Reaches for popcorn. :D
Reaches for popcorn. :D
No need for popcorn, it's a simple concept.
In reply to Knurled:
I should explain something. I've read Herb Adams and I undestand the concepts. I understand how you and Herb Adams think it works.
I could very well be wrong, but I think you and Mr. Adams are mistaken in this instance.
In reply to DaewooOfDeath:
If the car had no suspension whatsoever it would have the exact same amount of weight transfer as a car with suspension. It would just happen instantly. The only thing the suspension does is slow the weight transfer. And by adjusting how quickly that weight transfers you can effect the balance of the car. It would take a lot to convince me it works in any other way.
Speed of weight transfer can affect grip. If you shock-load the tires with the transfer too quickly, that can make them break loose suddenly while loading them up more gently might let you get more total load on the tires.
Nick (Bo) Comstock wrote: In reply to DaewooOfDeath: If the car had no suspension whatsoever it would have the exact same amount of weight transfer as a car with suspension. It would just happen instantly. The only thing the suspension does is slow the weight transfer. And by adjusting how quickly that weight transfers you can effect the balance of the car. It would take a lot to convince me it works in any other way.
I agree with what you just said, unless that suspension has swaybars and/or something that moves relative stiffness side to side in response to roll forces.
In reply to DaewooOfDeath:
Maybe I'm getting confused by your terms. You keep saying "adds to weight transfer" maybe what you are meaning is "increases the rate of weight transfer"? Because sway bars can only do one of those things.
DaewooOfDeath wrote: In reply to Knurled: I should explain something. I've read Herb Adams and I undestand the concepts. I understand how you and Herb Adams think it works. I could very well be wrong, but I think you and Mr. Adams are mistaken in this instance.
Herb Adams is wrong about some things too. Antis in geometry don't work the way he seems to think they work.
Ignore touching the control arms with anything, and just grab the roof and push on it with a given amount of force. That is your cornering load. At a given speed and turn radius, that will always be the same. The load change on the tires will always be the same. If you make the springs stiffer, or make the sway bar stiffer, you will change how much you push the car over, but the load on the tires will not change. The springs or sway bars merely resist what your hand is doing. If you push the car hard enough to lift the inside tires completely off the ground, it will take the same amount of force to do so, no matter if the suspension goes through its whole travel or stays completely motionless.
See now?
The sway bar is doing the exact same thing as a stiffer spring. It is helping the suspension resist side loading to keep the chassis more stable. It isn't actually increasing or reducing side load relative to a set of springs that have the same amount of roll resistance.
I can show this with math if it still doesn't make sense.
In reply to DaewooOfDeath:
I guess I was 1 for 2 on trying to read between the lines on your original two points.
So if I'm understanding you correctly, you're saying that a car with stock springs and infinitely stiff sway bar will load the tires differently than a car with stock sway bar and infinitely stiff springs? Either way, roll stiffness is infinite, but you're saying the former puts more weight on the outside tire than the later. In fact, it seems to me that in order your argument to hold true, 100% of the inside tire load would get transferred to the outside tire with any lateral load via the infinitely stiff sway bar, but none none of it would via the infinitely stiff springs. I don't know about you, but hat doesn't sound very accurate to me.
Have you tried doing a free body diagram of the two scenarios to see if that's true? The system in equilibrium (steady state cornering) is pretty straight forward math. The center of gravity is the peak of a triangle where the force is being applied, and the tire contact patches are the base points of the triangle where the counter forces are being applied. All forces and moments must necessarily cancel to zero.
Unlike empirical observations, the math doesn't lie.
Driven5 wrote: So if I'm understanding you correctly, you're saying that a car with stock springs and infinitely stiff sway bar will load the tires differently than a car with stock sway bar and infinitely stiff springs? Either way, roll stiffness is the same and no body roll can occur. It seems to me that your argument implies that 100% of the inside tire load would get transferred to the outside tire with any lateral load, via the infinitely stiff sway bar.
However, if we make the pavement anything less than 100% glass smooth, flat and level, those 2 setups will load the tires differently.
In the infinite springs example, the suspension cannot move at all. In the infinite sway bar example, the suspension can move, but both wheels at a given end of the car must move together. As far as grip, that would end up being better (although still not good).
In reply to rslifkin:
That may be correct, but in order to validate the hypothesis being presented, the weight transfer needs to be considered in isolation from countless other variables (on a perfect surface) and still hold true even at extremes.
Keith Tanner wrote: Also, we're talking about a Miata here. It's a pretty well-developed platform. You may do just as well by saying "here's what I want to do, what do you suggest"?
OK, I'll bite.
I run a 15x8 wheel with 205/50 AD08s (might move to RE-71s or other ultra-high end street tire). HPDE fun.
FM1.5 kit with Tokikos. Stock sway bars for now.
949 racing suggests a large RB sway (54104, 1.125" with .125 wall) with a 14mm rear sway. I know FM sells their own kit with their own material/thickness as well. However, they also suggest a 700/400 spring, almost double the 318/233 that the FM kit got me. So how do they correlate? I'm fully aware that FM and 949 sell to different customers but for now I'm going to use my street (FM) suspension on the track, which is why the odd mix.
I'm gonna call you guys today anyway but it's too easy since there's already a thread :)
Blaise wrote:Keith Tanner wrote: Also, we're talking about a Miata here. It's a pretty well-developed platform. You may do just as well by saying "here's what I want to do, what do you suggest"?OK, I'll bite. I run a 15x8 wheel with 205/50 AD08s (might move to RE-71s or other ultra-high end street tire). HPDE fun. FM1.5 kit with Tokikos. Stock sway bars for now. 949 racing suggests a large RB sway (54104, 1.125" with .125 wall) with a 14mm rear sway. I know FM sells their own kit with their own material/thickness as well. However, they also suggest a 700/400 spring, almost double the 318/233 that the FM kit got me. So how do they correlate? I'm fully aware that FM and 949 sell to different customers but for now I'm going to use my street (FM) suspension on the track, which is why the odd mix. I'm gonna call you guys today anyway but it's too easy since there's already a thread :)
What is the car doing or not doing now that you'd like to change?
I'm actually not going to change the setup quite yet because I only want to change one thing at a time and before next track day my only change will be a more aggressive alignment. It was pretty understeer-y last track weekend.
I'm interested in understanding the relationship between spring rate and sway bar selection. I understand what the sway bar does mechanically but not how you make sway bar choices relative to spring rate. AKA can you choose too stiff of a bar like a 1.125 .188" RB bar with stock suspension.
Interest in better understanding of suspension voodoo was the motivation in starting this thread.
Intuitively, it seems to be that bigger bars and softer springs would be better for comfort, and smaller bars with stiffer springs would be a better choice for performance, as long as the springs are not wound so tight that you wheels lose contact with the surface.
Driven5 wrote: Unlike empirical observations, the math doesn't lie.
Empirical observations can't lie any more than math can, they both however can be misinterpreted.
Toebra wrote: Empirical observations can't lie any more than math can, they both however can be misinterpreted.
The math in question is fundamental and foundational mathematic principles...If that is being misinterpreted, the physics and engineering worlds have much bigger problems than the effects of springs vs sway bars.
DaewooOfDeath wrote: 1. A swaybar works by converting body roll into weight transfer. Since the tire's coefficient of friction decreases with load, this means the mechanical grip of the entire system decreases with bars.
The friction force increases with the "Normal" force. Friction force is friction coefficient x the vertical component of force, the Normal force in engineering terms. F= mu x N as in physics books.
I have never read that the friction coefficient of tires decreases with increase in load but, if so, it would have to decrease at a greater rate than the increase in force for the total friction force to be less.
I would be interested in a reference to the decreasing friction coefficient with increasing Normal force phenomenom in tires.
jharry3 wrote:DaewooOfDeath wrote: 1. A swaybar works by converting body roll into weight transfer. Since the tire's coefficient of friction decreases with load, this means the mechanical grip of the entire system decreases with bars.The friction force increases with the "Normal" force. Friction force is friction coefficient x the vertical component of force, the Normal force in engineering terms. F= mu x N as in physics books. I have never read that the friction coefficient of tires decreases with increase in load but, if so, it would have to decrease at a greater rate than the increase in force for the total friction force to be less. I would be interested in a reference to the decreasing friction coefficient with increasing Normal force phenomenom in tires.
Tires work via traction, not friction. That's a big part of the puzzle 
That's why the amount of slip increases with the amount of grip, both laterally and rotationally. And tires' handling characteristics can be seen easily if you look at a slip angle graph. Old bias-ply tires have long, lazy curves, radials are much shorter and pointier, and low-profile rubber is even shorter and pointer. Makes for "responsive" feeling handling (less slip required for a given grip) but the pointiness means once you exceed max grip, it falls away fast. A lazy long curve makes for a much more forgiving tire to overdriving.
The amount of grip is indeed nonlinear relative to the vertical loading on the tire. Since the mass over a given axle is constant, we can utilize this nonlinearity to change handling by varying how much of the roll resistance comes from each axle...
Knurled wrote:jharry3 wrote:Tires work via traction, not friction. That's a big part of the puzzleDaewooOfDeath wrote: 1. A swaybar works by converting body roll into weight transfer. Since the tire's coefficient of friction decreases with load, this means the mechanical grip of the entire system decreases with bars.The friction force increases with the "Normal" force. Friction force is friction coefficient x the vertical component of force, the Normal force in engineering terms. F= mu x N as in physics books. I have never read that the friction coefficient of tires decreases with increase in load but, if so, it would have to decrease at a greater rate than the increase in force for the total friction force to be less. I would be interested in a reference to the decreasing friction coefficient with increasing Normal force phenomenom in tires.
I see what you are saying, that tires have a non-linear response but if Normal force did not make a difference then the whole science of downforce is false. We know this is not true. I think where we are going apples and oranges is that there is static friction and dynamic friction. A tire that is not slipping is in constant static friction. As soon as it starts slipping it is in dynamic friction and that is where, I believe, your peaky curve comes in.
see this link: http://curriculum.vexrobotics.com/curriculum/drivetrain-design/friction-and-traction
In reply to jharry3:
This might help: http://racingcardynamics.com/racing-tires-lateral-force/
So a tire with 500 pounds of normal load might be able to produce 500 pounds of lateral load. But the same tire with 750 pounds of normal load might be able to 'only' produce 700 pounds of lateral load. So the lighter one would actually produce a slightly higher 'G' cornering limit. However, if that extra 250 pounds of normal load comes from downforce, it can still exert that additional 200 pounds of lateral load over the non-downforce scenario, producing a substantially higher 'G' cornering limit yet.
jharry3 wrote: I think where we are going apples and oranges is that there is static friction and dynamic friction. A tire that is not slipping is in constant static friction.
That is just it - tires work via traction, not friction. The rubber deforms to the contour of the road. Friction theory can't be applied to tires.
Tires that work strictly by friction, are UHP tires in the winter when they are below their "glass" temperature. Doesn't work so good.
Also, ask a cyclist if they'd ever apply hard braking or cornering over a steel plate No traction there, just friction.
Driven5 wrote: In reply to DaewooOfDeath: I guess I was 1 for 2 on trying to read between the lines on your original two points. So if I'm understanding you correctly, you're saying that a car with stock springs and infinitely stiff sway bar will load the tires differently than a car with stock sway bar and infinitely stiff springs? Either way, roll stiffness is infinite, but you're saying the former puts more weight on the outside tire than the later. In fact, it seems to me that in order your argument to hold true, 100% of the inside tire load would get transferred to the outside tire with any lateral load via the infinitely stiff sway bar, but none none of it would via the infinitely stiff springs. I don't know about you, but hat doesn't sound very accurate to me. Have you tried doing a free body diagram of the two scenarios to see if that's true? The system in equilibrium (steady state cornering) is pretty straight forward math. The center of gravity is the peak of a triangle where the force is being applied, and the tire contact patches are the base points of the triangle where the counter forces are being applied. All forces and moments must necessarily cancel to zero. Unlike empirical observations, the math doesn't lie.
My understanding of the issue is that a swaybar effectively functions by "borrowing" compression stiffness from the unloaded side.
This means that while a car is experiencing body roll, the wheel rates on each side of the chassis diverge. A 500 lb spring, coupled to a swaybar, becomes a 700 lb spring on the outside and a 300 lb spring on the inside.
Since this happens by torquing the chassis relative to the suspension, the energy that would otherwise be rolling the car becomes weight transfer.
I don't think this would hold true for either infinite springs or infinate bars because there'd be no roll.
jharry3 wrote:DaewooOfDeath wrote: 1. A swaybar works by converting body roll into weight transfer. Since the tire's coefficient of friction decreases with load, this means the mechanical grip of the entire system decreases with bars.The friction force increases with the "Normal" force. Friction force is friction coefficient x the vertical component of force, the Normal force in engineering terms. F= mu x N as in physics books. I have never read that the friction coefficient of tires decreases with increase in load but, if so, it would have to decrease at a greater rate than the increase in force for the total friction force to be less. I would be interested in a reference to the decreasing friction coefficient with increasing Normal force phenomenom in tires.
Knurled nailed it with his response, but I figured this might help.
Each one pound increase in normal force improves traction at a lower coecoefficient than the previous pound, but since downforce doesn't increase the system's sideload/inertia, it still makes you faster.
DaewooOfDeath wrote: My understanding of the issue is that a swaybar effectively functions by "borrowing" compression stiffness from the unloaded side. This means that while a car is experiencing body roll, the wheel rates on each side of the chassis diverge. A 500 lb spring, coupled to a swaybar, becomes a 700 lb spring on the outside and a 300 lb spring on the inside.
How to describe this... The wheel rates are not changing. The preload on the suspension is changing. The stiffness of the swaybar determines how much preload gets shifted from one side to the other.
Infiniti has some models with "hydraulic swaybars" that make this plain to see, they add hydraulic pressure to the top of the spring on the outside and remove it from the top of the spring on the inside. The spring rate isn't changing, the but the spring load is changing. Of course, it's all computer controlled so they can do whatever active handling voodoo that they felt they needed to do.
And, in retrospect, since it is done with fluid instead of spring steel, it probably adds its own level of damping rather than its own level of... anti-damping?
DaewooOfDeath wrote: My understanding of the issue is that a swaybar effectively functions by "borrowing" compression stiffness from the unloaded side.
Your understanding seems to only be looking at one half of what's going on. Remember Newton's Third Law. The droop from the unloaded side trying to prevent compression on the loaded side, is equal and opposite to the compression from the loaded side trying to prevent droop of the unloaded side. If this is viewed as transferring weight, it's doing so in a circular fashion, such that it cancels itself out.
.
DaewooOfDeath wrote: This means that while a car is experiencing body roll, the wheel rates on each side of the chassis diverge. A 500 lb spring, coupled to a swaybar, becomes a 700 lb spring on the outside and a 300 lb spring on the inside.
Lets start off with the assumption that you mean lb/in, not just lb, as there is a big difference in meaning there. So you're talking about a 500 lb/in spring coupled to a 200 lb/in sway bar. I think we can all agree that this would result in both less compression on the outboard side and less droop on the inboard side vs just the 500 lb/in springs alone. However, remember that even in droop, less travel means a higher effective rate. So while a 700 lb/in effective rate on the outboard side would agree with this, a 300 lb/in effective rate on the inboard side would actually result in more droop relative to the 500 lb/in springs alone...Which is not what actually happens. Thus the real world reduced droop travel would also indicate a 700 lb/in effective roll rate on the inside as well.
That's one of the most common misunderstandings about constant rate springs. It doesn't matter how much load is on them, their rate does not functionally change within their range of motion. In the case of the inboard side, the upward force from the outboard side acting through the sway bar is already fully reacted by the inboard spring, such that any change in load still necessarily act via the same effective rate as the outboard side.
.
DaewooOfDeath wrote: Since this happens by torquing the chassis relative to the suspension, the energy that would otherwise be rolling the car becomes weight transfer.
A sway bar is not 'torquing the chassis' any more than the same amount of total roll resistance applied to the chassis through the springs alone would be. See below.
.
DaewooOfDeath wrote: I don't think this would hold true for either infinite springs or infinate bars because there'd be no roll.
This is probably the other most common misconception about springs. The rate/deflection of a sway bar (or any other spring for that matter) does not in any way affect the amount of force being transmitted through it. Think about it: If you apply 1000 pounds to one end of a 250 lb/in sway bar, it will transmit 1000 pounds to the other side via a 4 inches of relative motion. If you apply 1000 pounds to one end of a 500 lb/in sway bar, it will transmit 1000 pounds to the other side via a 2 inches of relative motion. If you apply 1000 pounds to one end of an infinite rate sway bar, it will transmit 1000 pounds to the other side without any relative motion. Replace "sway bar" with "springs" and the statements all still hold true, because a sway bar is just a spring. In fact, a coil spring is functionally nothing more than a (helically wound) torsion spring...So it actually even operates on the same principle as a sway bar.
So again, if your claims are accurate, they would be just as applicable with infinite rate springs/bars as any other rates. I was just offering an option that simplifies the math you'll do worrying about deflections and rates, since infinite spring rates and infinite bar rates will both still transfer weight in the same manner as they each would at less rates, even though there is no suspension deflection. If you'd rather do the math with any other series of randomly selected bar and spring rate sets to compare the weight transfer against each other, you are more than welcome to do so, as the end results will always be the same regardless of the chosen combination of rates.
.
Edit: I also found an old post of yours with the Steve Hoelscher 'manifesto' and he does appear to achieved good results by following observed trends through trial-and-error experimentation. Which in and of itself is fantastic, especially when it's repeatable within at least a small sampling. And he does certainly understand a lot (most) of the mathematically verified principles. Enough so to present convincing sounding arguments either way. However, he also offers his completely unverified opinions as equivalent facts. Like this notion of sway bar weight transfer that, in the absence of mathematical models, mysteriously occurs outside of our existing mathematical understanding of how simple springs work. This gets bought into on blind faith on his word simply because his alternative explanation intuitively seems legit...Of course, completely disregarding that our mathematical understanding of such (apparently) counter-intuitive physical concepts that run contrary to (i.e. are able to disprove) his claims are based on principles and equations whose derivations have been validated countless times over multiple centuries. If one were so inclined, they could follow the derivations and see this for themselves.
It became quite evident in his explanations of how and why some of his ideas work, that he has been using correlation to determine causation. In a complex multi-variable system that's not even a SWAG, it's really just a WAG...Which substantially undermines his credibility, in my opinion. It's unfortunate, as he does seem to have some interesting (and otherwise seemingly successful in practice) ideas to share and learn something from.
You'll need to log in to post.
