Semi-trailing arm rear suspension - roll center change

[sierra4000]

How can I change the roll center semi-trailing rear axle?
for stiffness rear end car (for understeer limitations )

Thanks for ideas!

 

[BrianPetersen]

Only possible in theory by changing the angle of the pivot axis. If you want a higher roll center, the inner pivot needs to come up or the outer one needs to move down. Most of them are not designed to allow this. Most of them have pivot points that are rather solidly designed into the chassis or subframe, and even if you do find a way to change the pivot axis, generally changing it will also screw up the camber at a minimum, because the trailing arm is one piece with the stub axle.

If you fabricate new subframe mounts with a redefined pivot axis and new trailing arms that correct the relationship of the pivot axis to the wheel axis then the geometry is completely up to you!

 

[cibachrome]

If it's a twist axle type, cut the cross beam off the trailing arms and weld it back on with the shear center of the crossbeam raised by elevating it or rolling it upward. Also, be sure to keep your welder in the trunk because you'll probably need it (often).

 

[sierra4000]

Thanks for interest!

I have inner pivot modified for this change
movement of the inner pivot up is correct for increased stiffness roll?

any other consequences of this change?

 

[BrianPetersen]

It will change your bump-steer / roll steer curve.

 

[dynatune]

Sierra4000,

There is another solution to change roll-center heights on a semi trailing arm suspension. The pivot points is the best known and obvious solution. However, in the early days in the 70's & 80's all BMW's series 5, 6 and 7 were equipped with a special "modification" to the standard trailing arm suspension. The suspension was called "Schraublenker Achse" which can be translated in "Semi Trailing Arm with Thread movement". Basically they have added a short link from the subframe to the trailing arm (more or less the same axis but parallel in top view to the trailing arm axis). Unfortunately a picture is difficult to find on the net, best to have a look directly under one of those oldies....
Being a such a short link any movement of this link out of it's static position causes an movement of the whole trailing arm along the axis of the trailing arm (like a bolt in thread) which was possible since the bushings were in axial direction very soft. This lateral movement is obviously also imposed on the contact patch and since the "normal vector" (=90° rotated line) to the lateral contact patch movement defines your rollcenter this feature was a very nice and clever way to make roll center height on a semi trailing arm tunable independently from camber gain and rollsteer (which the pivots only do not allow). Obviously this is not a simple matter and requires some pretty profound understanding of cinematics, but it could be worthwile to learn about it. If you feel up to it, you could think of doing the same thing and tune the beast to your likings. You would have to make sure that your trailing arm joints permit axial movements since all axial loads would go to the link.

Cheers,
dynatune,

http://www.dynatune-xl.com

 

[sierra4000]

Thanks dynatune!
I know you are probably thinking,
roll center position is determined by the angle of the link?

If I understand so right - is required axial move in their original bearings.
This movement, however, my current spherical bearings does not allow.
I'm sorry did not know this before, it is very elegant solution.

 

[dynatune]

Correct ! The roll center height will be determined by the angle of the link. It is indeed a nice trick from BMW and during my days in Ford Motor Company the Sierra people were quit envious about it, but they could not do much against it since it was a Patent from BMW which was very heavily protected.

You can keep this solution in the back of your head in case you ever want to test it, you would probably need "just" some nylon bearings that would allow axial movement and the "famous" small link.

Cheers
dynatune,

http://www.dynatune-xl.com

 

[sierra4000]

I tried to imitate the rear axle design for my Sierra (as They once had BTCC Sierras) Including the modification the swing angle for rollsteer reduce

I have a high spring rate rear and big rear swaybar used to reduce understeer
It causes much lighten inside wheel- following acceleration slip of slow corners
so I hope upper rollcenter help me to return press on the inside wheel for better acceleration

 

[dynatune]

Trust me the last thing you are going to gain with a higher roll center is traction or laptime. A higher rear roll center will increase "jacking" which will start lifting the inside wheel even quicker and further up. That will not help you. I have made that experience very recently unvoluntarily and as expected it did not work.

I am a bit puzzled why you must run such a big rollbar on the back, this would to me indicate that you create a lot of understeer at the front of the car. I assume that your car is rear wheel drive which typically would not need high rollrates on the rear. Are you sure that you are not trying to compensate on the rear for something that is very wrong on the front ? If I can make a well educated guess I would immediately lower the front roll center to 0 (or below) and position myself on the rear between 70 and 90mm and start working from there on with my front and rear rollbars (assuming that my vertical springs are OK)

Cheers
dynatune,

http://www.dynatune-xl.com

 

[sierra4000]

Car is 4X4 with very weight V6 engine - In the first phase was very lowered body
thus both front and rear get into the wrong position car is struggled with a very significant understeer
I tried to fix the position of the front arms into original position for better dynamic camber curve
but thus I also raised the front roll center back toward understeer

Although I copied the RS500 arms but I misunderstood the the principle of moving rear rollcenter and moved more down!!!

this I managed to partially compensate large swaybar but with all the other effects
Now I try to also the rear axle in the correct position

my thoughts are is correct?

 

[dynatune]

Okay, so you are running the old V6 "cologne engine" used also for the for explorer made from "solid steel" all the way. Very nice picture by the way your rear suspension, compliments!!. Yes the engine is indeed very heavy and given the fact that you cannot change that, I would assume that you should run rather stiff front springs to get a decent ride frequency out of the car. This should in theory permit you to run a softer rollbar favouring less understeer. I assume that you did put the roll center back-up for camber recovery but why not try to do this with more caster ? A McPherson strut, even with a rather high rollcenter has very little camber gain recovery so this could be "easily" achieved in a different way. You need camber in cornering and increased caster (have a look at the WRC Focus world rallye car with a very rearwards inclined strut) will give you that. That would make a raise of front roll-center not necessary and cause less trouble at the back end of the car. You could also modify the front suspension to have a more significant amount of anti-dive by lowering the front A-arm joint/ raising the rear A-arm Joint which also will increase caster on the outside wheel during cornering. The very nice thing about caster is that it will make the car roll "more" without activating the front rollbar (just think of a go-cart wheel jacking with steering) but it WILL load the rear rollbar up, so it is a double whammy for reducing understeer. Beyond that I also fear that your AWD setup might be biassed itself to much towards "understeer". In general every AWD tends to inherently understeer more, so if there is a way for you to split your torques front to rear diverse this might help you more than anything else (running any LSDs ?). OK lets recap .. you also said that you are suffering from traction at the rear. So ... you have understeer and not enough traction at the rear ... actions .....
1) better traction rear can be provided by less roll bar rear ... ok, going for this, lets assume this resolves the traction issue
2) less rear rollbar causes more understeer ... should i go up with rear rollcenter ? To some extend yes but carefull about jacking, raising rear rideheight in cornering will not make my car much faster .. probably.
3) ok ... I cannot go up too much on the rear rch then. Can i go down on the front ?
4) sure, yes I could but I probably loose a lot of camber gain on the front ...hang on this is a McPherson strut, they do not have a lot of camber gain anyway .... could this be an idea ?
5) sure, but how can I fix my camber gain ? I could also increase static negative camber ?
6) Static negative camber will help but I could also use increased caster to get more negative camber angle gain with steering
Food for thought ...

Cheers, I am going to bed
dynatune,

http://www.dynatune-xl.com

 

[sierra4000]

Yes,
Unfortunately, I also have to sleep sometime , although this discussion is very interesting
and instructive for me
thanks for the compliment
yes V6 cologne, But 4.0 V6 Mustang !! + diff, shaft etc. - really  much iron front end.



You're right them there stock viscous coupling even when the torque split 60/40 F / R


I have modified knuckle and upper tower MacPherson for continuous adjustment of the camber, caster and KPI
also I have a little antidive position of the lower arm.
and think that I finally reached a nice camber curve(I crossply tires
they do not want much)

https://fbcdn-sphotos-d-a.akamaihd.net/hphotos-ak-frc3/190902_195055187192477_2660907_o.jpg

it looks like,lifting the rear wheel is caused just due to caster vs rear swaybar
because it happens at a very slow corners (large steering angle)when the caster effect

So you say:
Stronger spring and a smaller front bar , help front end.
I can then reduce back a little rear bar?
Lower front rollcenter reach further lowering front ride height?
With a slight increase in rear rollcenter again I can reduce rear bar.

is that correct?
Thanks Radek

 

[dynatune]

Yep, more or less that would be a road I would look into, based on your comment if I understood them correct.

Let me know how it goes. Would be interested to hear if our thoughts made sense.

Cheers,

Paul

 

[sierra4000]

OK,
Thank you Paul!
this weekend I'll know more

One more question for the corner car balance:
If a neutral balance in the slow corners OK(for example skid pad )
will be OK balanced for high speed corners?

Once again many thanks!

Radek

 

[BrianPetersen]

Stable at low speed does not guarantee stable at high speed. You can get away with a lot of oversteer-ish tuning in autocross that you could never do in a higher-speed road-course car.

 

[sierra4000]

Is the rule-that low speed corner balanced car = high speed understeer?
What the relationship low speed vs high speed corner balance?

 

[BrianPetersen]

It will likely vary from one vehicle to the next, but my experience has been that if you set up the vehicle to have neutral-ish steering at low speed, it will oversteer at high speed under some conditions. A vehicle designed for everyman to drive needs to have a considerable understeer margin to avoid specific situations that could lead to a high-speed spin-out into the ditch - like lift-throttle oversteer on a rain-soaked motorway when a normal, everyday, non-race-skilled driver has to avoid an obstacle.

If it's a racing vehicle then ask the driver what he wants and what the car is doing that is different from what he wants. Even roadracing vehicles need a margin of stability (e.g. a smidge of understeer). Can't win a race if the car is so tricky to drive that the driver ends up backwards into the guard rail at the slightest provocation.

 

[dynatune]

I agree with Brian,

Typically at high speeds the mechanical balance from springs and rollbars are also affected by aerodynamics and another phenomenon which if you have a low speed neutral balanced car this can cause the car be too neutral for high speed. The high-speed stability is unfortunately massively influenced by this physical phenomenon called "de-minuishing yaw damping", there is a long formula available but the important thing to know is that the velocity is in the denominator of that formula making all things less stable with speed. I don't know whether you have ever been on an old 80s fast motorbike that above 200kph scared the hell out of you ... that is the thing I am talking about. The physical effect that occurs is that your natural yaw frequency from let's say 2 Hz (this means your bike is "autocorrecting" itself from steering inputs below that frequency and unstable for steering inputs above the frequency) can go down to 0 Hz which means are "unstable" and you get that horrible feeling that you are a passenger only ... This principle is also valid for cars and the ONLY way to counteract on that is increasing what experts call "linear understeer gradient" which will raise the frequency again since it is in the nominator of that very difficult formula. If you need an example of a car that got that formula all wrong look at the first Audi TT, that became infamous for high speed accidents on the german highways ..... they learned the hardway
Then finally do not forget aerodynamics since they can be difficult to understand if you do not have a windtunnel at your hands. Small changes can make big changes to the car balance. Try running some stringpots for logging your suspension travel in order to see what your car is doing in high speed. From the wheel travel you can relate back to the load and learn.

Cheers, I should go to bed now ;-)

 

[sierra4000]

Thank you all for your very interesting posts!!



Radek