41 replies to this topic

[Greg Locock]

How do you decide how much to use? Do you have prefered values for each, and then the horizontal offset is a result, or do you think horizontal offset is important as well?

Reason I ask is that production cars with decentish handling have castors varying from 3 to 10 degrees.

[Ben]

In the three times I've done it from scratch I've put the lower ball joint on the axle line (side view) and specified a caster value.

Ben

[phantom II]

I am shocked and dismayed., Greg. You don't know something. Race car mag has done some stories on that subject and various authors have covered it in their books, but I don't think that there is any real consensus. You of all people should know how to find this information and we should be asking you. Ask RDV, I don't think Peter Wright has even suggested a starting point as to how and why he arrives at his values. My new upright didn't have enough inclination and the car didn't feel stable at speed and had little spin back so I just added more castor 'till it felt right. That's how I got 7'. The line drawn thru the center of the ball joints and then to the ground (A)arrives in front of the vertical line drawn thru the wheel bearing to the ground (B)in a side view. The trail is ofcourse the distance between these two points. If you draw a line straight back from this point(A) on the ground in plan view, it must pass thru the inside edge of the contact patch which is on the center line of the wheel bearing. This is the offset. I have heard other definitions as to what it is, but that is how I understand it. I just have my own rule of thumb methods and I can't remember where they came from but they seem to work.
IRL cars have about 11' castor and the wheel bearing is on a line horizontally ahead of center of the axis between the ball joints by about 2". Never seen that before. Also, anti dive geometry and the position of the A arms can alter caster in motion. I keep the mount points horizontal. I draw it out on a black board before I put it into the computer.
Don't ask me.

Originally posted by Greg Locock
How do you decide how much to use? Do you have prefered values for each, and then the horizontal offset is a result, or do you think horizontal offset is important as well?

Reason I ask is that production cars with decentish handling have castors varying from 3 to 10 degrees.

[phantom II]

Another determining factor is the more caster, the more negative camber and jacking on turn in. Can your contact patch stand it? You may have to alter spring rate to stop a CG rise and God forbid, add an ARB. The lower the CG and the lower the roll center, the less need for ARB. In slow speed events you can let the air out the tire but generally, get most of the spin back with inclination and trim with castor. You can see my upper ball joint bolt on mount. Well, I'm going to make it shorter and take a degree out of the caster

[bobdar]

On formula cars, we adjust castor to minimize bumpsteer--in that context, there is really no desire to have equal castor left to right, just minimum bumpsteer. This also is how we minimize bump steer in the rear. One complication is that my arms get really tired driving a car with a bunch of front castor, so I try to have 3-4 degrees max.

[bobqzzi]

As little as possible on a race car. Edit: which for me means a FWD sedan

[Fat Boy]

Caster gives weight jacking to the suspension and camber changes due to steering angle. Neither one of these things are 'good' or 'bad'. I end up using less caster on tracks with more steering input. If you have a lot of caster on a tight track, then it's easy to get inside rear wheelspin and wear out your driver. On a more open track, the added caster can give better steering stability and more steering feedback for the driver.

I make no effort to run a minimum or maximum of caster. I don't find it to be a big player on the stopwatch, so I spend my time on things. I find it a very 'salt to taste' type of adjustment.

I do not adjust caster to address bump steer. I change the outboard toe-link pickup height to address bump steer. I do keep keep cross caster between +/- 0.1 degree.

[Greg Locock]

I /start/ with the castor trail. That is so fundamental to the limit steering feel that I can't go past it. Then I pick a castor number out of the air, and see where that takes me.

Weight jacking is helpful for returnability from high SWA at low speeds, but is bad for efforts and feel - I don't want the weight transfer to feed back directly into the steering wheel, I want the EFFECT of the weight transfer (ie the change in SAT) to be fed back.

[McGuire]

Final caster specs depend on a lot stuff...starting with wheelbase, track and geometry, thrust line, thrust angle, SAI and a couple dozen other factors. Essentially, anything that speaks to the vehicle's straightline stability vs. desired steering response, effort and feel. If there is a useful process for calculating a basic spec with no huge empirical component I would like to see it.

More on caster... on road cars, regardless of the factory specs most any good alignment mechanic will stagger the caster L/R...more on the right so the car will track straight on a crowned road (in LHD countries; in RHD countries the other way around, I imagine). Typically he will set one side toward the min spec and the other toward the max spec within the factory tolerances. But he may use more depending on local road conditions. To the consumer, a good wheel alignment is when the car tracks straight hands off on the roads he drives. He doesn't care about much else. Also, with IRS the rear has caster too eh.

In oval racing, running considerable caster split L/R is very common. On beam-axle cars this is kinda special as in order to make the caster adjustable in a practical manner, the axle must be cut in two and flanged, and the adjustment taken up there and at the four-links.

[imaginesix]

Originally posted by McGuire
Also, with IRS the rear has caster too eh.

Only if you have RWS. Caster is the angle of the kingpin axis relative to vertical when viewed from the side. A non-steering rear wheel has no kingpin axis.

[McGuire]

Originally posted by imaginesix

Only if you have RWS. Caster is the angle of the kingpin axis relative to vertical when viewed from the side. A non-steering rear wheel has no kingpin axis.

Of course it does.

[Greg Locock]

I was going to carefully ignore the rear suspension. But, seeing as we have opened that can of worms... the castor angle effectively controls the inter relationship of toe and camber in response to a lateral force, or self aligning torque.

[dc21]

What effect does castor at the rear have on handling (in laymans terms if possible)? I understand what it does at the front but I can't picture what it's doing at the rear.

Dave

[Greg Locock]

For high speed stability the rear geometry (etc) is considered to be more important than the front. It roughly follows the same rules, except, obviously , that the signs are reversed for steer effects.

So we spend a lot of time making the rear end as neutral as possible, tending towards axle oversteer =vehicle understeer.

[RDV]

Greg Locock-I was going to carefully ignore the rear suspension. But, seeing as we have opened that can of worms...

...wise man....

the castor angle effectively controls the inter relationship of toe and camber in response to a lateral force, or self aligning torque.

Unless I mistook the reference by assuming it meant rear caster will say only if mounted on silentblocks or other non rigid pivots (thus deflecting with loads), if otherwise will be a geometrical effect decided by the toe link.... in fact you can chose which is the caster... by defining which is the toe link...in the case of an A-arm on the bottom pivot and a drive shaft as a top link, where is the pivot point? .. worse even are things like twin link where pivot point is a point in space... the locus can have a strange trajectory as suspension moves up and down and we have any bump steer.

Several of these multilink systems in fact coulsd not work if they didnt have flexible mounts or elements, not geometricaly compatible.. a rigid , ball jointed system would lock-up.. as not true 5 or 6bar geometries.

Perhaps a better solution would be to see what is rotating about what and thus define castor.... no such problems with kingpin angle.. that one fairly evident...

On front independent suspensions a good example would be the Opel twin link front suspension, or even the Nissan350Z , where the bottom arm consists of 2 links... the bottom pivot is the locus defined by the geometrical movement of the two links... very complex geometry.

For more thoughts on this there was an earlyer thread=
http://forums.autosp...ighlight=castor

Idealy there will be a interaction between caster and kingpin angle, this is what will control weight-jacking, and relative camber on inner wheel and outer wheel...

Let us consider case where caster angle is zero and likewise kingpin (and assuming there is no trail or leading axle , which would give another type of caster...)= as wheel turns camber angle remains same , and there would be no steering feel related to cornering, as no self aligning torque would be produced (apart from pneumatic trail...but lets ignore that)

Folowing the thought experiment , we will now increment caster angle. On extreme case (I.E. 90 deg caster) rotating across pivots would only change camber, and give no steering at all..
so lets look at 45deg positive, there half the steering rods movement gets translated into lock, the other half into camber change.

Turn wheel towards center of car and you will get more camber, turn to ouside and you will get less.. secondary effect is that outer wheel jacks up, and inner wheel jacks down..

Doing same for kingpin angle you will get the same, but with one important difference... both inner and outer wheel lose camber... and both wheels jack down...

Corolary of all this is that caster and kpa are additive on inner wheel, deductive on outer wheel as regards to camber angle and jack down..

Judicious use of caster angle combined with kpa will give you many geometrical combinations, which contribute to turn in response, handling, driver feed back, tire life, brake feel ... and we havent even started talking about offsets, trail SLA or the rest....

And no, despite its usefulness.. it seems to still be an empirical choice, the main criteria on design side , is as Greg has mentioned, start out with caster trail....

[McGuire]

Originally posted by imaginesix

Only if you have RWS. Caster is the angle of the kingpin axis relative to vertical when viewed from the side. A non-steering rear wheel has no kingpin axis.

First let's refine the definition. "Kingpin axis" provides a handy illustration of caster, but not a true definition. The specific hardware and means of attachment are incidental, and it has been many decades since cars were actually built with kingpins. Modern cars use ball or spherical joints to allow their front suspensions to articulate, but as we know they still employ caster just the same.

So to be more precise about the property we are actually describing: Caster is the angular offset in degrees of the horizontal or rotating axis of the wheel from its vertical or steering axis.

So of course we can have caster in the rear. "Only if you have RWS." Well, that's the rub, isn't it?

To whatever extent a given suspension geometry has bump steer, thrust steer, deflection steer or any flavor of steering, whether indavertent or intentional, caster will be a factor. Front or rear.

Originally posted by dc21
What effect does castor at the rear have on handling (in laymans terms if possible)? I understand what it does at the front but I can't picture what it's doing at the rear.
Dave

It can sound complicated but it's really not so much, in principle. Imagine a shopping cart with four casters instead of just the two on the front...but the rear pair are very limited in their range. In fact we could say that rear caster is a simpler issue in one regard: we are dealing with only an incremental degree of steering (we hope) rather than an entire steering arc over a lock-to-lock range.

[imaginesix]

Originally posted by McGuire
To whatever extent a given suspension geometry has bump steer, thrust steer, deflection steer or any flavor of steering, whether indavertent or intentional, caster will be a factor. Front or rear.

I've been reflecting on this issue all day and I can see how you may choose to define a caster angle in a non-steered wheel based on it's compliance or bump steer behaviour, but I don't see the point.

On a steered wheel, the caster angle gives us the direct relationship between the steered angle and the camber change (though compliance and bump steer must also be taken into consideration).

On a non-steered wheel, compliance steer and bump steer have an effect on that relationship but I don't see how those considerations involve in any way the camber angle value. Compliance, for example, is either measured or simulated, and choosing to define any value for caster angle will have no effect on the results. Caster angle in that scenario is an end product, and I can't see the utility of ascribing it a value.

The same is true for bump steer; you can move the suspension through it's range of motion and thus determine it's camber and toe change. Then if you go and establish the relationship between those changes you can define a caster angle. But it seems to me to be a data 'dead end' since no further information can be derived from it.

I am only an amateur here (which is a nice way of saying that I'm wrong a lot), but this seems to be a rather nit-picky argument to make considering that it has no apparent practical value.

[Greg Locock]

I think you are being tripped up by terminology. When I talk about castor I mean the angle between the kingpin and the vertical in side view, like any normal person. But when my program measures it it does so by looking at the relationship between forces and deflections. And if I sound hazy about the details, well, there ya go.

Perhaps easier to visualise is the mechanical trail, which is zero when a lateral force applied at that point causes no change in toe for that wheel. If you did the same in some other horizontal plane you've just drawn the castor line.

These force based definitions are more robust that their geometrical counterparts, but also less easy to think about. I look at both.

A good example where they tend to agree is castor, one where they don't agree at all is scrub radius.

[Ben]

I'm learning more and more recently that calculating specific metrics from various bits of data is quite important when trying to understand a situation where a lot of parameters interact.

Given RDV's comments it seems that caster isn't, on its own, a reliable metric because KPI alters the situation a lot. Greg talked about trail, but the effect there depends on the tyre characteristics.

Given all that I thought the following metrics might be useful. (I'm designing a sports prototype suspension at the moment)

- Camber angle as a function of steer angle (captures the effect of KPI and caster)
- Steering torque as a function of lateral force (needs tyre data)
- Weight jacking as a function of steer angle (i.e. lift due to non-vertical steer axis multiplied by suspension stiffness)

On the second one I was thinking alongs the lines of selecting a load and slip angle, calculating the lateral force then taking the self aligning torque at this condition, calculating pnuematic trail, then combining with the mechanical trail to calculate the torque through the steering.

Greg - given that you said the force and kinematics approaches to caster are fairly close do you think this would be worthwhile? I do have access to ADAMS at work, but can't do big jobs like parameter studies for private work, so need to do the bulk of the work using a kinematic approach.

Ben

[imaginesix]

Originally posted by Greg Locock
Perhaps easier to visualise is the mechanical trail, which is zero when a lateral force applied at that point causes no change in toe for that wheel. If you did the same in some other horizontal plane you've just drawn the castor line.

That's a neat way to explain it, but it doesn't reslove my issue and it raises another question.

The issue I have is that there doesn't seem to be anything to be gained from ascribing a caster value to a wheel that only deflects when lateral force is applied. Unlike caster on a steered wheel, where the caster angle is used to calculate camber change from steer angle, there is no utility in knowing the caster angle on a non-steered wheel.

The question you've raised in my mind is that using the lateral force deflection method at different heights along the horizontal plane to determine the caster line, will more than likely result in a caster CURVE. Which just reinforces for me how useless the concept is when applied to a non-steered wheel.

[Greg Locock]

Ben, at home I use Wishbone.bas aka cosine.bas, it may be crude to look at but it is as good as anything for kinematics. If your targets are kinematic then there's no real danger in using a kinematic program. Certainly when I design a suspension from scratch it starts life as a mechanism, compliance effects are for fine tuning only. Then I chuck it into CarSim and have a play with it there. So far as tire data goes I look at the results with 0 30 and 60 mm of pneumatic trail, that progression tells me about feedback at the wheel.


imaginesix, bear in mind that these programs (or real life tests) are just reporting curves of toe vs lateral force, etc, it is human beings who need to reduce this mass of curves to concepts like castor angles and kingpin angles, they are simplifications, or abstractions or first order approximations. With production car suspensions the level of interaction is so great that we often end up looking at vehicle or system repsonses to manouevres rather than trying to use rules of thumb for the geometrical concepts.

[imaginesix]

I still don't get the utility of describing a caster angle to a non-steered wheel, but I think everyone's said their piece by now so I'll just hold on to my views with a healthy beetroot-sized dose of skepticism.

Moving along then, I am curious about the description you (Greg) gave of finding the castor line by determining the point at which no toe change occurs when a lateral force is applied. Could this method also be used to determine what I would best describe as the 'Center of Deflection' by applying force vertically or longitudinally and identifying the point of zero deflection for those forces too. Much like the CofG or CofP in concept, the CofD would be the virtual point about which the wheel moves when a force is applied (assuming it is fixed in steering and suspension travel). If the force is pointing directly towards the CofD then no deflection occurs.

It might be interesting to see the relative locations of such a CofD on different suspensions, like the Porsche Weissach(sp?) rear axle, and other passive rear-steer systems as well as a few known bad suspension setups. What would be revealed?

[Greg Locock]

Actually I agree, nobody worries about the rear castor very much, it is very low on the list of priorities. I don't remember ever discussing it, or reading about it, before now.

I'd call what you said the elastomeric centre of the suspension. It's not quite as helpful as you might think since in practice all forces are applied at the conatct patch.

However, it is the main reason why trailing arms tend to be frowned on, since you will get compliance oversteer, and the main struggle with a RWD production car is to get a nice neutral rear end. To some extent you can compensate for compliant effects with kinemtic ones, but it is better to have small numbers for everything rather than big numbers with opposite signs.

If you do some research on Bundorf cornering compliance models you'll find toe per unit lateral force, and toe per unit SAT are well recognised properties of suspensions. Those two numbers are directly related to the elastomeric centre and the trail, if you think about it.

here's the input deck for the kinematic and compliant part of a Bundorf model for a car I'm working on at the moment, using a custom front end with Crown Vic uprights, and a Mustang Cobra IRS, slightly modified. So as you can see, we do think about the toe compliance. (sorry about the formatting)


Kinematic Ride Motion F R
Toe Curve Gradient -4 U 2 U (deg toe)/(meters jounce)
Camber Curve Gradient -15.91 -29.33 (deg camber)/(meters jounce)
Actual Roll Steer 0.050 U -0.025 U (deg toe) / (deg roll)
Actual Roll Camber -0.198 U -0.364 O (deg camber) / (deg roll)
Wheel Inclination Angle Gain 0.802 U 0.636 O (deg inclination / deg roll)


Suspension Compliances
Lateral Force Toe Compliance -0.200 U 0.120 U deg/kN
Lateral Force Camber Compliance 0.200 0.200 deg/kN
Brake Steer 0.139 0.123 deg/kN
Acceleration Steer -0.060 deg/kN
Aligning Torque Compliance Steer 7.720 U 1.590 O deg/kN-m


Tire Compliances
Cornering Stiffness 1100 1100 N/deg
Aligning Torque Stiffness 35.9 35.9 N-m/deg
Camber Stiffness 39.1 39.1 N/deg
Tire Vertical Stiffness 218 218 N/mm


Spring and Bar Stiffnesses
Suspension Wheel Rate 24.2 39.4 N/mm
Roll Stiffness due to Stabar as Installed 600.0 0.0 Nm/deg

[phantom II]

Quote: How do you decide how much to use? Do you have prefered values for each, and then the horizontal offset is a result, or do you think horizontal offset is important as well?


So Greg,

What have you learned?

In summation......

Originally posted by Greg Locock
I /start/ with the castor trail. That is so fundamental to the limit steering feel that I can't go past it. Then I pick a castor number out of the air, and see where that takes me.

Weight jacking is helpful for returnability from high SWA at low speeds, but is bad for efforts and feel - I don't want the weight transfer to feed back directly into the steering wheel, I want the EFFECT of the weight transfer (ie the change in SAT) to be fed back.

[Greg Locock]

Haven't learnt much, directly, except that different people use different approaches?

Oh, my research assistant did find a reference to rear suspension castor. Here goes:

"On the IRSs the steering knuckle can be given negative caster to achieve lateral force understeer"

That's it.

Reimpell & Stoll Fig 3.90

[imaginesix]

Originally posted by Greg Locock
here's the input deck for the kinematic and compliant part of a Bundorf model for a car I'm working on at the moment, using a custom front end with Crown Vic uprights, and a Mustang Cobra IRS, slightly modified.
...

Too many 'U's

But seriously, that's a real nugget and it's motivated me to open up my Milliken again. Thanks. Made me wonder though, is this a personal project or do you work for FoMoCo?

[Greg Locock]

Check out the last chapter of Dixon, roughly speaking there's no such thing as too many Us so long as they are small!

Yes, I do work for Ford, but that project is nothing to do with them, it is a rolling chassis for a 55 Tbird.

[imaginesix]

Ah, well the reason I asked is because you made me curious about how manufacturers stipulate handling and ride performance criteria for their engineers to follow. I get the sense that the Bundorf model might go some ways to help define a car's handling characteristic, but ultimately isn't it all a very subjective target to be aiming for? And business HATES subjectivity.

Also, which Dixon publication are you referring to? I know of The Shock Absorber Handbook as well as Tires, Suspension and Handling from the other current thread on recommended reading.

[Greg Locock]

We have a short list of 12 overall objective measures of overall vehicle performance, eg u/s at 0.3 g at 100 kph, selected from a somewhat longer list, and a list of say a hundred subsystem level targets such as castor at design height, and castor gain in roll. A full deck of results for a car is about 300 graphs, but some of those are of no interest to anybody.

We also have a list of about 30 subjective criterion that are sort of used to generate the objective lists.

Let's just say the 'linkage' between those three lists is flaky in the extreme!

Here's a list of objective subsystem targets


1 WHOLE VEHICLE:

Sprung mass
Sprung CG X and Z
Moments of inertia
Maximum steady state lateral acceleration
Ground clearance
roll stiffness (degrees per g of lat acc)
ride frequency, including front to rear percentages
lateral weight transfer percentage distribution front to rear
Torsional stiffness of the body structure
Roll axis gradient and any gradient change relative to roll
roll couple distribution

2 AXLE and WHEEL:

2.1 Packaging and general
spindle length
position
track
tyre size, hence rolling radius
Jounce travel
rebound travel
wheel offset
unsprung mass

2.2 Toe
Static toe setting
Ride Steer
Roll Steer
Tractive Force Steer
Brake Force Steer

2.3 Camber
static camber
Ride Camber
Roll Camber

2.4 Castor
static castor setting
castor gain
castor trail

2.5 Roll centres
roll center height at design load
roll centre height gain
TLLTD (tire lateral load tranfer distribution)

2.6 Track
track gain in roll
semitrack gain, ie lateral displacement of contact patch per unit of jounce travel

2.7 King Pin Geometry and Steering
KPI
scrub radius
steering axis angle
hub trail
pin lead/trail
Ackerman percentage/change
steering ratios
Lateral force steer (deg/g)
Aligning torque steer (Nm/deg)

2.8 Compliances
Stiffness of the major suspension mounting points, ie on a McPherson Strut, do the strut towers move around
Aligning torque Camber compliance (small effect)
Lateral Stiffness (at ground w/o tire)
Fore-Aft Stiffness (at wheel center w/o tire)
Lateral Force Camber Compliance

2.9 Other
Anti-x
Side View Swing Arm Angle (SVSAA)

[phantom II]

If you look at the Evans LMP (converted for street) front suspension you can see the neat adjustmnt of the caster which doesn't effect camber or tow. Fine adjustment is done on the leading part of the upper and lower A arms mounted with left and right Heims. John Evans says trail must be kept to a minimum and that is why the lower ball joint is mounted just above the rim as low as possible. This way it keeps caster jacking in turn in to a minimum. The offset must be in the inner edge of the contact patch so that in camber gain, the scrub stays in the patch.
The wheel bearing must be ahead of the line between the ball joints as shown.
My car has a diffrent application so the design differs as it does on any car depending what it's design objectives are..



[QUOTE]Originally posted by Greg Locock
[B]We have a short list of 12 overall objective measures of overall vehicle performance,

[Supercar]

Originally posted by phantom II
[B]If you look at the Evans LMP (converted for street) front suspension you can see the neat adjustmnt of the caster which doesn't effect camber or tow. Fine adjustment is done on the leading part of the upper and lower A arms mounted with left and right Heims. ...

Actually, those holes in the knuckle adjust the caster *trail*, not the caster itself. Caster can only be changed if the steering axle inclination changes, which is done with the A-arm Heim joints in this case.

You have to move the upper ball joint from one hole to another and do the Heim joint thing at the same time to keep your toe and bump steer unchanged. Alternatively, you can raise or lower the steering rack to keep the toe and the bump steer unchanged.

Philip

[phantom II]

That is if you define caster as the line thru the ball joints but geometrically speaking, you can reduce jacking with more caster. This car has 7' included caster. Also changing the diameter of the wheels or moving the spindle back or forth will change trail..IRL has some strange looking set ups.

QUOTE]Originally posted by Supercar

Actually, those holes in the knuckle adjust the caster *trail*, not the caster itself. Caster can only be changed if the steering axle inclination changes, which is done with the A-arm Heim joints in this case.

You have to move the upper ball joint from one hole to another and do the Heim joint thing at the same time to keep your toe and bump steer unchanged. Alternatively, you can raise or lower the steering rack to keep the toe and the bump steer unchanged.

Philip [/QUOTE]

[Supercar]

Originally posted by phantom II
[B]That is if you define caster as the line thru the ball joints but geometrically speaking, you can reduce jacking with more caster. This car has 7' included caster. Also changing the diameter of the wheels or moving the spindle back or forth will change trail..IRL has some strange looking set ups.

Ah, I see. You will keep running into misunderstanding problems with this definition of caster. All automotive literature defines caster and camber as angles. The distances measured on the ground between the kingpin and the projected wheel center are usually called caster trail and camber trail (a.k.a. KPI trail, or scrub radius). I am not sure what "included caster" is. I have never heard of it and even Google does not have it. Please 'splain.

Philip

[Fat Boy]

Originally posted by Supercar
Actually, those holes in the knuckle adjust the caster *trail*, not the caster itself.

If you're not changing the positions of the upper and lower ball joints with respect to each other, you aren't changing the caster angle. You are changing the hub centerline w.r.t. the line that intersects the upper and lower ball joint. That's trail.

What those types of steering arms are good for is a couple things.

1. You can have various wishbones jigged up and ready to go. Unbolt one and bolt in another. You've change caster without changing anything else.

2. Adjusting steering ratio quickly.

Many higher end racecars use that kind of arrangement, and it's pretty handy to have at your disposal.

[phantom II]

If I have a 'misunderstanding' of problems, why would you want me to splain stuff. Modern suspensions have additional links and the loci they describe does not fit conventional suspension geometory in motion of conventional suspensions. Stick this Renault suspension into your Mechanical Desctop and normal definitions won't explain what is happening. Check out the rear suspensions of P 928s, BMWs Porsche, etc and front suspensions of Honda NSX and this Evans LMP. Normally, the caster is the inclination of the line drawn thru the center of the upper and lower ball joint or the kingpin. The combination of ball joint inclinaton and caster produce some very interesting results and engineers have been dabling with this stuff since chariots and chuck wagons. Move the spindle and or the upright and add a few links and with the aid of a computer, you can perfect the contact patch. Move this combination around on your computer and you have what is called included angles. Each new bit adds new terminology and it varies from country to country even England and America. The inovators create their own language such as this OPT configuration.
The caster trail is taken from the point on the ground that the projection of the line drawn thru the ball joints describes. Draw another line straight back on the ground in the cse of positive caster angle on the side view. Draw a line perpendicular to the ground from the wheel bearing. That distance is the trail. The scrub radius is not the same as the caster offset. In a plan view, the line drawn back from the point on the ground that you first described will intersect the wheel bearing center line on the ground. Ideal intersection should be at the inside edge of the contact patch and in roll it must not migrate outside of the contact patch. Camber trail is the extention of the caster trail line to the outside depending on camber. Scrub radius is the arc the contact patch describes from lock to lock. There are other definitions but it doesnt matter, because you have something to work with. All these angles work together and there is a black art to make it all work together. A Baja car will be different to a rally car and so on. RDV will get into this in depth if you ask him nicely.

Originally posted by Supercar
Ah, I see. You will keep running into misunderstanding problems with this definition of caster. All automotive literature defines caster and camber as angles. The distances measured on the ground between the kingpin and the projected wheel center are usually called caster trail and camber trail (a.k.a. KPI trail, or scrub radius). I am not sure what "included caster" is. I have never heard of it and even Google does not have it. Please 'splain.

Philip

[Supercar]

Originally posted by phantom II
If I have a 'misunderstanding'...

I meant it in the 'miscommunication' meaning of the word.

--------------------------
In the latest article on suspension geometry that I read, the author first spent several paragraphs defining the terminology before he started discussing the subject. Maybe we should make a "sticky" on this forum and define our suspension terminology. There are at least 2 or 3 terms for almost everything. Sometimes they are identical and interchangeable, but sometimes they are close but not exactly the same.

The scrub radius is not the same as the caster offset.

Actually I said "camber trail". But okay, if you say that Scrub Radius is SQRT(Caster_Trail^2 + KPI_Trail^2), where the KPI Trail is the distance between the steering axis at the ground and the center of the contact patch as measured in the frontal view, then I will buy that.

Please do not introduce "offsets". Then we will get completely confused. Those are usually measured at the wheel center and not on the ground.

Camber trail is the extention of the caster trail line to the outside depending on camber.

You probably meant to say the extension of the *KPI trail*. Gotcha!

Scrub radius is the arc the contact patch describes from lock to lock. There are other definitions but it doesnt matter, because you have something to work with.

BOOOOOOO... Let's not think we are always right. But, honestly, that's the best definition that I've heard.

We probably need to define the contact patch too. It moved around, you know, when you pump up the tires, when the car starts moving, when we steer, when camber changes, when the car rolls, and when lateral accelerations are applied.

Philip

[Greg Locock]

Just to add to the confusion, when Milliken (and me) say scrub radius we mean the front view distance from the king pin axis to the nominal centre of the contact patch. Your definition would make a negative scrub radius mathematically impossible. Check figure 19.1 in RCVD.

I agree the CP moves around, but unless you want to go to a force based definition for EVERYTHING then at least we can agree that there is an instantaneous steering axis, and an instantaneous centre of the CP, even if one millisecond later it has moved.

Incidentally one of the less appealing aspects of a dual ball joint arm for a suspension is that the scrub radius and the mechanical trail move around as you steer or raise the wheel.

[Supercar]

Originally posted by Greg Locock
Just to add to the confusion, when Milliken (and me) say scrub radius we mean the front view distance from the king pin axis to the nominal centre of the contact patch. Your definition would make a negative scrub radius mathematically impossible. Check figure 19.1 in RCVD.

But you have to agree, Greg, that the Phantom's definition of scrub radius is so unique and elegant! Next time I meet Milliken I will ask him to change his book and redefine the scrub radius correctly.

A little bit more on terminology. Recently an editor of a US pop auto magazine did not know what to call the steering axis and the ground intersection. So he named it after himself. He called it the "Dave's Point".

Philip

[desmo]

Is there a "proper" name for that? In quasi or non-technical discussions of trail on motorcycles I've only heard it referred to as "the point where the steering axis intersects with the ground."

[Supercar]

Originally posted by desmo
Is there a "proper" name for that? In quasi or non-technical discussions of trail on motorcycles I've only heard it referred to as "the point where the steering axis intersects with the ground."

Mark Ortiz calls it a little shorter: The Ground Intercept.

Here, I found his articles:
http://www.auto-ware...edgebase&page=0

This one is about "Ideal Geometry":
http://www.auto-ware...an=0&page=0#665

Here is some sticky-worty material: Steering Geometry Terminology by Mark Ortiz:
---------------------------------------------------------
To help readers who are less conversant with steering geometry, I am inserting some comments of mine from the August 2002 newsletter, which explain various steering geometry parameters and their effects:

The steering axis is a line about which the wheel steers, usually through the two ball joint centers of rotation in an independent suspension, or the kingpin axis in a beam axle. This line can be defined by the point where it intersects the ground and by its angular orientation. These are commonly described in terms of the X and Y coordinates of the ground intercept, with respect to a local origin at the contact patch center, and the transverse and longitudinal angles relative to ground plane horizontal.

The front view distance from ground intercept to contact patch center, or local Y, is called scrub radius, or steering offset. It would make more sense to call the top view distance from ground intercept to contact patch center the scrub radius, but most people use the term to mean the Y or transverse component of this. This quantity is generally considered positive when the contact patch center is outboard of the ground intercept.

The side view distance from ground intercept to contact patch center, or local X, is called trail, or sometimes caster trail or mechanical trail. It is positive when the ground intercept is forward of the contact patch center.

The front view angle of the steering axis from ground vertical is called steering axis inclination (SAI), or sometimes kingpin inclination (KPI). It is positive when the steering axis tilts inboard at the top, which is almost always the case.

The side view angle of the steering axis from ground vertical is called caster. It is positive when the steering axis slopes rearward at the top.

These parameters are controlled partly by the design and adjustment of the control arms, and partly by the design of the spindle, or spindle/upright assembly, together with the hub and wheel.

The term spindle can mean either the stub axle, or pin, that carries the bearings, or the assembly including this pin and the upright, especially when these are one piece.

The spindle or spindle/upright determines two important parameters: spindle inclination and pin lead or pin trail.

Spindle inclination is the front-view inclination of the steering axis, relative to pin or wheel vertical, as opposed to ground vertical. Spindle inclination approximately equals SAI minus camber. Spindle inclination is almost exactly identical to SAI when camber is zero. It is exactly identical when both camber and caster are zero.

The steering axis and the wheel axis do not have to intersect, unless we want the right and left uprights to be identical parts, with bolt-on steering arms and caliper brackets. The steering axis can pass behind the wheel axis, as it does on a bicycle. The perpendicular distance between the two axes is called pin lead. This is equivalent to the dimension we call fork rake on a bicycle. If the steering axis passes in front of the wheel axis, that's pin trail. So pin trail is negative pin lead, and vice versa.

Effective pin length is the distance, along the wheel axis, in front view, from the steering axis to the wheel centerplane. This distance depends on the wheel and hub as well as the spindle/upright.

[Greg Locock]

Yes that's reasonably clear and consistent. We call pin lead, horizontal offset, and don't really talk about that, or pin length, very much at all (if we did I'd call it front view lateral offset.

Pin lead is quite important in its own right, some people have criteria for it, I don't, I cover it by using scrub radius and kpi. Either works.

[desmo]

Thanks Supercar, clear and concise