Pretty much any strut design will have lots of dynamic positive camber unless you start out with a lot of static negative camber. The best of them are hard pressed to gain near -1*/inch of bump, which isn't enough to keep the wheel upright as the chassis rolls. I guess what I'm saying is that with any strut at any ride height your dynamic camber is going to be largely determined by your static setting, regardless of camber gain.

But when you start with the chassis side of the control arm already below the knuckel, the situation only gets worse as you compress the suspension, rather than at least staying neutral for the first inch or two of compression.

...course that can be combatted by simply going with extremely stiff springs, so there's is no compression. :lol:

That's really not how it works, it doesn't stay neutral and then go bad. The wheel angle at any given point equals the chassis roll angle minus any camber gain. As soon as the car develops any lateral force at all, there is chassis roll, simple physics. A couple inches of suspension travel will typically work out to about 3 or 4* of chassis roll. If you have camber gain that's only in the neighborhood of -0.4* to -0.9* per inch of bump, which is typical for production car-layout struts, you can do the math and see that if you start out with a static camber angle of 0, the outside wheel will always be at a positive dynamic camber angle.

My main point is that dropping the ride height way down will affect the camber gain, because it changes the angle of the lower control arm, but the camber gain is so small to begin with relative to the chassis roll and static camber setting, that it is not the main determining factor in the dynamic camber you will have when the car is loaded up in the middle of a corner.

This is not to say that the handling will stay hunkey-dorey if you slam the car to the ground! Altering the LCA angle affects lots of other things like roll center height, front view swing arm length, jacking forces, & tire lateral scrub... which in my opinion at least is usually the reason why a strut car will handle poorly if it's lowered too much - not the change in camber gain. It's somewhat of a subtle technical point, I know... but I'm just trying to shed some light on the subject.

W/o me knowing all the technical jargon, I'm gonna need a picture to understand what you're talking about.

The way I had it explained to me, as the LCA pivots, the further away from 0* horizontal it gets, the faster the suspension geometry "gets bad", i.e. if you pivot from -1* to +1*, things are better than if you pivot from 0* to +2*. Even though it's a 2* delta in both cases, +2* is worse than +1* for screwing up the geometry, which is why everyone recommends lowering the car no further than a horizontal LCA at static ride height.

Now, I don't know the details as to why things get worse, so I can't really back up what I'm talking about, but it sounds pretty good to a non-ME like me. ;) If the dynamic camber change is really a linear function of LCA angle, and not as I described it, then it sounds like the handling issues w/ going super low aren't actually due to the dynamic camber change.

This is not true. It's all just a question of application. If you have a heavy street car with a tall C.G., you might want a relatively high roll center to resist chassis roll, which would require angling the LCA's upwards towards the chassis. This then affects all those other parameters I mentioned before.

If you have a lightweight low-slung racecar, you might put more of a priority on minimizing jacking forces which requires a low roll center, so you go with a LCA angle near horizontal or even angled down towards the chassis. Which incidentally results in worse camber gain & more roll center movement, but that's not a big deal if you're mitigating it with static negative camber and stiff springs & swaybars. It's all a trade-off. Every single aspect of suspension design is a trade-off between one parameter vs. another.

It's not perfectly linear.. but like I said I believe it's usually the other parameters I mentioned that have a detrimental effect on handling when a street car is lowered, when they're coupled with the amount of chassis roll that street cars have.

If you want a working stick-figure model to play with instead of a picture, you can download PT's roll center calculator program here. :) As I recall they have some example cars you can load up and experiment with. It's a pretty limited capability program, but it shows some of the front-view geometry like instant centers, roll center, camber gain, etc...

Thread split.
 

Also I went back and looked at the RC calculator program and it pretty much won't let you change any values to experiment with changes. Suck. There are cracks floating around for that program though, so you could have full functionality...

Or you can just edit the MUSTANG file by hand w/ the correct numbers... and open it. :lol:

All I need now are the correct numbers, but I don't exactly have a WRX on hand to measure everything.

Keep that programmer-talk to yourself. :lol: That would take more time & effort for me than finding the crack...

http://www.skapunkandotherjunk.com/i...hap_crack2.jpg

Austin, in most cases, I would agree body roll is a bigger concern, in this case, it is just another contributer to the terrible geometry on the front of the Impreza.

The impreza dynamic camber is atrocious for all the roll reasons and amplified by the short lower arms, short overall height of the strut, arm angle at reasonable ride heights, and minimal castor availability. Oh and arm shape/connecting points suck also.

:lol: Whatever it is, I doubt it's worse than a 1978 Fairmont which is what I deal with every day and have formed a lot of opinions based on.

If you guys measured up all the pivot points on one of your cars I'd be happy to analyze it though, if you want to see what it's actual behavior is.

This thread is stealing my thunder. :(

what thunder?

I had never heard that the Impreza had terrible suspension geometry before. How depressing.

Its the bane of MacPherson suspensions. While it is possible to make the design workable and useful in motorsports applications (with enough money), the reality is, you are starting from an inheriently weaker design. It is an economy design, not a motorsports design.

Ah well, I find that I'm very happy with how my car handles. I'll just compare it to the stock suspension/tires setup and remain happy. I love my Eibach/AGX combo. I can't decide if I want to pick up a Progress Sway for cheap to acompany the solid endlinks that I'm running with a 20mm SOA bar from an 02 WRX sedan.

Also this weekend I added a rear strut bar...the Whiteline quickrelase one.

So do Evo's, Miata's, S2000's, GTI's, have the same problems? I'm trying to put this in perspective. I mean if other cars in our class are just as "inferior"...

Lets be clear... Matt's STI on Street tires ran a faster time than a Z06 Corvette on race tires with the same driver at the last autocross, so the impreza handling doesn't suck.

All I am saying is that the front end geometry is such that some significantly unusual tuning has to be done to overcome it's limitations, and amung those is not lowering the car as much as one might like to lower the CG.

EVO's are strut suspensions as far as I remember, so yes they do. Miatas have a dual A-arm design, so no, not necessarily. I don't know about the others...

Doesn't the Impreza suspension excel on uneven surfaces? I thought the Impreza was known for being able to hit a pothole while cornering without upsetting the vehicle?

S2000 has double wishbone suspension front and rear. While the GTI has strut suspnesion front and semi-independent torsion beam axle rear.

I don't know nearly enough about different suspension setups, but the S2000 double wishbone front and rear is supposed to be amazing.

hahahahahahaha

Sorry, just had to share my mirth. ;)

I'm curious about this as well, particularly how to best counteract it. With the Impreza so widely regarded as a relatively impressive and inexpensive (at least to start with) AWD rally/race platform, it's disconcerting to think of its suspension as being so inferior. Not that I expect it to be perfect without heavy mods, but still. . .

That will typically have a lot more to do with spring & shock rates (ride frequency) & bushing compliance design than suspension geometry like we've been talking about.

Yes and no. The primary gotcha on the Impreza is that it is a semi-heavy front biased sedan/coupe with relatively high CG compared at least to the Miata and S2000, and it has front drive.

The Evo and GTI also have similar issues, but I am not sure to what extent as I have not studied them in detail.

I thought it had to do with "long travel suspension" that Subaru advertises. Don't Subarus have longer suspension components with the pivot point near the center of the vehicle? Sorry I'm not familiar with suspension terminology. I'm looking for a diagram to illustrate what I'm talking about.

This article I'm reading is pretty cool.

This pretty much echos what nKoan was saying about Macphearson supsension attributes.
http://www.automotivearticles.com/up...macpherson.jpg

The article then goes on to talk about double wish bone suspension. Yeah, I like learning.

I wonder if any rally cars use double wishbone setups (or as the article describes it, "unequal length double A-arm suspension"). More specifically, is a Macpherson or the DWB setup superior for off road use? The article claims there is no downside to the DWB setup apart from cost and space requirements.
http://www.automotivearticles.com/uploads/a-arm_img.jpg
The image shows a typical unequal length double A-arm setup. Note the difference in length between the upper and lower arms. This is what gives this suspension its ability to generate negative camber in bump.

Well, "long travel suspension" would be spring and shock rates as Austin mentioned, and not a factor of "geometry". The term "geometry" more applies to the relative mounting points of the control arms, strut, axle, etc. and their relation to each other.

I just remember this flash demo on SOA's site that doesn't seem to be there anymore. It illustrated that the suspension pivoted from the center of the vehicle and therefore the camber was not as adversley affected by uneven surfaces...thus making Subarus safer. Sorry if I'm not making sense.

"Long travel" is a generic phrase for a spring & shock package with, simply, a lot of travel. Typical production cars will have about 6-7" of total bump & droop travel; if you look at a desert race truck, it might have 20" or more. This allows you to run fairly soft spring & shock rates which cushion the impact shocks from racing across dunes & gulleys at 80mph...

Pretty much no suspension design will ever physically pivot from the center of the vehicle, it's impossible due to packaging constraints with everything else in the car... engine, fuel tank, passenger compartment, etc. Even F1 cars don't have their suspension links mounted in the center of the car. That picture you posted is pretty typical, and you can see clearly where the inboard pivots are.

Now, maybe they were talking about the suspension instant centers, which would be a different story.. I don't know.

The reason you see MacPherson struts on rally cars (and stuff like monster trucks and off-road racers) is because they need a way to let the suspension travel long distances.

Imagine that Double A-Arm in a configuration that would allow say a foot of travel... the arms would have to be very long. So on a vehicle that doesn't have space for long control arms, you have to go with a MacPherson strut. Some cars can get away with it (look at the unlimited Baja buggys) but any sort of production vehicle that has stuff like a motor or passenegers between the wheels, there's just not space.

In addition, MacPherson struts are superior when it comes to quick repairs. They'ree more simple and more rugged than a double a-arm setup. This means if the suspension fails in the middle of a stage, the driver and co-driver can get under the car and swap out the strut and finish the stage, whereas swapping a control arm, and associated springs and shocks is a much more difficult job in the field.

But when it comes to road racing, the double a-arm suspension is the way to go. By selecting the proper control arm length, you can design a geometry that allows the tire to have an explicit camber at a specific amount of travel. It removes one of the biggest compromises we have to make, which is static camber.

On a MacPherson strut, we have to pick the right amount of negative camber, so when we corner the tire ends up flat. What that means is that on the straights while accelerating and braking, we're actually not able to use all the tire, since it's cambered. With the a-arm setup, you can have the tire flat on the ground in the straights, and flat on the ground in the corners. Not to mention how much easier it is to run wide rubber on that a-arm setup.

So yeah, the MacPherson setup is "fundamentally" flawed when it comes to road racing... but it allows for a pretty good setup within the packaging limitations of a sedan type vehicle. Plus it has some advantages in the context of rally and off-road uses. So it's not like it's total crap, it just means there's a bit more work to be done to find the optimal setup for conditions.

That makes way too much sense. :cool:

Overall good points, but the above doesn't necessarily ring true. A McStrut still has a lower control arm, so you have similar LCA length requirements if you're designing for a lot of travel.

True, but it can attach under the vehicle, whereas the UCA has to bolt higher up, usually right where the motor is on a front-engine car.

It all depends.. even LCA's have plenty of packaging constraints based on the engine & support frame, steering rack & mounts, etc. etc..

They make it work pretty well on trophy trucks. :)

Trophy trucks don't need a low CG like a Rally car does.

Rally suspension is designed to sit very low in the suspension's throw at static ride height. It's actually very much like a road suspension at static through max compression. However, if the car becomes airborn, the suspension has a *ton* of droop combined with a very low fast rebound damper, which allows it to drop quickly to be ready to provide a soft landing when jumping. See pic.

Of course there's also the chance that due to homoligation rules, rally teams are simply required by the factory to use the "inferior" MacPherson struts because the factory can't afford to put double a-arm suspension on the production car... But, considering how many things are build into the street Impreza just to make it a better WRC car, I would bet if there really was a big advantage to the double a-arm, that's what would be on the street car so they could use it in the WRC.

A low CG requirement only increases packaging conflicts with a very long LCA, due to the engine getting lower and lower.

My guess would be that is indeed a rules requirement, and Subaru just doesn't use it in the production cars because of the increase in cost/unit over a McStrut. Even a $1 or $2 increase per car gets real expensive when you're selling hundreds of thousands of them every year...