

Edited by 66Quinny66 on Friday 12th March 18:53

[Sam_68]

66Quinny66 said:

Sam_68 - it is power-on understeer the OP mentions, not power oversteer.

Doh! So it is! My apologies, I misread that completely, possibly because it's quite an unusual combination; corner-exit oversteer is usually preceded by corner entry understeer.

If it's serious, then it kind of points toward pretty fundamental problems with the design geometry (like too steep a roll axis inclination... too low a roll centre at the front and/or too high a roll centre at the back, or roll centre is moving as the suspension loads up), so that while the car will intially point in, the loads build up on the front corner as the car rolls into the bend and hence it washes out when you apply power on the exit.

If it is fundamental roll geometry, then fixing it would mean a pretty serious re-engineering exercise (moving suspension pick-ups etc), which you probably don't want to get into. In lieue of that, try:

• Reducing the chassis rake. Roll centres sometimes do weird and unpredictable things, but as a general rule, lowering the front of the car will lower the front roll centre with it, so if your front ride height is a lot lower than the rear, it might be resulting in too steep a roll axis inclination. You want to be running a slight nose-down rake (for aero reasons, if nothing else), but unless I knew that the chassis designer had intended otherwise, I'd be looking the front ride height only maybe 10-15mm lower than the rear.
• Stiffening the front dampers, so that there is more initial roll resistance at the front, to promote initial understeer.
• Stiffen the rear springs (or soften the fronts) a little, so that it leans more on the rear corner (hence more toward oversteer) as the cornering loads build.
• Reduce the toe in (or try toe-out), if the above modifications have killed its current keen turn in (which I assume you will want to retain?)

But don't overdo any of the above, because ultimately a car that turns in well but which is difficult to provoke into oversteer under power is a safe car, so you're effectively wanting to move your car's handling towards the 'dangerous' end of the spectrum!

...And the basic rule is only to make one modification at a time (I'd be inclined to go in the order listed above), so that you can test the effect of each in isolation. And obviously make sure you have accurately measured a setting before you change it, so that you can revert to the previous settings if it makes things worse.

[66Quinny66]

Sam_68 said:

Doh! So it is! My apologies, I misread that completely, possibly because it's quite an unusual combination; corner-exit oversteer is usually preceded by corner entry understeer.

Robinm - can you confirm that Sam's interpretation of this is correct and that you are experiencing the opposite of what he describes above?

[robinm]

For clarification:

I started this post for 2 reasons; firstly to enter into a discussion and to learn more about suspension dynamics and secondly to get advice regarding the setup of my own car.

Sam, your contribution is brilliant, very relevant, interesting and thought-provoking. All other interest is also well received.

Regarding my Quantum; generally the handling through the corners on a neutral or trailing throttle is fine. The faster the corner the better the balance. Initial turn in is good. However, I have always hated understeer, anywhere, at any time. If I apply power whilst in a corner but prior to starting to exit the car will start to run wide (understeer). I know this is not unusual and I appreciate the inherent safety of this dynamic. But if I can reduce the tendency, in other words, improve the front end grip under power, I would be much happier.

With plenty of power available, power oversteer on exiting the corner I can induce easily. That's fine. I like it. (I guess it's not really oversteer, it's more to do with losing traction as the wheels start to spin, especially in the lower gears.)

Regarding chassis rake; I understand this to refer to the elevation of the rear in relation to the front of the car. In my previous post I questioned the effect of rear ride height being lower than front ride height. I believe this to be the case with my car. If so, because I can't lower the front any more, should I consider raising the rear?

What's the difference between fitting stiffer springs and increasing the stiffness of the dampers. As I mentioned in my previous post I had to reduce the stiffness of the rear dampers because the ride deteriorated too much. If I fit stiffer rear springs will that effect the ride in the same way?

I hope this is not getting too boring. I'm finding it extremely useful and already have lots to think about.

Thanks again to you all. Hang in there.

[singlecoil]

robinm said:

If so, because I can't lower the front any more, should I consider raising the rear?

I don't know much about this stuff, but I do know a bit about springs. If you can't lower the car att he front I presume you mean that you've turned the spring pans to their lowest setting, and the car is still too high at that point. What you need then is for the springs to be compressed some more. You can't make the front end heavier, so you need to change the springs. The new springs need to be softer (the weight of the car will compress them more, or shorter (i.e. the unfitted length is shorter).

Which solution will suit best depends on whether you need the front end to be stiffer or not, and I can't answer that. A good first step would be to take the springs off, measure them (unfitted length) and test their poundage. When you know those two things you will have a good starting point for ordering different ones.

[robinm]

No, the reason I can't lower the car any more at the front is because of ground clearance. The engine sump (the lowest part of the car) is already as low as I dare go.

Thanks for your comments.

[Sam_68]

robinm said:

Regarding chassis rake; I understand this to refer to the elevation of the rear in relation to the front of the car. In my previous post I questioned the effect of rear ride height being lower than front ride height. I believe this to be the case with my car. If so, because I can't lower the front any more, should I consider raising the rear?

Speak to the factory about design ride heights, but yes, it is most unusual to have a nose up rake, if only because it encourages air to get trapped between the flat bottom of the car and the ground at speed, causing lift.

If your rear is currently lower than the front, and the front is as low as it will go, then you should perhaps consider raising it, and this will help with your corner exit understeer problem because it will cause more weight to transfer at the rear (but is still doesn't square with the keen turn-in response, which is confusing me slightly). But do it cautiously and in moderation because you could transform the handling into sudden and lairy oversteer if you don't watch it.

Try setting the ride height level, first and if the handling is still benign, maybe move to the 10-15mm nose-down rake I suggested before.

When you say you can't lower the front any more, is it just that you have run out of adjustment on the spring seats (in which case, as Singlecoil says, you need shorter springs... I'd be inclined to avoid the 'softer' option unless you are sure it's what's needed, as it will affect other things), or simply that you can't lower the font any more for fear of grounding/taking your nose out on speed humps? (ETA: answered above - sorry, our posts crossed over, obviously)

It goes without saying (hopefully) that all adjustments to ride height should be made in conjunction with a proper corner-weighting session with you (or your equivalent weight) in the driver's seat.

robinm said:

What's the difference between fitting stiffer springs and increasing the stiffness of the dampers.

In very basic terms, dampers slow down the reaction of the springs, therefore damper adjustments can be used to affect transient response (eg. initial turn-in), whereas spring changes have a more fundamental, overall effect.

So, for instance, if you stiffen the front dampers, it will 'prop up' the front corner on initial turn in (loading the font outside tire more and hence giving a bit more initial understeer), but as the damper settles as the chassis takes a 'set' in the corner, it will allow the springs to compress and the rear will be allowed to take up more of its share of the load.

robinm said:

As I mentioned in my previous post I had to reduce the stiffness of the rear dampers because the ride deteriorated too much.

If I fit stiffer rear springs will that effect the ride in the same way?

Not necessarily, but this is where it becomes complicated, and if you've ever driven a current Audi A4 with the 'sport' suspension pack, you'll know that even major manufacturers with massive development resources can get it badly wrong!

In theory, dampers should be matched to the spring resistance: theoretical perfection would be what is called 'critical damping' where the damper stops the spring exactly at the deflection proportionate to the load imposed on it, so that it doesn't 'over shoot' and bounce back, but equally that the spring isn't 'stopped short' by the damper, leading to a harsh jolt. In practice, we don't do that (I won't bore you with the reasons why at this stage), but in any case if you're using the dampers to tune transient response as suggested above, then you have to accept some degree of mis-match.

A stiff rear spring with a well matched (softer) damper can often give a better ride than a softer rear spring with an over-stiff damper.

In reality, you'll never get it spot on; dampers that are set too stiff for the springs will give a very harsh ride, dampers that are too soft will feel bouncy.

One of the main problems you will be faced with, unfortunately, is that the Spax/Protech/Avo type single-adjustable dampers found on most kit cars are, when all is said and done, pretty basic devices. As the 'single adjustable' bit suggests, you can make them stiffer or softer overall, but that's it; you have no separate control over bump and rebound and no separate control over high speed and low speed damping.

Again in very basic terms (I'm conscious that a lot of what I'm saying here could be picked apart with 'ifs' and 'buts'), high speed damping deals with the rapid spring movements caused by surface imperfections (ie. ride), and low speed damping deals with the slower reactions to weight transfer due to pitch (braking/acceleration) and cornering. Clearly, if you only have single adjustable dampers, any change you make in an attempt to tune the 'low speed' pitch/roll transition will impact on ride quality, and vice versa.

Similarly, if you can't adjust bump and rebound separately, you have limited ability to adjust responses, because any change you make to the way the damper resists the spring being compressed also changes the spring's ability to extend the damper again, so you end up with either insufficient rebound damping (the spriung 'bounces back' an over-extends the damper on the rebound) or excessive rebound damping (which means the spring isn't strong enough to extend the damper again quickly enough after compression, which in the worst case can mean that a series of bumps in short succession can literally jack the car down onto its bump stops).

If you have the money, you can get Penske (or whoever) to build you a set of dampers with separate control over bump and rebound, and high- and low- speed damping which - in the hands of someone who knows what they're doing - can be a hugely powerful tuning tool (and in the hands of someone who doesn't understand them properly can screw things up beyond all recognition). But unless you're willing to run a set of dampers that are worth as much as the rest of the car put together, you're probably going to have to accept some limitations, otherwise.

Edited by Sam_68 on Saturday 13th March 10:36

[66Quinny66]

Sam_68 said:

If your rear is currently lower than the front, and the front is as low as it will go, then you should perhaps consider raising it, and this will help with your corner exit understeer problem because it will cause more weight to transfer at the rear (but is still doesn't square with the keen turn-in response, which is confusing me slightly).

I too am slightly confused (hence my lack of response so far ).

robinm - Unfortunately I know very little of the Quantum Extreme. You mention that the front ride height is as low as you dare go due to the sump. I know that the S2000 is a pretty "tall" engine but have you got it mounted fairly low for "packaging" reasons? How does your current front ride height compare to that suggested by Quantum? Failing that, can you just give some quick measurements (from floor to bottom of chassis at the front of the chassis and the same just in front of the rear wheels)?

[robinm]

Wow, Sam!!

Great explanation of what clearly can be a very complex subject. Many thanks for taking so much time and trouble to explain it all. I will need to read thro' your posts many times to get to grips with the detail.

I think my first move is to check the comparative ride heights and reset as necessary so that the rear is no lower than the front.

After that I'll re-read your posts and make a decision on which way to go with springs and dampers. As you rightly say; one step at a time.

I'm going to be even more boring than usual in the pub this lunchtime, especially when I introduce 'mass centroid axis' into the conversation. Pity I havn't got a clue what it means.

Quinny,
Yes the S2000 is taller than the Duratec that is usually fitted. And yes again, Quantum did say that they felt that ideally the front should be 10mm-15mm lower. I will measure the ride heights, hopefully this w/e, and report back. Thanks for your interest.

[robinm]

I've measured the ride heights. The car has a monocoque chassis so I've measured from points close to the wheels. I also measured to the centre-line of the lower wishbone mounting point. In both cases, and to my surprise, the rear height is between 15 and 20mm higher than the front. Does that seem reasonable?

[Sam_68]

Yes, it sounds reasonable, but was that with you in the car?

On a car as light as a Seven, the weight of the driver makes a big difference, so you really need to do ride heights and corner weights in full running trim, complete with the fatbd in the driving seat.

The weight of my Westfield increases by nearly a quarter when I step in!

[robinm]

I'll check again with driver included.

I need to redo the corner weighting. It was done 12 months ago but there have been many changes since then.

I've just dug out the last corner weighting report: with a driver 15 kgs heavier than me it had a rear weight bias 51.5 / 48.5. It recommended lowering the front by 20mm which I did, so in fact the ride heights are now as recommended. It also suggested that the rear rims (8") were too wide for the tyres (215/45)?

Edited by robinm on Sunday 14th March 10:55

[gtmdriver]

Get a copy of Alan Staniforths 'Competition Car Suspension'.

All will become clear.