but when the brakes are on, there are huge torques present due to the caliper being bolted to the steering knuckle. These are there based on the act of braking, and the force vector vs the bushing axis (if it has one) will be different than under normal loads of standard suspension motion when the car isn't braking. Now these torques pivot as well as twist, so the force on each suspension member can point all over the place. This is why I don't see how having some change in contact point can have that much a difference in the magnatude of the force felt by the bushing, both the magnatude and the direction may very somewhat, but not huge changes vs what's there already.

From my limited understanding, what's more at work here than the magnatude of the forces on the bushings is that the relavent moments are much shorter, and that small changes in the bushing location are much more significant, as a percentage, for the moments and hench how they couple.

Now, this is the "statics" version of force vectors. I'll do more digging to see what else is at play here.

In the circles that I travel in, this momentary lock is called "sticktion" (a bastardization of "sticking" and "friction") and is very much a property of the materials used. If you've been in a full race car with metal bushings everywhere and you hear that groaning, that is audible noise stimulated by this sticktion, as it rapidly binds and brakes free. While the race drivers ignore it, I can't stand the sound! In the case of elastic materials, you also have the deformation of the material to "soften" the effect of stiction.

But this is all interesting stuff, and I'm learning here as we go....

Matt

Matt

 

Not sticktion, I know sticktion.

I used the term side view roll center...maybe it's side view instant center...I think so and this may be where some confusion lay because an instant center (fulcrum) defines a swing arm and the length of the swing arm and its hieght define how 'things' will move on the other end.

While looking at a side view of the Mini, run a line thru the LCA arm pickup points - begin with the forward most point and extend the line towrds the back of the car. Now extend a line from the top of the strut tower (which is angled back) until it intersects the line coming from the LCA. Where these two points converge is the side view instant center - SVIC. If the rear most pickup point of the LCA is lowered - anti-dive, the side view instant center is lower and farther away. What does this tell you?

The scenario is just the reverse when analizing anti-squat, the front pickup point is lowered, and the SVIC is now higher and closer to the front wheel.

 

I still don't see any difference to the forces from braking to anti-dive. I don't see why one force would lock a suspension and the other wouldn't. So this is why I don't buy the quoted expanation as to why it's a bad idea.

The fact that there is a geometry that has no anti-dive is a perfect example of how small changes to something will only have small effects at one place and large effects at another. A suspension that doesn't go up or down when under braking has a coupling moment of zero. If you move a suspension point a few mm, you've got to grant that the forces at play won't change that much, maybe a few percent, but now you have a non-zero moment, and the ratio of before and after is huge.

I didn't mean to say that you're spouting BS by using the work magic. I guess I'm just referenceing that the explanation requires information that's not currently in play to make any sense.

Matt

FWIW, there is physical lock, like the teeth of a gear. There is adhesion, like glue. There is static friction, like your foot on the sidewalk, and there's kinetic friction, like sliding on ice. If the lock you are speaking about is the first type, there is bad design. If it's not that, then the momentary lock IS sticktion, where things don't move relative to one another, due to a combination of static friction and adhesion. When it snaps free you go to sliding motion, until the forces are low enough that the surfaces stick again. Whether this happens just once, or over and over (making the bushing groan) it's the same thing.

 

I don't think magic is necessarily a bad explanation because it's all suspension voodoo, isn't it?

While ladder bars sometimes do bind, that certainly can't explain how they can raise the back of a car when under power with their anti-squat. There's simply no way binding can cause motion in the opposite direction.

Maybe I'm being overly simplistic, but I'll consider any force that acts to pick up one end of a car (and simultaneously push the tires on that end harder into the ground) to be effectively additive to instantaneous spring rate. Even if it is just binding... well bump stops also resist compression and those have a spring rate too. I think we can all agree that being hard into a bumpstop results in poor traction because the spring rate is just way too high. Too high to follow the bumps since it's far higher than what the shocks are tuned for and travel is so limited anyway.

Is it any surprise the considerable torque input from braking or acceleration can be used to generate enough force to lift a car, when even YOU can lift one end of the car just by turning the steering wheel? The strong self-centering effect of both caster and steering axis of inclination is due to dead center being at the top of an arc. The farther from center you turn the steering, the more you lift the front of the car. The trick is to somehow convert rotational torque into a wedging effect using geometry, but I don't think it's too difficult given the magnitude of the forces available (the power of hundreds of horses...).

The movement of the pickup point deflects the input forces into a different vector. Think of it like Judo

 

because it has me baffled. I've come to the conculsion that for our suspension geometry, small changes in contact point have little effect on the magnatude of the forces at play there during braking. That's not to say that they can't have large effects on chassis dynamics. But one way I found to think of this that was usefull was the following.... Think about the front strut system under braking. You put a torque onto the steering knuckle with the brakes, and this torque is resisted by the ball joint and the strut (firmly clamped by the pinch bolt, but with a very long lever arm). The torques present are on that's in the direction of the car's axis at the top of the strut, and there are the two at the lower control arm contact point. One of these points is pulled away from the car, and the other pressed into the car. Now if you move the rear control arm contact point, you've changed the axis of rotation of the lower control arm a very little bit. This means that the forces on the lower ball joint are in a slightly different plane than they were before. For very small changes the force change is very small, maybe a degree or two. This would be a maximum force change of 3%-5%.

Now, if I think about that perfect car, where the Cg is in the plane of the lower control arm. While there may be little or no torque present to rotate the car (no dive), there are still the forces pushing one side of the control arm and pulling on the other. So the fact that there is or isn't anti-dive in the suspensions isn't what determines that there's a load on the bushings. Adding anti-dive will only change the forces that are already there. And it will take large changes in geometry to make large changes in force.

So, that's what I figured out. I'm sure it's wrong and sometime later when I have the "ah ha!" moment I'll be posting how clueless I was!

Matt

ps, not all day. It was beautiful out so a neighbor and I put Alison in the back of the Mini and we drove to the coast to feel the new Alta PSRS bushings and stiffer front springs. Could see down the coast for miles! It wasn't all hard work.

 

...I don't have lots of time today, but here is another picture - I use visuals rather than math since I draw for a living.

If the instant center is the swing arm, its position must have an affect on how brake torque tries to pull the front tires backward...think about the action with the instant center in two different locations.

Matt, also, I read the cahpter in Miliken last night again, but quickly. It seems that anti-devises only work on the axle with the driven wheels. Do you understand this to be true as well and what does that tell us? I'm asking a question I do not have an answer to. Milliken is a good book, but the author seems to take snapshots rather than a wholistic approach...which might make for a book twice the size.

 

I think anti-squat from a driven wheel is much easier to visualize than anti-dive, so fundamentally how do traction/ladder bars or torque arms work at all, especially when they have hinges on both ends (traditional traction bars have snubbers and do not even need to be attached at one end)?

This traction bar is for a FWD car with MacPherson struts. It has hinges on both ends and appears to work similarly to weight jacking from an improperly located Panhard bar (imagine a lifted truck going around a corner, where the owner has neglected to relocate it). If it caused binding at all it likely wouldn't work too well to improve traction and they'd have lots of irate customers.

Anti-dive can be produced by the mirror image of a RWD anti-squat scheme (just imagine a RWD car with traction bars screeching to a stop while going backwards--torque is supplied by brakes instead, in the same direction as when powering forward). That seems easy enough to understand, but the million dollar question is what about a stock MacPherson setup with just three attachment points?

Every bolt-on traction bar I've seen works by lowering the attachment point of a lower control arm or raising the attachment point of an upper arm (or adding an auxiliary arm that does the same) so angling the single MacPherson control arm's rear attachment point higher than the front one may provide a similar jacking effect, particularly if the rear mount is much farther from the centerline of the wheel than the front one. Considering the huge 200% effect on anti-squat from just moving the attachment point a couple inches with those bolt-on devices (the car rises as far as you would have expected it to dive) it's probably not unreasonable to expect those Whiteline bushings can achieve their claim of 30% reduction by just moving the attachment point by ~1/3rd of an inch.

If this simplistic reasoning is all wrong, I'm going to go back to thinking it's just magic.

 

So the driven wheels under accleration, and all wheels under braking.

Matt

 

I've been thinking about this all day and I have a way to describe what I think is happening, but my descriptions are almost always in non-scientific terms...but here goes anyway.

We agree that lowering the LCA's rear pickup point establishes a new lower side view instant center that is also farther away front the control arm? And for all this stuff to work and make sense, the car needs to be in motion and a brake torque applied.

Now, due to the new position of the instant center and the LCA's rear pick up point, as the brakes are applied, the new 'better' leverage created at the rear of the LCA arm allows the rear of the control arm to literally lift up the car and in doing so attempts to spin the rear of the car over the front (with greater leverage) driving the front wheels into the ground. This is what I meant previously when I wrote the wheels move backward, which they would do if the ground wasn't there. The term "lock" that I used previously is a subjective term and there must be s better term, but! if you think about the leverage and the rear of the LCA literally lifting up the car, the suspension is sort of locked out of operation...as the body lifts under braking, the springs and dampers offer no resistance to dive, they cannot. This is also why these forces move thru bushings and control arms first.

And as I stated previously, as the wheels are driven into the ground upon intitial braking, the tires can become abruptly loaded and more so with 60% of the weight over them. Further, once the anti-dive affect lessens, the springs and dampers begin to work and this causes a load, unload, then load condition all in an instant at track level braking threshold. Unfortunately, anti-dive can over load the front tires to a point of saturation at which point they slide.

I am sure anti-dive has its place in some applications if not over-done. But as J. Petrich notes, there aint' no free lunch. At some level, anti-dive makes things much worse.

This is the best I can do. I do not know how to apply the proper engineering terminology. Matt and BFG9000, I am satisfied, I feel I know what is going on here, but perhaps you two can put this into a more scientific or mathmatical environment.

As a note, Milliken says we call anti-squat on a front driver anti-lift because the anti-effect is designed to work up front. Splitting hairs again in the already thin air.

 

but the point is that the torques are present at the bushing already, due to braking or acceleration. Moving the contact points only changes how the torque is coupled to the chassis. This is why I can't buy a bushing based description of the downside of doing something like this. That's not to say that there aren't geometric trade offs in doing this... I'll have to think further and my head hurts already! But friction or forces at the bushing doesn't hold water from where I sit.

FWIW, the rear suspension I have on the mustang is a "three-link" trailing arm set up with a torque arm. The shock and lower trailing arm contact points are way low down, looking very, very funky, but boy does it hook up with no lift or droop! I actually didn't understand why it was the way it was till I thought about how the torques coupled to the chassis under braking and power.

And to the guy who mentioned the "slapper" bars that only had rubber snubbers. These were popular on 1/4 mile cars and street muscle cars as they only acted in acceleration. No one cared about slowing down.... If you've owned a 60 muscle car you'll know exactly what I mean!

Matt

 

Matt,

The bushing(s) is not the downside. The lifting in effect do not allow the springs/dampers to work. Think about slamming the brake pedal...the car does not dive...so the springs/dampers are not performing their funstion - intitially anyway. So, the forces are going somewhere...the bushings and control arms intitially do the resisting.

 

it just occured to me that this very same effect happens as the rake of the car is changed. I wonder how much the rake effect is compared to moving the rear bushing point? Also, as the car leans forward under weight transfer, this effect is somewhat induced as well.

Now my brain really hurts.....

Matt

 

...interesting...the instant centers remain the same distance - the swing arm length is not altered...but, the rear of the control arm is now raised, or the same as lowering the front of the control arm which in effect is anti-lift or anti-squat because this raises the SVIC...although the length of the swingarm is not altered.

 

turns out that not only is it built into pretty much every front driver with a soft supension to counter a plunging nose under braking, but it's built into both the Gallardo and this rather interesting ride.

I also found a couple of neat references here and here.

Matt

ps, I also found this link where the author describes removing some anti-dive from cars that come with too much from the factory.

 

I've read about pro-drive as it is anti-dive's counterpoint so to speak.

I'm not suggesting the Alta unit is bad, just that most reactions have equal and opposite reactions; the lift associated with anti-dive is much like using your hands to push up at the ceiling...this forces your feet into the ground. The lifting associated with anti-dive does just that to the tires.

But your point is well taken Doc; Every car is different and a little of this and a little of that maybe what the Dr. ordered - couldn't help myself.

I guess that every manufacturer plays a game with ordinary road cars and though some of these devises may make up for compromises made elsewhere in a design. If the overall affect is positive, it (anti-dive in this case) is acceptable.

I dont't know if the Mini incorporates anti-dive. The Mustang/Pinto article appear quite old, but may not be. If this article is from that period, fwd was not a ubiquitous drivetrain layout. If the mini incorporates some anti-dive, I'm not sure any additional anti-dive is a good thing or a bad thing. I like the notion of the springs performing their function without other contrivances, but the mini isn't perfect.

If one reads all the links you provided, we can come away with, what one does ultimately depends upon where one drives. I wouldn't consider applying anti-dive to one side and pro-drive to another. But I don't race NASCAR either.

My goal everytime I contribute here is to shine a little light on some detail so that we all know what may happen. I also learn as I write...helps me to rationalize my thoughts. If we understand that anti-dive at some extreme is counter productive to good handling, that thinking will force us to recognize what anti-dive does and what to expect.

Chance favors the prepared...being caught off-guard at track level speeds will eventually hurt us.

Ask yourself why you want to and anti-dive...what will be cured? As we have discovered, anti-dive may potentially add to tire saturation during threshold braking. So at some level anti-dive is worse than dive. If, however, brake dive causes unwanted geometry changes and the compromise is anti-dive, then I say it is a good thing. I don't know that the mini in particular posseses poor bump geometry...toe change is negligable so I cannot imagine that anti-dive will help that much. What doesn't work, can I live with it, does the cure complicate matters or make them better???

Will the Mini benefit from a better bushing...probably. I removed the outter ball joint today to check for any freeplay or binding in my lower control arm...none. The P-flex bushing appears to work and is still tight.

Thanks for the links Matt!

 

I have been running the P-Flex from bushings (with motorsport kit) and Ireland rear bushing kit now for about 12k street miles and about 4 track days. I am very happy with the setup and highly recomend it. I have Ireland fixed camber plates in front and Irlend lower control arms (PU bushings) in the rear. I run Koni FSD shocks and Wilwood brakes. On track I use 40x16 tires and street is 45x17. This combo gives me tight control on track with little squirm or movment and adequte street ride quality.
Road Atlanta is getting repaved so maybe the brake zones will be smoother in May the next time I go play. Before there were significant ripples in the high speed (122 down to about 65) 10A zone and the P-Flex bushing really help keep the front planted. It moved around a lot more with factory rubber. The last time I ran I had fast time of class in Group B (mostly vettes, Zs, a few vipers and one GT40). I also am a past multiple time SCCA champion - so I do know what I am talking about. I did run a mix of solid and Delrin bushing on my IT cars (VW Scirrocos) and while good on track these were totaly unaceptable on the street. I feel the Ireland and P-flex parts are a good comprimise.
Wes

 

I mis-wrote before. Anti-dive is in all front drive cars and all cars with soft springs..... But it's a matter of degree for sure.

one thought I had was that anti-dive should help with those with short suspension strokes from lowering in that the car won't compress as much under braking and reduce travel as much.

Another thought is that since the weight transfer is a fucntion of the Cg height, roll stiffness and wheel base, that the forces on the contact patch are the same no matter what. Basically you're using geometry to shift the load from the spring in the suspension to using the car chassis as a torque arm. From a spring theory point of view, this is fine. But it sure is a highly coupled complex interaction.

It would be interesting to know how much factory anti-dive is in the suspension, and how much change the Alta stuff represents. Because it's an off center insert, you really have only two choices for no changes in anti-dive. What's interesting is that if you were to place the index mark at about 4 o'clock instead of 10 then you'd get a bit caster increase with a decrease in anti-dive. Makes one think....

Too bad it's such a biatch to get them in and out!

Matt

 

...But I think the action of lifting the car causes, if momentarily, an infinitley stiff spring (spring is the wrong word, but like an infinitely stiff spring)...since the springs are not working??? I didn't word this correctly. Boy this is tuff to get my brain around...

 

"Another thought is that since the weight transfer is a fucntion of the Cg height, roll stiffness and wheel base, that the forces on the contact patch are the same no matter what. Basically you're using geometry to shift the load from the spring in the suspension to using the car chassis as a torque arm. From a spring theory point of view, this is fine. But it sure is a highly coupled complex interaction."

Matt,

Right on ! What you are describing for anti-dive in the longitudinal direction is the same for roll center action in the transverse dimension. The moment to the wheel is transferred thru the suspension components and not thru the spring. An important difference with anti-dive, I think, is that the geometry results in a certain "binding" of the suspension links which may interfere with wheel motion. Using the chassis as an arm in roll doesn't seem to "bind" moveable links.

Just my thoughts,
John Petrich in Sammamish

 

This is where I get baffled. I don't see why the suspension has to bind, as just transfering the torques of braking get the forces to be very large and torquing the control arm. It's just that without the tilt to the arm, these forces result in no net torqu on the chassis (At least with regard to anti-dive) I still don't see why changing the angle of a control arm a few degrees will make it go from standard motion to "momentary bind".

But I'm just starting the learning curve.....

Matt

 

Doc,

I can see this working, but I cannot describe it in a way that is meaningful...yet. But, it may totally based upon where the side view instant center is and how this new swingarm - lever arm - transfers torque...where the torque is tranfered relative to all else which has not changed.

Unfortunately, the engieer with whom I started to consult with about a month ago is from Sweeden. His accent is one barrier and his terminaology is so esoteric as to be yet another language - two language barriers. I left him a message about this last Thursday and asked him if he could explain in layman terms...dunno if this is possible.

But...

if we take a curcle and run and x/y axis thru it. Bottom is the tire, left axis points to the stock IC or is the lever arm at 3:00. Now, move the lever arm down to say the 7:00 postion still using the same center point at x/y. I don't know if this is the correct analogy because it oversimplifies many of the things happening here, but to me the lower lever arm will impart a lifting action until it reaches some point beyond 3:00...

Matt, John, tell me what you think about this...

 

Michael,
Got confused reading your analogy. Give it another run. Want to think about what you are saying. And, provide some context. Maybe, I missed why this might be important.

John Petrich

 

alta says their bushing is anti-lift if that makes any difference. great discussion BTW.

 

Wes,

Thanks for the encouraging information. A good example of the benefits from a combination of high level of driver skill and some common sense chassis modifications. Check you PM's tonight of tomorrow. I have a few questions.

Regards,
John Petrich in Seattle

 

...anti-lift is another animal alltogether and although this is the reverse of anti-dive (no free lunch) it can aid acceleration in a fwd car. I used this quite successfully in my 1981 Ford Fiesta. It's all a metter of how much in either direction. Anti-lift lowers the front LCA pick up point or raises the rear pick up point.

John, I'm short on time this morning but I can clean up that analogy

Michael