Giovanni Tifosi's guru meditations
Giovanni 'drive it like you stole it' Tifosi explains the difficult stuff
|
|
Giovanni Tifosi's guru meditationsGiovanni 'drive it like you stole it' Tifosi explains the difficult stuff |
Theory
Let us begin this little exploit by defining crossweight. It's purpose, use, and designed intent is to keep the chassis balanced in such a way that all four tires have optimum grip and weight distribution after the chassis has gone into roll transition in a corner, by making the weight distribute in such a fashion that one front or rear tire has more static weight upon it than its counterpart.
There are 3 common ways to adjust this:
Depending on how much you want to "put in", any combination of the three can be used. You can change the corner entry traits, mid corner handling (under/oversteer), and exit characteristics of a given chassis depending on what combinations of method used, and what wheels the adjustments are made on.
For example: one of the left-turn-only karts I work on has a weight distribution of 47% front, 53% rear and a crossweight of 54% taken by the combining of the LR and RF wheels. This works out to be something along the line of the RF wheel having 2-5 lbs more static weight on it than the LF, and the LR having 15-24 lbs more than the RR. Once the chassis is in full sideload at mid-turn, the weight distributes in such a fashion as to keep all four tires in full contact with the ground (with the exception that the LF sometimes lifts completely off the track, depending on chassis flex, and track grip conditions ... dirt is not a consistent thing to race on ... but on asphalt the weight distribution works marvelously).
Application to GPL
Ideally, you would start with a fully spring balanced and symmetrically air pressured chassis and make adjustments through spring weight first, followed by damper adjustment, then fine tuning with air pressure. But, due to the variety of turns involved at each track and the drastic effects that changing spring weight has on the handling characteristics of said chassis, we should try to avoid using that method in all but the most extreme cases, and then keep spring adjustment to no more than 5-10 lbs. (Real life NASCAR utilizes a "jack screw" system that adjusts in 15 lb per full turn on the left or right rear, sometimes making up to a 45 lb adjustment in 1 pitstop!) Primary adjustments should be made in the dampers, secondary in tire pressure.
For starters you need to determine what is irritating you about the way the car handles so that you can decide which wheel(s) need to have adjustments made. For reference we will use Curva Grande for the test curve, it has appropriate characteristics for supplying data, and is fast enough to see the full effect after each adjustment.
Problem, car understeers on entry of corner:
Solution (#1) increase crossweight on RF wheel by increasing spring weight on LR wheel. Benefit, RF wheel will have more initial grip on corner entry due to "extra" weight placed upon it.
Solution (#2) Stiffer damper setting, (but symmetrical for bump and rebound), on LF wheel. This will "slowdown" the chassis roll a bit in the loaded direction, (especially under braking), but maintain the desired preset spring weight.
Solution (#3) Increase damper and/or spring weight of RR wheel. This will help "force" a more rapid roll transition, thus causing the front to "dive" into roll and since the RR will now be firmer the LF will be the first to go into deflection pinning the nose.
Disadvantages: in (#1), if crossweight on unladen front wheel is increased to much, there will begin to be a more pronounced oversteer early and mid corner, and potentially severe understeer once the car begins to drift since the weight transfer will be lost once all wheels start to slide. In (#2), if damper is to stiff, (or conversely, the LR setting is to soft), the car will have decent entry, but once laden the RF will lose grip due to the "tricycling" effect of the LR compressing more rapidly than the LF thus lifting the wheel up off the track. In (#3), potential for mild initial understeer, (when you very first start to turn in), followed by massive oversteer, (once the front "bites"), due to turning the wheel to far to correct the first part.
Recommended solution: (#2); easiest to correct as the dampers have an advantage of being "small" adjustments vs. what spring changes would do and can be further tweaked by stiffening/softening bump and rebound of opposite corner wheel. Minimal overall handling/braking characteristic changes due to the spring weights being unaltered.
Problem, mid-corner understeer:
Solution (#1) Stiffen LR and/or RF dampers, (LR adjusted 1st, RF to tune). Benefit, slower weight transition to LR corner of chassis thus keeping RF "pinned" longer under initial acceleration.
Solution (#2) Soften RR spring. Benefit, under acceleration the RR corner of car sinks quicker than LR thus balancing weight across front wheels for more "equal" grip.
Disadvantages: in (#1) chassis somewhat insensitive to small changes in damper setting, sometimes requiring the RF to be increased also to aid in the effect (RF spring will expand slightly more quickly, pushing more weight to LR). Slight "wobble" under braking due to front end settling unevenly (amplified if front end is also altered) in (#2) requires very hard acceleration to work correctly, and can cause severe oversteer if clutches are improperly set. Detracts from right turn handling characteristics, due to the spring rate being changed.
Recommended solution: (#1); spring rates remain symmetrical all round, just the speed at which they compress/expand is different. Handling will remain generally the same with a minimum of detraction in left hand turns. The initial "wobble" upon applying the brake is quite controllable if you know it's going to happen, and stops once the chassis goes into full front weight transfer (it does not wobble when going into rear weight transfer, oddly enough).
If the opposite, (oversteer), is occurring, then just invert the above solutions to get the opposite effects.
Tire pressures can be used to fine tune all of the effects. In increments of 1-2 lbs decrease pressures where dampers are softened, increase where stiffened, (this is due to the grip characteristics employed when the dampers are altered). Where springs are altered decrease pressure where springs are stiffened, increase where softened (due to stiffer springs promoting more slip, thus more heat).
The forum is now open for questions/explanations...
Stop the car at the beginning of a long straight, and do a practice start. Does the car just sit in a cloud of tyre smoke? Does it "step out" and doughnut itself into a frenzy? This is usually a good indicator that either: (1) The gear ratio (trans or diff), is allowing the engine to reach to high an RPM to quickly at launch to make the torque "manageable", or (2) there are not enough clutches in the differential to lock both wheels together quickly enough (this is usually indicated by the rear end "stepping out"). Now take into consideration that there will probably be wheelspin and squirm anyway, but it should be controllable through the the use of some throttle modulation and maybe a touch, a very small touch at that, of steering correction. If you get NO wheelspin at all, you probably have too "tall" a gear and are wasting precious power trying to make the car get going. (Practical exercise: get out in a real car, with a manual transmission, and put it in top gear. Now from a dead stop, try to accelerate as quickly as possible, anyway you can barring a downshift.)
Once you get a few starts in, do one more, only this time don't stop after the launch. Run the car up through 1st gear, and grab yourself a piece of second. Pay attention to the tach when you make your shift. Does the needle fall into the beginning of the torque band range, as indicated by the chart found elsewhere on this site? Does does it stay right up there near redline (a good indicator that the gear is too "short")? Or does it pretty much scrub all RPM out of the motor and act like somebody just tied your rear bumper to a tree (quite possibly, this gear is too "tall")? You should see the needle fall into the mid to low side, around 2-500rpm from peak, of the torque band to get the optimum power to the wheels as rapidly as possible. Do this for each gear until you are satisfied that each gearshift will result in a good feed of power at the bottom of each gear.
There are some exceptions to the above though:
Keep the lower gears (1 and 2) wider in ratio, (bottom of 2nd falls to low side of torque band), 2 and 3 a little closer (3rd comes in a bit to the middle of torque band), 3 and 4 more yet, (4th hits toward the top of torque band), and 5th should start within the last little bit of 4th to squeeze that last couple MPH out before the turn (5th usually should come in at the top of torque band within 2-300rpm). Remember that 5th is used to get the last little bit of speed that 4th won't supply, you've already done all the hard/rapid acceleration you're going to in the first 3 gears.
When downshifting, using the brake with throttle lift, a good transmission will allow very rapid downshifts without scattering the engine because of over-revving.
The purpose of bump rubbers is to prevent the suspension control arms from impacting the frame, when it is moved through its range of motion. Due to the lower control arms mounting point generally being further inboard on the frame than the upper (it's a geometry thing), there is a point at which the lower control arm will come into direct contact with the frame of the vehicle.
To prevent this from happening, bump rubbers are placed on the outboard side of the frame in such a fashion that they "catch" the control arm and stop it before it can do itself harm on the frame. The rubbers come in different lengths to aid in this proccess, longer stops will more gradually they will slow a rapidly moving control arm, and shorter ones will allow for less ground clearance.
Short bumpstops are used in aiding setups that use very low static ride height, since taller stops would allow the chassis to rest upon them, not the springs. The major detraction of short stops is the fact that they very abruptly "catch" the control arm, so there is a bit of a jolt when coming into contact with them. Unfortunately this usually happens at the most inopportune times, like at the point of full chassis roll mid-turn. If you notice the generally smooth handling traits of you well thought out chassis going to crap VERY suddenly when rolling through turns or zagging about, then you have probably achieved full suspension travel due to the control arm hitting, then fully compressing, the bumpstop. Suddenly you have a full rigid frame with no suspension to work with at all.
Long stops are more in tune with the higher ride heights. This is due to the suspension having more time to accelerate (once chassis roll begins, the speed at which the control arms move increases with distance due to leverage overcoming spring weight), and the longer stops provide a more gentle "catch" by giving gradual deceleration not offered by shorter stops.
Think of the above examples in terms of catching an egg tossed into the air: if you toss it a short distance into the air, you can basically hold out your hand and make the catch with no need to move your arm in the direction of the fall (low ride height, short bump rubbers). If you toss the egg quite high, you will be forced to move your arm in the direction of the fall to "cradle" the egg as you catch it, lest it break and mess up you new Nomex tuxedo (high ride height, long bump stops).
There seems to be a lot of confusion as to what is the "right" length stop for a given chassis/ride height. I myself believe in trying to get the springs, dampers, and swaybars set in such a fashion so as to not even really need them. Swaybars and dampers seem to have the most dramatic effect in these configurations. Examples: if the front dampers are too soft then the chassis will tend to slam forward on the front bumpstops because the forward transfer of weight is to rapid for the springs to compensate for on their own. If the anti-rollbars are to soft, the chassis will have an overabundance of roll and it will allow the laden side of the suspension to bottom. Remember, the design of the anti-roll bars is to pull the chassis DOWN (by pulling the unladen wheel up) as squarely possible, thus maintaining fairly even tyre contact pressure, instead of letting it heel over.
To determine if your bumpstops are too long, drive about in the setup in question. If the chassis exibits radical changes in the way it handles, such as rolling through a turn and somewhere near the apex the car very suddenly throws itself around for no rhyme or reason, you've probably settled the chassis down on and compressed the bumpstops, robbing the car of suspension feel. Basically it takes your softish spring settings and turns them to stone. (bump rubbers are MUCH firmer than even the stiffest springs)
Bump stops that are to short are a little easier to detect. Generally most suspension failures (not accident induced), are a result of slamming the control arms against the frame one to many times (a result of having dampers to soft, too much chassis roll, baffing curbs repeatedly or that really cool looking high jump).
Placement of the bumpstops, like the placement of the control arm mounts, is fixed. The only variation available is in the length of the rubber cone. Be aware, though, that suspension travel will generally be capable of exceeding ride height. This means that if you bottom the chassis on the road surface, there is still some suspension travel still available. The chassis' have this designed into them to aid in suspension travel during dynamic roll, and in situations simular to when you put one wheel up on a curb. If the suspension were to stop travel at the ride height limit, you would be snapping control arms left and right in the aforementioned situations.
If you must use bumpstops longer than the 1" minimum, I would recommend keeping them at 1/2 the static ride height in length or less. This should give adequate length for various ride heights, without causing interference with normal suspension operation. Although if you tend to like very soft springs, dampers, and swaybars it would be wise to shorten the length even more.
I've been doing some practical research on the question of "how is toe affected by chassis/suspension", so here's the poop:
First off you have to know where on the wheel the tie-rods connect to the hub, (out front up high, down low in back.......). The Ferrari's show them being out to the front and above the centerline of the wheel, for example. [The Lotus too - ed.] Okay, with that established, let's get down to the proverbial brass tacks.
If the tie-rod end is above the wheel center, and to the front, the outside wheel will start to be pushed outward, (less toe), as the chassis rolls over on the the suspension.
This is caused by the fact that the chassis, rack and peanut, and inboard ends of the control arms, are fixed solid, whereas the wheel and outer ends of the control arms are not. The tie-rods must bridge this movable, constantly fluctuating gap and since they do not actually have any means of changing their length the result is that the wheel is turned a bit in or out as the tie-rod pushes and pulls at them during roll cycles of the chassis. If you are going down the straight, using the above configuration, the wheels have the toe set where you put it, (although they should theoretically toe in and out variably as the front end dips and rises, but that is another story), you start to roll it into a high speed right, (quick!!, where are we?), the chassis begins to roll to the left on its springs and in doing so the left side control arms are pivoted up on there fixed frame mounts making the distance from the wheel hub to the rack and peanut shorter and since the tie-rods are of fixed length they are at that point actually pushing the front of the wheel outward, thus increasing the degree of toe.
If the tie-rods were mounted to the rear of center and above, the opposite, (toe in), would occur due to the back of the wheel being pushed outward instead. Now, if the tie-rods are mounted BELOW center and to the front the toe would be DECREASED instead of increased due to the tie-rod effectively being shortened, suspension goes up, tie-rod exerts pull. And, as you may have already guessed, if it is mounted below and behind you get toe out.
As for the inside front wheel, the same thing happens but in the opposite direction, (suspension going down instead of up), but it seems, through practical experience, that the effect is not as pronounced on the inner wheel as the outer, possibly due to inequality in suspension motion form the inside to the outside, I don't really know why but it just does.
Now, if you are ready for a real headache, the above principles can also be applied to the REAR wheels also!! Since they are on independent suspension mounts and have the ability to be toe adjustable in the same fashion as the fronts, they just have a linkage rod in place of the tie-rod. Also the length of the mount, from wheel center to tie-rod outer end, (we just call them ears), can have an effect also, if they are on the short side the effect is amplified due to a shorter radius from center, longer just the opposite.
In fact, one of the karts I work on is a left-turn-only and we have determined that the left "ear" should be shorter than the right. The reasoning is that since the driver theoretically only turns left, the left front can be used to help "pull" the front end into a turn, as in when the front wheels are turned left the degree of toe opens out quite dramatically, like about 3/4 inch, yet when going down the straight it is generally set at about 1/16 toe in. I believe the politically correct name for this principal is "Akerman Angle Steering".
There, that pretty much explains it in the best terms I can come up with, and I hope it clarified, (yeah, right, I barely grasp it myself let alone try to explain it!), your query about how toe is affected by chassis movement. Hope this is of some bit of use to you, although I REALLY doubt GPL has it modelled, but hey, you never know...
Ricardo mentioned to me that he thought that the Ferrari had relatively narrow rear tires (front 4.75-10/30-15", rear 6.00-12/30-15") and handled a bit like a rally car. This leads us to yet another discussion of racing physics, which may or may not be modeled in GPL...
If a given vehicle has narrow tyres the contact patch is much smaller than with the standard steamrolling meats, but it also means the potential for traction is GREATER due to there being more PSI on the given contact patch on the road, within reason of course, than there would be on wide wheels. It's a matter of using a tire just wide enough to overcome the torque with a minimal amount of wheel spin. Although the wide tyres have a greater potential for more top speed due to their ability to "stand up" through centrifugal forces applied to the wider tread surface under high rotational speed, (look at a top fuel dragsters wheels late in a burnout or at the end of a run), taller tyres make for more speed, (if you do it right, it can be used as a form of a taller "gear" than what actually is in the driveline, variable ratios without variations...). But, on the other hand narrow tyres create less rolling resistance, therefore they are creating less mechanical drag on the engine, so they are capable of producing higher speed also, just not as much as a tyre that actually changes to a larger diameter. That is why you can get slightly higher speeds with overinflated tyres, less rolling resistance plus a taller, diametrically larger, tyre.
The narrower tire also has more stable handling characteristics due to less "flex" of the sidewall under lateral acceleration of the vehicle, although the traction loss at the "sheer point", (I assume this has something to do with a form of slip angles), is much more abrupt than when using wide tyres. This is due to the wider tyres having a broader contact patch ACROSS the tread, as opposed to the narrow tyre, which has a more square patch.
This would go a long way towards explaining why the car acts like a rally racer, (good comparison, by the way), when you are giving it the heavies through turns. Once the traction is broken, it takes more effort to stop wheelspin than it does to just keep the hammer down and drift it out.
This is partially the reason why F1 still uses bias-ply tyres instead of radials, you can get them hotter and the grip characteristics are more consistent than the radials that most other forms of racing use.
Ricardo asked me "since the roll center should be as close to the ground as possible to reduce the vertical chassis movement due to lateral forces, should we be tweaking the static ride height to achieve this"? I calmed him down, popped a beer, and explained:
Roll center height, IMO, does NOT have bearing on the chassis roll during cornering, CoG distance above roll center, how "top heavy" the chassis is, has EVERYTHING to do with chassis roll. For example, if a chassis has a roll center 25" above the ground and a CoG of 26", the leverage created by the CoG on the roll center will be minimal thus reducing roll. A chassis with a 10" roll center and a 26" CoG will probably fall over due to the INCREASED LEVERAGE of the CoG on the roll center.
The CoG is higher than the roll center because the top half of the engine is significantly heavier than the bottom. The cylinder heads, intake manifold, cam shaft(s) and carburettors have a higher sum weight than the base of the engine block and crank shaft. Also, physically, the CoG of an adult male is located at the center of the chest [when sitting - ed]. All of these points are generally above the roll center, especially in 60's era racers.
To compensate for this, engineers have redesigned engines with flatter angles between the left and right bank of cylinders - a wider engine creates a lower CoG - and the use of "lay down" style driver seating with physically smaller drivers. This moves more weight down closer to the roll center, thus reducing chassis roll. The chassis now tries to push sideways instead of heaving over.
Since there seems to be no real means of measuring CoG in GPL, I would suggest using the imaginary line from the center of the steering wheel (most racers and engineers set the wheel at center of chest, just short of arms length for better muscle use (pectorals are stronger than arms), and the point just at the top edge of the gearbox.
If you can get these two points level with each other you should have an equilateral, (or would that be "equalinear"), CoG. Now take into consideration that you cannot just lower the whole thing down too far because you will begin to get suspension bind due to the mean vertical distance increase between wheel center and the suspension arm center (suspension geometry and wheel center are designed together, since the suspension is designed to give maximum optimal wheel contact through the full range of chassis motion), the greater this distance the more potential for deviation from designed suspension geometry with the resultant loss in handling. But, if you keep the CoG LEVEL you will effectively reduce extra roll by means of having the narrowest fluctuations in CoG height. Hence moving the car up and down on the ride height in equal increments front to rear (not necessarily the same height, though), to keep the CoG level.
Ideally, if you can find the CoG of "x" chassis and level it, by setting front and rear ride height, you can then begin to raise and lower the whole thing in equal increments until you get a good balance of suspension; too low it will bind, too high and it will wallow.
This is more or less what I did to the Ferrari, if you want to check it out, change the ride height in some small but unequal increment. I think you will be shocked at how much WORSE it handles! Then raise and lower it in equal increments and note that the handling remains relatively stable regardless of height until you go so low that the suspension binds or you go so high as to overcome the counter leverage created by the sway bars and it begins to flounder about.
I believe that if you can get the CoG level, lower the chassis to just above the bind at full roll point then make the sway bars adequately stiff enough (not necessarily full stiff or the suspension won't be able to "work"), to compensate for difference in roll center to CoG you will have a very sweet handling chassis.
Since the rear of the car is heavier than the front, it will have greater inertia and so will be harder to start rolling than the front. But once roll is under way, the rear would be more difficult to slow/stop, requiring the front bars to be stiffer lest the chassis go into "tricycle" as the weight of the rear begins to accelerate through roll at a faster rate than the front. Specifically, if the rear rolls into a corner faster than the front, the inside/unloaded front wheel will tend to lift up.
By lowering the front, the roll resistance at the front would stiffen due to the decreased leverage that the CoG has over the roll center. The rear would remain approximately the same (an effect of crossweight from the stiffer front end).
You can shift the weight of the car fore and aft by altering the rake angle of the car. By lowering the front of the car (in relation to the rear) you effectively shift the CoG forward, because the CoG is higher than the longitudinal roll axis.
It has been my experience that the more static weight you have at one end, the softer the opposite ends ARB needs to be. So it you lower the front, you shift weight forward, and thus need to soften the rear roll bar.
I took some measurements of the Lotus front suspension and used them to construct a model to explore the way camber changes during chassis roll.
Assuming the static ride height stays relatively the same during roll, outside wheel goes up inside wheel goes down middle stays around the middle, (hey, they don't model camber thrust, so they probably don't model ride height to roll axis proportions either). The camber change is approximately 1 degree of wheel camber change for every 1 degree of chassis roll up to 8 degrees deflection. After this point the degree of wheel camber change amplifies to 1.6 degrees to 1 degree of chassis roll. This gives a total of 8 degrees camber change at 8 degrees and 13 degrees camber change at 11 degrees chassis roll.
The model showed that the difference in degree changes from 8 degrees down and 9 degrees up are the result of the suspension running out of room to move, the proportionate angle between upper and lower control arm changes drastically at this point and the lower control arm begins to push quite hard on the lower wheel mount.
Since I could not model lateral g-forces, I do not know how much effect G's would have on the outside wheel, but the current model shows that inner and outer wheel move propertionately and equally to each other.
Ideally, if we could find out the slip angles of the tyres we could determine the best camber to run for optimum contact patch use without having to make a hundred runs checking grip.
When we set up a chassis for a circle track we try to get optimum contact patches on the inside tyres as opposed to the outside, ( the fronts are all we have available to adjust camber). The reasoning behind this is that the inner front wheel aids more in pulling into the corner, as opposed to the outer wheel pushing it in. The same can be said of the inner rear tyre helping to keep the outer from slipping out.
One of the major prerequisites for this to work is a well balanced chassis, one that does not have a lot of roll. It is easier for the unladen tyres to help hold the laden tyres in place than it is the other way around. This is due to the fact that, if the laden wheels are producing more grip, (using equal temps as a guide), they tend to cause the chassis to lift to the outside, not roll, thus raising the unladen wheels off the road surface and actually reducing overall grip. If the unladen tyres are heating evenly this is a good indicator that they are providing optimum grip to the laden tyres, ( with their unequal temps).
The physics behind this are that if the unladen tyres are producing more traction than the laden they tend to cause the chassis to be forced straight down instead of rolling, thus increasing weight and weight distribution in a more equal fashion across ALL the wheels, as opposed to just the outers. You need some slip in the outer tyres for this to work efficiently, or the grip of the outer wheels will effectively reduce overall grip by lifting the inners off their optimum contact patches. (the total sum of inner tyres on full optimum contact patch and outer tyres on say 50% of optimum will produce 50% more grip than having just the outer tyres at 100% optimum, with the inner tyres doing nothing more than stopping the chassis from falling over. It's easier to drag a tyre sideways than it is to push it, try the pencil eraser, like the camber thrust experiment, for an example of this)
The way to achieve this effect is to use camber as you have to optimize inner tyre contact patch, and through the use of stiffer roll bars, ( hmmmm, you've done that too...), and softish springs that allow the chassis to be pulled down, not over. I don't believe bump and rebound have an effect on this, just overall spring tension.
Ideally, you should look for equal temps across the unladen tyres, with a temp approximately equal to the unladen tyres at the inner edge of the laden tyres. This should provide the best usable contact patch area over all 4 tyres. It'll also drift quite nicely, as the weight is balanced over all 4 wheels, not just forcing the outer tyres to produce the grip, drift, steering and stability, alone. Basically, the car should turn FLAT, not rolled, or as close as possible to it.
Ricardo mentioned that he had heard of some GPL drivers who were trying small amounts of front toe out to get an 'Akerman effect'. What's that?
Akerman steering is the principle of matching the individual front wheel tracks to the radius of a turn through the use of toe. The actual pivot point for a vehicle is the centerline of the rear wheels, this causes the front wheels to actually need a different radius from one another due to the width of the chassis, the inside front wheel will have a different turn radius than the outside front.
The angle can be measured by having an imaginary radius line from the center of the turn through the rear axle centerline. From this point you would draw an arc, that matches the radius of the turn, from the center of the inside rear wheel through the center of the inside front wheel, and then another from the center of the outside rear wheel through the center of the outside front wheel. Now, this is where scale drawing comes in handy, (could you see the piece of paper for a 150m radius turn!?!), go to one of the front wheels and pivot it on its center point until the radius line from its respective rear wheel is aligned with the fronts linear centerline, then do the same to the other front/rear wheel. Now we have a graphic that actual measurements can be made on.
From the radius point of the turn, draw a straight line through the exact center point of one of the front wheels, now draw another line from the same radius point of the turn through the other front wheel center. This must be done VERY carefully, and ACCURATELY, as there WILL be a difference in the path of the respective lines! Once lines are accurately in place you can take a protractor reading, using the axis point of the turn as the alignment point for said protractor, on each of the lines that go through the centerlines of the front wheels. The difference in angle of these lines is the Akerman angle. If you can cause the front wheels to turn at this angle through the use of toe, (it's actually a form of a ratio), the front end of the car will have the smoothest, most accurate angle for ideal trajectory around a given turn. The only problem is that, you as a driver, must be able to consistently drive this radius to get the full benefit of the now very accurate front end geometry.
In GPL, if you desire to use this technique, you would need to know the average radius of all the turns on a given track, and set the front end to match this average. This would give the best all round front end handling characteristics for the given track. NOTE: hairpins don't count in the calculation of average turn radii on the track.
Now, as an aside, if you are driving a circle track, (by the way, are any of them actually "circles"?), you can also calculate stagger, the difference in rollout of the rear tyres, by just knowing the radius of the turn ( or the radius of your preferred line), your inside rear wheel circumference, and the width of your chassis' wheelbase. The outer wheel will have a larger base circle than the inner, due to it's being farther from the center of the turn, (difference in wheelbase). You need to calculate the difference in rear tyre rollout by finding the difference in the base circles of the inner/outer tyres. pi * radius of turn squared for the inner wheel, radius + wheelbase width * pi squared for the outer. With these numbers you can find the ratio if inner wheel revolutions to outer wheel revolutions and apply it to the wheels, ( you want the wheels to end up with a ratio of 1:1, wheel revolution wise, by making the outer wheel larger). This will cause the rear of the chassis to want to follow this optimum path with no wheel bind through the arc of the turn. When used in conjunction with the Akerman steering angle above a circle track driver will notice a dramatic increase in the ability of the chassis to roll into, and hold at speed, the desired line through the corner.
It must also be noted that the LARGER a given turns radius is the LESS rollout difference there will be, as the turn is getting closer to a straight line the larger the radius becomes.
EXAMPLE: If you have a 100' turn radius, 150" inside wheel circumference and a wheelbase of 6'...
All wheel circumferences are imaginary and are in no way meant to imply that you'll be using wheels this size, ( they are about tractor tyre sized, actually), they are merely meant to be used as an example........
Dampers are used to control the speed at which a spring expands/contracts, by robbing the spring of energy as it moves, which aids in preventing the "wallows", kind of like an old fashioned baby carriage does with it's rolling and pitching. If you were to make the rebound harder than the bump, the effect would be that laden springs would compress at a proportionately faster rate than the unladen would expand, and effectively raise the chassis vertically, thus causing a higher CoG and adding leverage to the roll center and defeating the extra tension that the heavy laden spring, (with a soft bump), would have. If they were set soft rebound/hard damper you would also get a certain amount of "lifting" due to the laden spring not wanting to compress at the same rate as the unladen spring was expanding.
Example: Pro-Street drag cars generally use 70-30 "uplock" shocks on the front, (similar to 5b and 2r), and 50-50 " downlock", shocks on the rear,(any equal b/r set, preferable softish though). This causes the front end of the chassis to go to full front spring extension very rapidly, and the rear to squat faster, (but not as fast as the front expands,which are at full extension almost immediately. Ideally the chassis will continue to transfer weight through the slower compression of the rear dampers, all the way down the track), for more traction, due to the rapid weight transfer to the rear. The rear is left around 50-50 so that when you hit the end of the run, and jump on the binders, the rear does not let itself expand so fast as to throw all the weight onto those little bicycle tyres you've got on the front.
Basically, you want the dampers to let the spring expand and contract at the same rate, (2+2, 4+4 etc.), because the style of racing puts chassis loads in ALL directions not just linear. With the dampers set at such a strength so as to not defeat the sway bars. If dampers are to stiff, the swaybar will not be able to "lift" the unladen wheel quickly enough to keep the chassis level, if to soft then the bars will lift too rapidly, throwing off the roll center. Remember, you are trying to use the weight of the chassis to benefit handling, not defeat it. Ideally you should get a minimum degree of roll, controlled by the sway bars, not the springs/dampers.
As far as having the dampers set proportionately to weight bias, I don't believe this is necessary. It is more important to have the dampers set soft enough to let the sway bars work but not so soft as to let the chassis wallow about on the soft springs, (you know, the ones that were put in to aid in grip). The springs are still what hold the chassis up, and are set with the weight bias already accounted for so weight bias on the dampers would defeat the purpose.
I would recommend , once you have spring tensions set for weight bias, using dampers in the 2-4 range. Anything softer and the chassis will begin to wallow because the springs are moving to freely, any higher and the dampers will defeat the ability of the sway bars to do their job in a timely fashion because they are fighting the damper, not the spring. Just get your spring tension where you want it, and starting with the lower end of the damper range, move up the damping until the chassis stops rolling and wallowing about. If you go to far the respective end of the chassis will begin to lose bite in the turns because the sway bars are not being allowed their full range of useful motion. As far as having heavy front vs soft rear dampers, or any combination thereof, so long as the dampers do not interfere with the sway bars or increase chassis roll to any great degree you should have no problem.
Remember: The springs hold the weight, the sway bars balance the weight, the dampers control the springs expansion and contraction, and the efficient motion of the sway bars. They do not have bearing on actual weight, just how it's moved about. Balance and timing, balance and timing........
|
|
