Question about left front geometry

[Mr.Kennedy777]

So with the winter season in place here I decided it would be a good time to hit the books and do a little bit of homework. One of the things I really started to learn about was tires and the forces they put into the chassis. The one thing that seemed to stick out to me when I was learning this was the geometry of left front suspension components. When the track begins to shift from a more longitudinal to increasing level of lateral application of force on the tire, the tires create a torque that tries to stand them up. The current geometry lay outs actually cause the left front to generate an increased roll resistance, by putting the left upper control arm in compression. More angle actually amplifies this effect, and in extreme cases can even cause the left front to lift. While it’s true the overall “moment center” is moved laterally to the left, it is raised when this change is made. Another part is the camber curves aren’t exactly great in comparison to a well designed conventional RF set up, or a RF on a bump device of some kind. To me this configuration seems to be a little more, flat, to be honest. If aero is the purpose of this, then it’s understandable but if it’s to try and work the left front, as well as aero, I think it’s only looking at part of the issue at hand. Unless you were running incredibly strong springs or on bumps on both fronts with a strong ARB, your chassis would deflect a lot of this force through its relatively soft springs and you would waste a lot of the jacking component by distributing it across the chassis when you change it from a mechanical, to elastic, to mechanical loading again as the spring and shock pushes down on the lower control arms which push down on the tire again. So why do people assume moving the moment center further to the left loads the left front more, if in the extreme case of rather large a arm angle, under very hard cornering we can cause premature tire lift because the tire is trying to compress the chassis down while forces are lifting up on the tire? If we keep increasing the angle this problem is only exaggerated. If we’re doing all these really funky things trying to gain on the left and do all this weird crap, why not try running the upper control arm only, uphill from the spindle into the chassis on the left front. With the right configuration this places the IC of the left front suspension outside of the tire, and turns the jacking component into a downward force on the tire directly, it generates traction from the control arms much like how the rear trailing arms do. This would be especially helpful when you want the feel of a heavy spring on entry into the middle and a soft spring on the left front for exit. Another suprising characteristic is that the camber curve can be designed to be way more stable using the right combination of lengths and angles. Also, moment center location can be easily placed at ground level, with minimal movement from chassis motion, both vertically and you can achieve a more left lateral location without increasing roll stiffness. This also helps promote RR loading with the introduction of lateral forces into the chassis. It might look a little weird putting a design on the car, but with the way the left fronts move, when the left front gains loading it loses positive camber. A lot of the movement in the left front on a dirt car these days isn’t so much from roll but vertical movement, since most cars hold their platform a lot longer than they used to. Just throwing this idea out there. Maybe someone is willing to try and look at this and see if there would be any serious draw backs that I am missing here, or if it might actually make sense to try and incorperate a different design for left front suspension in cars that aren’t running bumps on both fronts with an ARB, and are trying to see gains in both LF and RR loading, as well as better stability with increased aero loading? Maybe I am way out to lunch here, but I think if you would want to work the left front and make it more stable this would be how you would do it.

[King1]

I think we are becoming more in the same to asphalt cars. Lf corner being the lowest then rf, lr, rr. Dirt racers or chassis mfr's have been blind or used the dirt late models are different that asphalt cars (which they are) but in all reality we are all turning left and we all have 4 tires. Since the coming of the new longhorn and rocket i think the creators of these machines realize that we arent so different and the whole asphalt thing is leaps and bounds ahead of us and the winners are taking some ideas from them. My buddy that worked for morgan mcclure knew exactly what i was talking about when i asked about leas bolted to an axle. His respons was "you dirt guys just figuring out something that has been around for 25 years?". Dont have an answer but i would love to hear mb racer or matt49's take on it.

[Brian Gray]

First off the wheel is almost always pointed to the right. So that lifting force is moderated buy antidive angles caster and don’t forget pin inclination.

As far as dirt cars and pavement being the same. Yes the principals or physics are but the rules are not. There are much more mechanical items at play. The suspension system particularly the rear asphalt guys wish they could use. I think you have some insight into the dynamics but don’t forget to take into consideration the effect that the antidive angle has on the scenario.

[7uptruckracer]

Asphalt isn’t ahead of Dirt or vice versa. Things trickle back and forth between them
Both I’m heavily involved in both and see a lot of knowledge trading sides. You can’t forget road racing either. The whole tethering the LF is a concept from road racing’s Zero droop setups. I’ve tried running dirt things on asphalt tried the LR axle tube lead on an asphalt car it doesn’t even come close to working. Drive isn’t the word to describe what it did lol. We also use to put concrete or molten lead in the LF lower control arms on asphalt cars same concept as the lead unsprung weight to keep the LF in the track. They key to some of this stuff is to not put yourself in a box take any and all info and see how you can or can’t apply it. Yes the front end is becoming more dominant but it’s far from how asphalt is. I’m not sure a ton of actually geometry has changed. It looks to me like it’s more the bolts ons like shocks and materials used. Then actual roll centers etc.


I think we are becoming more in the same to asphalt cars. Lf corner being the lowest then rf, lr, rr. Dirt racers or chassis mfr's have been blind or used the dirt late models are different that asphalt cars (which they are) but in all reality we are all turning left and we all have 4 tires. Since the coming of the new longhorn and rocket i think the creators of these machines realize that we arent so different and the whole asphalt thing is leaps and bounds ahead of us and the winners are taking some ideas from them. My buddy that worked for morgan mcclure knew exactly what i was talking about when i asked about leas bolted to an axle. His respons was "you dirt guys just figuring out something that has been around for 25 years?". Dont have an answer but i would love to hear mb racer or matt49's take on it.

[billetbirdcage]

I think your over estimating the force on the upper control arm. About 10 years ago I asked one of the well know engineers that races LM's himself if he had info or could run a simulation on the RF upper control arm to see how much compression or tension was on the arm itself so I had an idea of how much it was adding to the lifting or compressing of the RF suspension. Now granted the RF is pulling in both directions: Weight on that corner trying to compress the RF and lateral traction trying to pull on the arm thus lift the car on the RF.

I didn't know which on over came the other or was it mostly one direction. He had already done it and the tension/compression on the arm was minor as they basically canceled each other out and the forces there where what he called minor. Sure they spiked one way or the other, but for the most part he called them a non issue.

Obviously each car is gonna be slightly different as the lower ball joint position to the bottom of the tire contact patch is not the same on every car and will have a large influence on how much compression force is put on the upper arm do to corning force. However many times the force on the LF can be very minimal and therefore apply very light force on the arm to do anything.

Not sure if that answered you question or not, but I see you could be creating a ton of other problems for what may be a very small force to try to use and may have very little to any effect on the LF.

[Mr.Kennedy777]

I think your over estimating the force on the upper control arm. About 10 years ago I asked one of the well know engineers that races LM's himself if he had info or could run a simulation on the RF upper control arm to see how much compression or tension was on the arm itself so I had an idea of how much it was adding to the lifting or compressing of the RF suspension. Now granted the RF is pulling in both directions: Weight on that corner trying to compress the RF and lateral traction trying to pull on the arm thus lift the car on the RF.I didn't know which on over came the other or was it mostly one direction. He had already done it and the tension/compression on the arm was minor as they basically canceled each other out and the forces there where what he called minor. Sure they spiked one way or the other, but for the most part he called them a non issue.Obviously each car is gonna be slightly different as the lower ball joint position to the bottom of the tire contact patch is not the same on every car and will have a large influence on how much compression force is put on the upper arm do to corning force. However many times the force on the LF can be very minimal and therefore apply very light force on the arm to do anything.Not sure if that answered you question or not, but I see you could be creating a ton of other problems for what may be a very small force to try to use and may have very little to any effect on the LF. I’m just wondering... what problems? The right side is usually set up to equal out because a lot of attention has been paid over the years with chassis design to reduce the net jacking component on the right so that it would travel more freely and be more compliant with the track. But then on the left guys have been adding these bar angles to try and move their MC further left, trying to gain more leverage to roll over on the front and use jacking to work the left front more, but this increased jacking in the left front from the bar angles is actually adding roll stiffness.. So what I’m actually asking is for examples of how this would be worse? Or what might be the drawbacks in using this design? Camber curve is improved, at least can be improved using the right combination of lengths and angles with an inclined upper from spindle. Even though the left front jacking effect might not amount to a whole lot, both arms configured properly would yeid a lower net positive camber loss, which is ideal is it not? The idea in racing is to reduce roll stiffness in the front, this is how you would do it.. Designing a spindle that would use these mechanics that still had good inclination isn’t that difficult, not to mention using the lengths and angles to mitigate anti dive angle change with chassis movement, because it does happen. Not to mention, the overall net change in the lever arm of the suspension by distributing forces can be stabilized. This is what I’m asking like, what would be some of the actual draw backs? Yes the camber curve is different than what is normally used, which is better because the more stable you make the front end the easier it is for the car to find an equilibrium when under load. Which makes the car more consistent.. Maybe I’m overlooking some things but if someone would make a list of things they think would be problems that would occur, I would be quite happy to see them. Like I said maybe I’m not seeing something and I’m hoping someone could point out issues with it. Not building it and throwing it on a car, yet, lol, like I said, hoping for some issues that might occur so I can look at them if anyone sees them.

[Austin34471]

My 2cents: Try it and let us know what happens. I don't think it will make or break your setup. I know we like to think we are working the LFs harder these days and using all 4 tires that we are given, but the truth is that there just isn't much weight there like Billet said. But that's just my opinion.
It looks like you've done your vehicle dynamics reading.. but generally understood vehicle dynamics don't apply to dirt racing... sometimes (increasing front roll stiffness is the "idea of racing"?). That's what makes our sport great. It's the last unexplored form of Motorsports left. "No one else does it" isn't a good reason not to try something.
There is no such thing as a perfect suspension. There is a best compromise. Prioritize what you want to accomplish based on all of the data that we DONT have (lol) and build accordingly. I genuinely believe that almost all chassis builders really don't care about the LF "IC" and jacking. If you think it's important (which you are entitled to think that), build a suspension to optimize it and drag it to the track to race the rest of us. Not many people will have anything to counter argue with other than "back in 19__ I tried ____ and it worked so I've left it alone since" or "I saw ___ ____ do it this way so I did too."
Final thoughts:
Only 3 things effect lateral weight transfer: CoG height, track width, and Lateral acceleration. None of those variables include roll center location, IC location, or jacking.. Roll centers are fun to think about, but have negligible importance to lap time speed. The simplified way that we look at roll centers is good for very very general generalizations, but roll centers, N lines, force application points, ICs, blah blah are all all 3D points acting in a 3D space on 3D body. A 2d screenshot doesn't tell a lot of the story.. and the story isn't that important anyways. Jmo

[Mr.Kennedy777]

My 2cents: Try it and let us know what happens. I don't think it will make or break your setup. I know we like to think we are working the LFs harder these days and using all 4 tires that we are given, but the truth is that there just isn't much weight there like Billet said. But that's just my opinion. It looks like you've done your vehicle dynamics reading.. but generally understood vehicle dynamics don't apply to dirt racing... sometimes (increasing front roll stiffness is the "idea of racing"?). That's what makes our sport great. It's the last unexplored form of Motorsports left. "No one else does it" isn't a good reason not to try something. There is no such thing as a perfect suspension. There is a best compromise. Prioritize what you want to accomplish based on all of the data that we DONT have (lol) and build accordingly. I genuinely believe that almost all chassis builders really don't care about the LF "IC" and jacking. If you think it's important (which you are entitled to think that), build a suspension to optimize it and drag it to the track to race the rest of us. Not many people will have anything to counter argue with other than "back in 19__ I tried ____ and it worked so I've left it alone since" or "I saw ___ ____ do it this way so I did too."Final thoughts: Only 3 things effect lateral weight transfer: CoG height, track width, and Lateral acceleration. None of those variables include roll center location, IC location, or jacking.. Roll centers are fun to think about, but have negligible importance to lap time speed. The simplified way that we look at roll centers is good for very very general generalizations, but roll centers, N lines, force application points, ICs, blah blah are all all 3D points acting in a 3D space on 3D body. A 2d screenshot doesn't tell a lot of the story.. and the story isn't that important anyways. Jmo How is it not important? Lol! Why even bother setting up a car then?

[Mr.Kennedy777]

You take away the forces coming into the chassis from the tires, and you literally have no such thing as racing, or even movement lol

[Austin34471]

You take away the forces coming into the chassis from the tires, and you literally have no such thing as racing, or even movement lol
Put suspension/shocks and springs on a gokart and let me know how much faster it goes around a racetrack

[Matt49]

Put suspension/shocks and springs on a gokart and let me know how much faster it goes around a racetrack

Ha ha...so true! And nobody in go-kart racing cares about roll center yet they win tons of races...have trophies to prove it.
What's funny is that we're only figuring it out one step at a time. We think the RF steers the car and LF gets it going straight again. So look at how rigid those corners are when the car is in business mode. RF nailed down on a rigid bump stop. LR jacked up and supported by rigid bars. Imagine how much faster we would be if we figured out how to apply some of the same principals to the other corners of the car instead crutching the two corners we THINK we should care about the most.
Look how rigid the suspensions on an F1 car are?
And then the peanut gallery says, "But Matt, they are so aero dependent, they don't care about mechanical grip."
Matt smiles.
The only reason nobody has had the balls to build a big go-kart and put a late model body on it is because so many tracks are so poorly prepared that the car would disintegrate after one night of racing unless it was built like a tank and then it would likely be too heavy.
We have so overly complicated these cars that it is really ridiculous when you stop and think about it.

[billetbirdcage]

First I want to say there really isn’t right or wrong answers or ways to do things on a race car, they are just different. Many times things work or don’t work that really can’t be explained by any theories, I’ve ran into tons of them that still defy me to this day. Then as technology continues things that didn’t work before now can or vice versa.

Some of this below is going to depend on what or how important you believe roll centers are and how much other things like springs effect where this imaginary roll center is.

Take an older popular car several years ago:

Roll center 1.8” high and 2.5” left
5 degrees camber on both LF and RR static

I’m going to use a certain amount of dive and roll on the car that I know is correct for this type of car approximately half way to the center of the turn. This is for an off the gas situation where you slowing and not on the gas yet.

The dynamic roll center is: 1.3” high and 29.1 right
RF -6.83 camber and LF -1.3 camber

Now doing what I think you described by making the LF upper arm run uphill to the chassis. I simply moved the arm up on the chassis and didn’t alter the spindle or the lower arm.

The arm is now 10.5 degrees uphill to the chassis

New static roll centers: 4.0” up and 50.2” left
Dynamic same roll and dive: 24.7” high and 92.4” left
Dynamic camber: RF -6.83 LF -6.7 (you have lost a ton of camber)

Only way your not going to lose more camber on the LF with the arm uphill to chassis is to really change the lower arm angle also and probably it’s length to or some combo of the two. I didn’t take the time to plot or simulate any of those changes.

As you can see LF camber is worse in your scenario (more negative) in the dynamic state of the LF typically is on a dirt LM. Now is this huge disadvantage? (Billet shrugs) again back to what you believe is more important camber curves, thrust’s or roll centers. On some tracks the LF will compress some and then increase positive camber, but it’s generally only during straight line deceleration and on other rounder tracks where you’re turning in before deceleration the LF may never compress below ride height while racing.

Even forgoing all of the above: Let’s say the car is on a bump stop and it’s compressing the bump stop to or close to solid. Where is the roll center at that point if the RF is basically solid? Is it where the roll center calculator shows it should be or is it more or less at the bottom of the RF tire? So how important is that roll center now? (shrugs).

Here another thing on that same front end above: If you think moving the roll center left or right has a pretty big effect on the car, then moving it say a few inches left or right should be very profound, correct? So how much of a change would 3269.8” be?

Again STD: 1.8” high 2.5” left
Move LF up 1” = -.1 high and 36.7” Right
Move LF up 1.5” from STD = -5.9 high and 160.7 Right
Move LF up 1.8” from STD = 147.9” high and 3109.1” left (that’s a swing of 3269.8” left in moving the arm another .300”)
Move LF up to 2.0” from STD = 15.0” high and 284.6” left (another swing of 2985.2” to the right with only moving .200”)

Things that make you say HMMM!

[Mr.Kennedy777]

I did adjust spindle height, lower geometry and tie rod location as well to eliminate bump steer. From my design the moment center was located underground, 3”, and was 29” left of center line. With 2” of dive, and 3.5 degrees of roll, it was 3.5” below ground, and 25” left of centreline. The whole roll center location is basically the sum of the net jacking forces on the chassis. Your moment center gives you a good idea which corner is utilizing jacking forces more than another when it comes to roll resistence.. If both sides utilize the same magnitude of jacking forces, this puts it in the middle of the two tire contact patches.. If one side utilizes more jacking forces, then the moment center will move that way. If they generate more overall roll resistence, they move up. Etc. However these are just general blue prints. What really matters is how the forces on the tires effect these numbers. Your roll center you’re looking at on your computer doesn’t mean squat once you have varying tire loads. Now you’re getting into force action points. The force based moment center is similar to the regular roll center in the sense it uses the same mechanic design to apply leverage but at varying levels between the two tires. If your RF tire sees the most loading in a left hand turn, in comparison to the left, this changes the magnitude of the forces, and since the right side tire experiences increased load, the force based moment center, which is the sum of the real time loads acting on the chassis through the geometric construction, will exist usually at a different spot than your roll center you measured in the shop. Now even with a stiff bump on the right front, your force application points are still the same, you’re still transferring the same amount of weight, through the spring, which now is just stiffer, and the same amount of weight is being transferred through your suspension linkages. The thing with weight from the bump is the shocks are designed to control it. On the right hand side you want this. There is a lot of demand on this tire, it sees a lot of loading a lot of thrust and it must assist keeping the car pointed the ideal way, which ever way that is, it must still do that. Any time you can use suspension to control spikes in wheel loads even when using a bump, this is better for the tire. Remember we race on dirt on pavement so taking into account inconsistencies in the track surface are also a lot more important. This is also another reason why the tendency although using stiff RF bump stop devices to stabilize front end, gain wheel loading sooner, and get the car lower to the ground passing ride height rules, we still want to increase lateral load distribution to the right rear, and one way of doing this is reducing front roll stiffness through the chassis. This has two significant benefits. One is that it makes setting up a bump device more predictable because the less roll resistance we have on the front of the car, from net jacking forces left and right, the closer we are to being able to set up bump devices more accurately because we eliminate the variable of jacking forces causing roll stiffness. The second is that when the car tries to roll at the rear, from the right rear digging in, the front is more compliant than the rear and this places more loading on the RR, naturally, no different than the j bar pushing the rear end down. Any little gains that can be taken to reduce roll stiffness on dirt at the front of the car will work both front tires more evenly and the RR harder. This is more ideal because 1. Dynamic cross weight of the car is insane based on wheel movement in relation to static. 2. Dynamic cross weight goes up based on weight transfer because of the stiff bump devices we are trying to run. The LR is doing so much work that we need jacking forces to assist the RR. This is why we have the j bar the way we run them. If you run a design that has less camber change during wheel travel, you’re going to eliminate unwanted contributors to wheel loads. I also think this is another reason why bump devices and anti roll bars are a big deal on pavement. A car that has less camber change during travel is going to have more consistent forces working through the tires. A car that changes 3 degrees versus a car that changes 0.3 degrees, big difference when you’re putting a large vertical load on the tire through aero. The trust forces and front end stability can be a big deal to a driver. Now all things considered, cars have been using LF a lot more and as time goes by aero does play a huge part. This is MORE reason to stabilize the front, not a reason to ignore it. You think an F1 car has 3 degrees of camber change? Lol. The RR points one way and doesn’t have to turn about an axis so this creates a lot of ability to gain traction on this tire. And even though the tire is turned right, it’s still rolled right out on the sidewall so that tells you there’s some loading on that tire. This isn’t really over complicating it. F1 cars have tons of problems dialing in low speed turns because they’ve compromised a lot of their suspension in the front for aero. Aero is a big deal on these cars, but not so much yet that front geometry can get thrown out the window. At 200mph with extremely low drag co efficient, and insanely low weight and low CG height, aero begins to trump everything. The car HAS to handle in those conditions. So they make the suspension insanely stiff to maintain compliance at those vertical loads from aero. Not for low or moderate speed cornering gains. Cars these days on dirt aren’t close to that yet. They’re a lot further than they used to be and aero does play a big role, but weight transfer is still very critical. But the thing is we are always driving the same direction. Throw left and right hand turns at these cars, some 40 mph turns, some 150 mph, some 85 mph... Our cars need to handle at the speeds they take, capitalizing on the fact the forces acting on the car are fairly similar for oval tracks driving one direction on them. Right now, this makes teams get the most out of both, aero and weight transfer.

[Mr.Kennedy777]

To answer the last part of your post Billet, depends on the corner speeds and forces on both tires. Just looking at the car and doing math it could be a fair bit of change, but then add aero on both tires, and then you get even more change going back up into the chassis because now the track is pushing up harder on the tires (car is pushing down on the track). The track pushing up on your tires is probaby the most important part of racing. This goes from the forces going into the chassis, during suspension movement and load changes, right down to the tires ability to last under those loads and even maintain its ideal level of traction. You take an F1 car and put the wrong tires on, at 200mph with all aero as a variable, you could be putting one set of tires way outside their ideal friction threshold and the same car with more ideal tire properties would far outperform you. This doesn’t always mean softer either. And this goes back to front end geometry. The tires maintaining stability on the track amongst all conditions makes the car have a stable equilibrium, so it’s easier to drive on hard or low speed corners.

[Matt49]

Don't get me wrong, I'm not saying that we can apply F1 principals to dirt late models. But it is worth noting that dirt late models probably have some of the softest wheel rates (at ride height) when compared to overall vehicle weight of any race car out there.
Here's something to consider or think about. For the sake of argument, let's assume we're back on LR spring ahead of axle and obviously no RF bump stops. Some standard spring setups in those days might have been some thing LF 550 RF 450, LR 250, RR 225. What would happen if you took that same car and doubled the spring rates all the way around the car and raised the center of gravity 7 inches? Would it be faster? I'm not saying it would or wouldn't but it does get a person to thinking. Go karts (of the racing variety) handle EXTREMELY well and will out run a dirt late model on a track of the same scale and conditions (by lap time I mean) and with FAR less power to weight ration. We should be asking ourselves why that is.
Maybe before the LR spring behind, there were other ways to get the LR to drive harder that nobody thought of. But the LR behind was easy to do, and it worked, so everybody jumped on board. And ever since, we've been coming up with one crutch after another to tame and balance the cars (even from an aero perspective) to work around this wonderful LR behind idea. Maybe there was a better way but now it is so far behind us that we've stopped considering alternative solutions.

[Mr.Kennedy777]

But you also need to take into consideration that go karts handle so well because in relation to geometry and change due to chassis movement, and forces on the tires, you’ve essentially removed camber change when it’s undesirable (vertical movement and chassis roll are almost nil)

[Mr.Kennedy777]

This means the loading on the chassis comes down to changes on forces on the tires from the track, and not suspension geometry changes. This makes it have a better equilibrium on its surface. I guarantee you put that design on a late model you’ll have your hands full lol!!

[Mr.Kennedy777]

And I think was it you who mentioned the cantilever idea? It’s still going the same way.. but it all comes down to tires and the forces going through them. I don’t think it would change the ride quality in full hike because your bar angles would dominate anything coming through the cantilever, but it would allow for more hike, I think you might run into problems with steer, and it would be tight as hell. I don’t think it’s going the wrong way. Look at the weight distribution of a sprint car. They’re insanely fast on dirt. And considering they also run soft wheel rates (comparatively, to asphalt or F1). The biggest problem is the surface of the track. That’s why I think LR devices used to cushion the forces and control the loading on the LR somewhere along the connection of links to chassis and rear end are a good idea. But the one thing that’s missing from a dirt late model I think is the next step, is similar to a sprint car. A change in tire for RR. With how much further it keeps going to increase rear loading through LR and j bar, eventually these RR are gonna run out of steam. A wider tire would allow for more room to work I believe. A wider tire would take a bit of time to figure out, but would have its obvious benefits. Since that’s likely to never happen, we need to work on ways to make the cars more compliant, such as LR devices, and optimizing geometry. As aero becomes more important, front end geometry with added loading becomes more important because jacking forces increase proportionate to the vertical load, due to the relationship of overturning moment, pneumatic scrub, and vertical load (Mz=Fz x Pt) moment is torque, so that’s force across distance, N•m, lb-ft etc. When changing camber during travel you change pneumatic scrub and thrust characteristics, left front isn’t AS important, but at the same time, what you do to the left has an effect on the right because it can either move it to the right front or right rear depending on how changes were made in relation to the rear.

[Matt49]

Sprint cars are certainly fast but it sure isn't because of their complex front end geometry. It likely has a lot more to do with the fact that the driver has about 3/4 HP per 1 pound of weight at their disposal.
I think you're taking a lot of this feedback the wrong way. Nobody is trying to argue. You've posted an idea and we would assume you're looking for some feedback and some of that could come in the form of poking holes in your theory. I think you've defended it to the point that I believe you should (as someone else suggested) just try it and let us know what you learned.
You've clearly done your homework but you will find (as was previously mentioned) that many things discussed in great detail in vehicle dynamics books must be viewed with skepticism when you add in variables that were not considered AT ALL when the engineers that wrote those books wrote them. Unequal spring/wheel rates left to right, unequal static weight distribution, EXTREME suspension travel, extremely dynamic vertical center of gravity, etc., etc. We've introduced all of these things into the equation in our cars these days. These things throw all kinds of wrenches in even the simplest things including the roll center calculations. Take the classic front roll center drawings that we have all seen and used to calculate kinematic roll center. But it doesn't take into account spring rates because it assumes equal spring rates. That's a bad assumption to make on an oval dirt car isn't it? Put a 40,000 pound spring in the RF and tell me where the roll center just went. The vehicle dynamics books (at least not one I've seen) can really tell you this but the answer is pretty obvious. Bottom line is that if you take that stuff as complete gospel, you'll often find yourself scratching your head as to why what works in theory often doesn't work in practice.
But give it a shot. Don’t let anybody on here stop you from trying something you believe in. I imagine a lot of people told Skip Arp he was crazy when he was just talking about putting the spring behind the axle on the LR. But at some point he had to actually do it.

[Mr.Kennedy777]

what about reversing the connecting points of the suspension? Like top to bottom and bottom to top? It would push out the left front tire and lose positive camber under jounce, and then keep the spindle moving in favor of stable camber curve during roll? Does anyone have an idea what that would do to your caster and anti dive?

[Mr.Kennedy777]

Another thing I was thinking of, was if the points were reversed, so upper control arm attached to the bottom ball joint and vice versa, if the lower control arm had some type of pull bar or shock or combination of the two that would allow just enough movement to maintain camber during jounce, and was basically solid under compression, so that when the chassis rolled to the right, you could at least maintain camber, if not gain positive camber. It would put the rol center far left and low to the ground, it would be fairly stable and wouldn’t move much.