Technical lowering Info

[bizzyboi]

( i did not write theses articles. they are te property of their respected owners. in my quest for knowledge in suspension systems i found theses, and though i would share this information with other cb7 owners)

During high laterally loading (going around corners) a car leans not over the
contact patches at ground level, but around something called the roll center, a
point in space which is determined by how all the various suspension parts move
relative to the chassis. A car with double wishbone suspension can be designed
with the roll center at the CG, below it, or even above it (the car would lean into the turns).

The tendancy of a car to lean is determined not by how far the
CG is above the ground, but how far the CG is from the roll center.
That is a simple statement of fact. Look it up, if you don't believe me.

Here's the problem. When you lower a car with MacPherson struts (our cars)
the roll center lowers farther/faster than the CG. So if at the stock height the
roll center is 4 inches below the CG, when you lower the car 2" the CG
drops 2" but the roll center drops say 6". The actual tendancy of the car
to lean during cornering then increases. To counter that tendency you need
MUCH larger springs. Not namby pamby 300-400lb springs but
500-600lb race springs. Are you willing to run 500lb springs on
your street car?

The reason MacPherson strut equipped cars have this problem is
because the angle of the lower A-arm relative to the chassis changes
as you lower the car. Lower too far and the arm tilts
the wrong way, lower on the chassis side and higher on the strut side
and the rolling of the car has a greater mechanical advantage on the
strut causing it to compress the spring easier as the weight
shifts to the outside. That change in geometry is what changes the
roll center.

It has other negative side effects too. At the stock height, the
lower A-arm is higher on the chassis side, and lower at the hub.
When the suspension compresses under cornering the arm moves
both up toward parallel with the ground, and because it pivots
on the car side, the outside end moves out a little as well. That changes
the angle of the strut itself pushing the bottom out, causing a bit of added
negative camber relative to the static position. That is a good thing,
especially for our front heavy cars.
You want the added negative camber to give you better laterally grip
as the car leans and as the tire sidewalls roll over a bit. (Race tires
can compensate for decreased negative camber by having stiffer/higher outter
sidewalls.. see R1s.. )

As you lower the car moderately the A-arm starts out parallel to the ground
and now with suspension compression during cornering, the arm swings
through the part of the arc where there is no change in camber. That'd
be fine if you could totally eliminate body lean while still having usable
suspension travel (you can do this with a double A-arm suspension but not
with a MacPherson strut). But because you're lowered, you're eating up
your suspension travel and are liable to run into the bump stops at which
point your suspension STOPS WORKING completely.. You're no longer
"suspended".

If you continue to lower, you experience the problems above, but
in addition now the A-arm is above parallel so you have to dial in
lots of static negative camber, and ANY suspension compression causes
the arm to swing up and back inward tilting the bottom of the strut inward
as well causing a positive camber change. A positive camber change
during hard cornering is *NEVER* desirable. The tire is leaning
the wrong way. So add it all up; the roll center lowers faster
than the CG, so the car actually wants to lean more, you have less
suspension travel so that lean causes you to come down on the bump
stops causing instantaneous weight transfer, and you push into positive
camber, all of which decreases your lateral grip, sometimes suddenly.

There is actually a way to fix a lot of these problems, and it's
called a drop spindle. Basically you drop the car with springs, decreasing
the distance from strut top to wheel, but then the hub has a lower A-arm
attachment point that sticks down below normal. That allows
the actual geometry of the strut and lower A-arm to remain the same
but the car to sit lower. Another way to picture it is if you leave
the stock looking strut in place, and simply moved the hub
that the wheel attaches to upwards. That lowers the car keeping
everything else the same.
The bad news.. They don't make drop spindles for our cars. Period.

Lowering an 25-35mm (1-1.5") won't cause great harm (1.5 is pushing it).
Going below that is for looks only unless you're running on a race track with
3 degrees of static negative camber, 500 lb springs and race rubber.

ian

[Modified by Daemon42, 1:08 AM 4-10-2002]
Daemon42
Moderator



Offline

Member Since
2-9-2001
17556 posts

Lakewood CO
2004 R32 Blue


Re: problems with lowering (Daemon42) « » 5:14 AM 4-10-2002

BTW, I've been wanting to do this for a while.
Here is a diagram I created showing the negative camber change that occurs
during suspension compression of a MacPherson strut. The red lines are in the car chassis
frame of reference and are fixed. The lower strut, wheel and tire all maintain a
fixed relationship to each other and the only parts that move are the spring, upper
perch, and lower A-arm, and the angle of the strut relative to the chassis changes.

It's not too hard to see that once the A-arm angle goes much above this point, the
camber will start to go back positive. I'll make a diagram showing that next,
and then perhaps I'll work on one that shows why the roll center drops
faster than the CG. If you want to think about it some, imagine if the A-arm angle
were at an extreme angle up or down, and a lateral load was applied from
the right to the left on the tire. If the arm is way down, it would cause
jacking, and if it's way up, it basically collapses the suspension.

ian

http://forums.vwvortex.com/zerothread?id=310880#2755391

[johnl]

There is some good information here (along with some misconceptions).

Note that in order to address the dynamic camber issues outlined above, with Mac Strut suspensions (when used on high performance or racing cars) it becomes necessary to use substantial static negative camber angles, and / or excessivly roll stiff suspension set ups. This is to address the problem of insufficient dynamic camber change in roll (with Mac Struts), and is in itself a bad thing, i.e. a compromise, a necessary evil...

However, CB7 owners are lucky, it's not a problem for our cars, so we don't need to use substantial static neg camber as Mac Strut cars are forced to do, in fact we should avoid it...

Cars such as CB7s with well designed SLA suspensions (i.e. 'short / long arm' aka wishbone) have superior 'camber curves', and don't suffer from the same geometric deficiencies that Mac Strut suspended cars do, so substantial static neg camber angles actually create more problems than they solve...

[johnl]

( i did not write theses articles. they are te property of their respected owners. in my quest for knowledge in suspension systems i found theses, and though i would share this information with other cb7 owners)

During high laterally loading (going around corners) a car leans not over the
contact patches at ground level, but around something called the roll center, a
point in space which is determined by how all the various suspension parts move
relative to the chassis. A car with double wishbone suspension can be designed
with the roll center at the CG, below it, or even above it (the car would lean into the turns).

Roll theory isn't as simple as this suggests. For starters the geometric roll centre (GRC) is only the point around which the sprung mass rolls (i.e. the centre of roll) in very limited circumstances:

1) At the infinitely minute moment of time when roll motion has begun but no weight transfer has yet occurred (don't think too hard about this, it's a chicken and egg type of paradox...).

2) When the GRC is at the exact same height as the CG of sprung mass.

Any time the GRC is below the CG (nearly always in nearly all instances) the GRC is not the point around which the sprung mass rolls, but it is a point that helps define the location of the point around which the sprung mass rolls.

GRC location is also a parameter that helps to define (along with effective wheel rate, i.e. spring and ARB stiffness as seen by the CG of sprung mass) just how much weight is transferring geometrically (via the vector of the suspension linkages) and how much is transferring 'elastically' (i.e. through the vector of the springs and anti toll bar).

The other point that helps to define the location of the point around which body roll actually occurs is located at the outside contact patch. My own understanding of the behaviour of the actual point around which roll occurs (my term: the 'actual roll centre' or ARC) is that it is a mobile point located somewhere along the line defined by the GRC and the outside CP.

Just where on this line at any moment in time the ARC is located being dependant upon the strength of lateral acceleration (force), the stiffness of the springs (and anti-roll bar), and the manner in which the GRC migrates with body roll motion (there will be transitional influences from the dampers).

Just how much weight might be transfering geometrically vs how much is transferring elastically at any moment in time is quite important and has substantial affects on the car's handling characteristics. This is because weight that transfers geometrically does so through an effectively 'rigid' vector, and thus occurs instantaneously. WT that occurs elastically occurs through a flexible vector, and as result there is a time lag involved with elastic weight transfer. The more you think about that the more important you should realise it is.

Roll theory is very hard to get one's head around, and it's just as well for us that very bright suspensiojn engineers have very carefully designed our suspension geometry for us. When we make substantial changes to this without understanding the complexity of the geometries we can cause problems that are difficult to correct.

Be aware that 99% of what you're likely to read concerning 'roll theory' will be so simplistic that it may as well be wrong (that is if it isn't actually wrong...). It is also the subject of much misunderstanding and disagreement, even amongst proffessional suspension engineers. My own understanding of it should to be taken with a grain of salt, I can't guarantee that it's entierly correct, but I have given this topic a lot of thought over the years, and I've only scatched the surface above.

The tendancy of a car to lean is determined not by how far the
CG is above the ground, but how far the CG is from the roll center.
That is a simple statement of fact. Look it up, if you don't believe me.

Partly. CG height above ground is important because it directly affects the degree of elastic weight transfer. CG height above the GRC does behave more or less as described, but only as far as the geometric component of roll motion and weight transfer is concerned.

The degree of roll is only partly dictated by the GRC location. It is also influenced by the elastic roll stiffness. If it weren't then changing ARB and / or spring stiffness would make no difference to roll motion (or weight transfer).

Having said that, if the GRC and the CG are at the same height, then elastic roll stiffness will have zero influence on roll motion or weight transfer, i.e. the greater the geometric roll stiffness the less affect the elastic roll stiffness will have on roll motion and weight transfer. This is also the case if geometric roll stiffness is low or zero, ie. the elastic roll srtiffness will be much more dominant.

Note that so far I've only been considering roll theory in 2 dimensions at a single axle line. In reality all this occurs in 3 dimensions at both the front and rear axle lines, with what is occuring at the front axle line affecting what is occuring at the rear axle line and vice versa.

I'll stop now, or I might never stop...

[C91BLX7]

Too bad we don't have McPherson setup...

[bizzyboi]

Be aware that 99% of what you're likely to read concerning 'roll theory' will be so simplistic that it may as well be wrong (that is if it isn't actually wrong...). It is also the subject of much misunderstanding and disagreement, even amongst proffessional suspension engineers. My own understanding of it should to be taken with a grain of salt, I can't guarantee that it's entierly correct, but I have given this topic a lot of thought over the years, and I've only scatched the surface above.

You have given this concept YEARS of thought. I have given it roughly a few hours of thought. My purpose of posting this, and the other post i have put up is to share some of the information that i have come across. What i am trying to achieve is an understanding of how and why suspension components work. How they react to outside forces and so on. This is my approach to building the setup on my vehicle. I research, learn, learn, research some more, and ask opinions. 'Note to anyone reading' John L has given this topic YEARS of consideration, and admits to not having the greatest understanding of it. Therefor in this post and others i have or may post in the future are for informational purposes only. Do your own research on these if you like, learn and ask opinions. I appreciate any and all comments and corrections.

[johnl]

Too bad we don't have McPherson setup...

What CB7s (et al) have is much better, especially if you don't try and set it up as if it were a Mac Strut (i.e. with heaps of neg camber).

With SLA suspensions (i.e. double wishbones) such as Honda uses, the forces acting in the upper suspension are still reacted into the tower (and into any tower brace) in a very similar manner as they are with Mac Struts, except that the lateral forces are reacted through the upper wishbone rather than through the Strut damper's body and shaft (which places substantial lateral loadings into the Mac Strut thus placing a substantial bending moment into the damper body and shaft).

When the suspension is laterally loaded with hard cornering (and when the Mac Strut damper is also being asked to move vertically with roll and bump motion), these lateral loadings acting within the Mac Strut damper cause some degree of accelerated wear and 'sticktion' in the damper.

If nothing else, reacting these forces through an upper wishbone eliminates the damper binding forces that act within a Mac Strut damper.

[9threeEX]

.......guess everything works in theory..........

[johnl]

.......guess everything works in theory..........

That's the wierd thing, in theory Mac Struts should work much worse than they actually do. Still rubbish though...

[9threeEX]

That's the wierd thing, in theory Mac Struts should work much worse than they actually do. Still rubbish though...

There's too many variables to apply an emcompassing explanation. I've seen strut type suspensions work very well, and some that work as expected. I think a lot of it has to do with material quality and design of the strut. These factors actually seem more important in struts than in coil-over shocks.

[johnl]

There's too many variables to apply an emcompassing explanation. I've seen strut type suspensions work very well, and some that work as expected. I think a lot of it has to do with material quality and design of the strut. These factors actually seem more important in struts than in coil-over shocks.

A well designed Mac Strut will never be as good as a well designed SLA suspension, the design restrictions are just too great.

The only advantages of a Mac Strut are;
Light weight.
Easy to package (doesn't take up much room in the chassis).
Fewer parts to maintain.
Cheap.

Disadvantages:
Poor camber curve (camber doesn't change enough with suspension travel).
Unstable caster angle (caster unavoidably changes with suspension travel unless caster angle is zero).
Lateral loads reacted within the damper, creating internal friction = 'sticky' damper action and wear.
Braking forces reacted within the damper = as above but more so since forces are higher.

Setting up Mac Struts for 'high performance' applications involves far more compromise than is the case with a good SLA, but they can be made to work quite well.

[gloryaccordy]

OP has nothing to do w/our car's design... SLA operates through totally different principles

I've found that the most important component in how our car handles is the quality of the shock. I'm no expert on suspensions but I've had a few modified cars and the CB's suspension seems to take care of itself geometry wise at reasonable static heights.

[9threeEX]

OP has nothing to do w/our car's design... SLA operates through totally different principles

I've found that the most important component in how our car handles is the quality of the shock. I'm no expert on suspensions but I've had a few modified cars and the CB's suspension seems to take care of itself geometry wise at reasonable static heights.

Quality of the shock is absolutely crucial and IMO the best spent money on improving the overall feel of how your car drives....unless you adjust a true coilover out of useable range for looks then you have wasted your time and money.

[gloryaccordy]

Quality of the shock is absolutely crucial and IMO the best spent money on improving the overall feel of how your car drives....unless you adjust a true coilover out of useable range for looks then you have wasted your time and money.

Yea Hondas are prob among the few cars on the streets that don't have geometry issues when lowered (outside of the UBJ banging, which should be an indication that your shocks are too weak + your car is too low). Outside of maybe the discussion for a race car geometry is a non-issue.