23 replies to this topic

[meb58]

I was skimming thru Milliken over the weekend while pondering the need for caster. I remember once a long time ago adding caster to a honda I was driving in an effort to help the car track stright, but also at the time, to add a little feel, right or wrong. My experience taught me that caster carries some penalties and I wonder why we use it all? The car was actually much better with stock caster...a better compromise.

I guess I see caster as a band aid...perhaps caster is expedient in the world of auto design and manufacturing. But from a pure suspension sense - theory - do we actually need caster?

[Ben]

But from a pure suspension sense - theory - do we actually need caster?

It gives some camber compensation with steering, which is useful.

Ben

[Greg Locock]

and it damps the self-osciallting mode of the wheel, which is useful.

But it allows variations in vertical force at the contact patch to create steering torques, which is not useful.

[Fat Boy]

It's like oregano. You can't make an entire meal out of it, but it's nice to throw a bit in the soup.

[Billzilla]

If your suspension has King Pin Inclination - which they pretty much all do to varying degrees - on the front wheels, then you need some castor to work with the KPI to control the camber change with the wheel turned.
I typically run about half the KPI angle for castor.

[johnny yuma]

Power steering has made it possible for manufacturers to set large positive caster settings which otherwise make steering heavy, because as you lock over, negative camber is generated which means you are lifting the car's weight slightly (with your arms) on that side of the car on the outside of your turn. The good aspect is the negative camber generated by the positive caster overcomes the lean of the car into the corner and the flex of the tyre sidewall to keep a flatter road contact patch.
However for my money if you have toe-out instead of a little toe-in,and lots of positive caster and a big static negative camber setting in the straight ahead position ,you will have a car with heavy steering which wears tyres out quickly and won't go in a good straight line on a straight road. They seem OK on a racetrack but crap on the highway.

[cheapracer]

It's like oregano. You can't make an entire meal out of it, but it's nice to throw a bit in the soup.

You eat folded paper?

I like caster, if you brake hard with a late turn in, run a little toe out and have strong forearms caster is your friend!

[meb58]

Is there another way to eliminate self oscillation? I understand this phenomenon pretty clearly.

Also, is there a clear relationship between caster angle and SAI/King pin from the perspective of washing out the self centering forces we get from SAI?

[Lee Nicolle]

I have used a fair amount castor in my tank 71 Galaxie cruiser, it has given the overassisted steering SOME feel, it self centres better and turns in well. Used with a small amount ofneg camber it handles quite well for its size. But the overassited steering is still a big let down.
The Aussie RHD cars used a Saginaw power steering from a mid 60s Chev and are not variable ratio. Strong but crap steering!
On tintop racecars castor is essential as it makes the car turn, though you have to work out how much or little suits the car and your style. But always more than original. And more camber for radials than crossply type tyres.And toe out is essential though again how much as more than a little makes the car very unstable. But they turn in great!

[gruntguru]

Is there another way to eliminate self oscillation? I understand this phenomenon pretty clearly.

Cold showers work well for me.

[Tony Matthews]

Cold showers work well for me.

So that's what you do all day!

[meb58]

No, not that caster angle silly! ...a long chopper fork comes to mind...but then, I've always been a dreamer...

Cold showers work well for me.

I don't understand how caster helps a car to turn...my experience with caster is that it can ruin turn-in repsonse at 'some' extreme.

[johnny yuma]

I don't understand how caster helps a car to turn...my experience with caster is that it can ruin turn-in repsonse at 'some' extreme.
[/quote]
Think of a big yellow road grader.The front wheels are upright when it goes straight ahead,but when it turns the wheel on the load side lays over at a big angle(negative camber),the unloaded one leans the other way(positive camber)).This is done with a big positive caster angle built in to the grader's front end geometry.That is the top pivot of the steering is well behind the bottom pivot.On a car, the upper steering ball joint or strut top will be some small distance behind or in front of the lower ball joint.This is the caster setting.In a decent car the top point is adjustable for fore-aft (CAMBER SETTING) and side-side (camber setting). A car with zero camber setting ,but some positive caster,will geometrically develop negative camber as lock is applied,the more lock,the more negative camber kicks in.This naturally helps turn in, especially on tight corners, as it fights the bad geometric effects of body roll ,and tyre distortion.

Edited by johnny yuma, 17 December 2009 - 04:58.

[cheapracer]

,.... as it fights the bad geometric effects of body roll ....

Thats very wide - caster can also be a double whammy as it actually assists the "bad geometric effects" as the front outside corner falls (and the inside lifts) as the wheel climbs around the caster angle and transfers weight to there.

Everything in moderation.

Edited by cheapracer, 17 December 2009 - 07:14.

[meb58]

I understand how much of caster works...and to my way of thinking, camber gain is somewhat offset by weight transfer to the wrong rear wheel in an opposite lock condition...and this is caused by the raising of the diagonal front wheel...and lowering of the other...which can wash-out SAI self centering affect.

But what I don't understand is how caster helsp a car to turn...the virtual point on the road ahead of the tire helps remove the tire's ocsillation tendancy, but also, adds to steering resistance as the wheel is turned. I'm sure we could measure the increase in steering resistance for a increasing caster angles at various steering wheel angles and speeds...right up to saturation...not sure how caster affects saturation...

Zero caster angle but postive caster...do we have cars like this on the road? The verticle strut would simply be forward of the tire's center line...

I know I throw a lot of crap on the wall and I apologize for it...but the more I ask the better I understand...I was that kid in class.

[Ben]

I understand how much of caster works...and to my way of thinking, camber gain is somewhat offset by weight transfer to the wrong rear wheel in an opposite lock condition...and this is caused by the raising of the diagonal front wheel...and lowering of the other...which can wash-out SAI self centering affect.

But what I don't understand is how caster helsp a car to turn...the virtual point on the road ahead of the tire helps remove the tire's ocsillation tendancy, but also, adds to steering resistance as the wheel is turned. I'm sure we could measure the increase in steering resistance for a increasing caster angles at various steering wheel angles and speeds...right up to saturation...not sure how caster affects saturation...

Zero caster angle but postive caster...do we have cars like this on the road? The verticle strut would simply be forward of the tire's center line...

I know I throw a lot of crap on the wall and I apologize for it...but the more I ask the better I understand...I was that kid in class.

Do you mean zero caster angle and positive caster trail?

Greg will know more, but I assume having the strut angled back has a big impact on recession rate and is therefore to vehicle ride.

Ben

[meb58]

Yes...but to have zero caster the strut must be upright, no? Just like a shopping cart. If we have even an infinitesmal radian we have a caster angle...? Positive falling in front of the tire's center line and negative behind it...

And, we have caster angle, mechanical trail and pneumatic trail. I understand the first to be the literal angle fromt strut tower to ball joint in side view...if we extend that line to a point on the ground...the distance from that point the the tire's centerline in side view is mechanical trail and pneumatic trail is determined by the distacne the contact patch moves backward under acceleration? This was a question.

Ben, aren't all caster angles defined by the strut leaning back?

And another thought...as SAI precesses the strut tower while turning, big caster angles must place some interesting loads on steering? Is this related to recession rate?

Edited by meb58, 17 December 2009 - 15:35.

[Fat Boy]

Yes...but to have zero caster the strut must be upright, no? Just like a shopping cart.

What happens when you push a shopping cart fast?

[gordmac]

Claude Roelle reconed you could pick up grip running a lot of caster angle. Reduce the caster trail though.

[meb58]

So, in this case we have lots of mechanical trail and no caster angle? What if we could establish trail at the center of the tire's contact patch...feels lik an interesting, if not impossble, proposition.

What happens when you push a shopping cart fast?

[Greg Locock]

Caster trail doesn't damp the mode (so much), caster angle does. All of this discussion is ignoring the pneumatic trail. I suspect kpi has a damping effect as well. Mech trail helps to self centre the steering a bit (why we don't use springs for that I don't know)

The way I think about trail is that if the steering axis hit the exact point at which the lateral force acted, you'd have no feedback relating to the lateral force vector. In practice of course the pneumatic trail moves forward as you approach saturation, so you could have the case where the rack effort was increasing as Fy increased, but was dropping due to the reduced trail moment, so giving a car that gave little feedback about Fy in the important region.

SO, if you increase the mechanical trail a bit, you get better discrimination of Fy, but are less sensitive to saturation. If you increase trail too much then you mask changes in pneu trail. You also start to create big problems with kinematic understeer, the car tends to follow the road crown.

[cheapracer]

Ben, aren't all caster angles defined by the strut leaning back?

From where the top pivot point starts and then intersects through the wheels center we all know is the caster centerline but don't go thinking that all struts are positioned along that line.

Most of course don't stray to far from it although I see one daily that the clamp for the lower strut is well on the opposite side of the ball joint to the wheel having the strut nowhere near the KPI line.

Edited by cheapracer, 18 December 2009 - 09:11.

[Fat Boy]

So, in this case we have lots of mechanical trail and no caster angle? What if we could establish trail at the center of the tire's contact patch...feels lik an interesting, if not impossble, proposition.

Well, you're the one proposing a change from the norm. If you've thought it out and have some good logic to present, then get on with it.

[meb58]

At precisely this point as a child my mother would say, "don't encourage him." I suggested to her last week, only recently have Italian women figured out they could use a rolling pin for baking and not just as a weapon...she hit me with the rolling pin. Every raviolli will now have a dimple shaped precisely like my forehead.

But I will get on with it...

Well, you're the one proposing a change from the norm. If you've thought it out and have some good logic to present, then get on with it.