Hello,
I've been debating this for a while. Due to ohio weather and rust... Money is going to limit and postpone my suspension wants for an unknown amount of time.

Car is a 1999 Civic SI, suspension is stock so far with the exception of a rear camber kit, tires Kuhmo SPT 195-55-15

With a koni and GC setup, I figure I will need about 1,200 when all is said and done. (shocks, springs, lower rear bar because of rusted seized bolts, random other bolts, an alignment and overall work since I am tool and space limited.)

However I autocross and I have noticed that I need something in the way of suspension bad... I often go up on 3 wheels or roll up on the side so much that traction is taken off of 4 wheels and onto 3, or even 2 depending on how you look at it... (mostly 3, with inside rear tire in the air, and outside front drilled into the ground, possibly bottomed out) Needless to say I lose all kinds of time because running slower or faster does not seem to help the situation. I have also seen a hatch with bad all season tires but good suspension set up completely beat me by no small amount in the turns.

Would the ASR sub-frame kit with the 24mm bar help any? Or would it be a complete waste of my time and money? I can find it much cheaper than the full setup, 350-380.

Thank you for your time, and please offer any other advice you feel could help.

 

autocross people tend to run pretty stiff rear springs. The stiffer the rear springs, the less difference the rear swaybar makes. Normally, I would say that a rear swaybar helps in increasing oversteer, which is good for autocross. But you said that you are already lifting the inside rear wheel.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beanbag &raquo;</TD></TR><TR><TD CLASS="quote">autocross people tend to run pretty stiff rear springs. The stiffer the rear springs, the less difference the rear swaybar makes. Normally, I would say that a rear swaybar helps in increasing oversteer, which is good for autocross. But you said that you are already lifting the inside rear wheel.</TD></TR></TABLE>

Which I believe is because I am bottoming out the front Driver's shock lol... I mean it grips, and DIVES that corner hard.. then slips.

 

Then you may as well get stiffer springs all around

 

With such a stock set up (which will be biased in some degree toward understeer, disregarding whatever you may have done with rear camber) I can't see the IR lifting with the only substantial acceleration being a lateral acceleration, but keep in mind that in Autocross you're turning tight corners pretty hard, so as you steer into the turn you'll also be generating a significantly strong effective braking force (even if you aren't actually using the brake) due to the front wheels ‘scrubbing’ with the severe direction change, causing a fairly strong forward weight transfer in addition to a lateral weight transfer.

So when you turn hard into a tight corner at some speed you will have two strong accelerations, causing substantial lateral and longitudinal (forward) transfers, which combined will cause the OF to load substantially and also for the IR to unload substantially (much more so than in a faster more open corner).

This might not be so bad, but if the OF suspension hits the bump stop, it will cause an instantaneous lateral weight transfer from the IF to the OF and an opposite lateral transfer from the OR back to the IR, as well as forward weight transfer from the IR to the OF. I would expect this to cause a sudden understeer when the OF suspension bottoms out. Overall, a lot of opportunity for non linear weight transfers to occur causing unstable handling.

I think you're going to need more front roll stiffness to keep this under control, though this may not be such a good set up for road use or race track use as it will tend to generally increase understeer. Front roll stiffness can be increased with a stiffer front ARB and / or stiffer front springs, though stiffer front damper settings may also help by increasing transient front roll stiffness (i.e. it will take longer for X degree of roll motion to occur, and weight transfer will occur more quickly, but won’t ultimately be stronger, in theory).

Increasing front roll stiffness will tend to increase steering response (good thing for tight turns), but also increase lateral front weight transfer from the IF to OF (and reduce lateral rear weight transfer from IR to OR) which typically will increase understeer. However, in your case you may actually pick up more front grip simply because you decrease body roll (which is likely to be substantial with a more or less stock set up in tight Autocross type turns).

If so this would be because you reduce the undesirable camber changes (relative to the road surface) caused by excessive chassis roll, meaning the contact patches will effectively be larger and capable of generating more grip / traction, despite increased front weight transfer (i.e. the effectively larger but more loaded OF contact patch but less loaded IF will generate more grip than both more equally loaded but effectively smaller front contact patches).

This may be the case despite front roll stiffness having been increased (which will be a negative), but I think it would most probably be better still if rear roll stiffness were also increased simultaneously with front roll stiffness (to reduce front lateral weight transfer / increase rear lateral weight transfer that will occur with an increase only in front roll stiffness), so you get the benefit of both the decrease in roll / camber change without the unwanted side affect or increasing front lateral weight transfer and decreasing rear transfer (you still want the IR to lift, but only just).

Having said that, if you're serious I think you need to do some more serious suspension work all round. The fact that other cars with crap tyres but decent suspension set ups are beating you (with I assume better tyres but poorer suspension set up?) shows how important the suspension set up is. Good tyres are more or less wasted with a poor suspension, but a good suspension can make poor tyres work surprisingly well...

 

JL,

You sure know a lot about suspension. Every time you post, I learn several things.

As regards your advice to this particular fellow, you make a good point that increasing the front spring rate or roll stiffness can actually decrease understeer due to bottoming of suspension. So it seems that in this case, simply getting a stiffer rear sway bar will not help if the inside rear wheel is already lifted.

The problem I have seen with getting new springs, and Ground Control in particular, is that they tend to spec springs that are too short. I think they expect that you "slam" your car. If you run it at "normal" height, you are at risk of coil-binding the springs. This is obviously worse than hitting the bump stops, although I have actually never heard anybody complain about it.

To the OP, I have some other suggestions that you might try:
Increase front tire pressure
Cut the front bump stops. This might gain you a little extra travel
Change your driving style a little. This is kind of hard to explain, but during a turn, I've felt that if I use a little more gas, it feels like more of the lateral load is placed on the rear wheels. In other words, there is some more weight transfer to the back, and unless you over-gas it, the car still doesn't understeer horribly.
There is also this driving technique that I learned from the BMW folks that says that the optimal corner entry braking and turn-in is such that the outside front stays compressed the same amount, i.e. if you turn the wheel you have to lift off the brakes.

If you go the route of getting new springs all around, you can save money in the following ways:
Get some koni yellow shocks, for about $450 total from honda-tech sponsors.
GC springs and adjusters for another $315.
Keep the car at stock height. This will save you from having to do an alignment. Or you can assume that if you lower, your toe changes slightly, and tweak it back by some appropriate amount.
No sure how much labor you are willing to put in, but money can be saved there as well.
Not sure if this works for your car, but there is this thing called the "shock fork trick" where you mount the front shocks lower into the shock fork than you normally would. The effectively buys you more front suspension travel. I might consider doing this myself since it seems that on hard turns, I might be into the front bumpstop as well.

I cannot overemphasize that if you get custom springs, make sure they are long enough and do not coil bind!

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by johnlear &raquo;</TD></TR><TR><TD CLASS="quote">

If so this would be because you reduce the undesirable camber changes (relative to the road surface) caused by excessive chassis roll, meaning the contact patches will effectively be larger and capable of generating more grip / traction, despite increased front weight transfer (i.e. the effectively larger but more loaded OF contact patch but less loaded IF will generate more grip than both more equally loaded but effectively smaller front contact patches).

</TD></TR></TABLE>

Depending on the suspension type, it may be the case that the dynamic camber change from compressing the suspension may outweight the roll of the car, i.e. even more negative camber relative to the road surface. I don't know if that is the case here.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beanbag &raquo;</TD></TR><TR><TD CLASS="quote">Depending on the suspension type, it may be the case that the dynamic camber change from compressing the suspension may outweight the roll of the car, i.e. even more negative camber relative to the road surface. I don't know if that is the case here.</TD></TR></TABLE>
Even with double wishbones (aka 'short / long arm' or SLA suspension), which has a better 'camber curve' than say Mac struts, body roll will typically cause the OF in particular (the outside tyres are always the more important for cornering grip) to gain less camber change than the body roll generates. Neg camber (relative to the road) will still reduce or pos camber increase, but if the degree of body roll is substantial then camber will still go past zero into pos camber (unless the static neg camber is fairly extreme).

An SLA suspension could be designed to have just about any camber curve you like, but typically won't be designed so that the camber curve equals the degree of body roll because if you make the camber curve too steep you tend to end up with 'virtual swing arms' that are too short, and because of this the geometric roll centre then moves around a lot with suspension travel in roll, which has significant non linear affects on the front / rear distribution and speed of weight transfer, which is not a good thing as the car can become more difficult to drive at the limit.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by johnlear &raquo;</TD></TR><TR><TD CLASS="quote">
Even with double wishbones (aka 'short / long arm' or SLA suspension), which has a better 'camber curve' than say Mac struts, body roll will typically cause the OF in particular (the outside tyres are always the more important for cornering grip) to gain less camber change than the body roll generates. Neg camber (relative to the road) will still reduce or pos camber increase, but if the degree of body roll is substantial then camber will still go past zero into pos camber (unless the static neg camber is fairly extreme).

An SLA suspension could be designed to have just about any camber curve you like, but typically won't be designed so that the camber curve equals the degree of body roll because if you make the camber curve too steep you tend to end up with 'virtual swing arms' that are too short, and because of this the geometric roll centre then moves around a lot with suspension travel in roll, which has significant non linear affects on the front / rear distribution and speed of weight transfer, which is not a good thing as the car can become more difficult to drive at the limit.


</TD></TR></TABLE>

Thanks. This is good to know. So in terms of camber, less roll is always better?
And more generally, suppose you have a car going around a skid pad, with a fixed spring rate. Do stiffer ARB's give overall higher grip due to less roll?

 

Just trying to help.

I’ve read a lot (not enough), discussed a lot (not enough), and thought a lot (not enough) about the physics of chassis dynamics. I wish I knew more, there’s always more to know. I have a significant amount of practical experience setting up racing kart chassis, but relatively little with cars. However, the fundamentals of the physics apply equally to karts and cars, and a lot of understanding can be transposed from one to the other, if you have a handle on the differences in how car and kart chassis apply the physics to achieve optimum performance, and are careful in your thinking!

Having said that, even far more intelligent, knowledgeable and experienced people than I don’t fully understand all the complex interactions that occur in chassis dynamics, karts and cars.

You mean due to not bottoming out the front suspension. You’re right, I don’t think the basic problem is to do with roll couple (i.e. relative front / rear roll stiffness). It’s already unloading the IR, making the rear alone any stiffer in roll is unlikely to improve things.

I doubt the coils will become bound, unless a significantly shorter damper is used (and / or perhaps the bump stop is deleted). With short dampers, the likelihood of binding the coils will be increased ever so slightly by using thicker coil wire, and more so if using more coils, but I doubt more coils would typically be used for stiffer springs because more coils = a longer length of coil wire which makes the spring softer (and increase unsprung weight).

If you’re finding it hard to source ready made over the counter stock height (or near) uprated springs on your side of the Pacific, North America is a big place, surely there’s some company over there that can do custom springs in whatever height and rate you want? There’s at least one company here in Australia (K-Mac Suspension) that can do this, and for a very reasonable price (I have no affiliation!).

Won’t help bottoming out, and may increase the speed of weight transfer when / if it does bottom out. A softer case (i.e. less pressure) will help ‘absorb’ some of the ‘instantaneous’ weight transfer. Having said that, anything is worth a try with tyres, especially something as cheap as air pressure. Tyres don’t always respect theory.

Worth a try, but not too short, they still have to act as cushions.

I previously said that hitting the bump stops would create an ‘instantaneous’ weight transfer, but this isn’t quite correct. If no bumps stops were fitted and the suspension bottomed out there would be an instantaneous weight transfer (cushioned only by the vertical tyre compliance and any rubber seats or bushes in the suspension). However bump stops ‘snub’ the force progressively, so when the suspension bottoms out the weight transfer won’t be instantaneous only somewhat abrupt, depending on the length, shape and material properties of the bump stop rubber (i.e. how ‘progressive’ the particular stop is)

You might try the other extreme, i.e. try fitting two bump stops per front damper (or one full bump stop and a cut bump stop, depending on what will fit) so that the car is resting slightly on the bump stops at static ride height. This will effectively add some degree of progressive spring rate to the suspension (more so than with a short bump stop), i.e. the stops will act like auxiliary front springs. This might improve matters to some degree, but possibly might make them worse! Possibly worth a try since it would be cheap!

This will create some rearward weight transfer that might unload the OF to some degree (which in itself might make matters better or worse), but it can also ‘pull’ the front of the car into the corner. It’s a bit hard to describe, but what happens is that the tractive grip (not lateral grip) of the front tyres pointed into the turn with some power applied effectively ‘pulls’ (can’t think of a better word) the front of the car in the direction in which the wheels are pointed.

I know this works on dirt surfaces, as I drive on a fair bit of dirt. I’m not sure how it would go on bitumen surfaces, not as well I think, and I’ve never really tried it other than on the dirt (on the dirt in a FWD car I find it’s an instinctive reaction in response to the understeer, but instinct has never suggested to me that it might work on a hard surface). It goes like this; a bit too fast on entry and the car understeers, so with the front wheels steered ‘too far’ you apply power (not gingerly but don’t floor it), and the line tightens up (but too much power and you’re in the weeds).

This only works for understeer with FWD cars, but it’s somewhat the same principle in reverse to driving an oversteering RWD car on the dirt, i.e. when a RWD car oversteers on a loose surface it’s often best to apply some power which prevents the oversteer from becoming worse or even lessen it (but too much power and you’ll spin it in a instant).

You can see the principle work (in reverse to FWD) rally cars or with sprint cars on a dirt speedway, i.e. the car corners at a substantial angle to the direction of travel with the rear wheels spinning to some degree, which allows ‘longitudinal’ traction to be used to gain ‘lateral’ traction, if you follow.

I suspect the BMW folks have been giving out rather simplistic advice. By definition none of the suspension corners can remain constantly ‘compressed’ to the same degree throughout a corner, except in longer corners where the car reaches a stable ‘steady state and remains there for some brief or longer period. The “i.e.” is more or less correct.

The optimum technique is ‘trail braking’. Brake hard up to the corner using all the tyre’s grip to slow down, then as you turn in start lifting off the brake progressively as the lateral acceleration progressively increases up to the apex, at which point you’ll have zero braking, then start getting progressively on the power as the corner opens out and lateral acceleration progressively abates.

The idea is that you keep the tyres at their maximum grip capacity the whole time during the entire cornering process (or as near as possible), keeping the demands for lateral grip and longitudinal grip in balance so that at all times the sum of the grip being used for lateral traction and the grip being used for longitudinal traction adds up to 100% of the grip capacity of the tyre.

A tyre has X grip that can be proportioned in any direction, so if you’re braking at say 90% of the tyres available grip then you have 10% grip that you can use for lateral traction. This % ratio between lateral and longitudinal grip constantly changes in the corner (hopefully in a totally progressive manner if the chassis is well set up), and the aim is to always be using 100% of that grip at all times during cornering, including accelerating out from the apex. You’re a pretty good driver if you can do it consistently!

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beanbag &raquo;</TD></TR><TR><TD CLASS="quote">Thanks. This is good to know. So in terms of camber, less roll is always better? </TD></TR></TABLE>

The less roll there is the less static neg camber will be required (and typically less so with a SLA suspension than a Mac strut suspension).

However, it's not necessarily desirable to eliminate roll (or nearly) with springs and / or ARBs because the car will become very stiff in roll. So when you hit a bump mid corner you'll run the risk of losing lateral grip at that wheel (it'll be more likely to 'skip' on the bump as the sprung mass is bounced off into space). Also, the car will become less comunicative to the driver on the limit, and the limit of adhesion will tend to arrive with little warning. It will however become more responsive, possibly too responsive if roll motion is reduced close to zero.

The drivers of F1 cars and similar have to be very good for good reason, such cars are very roll stiff and hard to drive on the limit as they generally give little warning of total grip loss (and I do wish I was talking from personal experience!).

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beanbag &raquo;</TD></TR><TR><TD CLASS="quote">And more generally, suppose you have a car going around a skid pad, with a fixed spring rate. Do stiffer ARB's give overall higher grip due to less roll? </TD></TR></TABLE>

It depends on how long each of the peices of string happen to be. On a totally smooth skid pad the stiffness of the spring rates (including the roll 'spring' rate provided by the ARBs) become less important than the roll couple (difference in front vs rear overall spring rate, including ARBs), so long as the contact patches are presented 'squarely' to the track surface.

What I mean is; the overall spring rate at each end front /rear (which would include any rate contributed by either the springs or the ARBs) could be soft or hard (and on a perfectly smooth skid pad the dampers would theoretically be irrelevant), but so long as the relative difference in rate (front / rear) were such that provided the optimum degree of front vs rear weight transfer for the front / rear weight distribution of the car, then maximum lateral acceleration would be the same so long as the tyres were presented to the track in the same way when the chassis rolled.

To a very very slight degree the chassis with the softer rates would roll more, and this will to a very very slight degree move the CG toward the outside of the chassis, and this would increase weight transfer to a very very slight degree and to a very very slight degree reduce total lateral grip. Not enough to be a real consideration.

 

I mainly meant this as a way to reduce understeer.


That's an interesting idea. I suppose it depends on whether the OP's problem is with excessive compression (car leans too much) or excessive rate, caused by hitting a bump stop.

I've read that the mechanism why "drifting" works on dirt is because you build up a small pile of dirt on the outsides of the sliding tires, increasing lateral grip.

Now regarding the issue with keeping the OF tire compressed the same amount. I know that the whole point is to maximize grip of the car, but on a more detailed level, how are you going to distribute the grip between the tires? BTW I'm sure there are all kinds of factors involved, and the F1 people probably have it figured out. But let me give my simplistic reasoning as follows...

You start off with these two assumptions:
The amount of grip a tire is able to provide is related to how much weight you put on it.
The grip vs weight curve is sub-linear, i.e. if you put twice the weight on a tire, you won't get twice the grip.

Now consider a car going into a corner. 1. First you start by braking, which puts most of the weight on the front two tires. 2. At some point, you are cornering, which puts the weight on the outside two tires. 3. Finally, you accelerate out, which of course lifts the front wheels.
So the question is how are you going to transition from 1 to 2? Ignoring the IR wheel, you go from
50 50
0 0

to

0 50
0 50

Let's say that for a certain combination of steering and braking, you get

15 70
0 15

For another combination, you get

25 50
0 25

I'm going to guess that the second combination is better because you are loading the OF wheel by less. Due to the sub-linear grip vs weight thing, the way to get maximal total grip is to load the tires as evenly as possible. And so keeping the OF wheel loaded at 50, while smoothly transitioning the weight from the IF to the OR is one way to do it.
Anyway, this was the point I was trying to make.

No wait, I think I said something wrong...


Modified by beanbag at 2:39 PM 3/11/2008


Modified by beanbag at 5:31 PM 3/11/2008

 

I'm just loving this conversation lol. I have decided Ill have to save up for a proper suspension. But Perhaps I can apply a bit more advanced theory to my driving to make up a second or two until I can increase my car's limits.

But please, keep up this convo.