83 replies to this topic

[Limits]

My daughter wanted to drive her tricycle backwards today. She sat down facing the rear and wanted me to push her. While doing that I noticed how much eaiser it was to navigate around the toys and furnitures that was laying around in the living room than it was when going the normal direction. I tried to Google a bit about rear wheel steering but I did not come up with much else than fork lifts, tractors and trikes/bikes. There is a lot of 4-wheel steering, but it seems that experiments with rear-wheel-only steering is either very rare or very poorly documented. Maybe both.

As I see it there is good historical reasons for having the front wheels doing the steering and that is simply because transmission was difficult enough on a stiff axle that did not move. But then came front wheel drive and I would suppose that someone must have been tempted to try out rear wheel steering to keep the simple transmission? No?

In the air and in water it is pretty much unthinkable to have the steering in front, as land vehicles have. The rudder of a boat or an aircraft is always in the rear. What is the technical reasons for having the steering on the front wheels still in these days? For a family car I guess stability is a big issue, but a racing car have to be built in a quite extreme way to get rid of some stability in search for agility.

This question/topic is also in the light of Max's 2011 suggestions with active aero and such. Why not go advanced with the steering as well?

[desmo]

That sounds like a good non-obvious question to me. I'd like to hear someone who is versed in vehicle dynamics answer it. Is rear wheel steering inherently inferior to front in any likely above walking speed scenario from a vehicle dynamics perspective?

[imaginesix]

Yes, though with the given comparisons to boats and planes I'm not sure why any more.

I suspect the important difference is that ships stabilize themselves by leaning into the turn despite their 'rear-steer'. If that's right, then if they somehow couldn't lean (like a car can't) they would be unstable in yaw.

[phantom II]

Backing into a parking is easier than driving into it forward. Tires need a slip angle to produce lateral grip. By turning the rear wheels away from the turn at speed would eliminate side thrust from the tire and you would spin out. 4 wheel steering in the case of some Chevy trucks reduces turning circles at slow speed but at high speed, the rear wheels turn the other way and increase cornering abilities. A lot of cars are designed with roll understeer especially front drive cars by causing the rear wheels to turn toward the turn. Renault Cleo and Porsche 928. The higher the speed, oversteer becomes more difficult to control. This is why you never rotate your tires to increase life. The best tires always go at the back except for drifters..
Some aircraft have canards for pitch control at a penalty of greater drag. The rudder and elevator is therefore placed at the rear but would work in the front. F22 and Euro Fighter have canard pitch control devises up front but also have rear stabilators and thrust vectoring but the Wright brothers and Burt Rutan used movable canards in the front with two rudders at the back on their aircraft designs.
A rudder on a plane does not steer the plane, the banked wing does and the rudder is used to prevent adverse yaw.
A plane, a boat, an airboat and a car all turn differently. In all cases the rudder is submersed in a fluid whereas the tire has mechanical grip.. A small boat turns by causing a rotation about the center of buoyancy which is above the keel and depending on the shape of the hull it will turn better or worse producing much drag as it slips. The Titanic could not turn because of the turbulence from the propellers being in reverse and since it could not rotate about its center of buoyancy, its turns were sluggish anyway.. Had the captain kept full thrust the Titanic would have avoided the ice berg. If it had a rudder in the front, it would have avoided the iceburg but prop wash makes rear steer better in most cases.. New ships have thrust vectors. A flat bottomed airboat's rudder re directs airflow from a rear mounted airplane propeller mounted on the deck and sluggishly turns the boat.
It is possible to turn an airplane in horizontal flight by using the rudder and by using the ailerons to keep it level but like the boat it would slip or skid around the turn. In a banked turn the rudder is neutral. A propeller at the back of an airplane is more efficient than if it were placed in the front.

Originally posted by Limits
My daughter wanted to drive her tricycle backwards today. She sat down facing the rear and wanted me to push her. While doing that I noticed how much eaiser it was to navigate around the toys and furnitures that was laying around in the living room than it was when going the normal direction. I tried to Google a bit about rear wheel steering but I did not come up with much else than fork lifts, tractors and trikes/bikes. There is a lot of 4-wheel steering, but it seems that experiments with rear-wheel-only steering is either very rare or very poorly documented. Maybe both.

As I see it there is good historical reasons for having the front wheels doing the steering and that is simply because transmission was difficult enough on a stiff axle that did not move. But then came front wheel drive and I would suppose that someone must have been tempted to try out rear wheel steering to keep the simple transmission? No?

In the air and in water it is pretty much unthinkable to have the steering in front, as land vehicles have. The rudder of a boat or an aircraft is always in the rear. What is the technical reasons for having the steering on the front wheels still in these days? For a family car I guess stability is a big issue, but a racing car have to be built in a quite extreme way to get rid of some stability in search for agility.

This question/topic is also in the light of Max's 2011 suggestions with active aero and such. Why not go advanced with the steering as well?

[ToooHeavy]

I'll take a blind stab at it; I think the problem with rear wheel steering is that it is much harder to control the vehicle as it appoaches the limits of lateral traction, say you start to understeer (as front wheels generally have greater angle than rear when cornering), then to get the front back you have to bring the tail of the vehicle back into line with your motion, rather than just twitching the front wheels back into line. Oversteer may be near as damn impossible to recover.

Ultimately front and rear steering together could be the optimum, with clever independent computer control of the steering, braking and traction control at each corner you could ensure that each wheel is always running at the limit of it's friction circle, with optimised slip angles etc. But if the car has been perfected to that level then there may not be much left for the driver to do.

As an aside I believe that the Thrust SSC speed record car of a few years ago had rear wheel steering, and I believe there was a honda prelude? in the late 80's with rear wheel steering (in conjunction with front), not sure what advantage it conferred, but may have given a better turning circle.

[imaginesix]

Originally posted by phantom II
A plane, a boat, an airboat and a car all turn differently. In all cases the rudder is submersed in a fluid whereas the tire has mechanical grip..

Oh yeah, that's another important difference. The ships aren't affixed to anything at the front, so when the rear is kicked out the front points in as the whole thing simply rotates around it's center of gravity.

So, all that might explain the apparent difference in behaviour between rear-steer and rudder-steered vehicles but it doesn't explain why rear-steer is so inherently unstable.

[Greg Locock]

It is a great question.

My guess for keeled boats is that

(a) the keel protects the rudder structurally

(b) the disturbed flow from the rudder doesn't cause drag along the whole of the keel

but

© I don't know.


For cars, at high speed you don't point the wheels so much as establish a slip angle. If the car is near its limit in order to reduce the overall slip angle of bothe axles you first have to push the rear inwards, that is, you have to make the rear tire work even harder. Not exactly a great recipe for success!

well that's my quick answer, I need to set up a thousand run experiemnt before i get to the pub tonight

[McGuire]

A couple of well-known vehicles with rear-wheel steering that come to mind... Thrust SSC and Bucky Fuller's Dymaxion car. But those were the proverbial special cases. The Dymaxion killed a person or two when it crashed, which was attributed to the RWS. The secret to the SSC's success was in not placing any solid objects in its path.

The same geometry that makes RWS a good deal on fork lifts makes it a bad deal at road speeds. It is a simple matter to experience a rear-steering vehicle for yourself. Simply get in your own car, place the transmission in reverse gear and then drive backward as fast as you can. Voila, you have successfully converted your car to rear-wheel steering... except the driver's seat is now facing the wrong way, which is admittedly an inconvenience.

Handy Tips: Do not look over your shoulder to see where you are going: use only the mirrors. Also, place your hands at the bottom of the steering wheel instead of the top: palms down at 8 and 4 o'clock instead of 10 and 2. Then you will not have to process steering backward -- your brain will do it automatically. In case you were wondering, yes: I have done LOTS of backward driving. Not to brag but I am a very good backward driver... at least as competent as many motorists going frontward, from what I can tell. It's a gift I suppose.

[baddog]

Could a big problem be weight-shifting? when braking, which typically precedes cornering at speed, the weight is always going to be shifted forwards, resulting in more load and grip at the front, while the rear tyres are at their lightest loads?

[Gecko]

The main problem is indeed the limit behaviour. On a regular car with no oversteering tendencies, if you steer more than the front tyres can handle you will end up with terminal understeer which is a stable condition. If you experience oversteer in a regular car, you put on some opposite lock to reduce the force on the front wheels and bring the car in line.

On a rear wheel steering car, imagine driving at the limit and then turning in more than is necessary. This will reduce the slip angle of the rear wheels temporarily and cause transient oversteer and therefore introduce a potential instability. In the case that the car is experiencing oversteer on its own, then an opposite lock will simply wind on more slip angle at the rear, potentially causing further oversteer, depending on the behaviour of the tyre past its limit.

The plane analogies work less than ideally because in a typical car both front and the rear wheels provide comparable cornering forces, whereas on a typical plane the wing is used for providing most of the lift, with the elevators or canards used mostly for adjusting the attitude (providing the torque needed to rotate the airframe and keep it there).

[Limits]

I can for sure see a lot of problems with rear wheel steering in standard family cars. I remember all too well when I first got understeer on snow with an old SAAB 96. Stability is an issue. But a racing driver spend most of the time fighting the stability of his car. When turning in he desperately tries to make the rear wheels breake loose a bit to aid turn in. Jackie Stewart, I think it was, said something like "people think we use the brake to slow the car down, we don't, we use it for steering".
A rear wheel steered car would not have to break traction to turn in, it could stay within the optimal slip angles into higher speed? Or is it necessary to have the front wheels helping out?

McGuires example of reversing a street car is not very good I think. First of all, reversing a RWD car feels completely different from reversing a FWD. At least to me. Second, the steering/suspension geometry must be made completely different if you want to drive comfortably in reverse. Or rather pretty much the same, but moved to the rear axle. When you are reversing your car you get a negative caster angle and that is a good recipe for instability. Put negative caster on a normal, front wheel steered, car and you will think: "oh, this is not very nice".

[imaginesix]

Originally posted by McGuire
Simply get in your own car, place the transmission in reverse gear and then drive backward as fast as you can.

You forgot the disclaimer: You are to be held harmless for any ensuing damage including property and personal, without limitation. (In other words try it in a big empty parking lot).

But the comparison isn't quite fair either because by driving a FWS car backwards you end up with negative trail and therefore inherent steering instability.

[McGuire]

Originally posted by Limits
McGuires example of reversing a street car is not very good I think.

No, that is exactly what it is about. Visualize the thrust paths at the steering axle vs. the longitudinal CL of the vehicle. Why is it easier to back into a tight spot? Why do tow motors have rear steering?

[Wolf]

I saw this a long time ago, so am posting it here because it's...well...crazy.



It certainly does not have standard engine (650cc I2, 24BHP), but would be good for your tests...

[phantom II]

This type of behavior truly makes life worth living. I laughed from beginning to end. My wife came over to see what was going on and walked away. Don't get me wrong, she is a happy fun loving person but females lack certain faculties, don't they? Thanks for that.
It's amazing just how much laughter there is on this here BB.

[/B][/QUOTE

Originally posted by Wolf
I saw this a long time ago, so am posting it here because it's...well...crazy. :lol:



It certainly does not have standard engine (650cc I2, 24BHP), but would be good for your tests... :lol:

[McGuire]

Originally posted by Limits
When you are reversing your car you get a negative caster angle and that is a good recipe for instability.

No, think about it. If you turn the car around backward so the front wheels are now the rear and use the rear wheels to steer, "negative" caster is exactly what you want.

Originally posted by imaginesix
But the comparison isn't quite fair either because by driving a FWS car backwards you end up with negative trail and therefore inherent steering instability.

Ditto.

I don't know about you guys. It's like you have never driven a car backward at high speeds or something.

[McGuire]

Originally posted by phantom II
This type of behavior truly makes life worth living.

True that.

[desmo]

I realize there are probably very good reasons for front axle steering's supremacy but maybe if there had been a full century of clever developers optimizing for rear steering things might look different. The question raises interesting issues in any case, as revisiting near axioms can.

I like driving in R relatively fast too. I doubt I can sufficiently seperate my unfamiliarity from a reversed seat and controls from whatever else is going on though.

[Limits]

Originally posted by McGuire


No, think about it. If you turn the car around backward so the front wheels are now the rear and use the rear wheels to steer, "negative" caster is exactly what you want.

I do not following you here. Negative caster is what you get, for sure, but that is hardly what you want? Why would you want negative caster?

I don't know about you guys. It's like you have never driven a car backward at high speeds or something.

Not only that, but I have tried to bike with a "funny bike" with negative camber. It is difficult

The main problem with reversing at speed is that instead of the normal state were the wheels want to point straight, meaning autocentering of the steering wheel, you have a situation were the front wheels (now rear in reverse) will want to "rest" fully right or fully left. They hate staying in the center. Of course you get instability. It's like balancing an egg on top of a ball instead of inside a ball.

[RDV]

When near slip limit reducing rear steer angle to increase yaw will induce a positive feed-back loop....very unstable... that said could be used in a fly(steer)-by wire system.... have used it in a 4 wheel steer car and it does wonders for turn-in in ice racing....

[Greg Locock]

So what /did/ happen to four wheel steer on production cars?

[Moon Tricky]

Originally posted by Limits

I do not following you here. Negative caster is what you get, for sure, but that is hardly what you want? Why would you want negative caster?

You certainly wouldn't want it on a FWS car. You'd get exactly the opposite of autocentring.

I think the reason why RWS is a bad idea on cars is that when you go round a corner at speed, the front of the car gets to the corner before the back of the car does. The front of the car ought to follow the curve so that the back can follow it. If the steering is at the back, the back end is going to swing out away from the line you are trying to follow.

[Dragonfly]

Originally posted by Wolf
I saw this a long time ago, so am posting it here because it's...well...crazy.



It certainly does not have standard engine (650cc I2, 24BHP), but would be good for your tests...

Ha-ha!
I had a Fiat 126 P (for Polish) once. And it contributed a lot to my informal qualifications as a mechanic.
Best thing about it was that there wasn't any trouble finding a place to park.
The one on the video though has the powertrain changed, I presume from another small front WD Fiat and the trailing original steering wheels are more or less loose. Steering again is done by the (driving) front wheels.
The old Skoda 120 (rear engine, rear WD) is more convinient for conversion since one could easily turn the gearbox (in fact it happened when people repaired them by themselves) into 4 backword and 1 forward gears simply by assembling it the wrong way.

[Moon Tricky]

Originally posted by Dragonfly

The old Skoda 120 (rear engine, rear WD) is more convinient for conversion since one could easily turn the gearbox (in fact it happened when people repaired them by themselves) into 4 backword and 1 forward gears simply by assembling it the wrong way.

I heard the engine in the old East German Trabants was so simple you could start it running backwards by pushing it forwards in reverse gear.

[Engineguy]

Originally posted by Greg Locock
So what /did/ happen to four wheel steer on production cars?

Real or perceived:

Lack of interest.
Solution to a non-problem.
Nobody asked for it.
Geeky.
Scary.
Goofy.
Oddball.
Gimmicky.
Stupid.
Cost.
Weight.
Ferraris don't have it; why do I need it?
Rolls Royces don't have it; why do I need it?
Race cars don't have it; why do I need it?
Cost.
Nobody cared.
Fear the U.S. Gummitup would mandate it on all cars (like they're doing with ESC )
Little or no benefit.
Benefit not demonstrable.
Misunderstood.
Cost.

[Greg Locock]

Agreed. Here's the real problems it does solve

1) Turning circle

2) High speed stability (subjectively)- mostly yaw gain linearity

They are both odd requirements - 99% of the time 99% of drivers don't care about them.

2) can be approached in all sorts of ways - 4WS or ESC being one extreme, but more usually careful tuning of PAS, tires, and the general cornering compliance budget will get you there. ESC is a method of last resort - the number of ESC interventions per 1000 km is an interesting metric - people like me like to hear the ESC going off, but your normal driver does not.

1) has few solutions and none make anyone very happy.

[phantom II]

http://www.usatoday....13-qsteer_x.htm

[Limits]

Well. I have been thinking a little. A boat and an aeroplane are stable. They have stabilizing rudders and fins or whatever and you can never bring it into an "oversteer" slide. You can get an aeroplane into a spin but that is another matter. So in those case the rear mounted rudder can just push and push and push. As soon as you release the push the machine will stabilize itself.

A car is only stabilized by it's wheels. When you go to fast into a corner and start to exceed the maximum slip angle you will want to reduce it. You do that by steering a bit less. On a rear wheel steered car that would increase the slip angle... So the only way to bring it into safety would be to lower the speed. Another thing, and that is propably what RDV means with positive feedback loop, when you are on the limit in a corner and need to steer more you will actually decrease the slip angle, giving the car plenty of grip and it will happily turn just as you ask it to. But as soon as you stop increasing the steering angle you will increase the slip angle again and this time you are well beyond the limit and the tyres will simply let go immediately.

In short: With FW steering the car will struggle to turn and with RW steering it will never want to stop turning.

I guess that in theory you can put tail wings like aeroplanes and speed boats have to counter the loop effect but I think the only safe way is to have also the front wheels turning. But imagine how early you would be able to get in the throttle with RWS

[imaginesix]

Originally posted by phantom II
http://www.usatoday....13-qsteer_x.htm

They've ceased production of the 4WS version. Was a good try though, the most sensible application of 4WS I've ever seen.

The Japs tried it on the late 80's Prelude, the Mazda MX-6 and the Nissan 240SX. From what I was told you could hardly notice the difference even driving at 10/10ths.

[Slumberer]

If I recall correctly, when designing the Thrust SSC they mocked up a scale version using a mini with the steering locked off and a huge rear subframe to get the wheelbase they needed.
They said that so long as there was no slack in any of the linkages it was fantastic to drive, but any slack made it dreadful.
One issue is that to initiate a turn, the rear wheels turn in the opposite direction and the back has to go the "wrong way" before following the front in the direction you want to go, which, when you think about it can't be a good thing. (Especially in a 10 tonne car doing 700mph!)

They only went down that route as it solved their packaging issues.

[cheapracer]

The Honda was quite clever actually and was a mechanical system.

For the first part of your turning the steering wheel the rears turned inward (if you turned left so did the rears) then as you turned further the rears straightened then as you got to full lock the rears turned outwards.

This combination gave slow speed parking help with large steering wheel movements and added turn stability at speed, with small steering wheel movements, especially lane changes.

It was liked by most testers of the time, but then like most testers they probably got a free T Shirt when they landed in Switzerland/Germany/Austria for the fully paid for 1 week holiday, errr release, so what else would they say.

The Mazda was the same but was hydraulic.

4 wheel drive was just taking off and I think it may have been quietly dropped for being seen as a poor compromise to actually having 4wd, imo.

[zac510]

If we drove only rear wheel steer cars wouldn't we have honed a driving style to accomodate its peculiarities?

[cheapracer]

Originally posted by zac510
If we drove only rear wheel steer cars wouldn't we have honed a driving style to accomodate its peculiarities?

Yes, S-L-O-W.

[zac510]

Originally posted by cheapracer


Yes, S-L-O-W.

Maybe you miss my point - if you had to walk on your hands for your whole life you'd end up pretty good at it, wouldn't you?

[Catalina Park]

I have driven a lot of rear wheel steer vehicles and they are only safe when travelling backwards.

[F1Fanatic.co.uk]

Benetton gave it a crack in '93 - did any other F1 teams try it? Of course it's out of the question today.

[zac510]

Originally posted by F1Fanatic.co.uk
Benetton gave it a crack in '93 - did any other F1 teams try it? Of course it's out of the question today.

It's hard to find information about that setup but what I have gathered is that it was only used in conjunction with front wheel steering and only to change the attitude of the car upon corner entry, much in the same way the the ice racing/Andros cars use it. I 'investigated' the ice racing earlier this year and likened the application of the 4WS to that of a fly-off handbrake on a rally car. It is quite a unique case and successful due to the types of circuits the Andros cars race on.

Neither of these examples apply 4WS in the way Limits asks about. He is talking about the rear wheels being the sole steering wheels and being used to negotiate the entire corner or obstacle.

[McGuire]

Originally posted by Catalina Park
I have driven a lot of rear wheel steer vehicles and they are only safe when travelling backwards.

Exactly. At one time very large forklift trucks (harbor use etc) were built with dual controls or a swing-over steering column so they could be driven steering-axle-first if they were going any speed/distance, like over the road. I believe nowadays these very large lift trucks are usually 4WS or "crab steering" as they call it over there.

[Limits]

Originally posted by Catalina Park
I have driven a lot of rear wheel steer vehicles and they are only safe when travelling backwards.

Yes. A FWS car will have forces trying to push in the opposite direction of that what the front wheels are turned and therefore they will form some kind of a stability. If you turn right, the forces will push left. With RWS it is not so, if you turn right, you will turn the rear left and the dynamic forces will push in the same direction and there is no way to stop a disaster if you come close to the limits. Flop flop flip flip. Impossible. So 'cheapracer' is right, the only driving style that must be learned is S-L-O-W

4WS is a different matter though, it could give excellent turn in, stability mid corner and a wonderful corner exit. Would need plenty of computing power though to get it optimized.

[Greg Locock]

I disagree with the computing comment, if a purely mechanical system can be made to offer an improvement on the road, it seems to me that you wouldn't have to be especially clever to see an improvement in the limited context of a racetrack. What it effectively gives you is a way of coping with different speed corners - which, given the convergent layour of modern tracks is perhaps not as useful as it once was.

I had a bit of a look at it last weekend - the canned strategies in the modelling program I use basically give you a low speed (out of phase) gain and a high speed (in phase) gain, and the option of making them polynomials. However the default setup was appalling, so I didn't get much further than trying to make it as good as 2ws, and failed.

[Moon Tricky]

Originally posted by Greg Locock
I disagree with the computing comment, if a purely mechanical system can be made to offer an improvement on the road, it seems to me that you wouldn't have to be especially clever to see an improvement in the limited context of a racetrack. What it effectively gives you is a way of coping with different speed corners - which, given the convergent layour of modern tracks is perhaps not as useful as it once was.

I'm still rather fond of this kind of set-up.
http://en.wikipedia....i/Weissach_axle

[Greg Locock]

Big whoop. Why were they using an STA in the first place?

[Moon Tricky]

Originally posted by Greg Locock
Big whoop. Why were they using an STA in the first place?

I don't know. I just presumed it's just because it was ages ago, when double-wishbone wasn't quite so common.

Even still, I can't see why a multilink version of the same principle isn't plausible.

[imaginesix]

This was beautifully answered for me a while back in this post by one of our generously helpful BB members;

"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 kinematic ones, but it is better to have small numbers for everything rather than big numbers with opposite signs."

Perhaps it may help explain things for you too?

[cheapracer]

Originally posted by Limits


Yes. A FWS car will have forces trying to push in the opposite direction of that what the front wheels are turned and therefore they will form some kind of a stability. If you turn right, the forces will push left. With RWS it is not so, if you turn right, you will turn the rear left and the dynamic forces will push in the same direction and there is no way to stop a disaster if you come close to the limits. Flop flop flip flip. Impossible. So 'cheapracer' is right, the only driving style that must be learned is S-L-O-W

4WS is a different matter though, it could give excellent turn in, stability mid corner and a wonderful corner exit. Would need plenty of computing power though to get it optimized.

I would have thought a selected gear related sytstem would be simple and effective - rear turn out in first gear graduating through the gears to turn in in fifth gear.

I built a Honda Civic '4cyl bomber' speedway car for a guy (forgive me, its probably the lowest I've sunk) and wound the rears as far as they would go to offer (substantial) turn in at the rear and that car won every race it entered and gave him the Archerfield (Brisbane, Queensland) title. It looked terrible crabbing down the straight but was neutral in corner and stayed down low and tight. Some may think that turn out is whats required in that situation, I assure you it doesn't work!

[cheapracer]

Originally posted by zac510


Maybe you miss my point - if you had to walk on your hands for your whole life you'd end up pretty good at it, wouldn't you?

Give me a circus performer who has spent his/her whole life practicing walking on their hands and I'll put my money on a 5 year old child in a running race.

[Moon Tricky]

Originally posted by imaginesix
This was beautifully answered for me a while back ...

Perhaps it may help explain things for you too?

I'm well aware of why semi trailing arms don't get used anymore, but this isn't a normal semi-trailing arm. This set-up produces understeer, not oversteer. It's not about what sort of suspension it is, it's about the idea of passive rear wheel steering, i.e. rear wheel steering that isn't directly controlled by the driver but rather depends on the vehicle's roll in corners. At slow speeds it won't make any difference at all to the steering (although you'd get toe-in if you went over a big bump, but I've got ideas of how to prevent that) but provides understeer at high speeds by turning the wheels into the curve as the car leans to one side.

I think you could implement this quite simply in a double-wishbone setup just by making the outer pivots point inwards towards the front, so that as the wheel goes upwards, it turns outwards as well.

Like this:


EDIT: wait, I think I might have got that backwards. You get the idea though, I hope.

[phantom II]

Haven taken delivery of one of these pieces of S.H. one T.s in 87 in Stuttgart, I can assure you that awful axle had a mind of its own with very unpleasant manors. At 170mph, it is steady as a rock and high speed tight turns on the Autobahn felt quite safe. But get to a two lane mountain road in the alps and then the fun begins. Left right transitions require continuous steering inputs far beyond normal. If you are accelerating, your steering activity becomes even more lively. Now try changing up or down, in tight turns, you will find that it must be accompanied by an abrupt opposite steering input especially when accelerating in high torque gears and then a quick return to where the steering was prior to the gear change. Two spins and I got it right. The true value of the car is revealed when you try to sell it.
In 93, I took a Vette ZR1 to Europe and did much the same trip. What a pleasure even with a 1960 Jag rear axle.

Originally posted by Moon Tricky


I'm still rather fond of this kind of set-up.
http://en.wikipedia....i/Weissach_axle

[Limits]

Above is the reason why I think you need the assistance of the computer department to get a 4WS steering that can cope with all situations. It needs to be able to first assist in reaching the limit quickly and then basically invert itself to help staying at the limit. Obviously the key is have the system knowing were the limit is. I think modern fighter planes uses a technology that is somewhat similar. You build something that is really unstable (to be agile) and have computers balancing it.

[Greg Locock]

Your life in my hands...