Sam@TDi

Yeah as standard the EP3 Civic is very toe-in with the wheels in droop (the car jacked) and toe out as you put the car down, the issue we have with the ep3 is that the track arms are at a pretty acute angle even at standard ride height and get rapidly worse as the car is lowered, with very low ride heights the arm angles get ridiculous and cause a very aggressive toe-out on bump scenario. I found that it gave rise to poor turn in and straight line bump steer. The fix for me was to lower the track rod end so that the steering arm formed a proper Parallelogram with the lower arm. Infact I've gone a bit further and dropped the TRE lower still to give a gentle toe-in on bump scenario, the result has been razor sharp turn in but at the expense of fast corner stability. My choice and certainly keeps me awake on the motorway

Remember last year when we were discussing what exactly "kinematic toe" really ment? .... well this is pretty much it

Seriously that supra is the fastest accelerating thing I've ever sat in, and that's saying something

Cool

Hey Jon I agree with you, I think at the moment we are spending far too much time thinking and talking about DI, it's one aspect, only one aspect! I think the reason you don't understand DI's relivence to your sw20 or to any other track based car is because you've mis-understood DI. It looks to me as though you beleive that a car with a DI less than one will have a rear first departure. And that a car with a DI more than one will have a front first departure... this is WRONG .... (notice I avoided the terms understeer and oversteer, but thats a discussion for a whole other thread some time soon) Think of the car as a system, and as with all systems it has 2 main features COMMAND and CONTROL (we're talking chassis not driver here). The steering wheels are the Command, the rear wheels are the Control. The "moment of instance" that the chassis rotates about dictates it's DI number, simple. The moment of instance that the chassis rotates about tells you how Command and Control are related in that particular chassis system... "does the rear tyre slip angle dynamically increase as i turn the steering wheel or does the rear tyre slip angle dynamically decrease as I turn the steering wheel?" that is the question that calculating the DI answers... nothing more, nothing less.... this has nothing directly to do with which tyres will saturate or "let go" first ... (as your beetle example graphically points out) but, it does have a great deal to do with the way the car will actually try to move, and therefore it has a great impact on the cars emotional relationship with the driver. I consider that car/driver relationship to be one of the most important things to get right, especially in a race car, especially if you want to go quick

I can see why it might seem that way, but the gain from the rubber compound change far out weighs the loss from the reduced surface area

The trick is to fix the waste gate arm in place under tension instead of attempting to weld the waste gate flap itself to the cast housing

Seriously you would be amazed, the Rover groups MGF was a DI tragedy and they had plans in the pipeline to replace it with something way worse before they went belly up, simply because it looked good

Personally I see this as cheeting, most of the performance gain in this system is actually made by essentially shifting to a race tyre for cornering and a hard tyre for the straights, not from the actual adjustment of the geometry. Strikes me as

I'm hoping to treat the civic to a pair of JAS motor sport anti roll bars so that I can play with the cars departure charichtoristics And if I can save enough money then i'd really like to invest in a set of AP Racing brakes for the front

I still don't understand this. Providing you kept the balance optimal and if you lost a lot of weight, resprung and redamped as appropriately surely a 600kgs X would be faster than a 1600kgs X? I hear you when you say it's tough trying to lose weight (in the car). I saved about 40kgs doing the seats, wheels and suspension but much after that is proving really difficult. Maybe 20kgs for all of the inside and another 10kgs for various bits I've cut off the car. Next big drop is having you guys making a custom light weight exhaust, but that'll save 8kgs if i'm lucky. Then putting the cage in will add 15kgs I totally agree with you Jon that this concept as with so many aspects in chassis set-up is not at all intuitive. You are absolutely right in your observation that the lightest car should always be the fastest, and in terms of power to weight ratio, braking to weight ratio, grip to weight ratio you are of course correct, on paper there should be no question ..... but The point is that as humans we're actually pretty poorly equipped to drive fast cars fast!... only being able to make around 2 adjustments per second based on optical information and if very skilled around 4 adjustments per second based on inner ear information (g-forces, yaw rate gains etc) ..... and so performance car designers / chassis set-up engineers are faced with 2 major area's of concern. 1/ Firstly the main question is the vehicle suitable for it's intended application? 2/ Secondly is my human driver going to be able to exercise command and control over this chassis? Often in motorsport less experienced designers, owner/drivers and set-up engineers put far to much emphasis on question-1 and pay little no attention to question-2 My point is that having a car that is way too light for it's wheel base will emphatically tick all the boxes for question-1... BUT along with other problems the DI will be so poor that the answer to question-2 is likely to be NO (even if the driver doesn't admit it) and if this is the case the car will turn in poor lap times and no one will know why (unless the driver admits he can't drive the car... but in my experience this almost never happens as a matter of pride)

I would suggest taking a look at the "static" front axle toe and the way it moves as the suspension is loaded. At first glance I’d suspect that this car is toeing out a little at the front, it is just possible that the lag being experienced in the yaw response is the loading of the contact patch simply reversing itself before it winds up and begins to produce meaningful lateral acceleration in the required direction. You can obviously test this fairly easily with one simple toe adjustment and if you find it works as expected I’d be inclined to return a small amount of static "toe out" to the front axle and instead maybe move the front wheel camber slightly more negative, this will have the effect of pre-loading the contact patch a little

Just to add, good quality coilover kits (example, Cusco) MacPherson strut systems are designed with adjustable top mounts and camber snails at the base. This rather neatly allows you to achieve your desired camber setting whilst keeping the king pin inclination as correct as you like. The process of setting these 2 parameters simultaneously though is easier said than done

Yeah fair enough, it's a real shame no body markets a proper race anti-roll kit for the SW20... even having just 3 levels of adjustment on the whitelines is still better than nothing and it would definatley improve the "leaning like a bus" problem \/ If you fancy being really different, check this out \/ Cunning adjustable anti-roll set up

You know in this instance I still believe that forgetting math’s is the best way to go. @ Jon …. I still very much recommend a pair of totally adjustable anti-roll bars for a few reasons, Firstly and most importantly if you follow this route any changes you decide to make to the balance or the overall amount of anti-roll the change can be reverted in the pit lane in a matter of seconds Secondly it gives you the ability to adjust the car as your driving style evolves, please don't take this the wrong way but you'll find you've been sub-consciously altering your driving style and your lines to panda to the cars weaknesses. So as the car begins to work as it should you might well find that the goal post’s move from where they seem to be right now. The other major aspect is that whilst in a racing situation it’s simple fact that it’s faster to have the car set up on a knife edge responding to every blink and breath, it’s not actually that pleasant for the driver and tends not to inspire confidence. In your case the whole idea as I see it is to have maximum fun and enjoy driving around flat out.

Here are couple of nice ones we built recently R32 Skyline GTR, 2.7 Jun forged steel rotating assembley, HKS cylinder head hardware, Full engine managment... produced around 600bhp @ the hubs using 1.7bar if memory serves Subaru Impreza GC8 UK spec, 2.2 engine using all HKS products. Set-up primarily for responce it made a shade over 400bhp using 1.3bar with a lovely torque curve

Agreed, an excellent explanation. Thanks guys, glad it made sense Yes definitely, try it, in a pick up normally you'll start with a low DI when empty (apprx 0.90) the DI would then get higher as you add mass, you'll normally feel the centre of rotation move rear-ward as you approach the chassis weight limit

Ok sorry for the confusion I’ll try to clear things up A Dynamic Index is the output of Inertia Match Theory Inertia Match theory describes the relationship between a vehicles Centre Of Gravity position (COG), the vehicles wheel base and then it's yaw inertia (mass squared). There is no voodoo, the hard truth of it is that a 50/50 weight distribution will always give the best Dynamic Index figure for a given yaw inertia and i'll cover the various techniques and tricks used to attain this in a moment. There's no doubt about it that Inertia Match Theory is much more important for a vehicle design engineer who is working for a OEM manufacturer than it is to us working in the after market, this is because vehicle wheelbase will always be pre-defined and indeed set in stone by the time we're actually dealing with a car. However simply by being aware of Inertia Match Theory and its various mechanisms we can gain valuable knowledge about a chassis by calculating it's dynamic index. The dynamic index or DI gives us a unique insight into the way the chassis will behave when maneuvered on a ground plane at speed, armed with this knowledge we are better equipped to then go forward with tuning the chassis to suit any specific needs. Below is a diagram showing the output of Inertia Match Theory after the theoretical length C has been calculated. Below is a diagram showing a car with a DI less than 1.00... a classic example of a car like this would be a classic VW Beetle which has a DI of around 0.48, these are so bad and so far past being "nimble" that they are pretty much un-controllable and un-drivable at the grip limit This is a diagram of a car with a DI greater than one, most modern production cars fall into this category in varying degree's, this makes the car feel sluggish to respond to steering wheel commands In many cases like my Civic it's possible to adjust the vehicles front and rear ride heights and therefore weight distribution in order to change the DI to something closer to 1.00, it's an everyday road car so certain ideal options listed below are unrealistic. Under motorsport or track day circumstances it becomes possible to seriously alter the weight distribution not only by a adjusting ride heights but also by relocating heavy items such as fuel tanks and batteries etc. The other major benefit under these circumstances is the ability to genuinely adjust the chassis mass (making the car either heavier or lighter) as this is a serious factor in calculating the DI. Hopefully this reads ok and makes sense, hopefully you can see that whilst it's important to understand Inertia Match Theory it only makes up one aspect of chassis dynamics... suspension design and calibration, static and kinematic geometry, braking and driving force variation are all equally important. And remember we only do all of this in order to control the tyre contact patch... the tyre is king!

Well shedding meaningful amounts of weight can actually be pretty tricky if the car was designed as lightweight or was very basic spec. And it totally depends on the vehicle, in the case of our demo car the fully electric and heated front seats alone weigh a great deal so replacing these and removing the rear seat would definately save 80-100kg, replacing the thick window glass with plexi-glas would be another good step... it all depends how serious the cars diet is, in the past i've spent 2days or so removing all the underseal and bonded sound deadening from a totally stripped body shell before the cars construction properly began It is worth remembering that it's completely possable to go the other way and be too light

Can we play a theoretical wim game? I have a front engined rear wheel drive car with a poor manufactured index of 1.7, how would you suggest improving this? Yep, ok Option one would always be to shed some weight! having a DI soo far above 1.00 can really only be caused by a mis-match in wheelbase to the mass, basically the chassis has far too much yaw inertia for the short wheelbase to influence easily. If we could take just 10% weight out of our demo car we'd have a DI of 1.39, much better! Unfortunately to take 10% weight out we'd have to seriously comprimise the car's very comfortable characteristics so for some (most) people this would defeat the object. The other option is to simply accecpt that the car will as a result of it's high DI always make poor use of it's rear axle in terms of producing lateral acceleration and then to try to work around it. A good example would be the Mitsubishi Lancer Evo-5, it's actually a car with a DI of 1.35 so this would make it pretty vectra/modeo'ish to drive... But Mitsubishi's chassis dynamics team have cunningly swerved this potential marketing disaster by employing an overly pointy geometry and suspension sollution giving the car the ability to change direction light a housefly, then in order for the average joe customer to be able to control and exploit this very odd and naturally quite unbalanced set-up they've been forced to add an active yaw control system in the form of an electronically controlled torque biasing center diff. This solution works so well that not only is the car acceptable to it's intended market it's actually now considered as one of the worlds best handling cars.

Hi Derek good to have you onboard!

An interesting dilemma for sure, and one faced by many people at clubman motorsport level. The accepted "norm" is for close circuit racing track cars to have massively stiff spring-rates and often no anti-roll at all, these are the pros & cons as I see them • + Body roll will be almost totally eliminated, this tends to have a strong positive emotional effect as large roll events can “un-nerve” amateur drivers • + Vertical load variation from the chassis will be very quickly transmitted to the tire’s contact patch and visa versa, so on a smooth surface the car will feel to respond faster and react more quickly to balancing corrections with for instance the throttle pedal. • + If the car is using any meaningful aerodynamics it is imperative to keep the aerodynamic devices horizontal to the track surface and often important to keep them at a specific separation from the track, the only practical way to do this is to stiffen the spring system. • - As the total stiffness goes up the systems natural frequency goes up, so the stiffer the chassis as a whole, the more inputs per second may be required to keep it on course, people using an open loop driving process (the playstation generation) generally can’t cope above 2hz …. The BAR F1 car is approx 4.8hz • - If the track surface has many undulations and the spring rate is too stiff to yield the tyre can be forced momentarily past the saturation point. If this happens a net reduction in grip will occur at the saturation limit. This is due to the simple fact that a loaded tyre (one that is producing lateral acceleration) will “un-wind” in one forth of the time it takes it to “wind” back up and start producing lat acc again. • - Having zero body roll means that you lose the chance to take advantage of kinematic geometry changes, such as “toe-in/out on bump” or “camber in on bump” Personally I would strongly recommend option-3. Being able to manipulate roll as a handling modifier is a superb tool to have in the box, you can quite literally decided whether you’d like a push or an oversteer departure, the trick is learning how to drive with the chassis moving and not let the roll events put you off

Nice Mat!

I think that it's admirable that you do this for free as I would of thought most customers would almost expect to pay a little extra for the follow up, and in a perfect world I would say it was totally unnecessary. But this isn't a perfect world and no matter how hard we try there will always be some unquantifiable variables sneaking into what we're doing and throwing up spurious results. In reality I can see 4 major points that this post inspection will check and compensate for, 1. Coil spring sag (mostly only applicable to new coils) 2. Bushes "settling" (mostly only applicable to new bushes) 3. A bump in the subsequent 2000 miles that may have messed up your original lovely geometry job 4. It might catch any earlier mistake in the geometry, we're all human and also we should remember machines can be subject to external influences that cause temporary errors

I think I agree ... As for how to measure a suspensions kinematics you've got 2 real options, if you want scientific data accurate to a few decimal places then you'll need a 4-wheel full chassis suspension dyno. But if you live in the real world like 99% of us you'll just have to take a best guess at how a system will behave dynamically simply by looking at it, one of the most important things to aid your summary will be to concentrate on moment leverages versus the rubber bush sizes. A soft bush that must deflect / conform will invariably have to absorb more energy than a hard bush which is designed to offer stiff resistance to load. When using rubber this energy is transformed into heat, in order to keep the compound stable it must have enough thermal capacity so as to cope with the energy that the bush is being required to accept. The easiest way to give a rubber bush more thermal capacity is to make it bigger (more rubber = more capacity) Big Bush = SOFT ............ Small Bush = HARD So with this in mind you can normally take a look at a multilink suspension and have a fairly good idea which way it's going move under certain forces simply by noting the position of the bushings and there relative sizes I hope that helps a little and I’m not teaching people to suck eggs