26 replies to this topic

[sosidge]

Flicking through this weeks Autosport mag, there is a quite from Matt Neal along the lines of "we had the best rear-wheel steering system on the car this year" (re: the 2011 BTCC Civic).

Does anyone have any insight into what systems the touring cars are running, or was it some kind of in-joke by Neal?

[BRG]

Flicking through this weeks Autosport mag, there is a quite from Matt Neal along the lines of "we had the best rear-wheel steering system on the car this year" (re: the 2011 BTCC Civic).

Does anyone have any insight into what systems the touring cars are running, or was it some kind of in-joke by Neal?

Surely not, or Jason Plato would have been whining about it.

I can't see anything in the BTCC Regs about this.

[Tony Matthews]

Isn't a case of compliance in the rear suspension bushes to allow a bit of movement?

[cheapracer]

Isn't a case of compliance in the rear suspension bushes to allow a bit of movement?

For sure, better known as roll steer.

I rallied a Civic for a short stint and had toe in on the very adjustable rears and looked after a speeedway Civic that won the Queensland 4 cylinder titles a couple of times mostly due to having the rears pointing well left - crabbed terribly down the strights but tracked like it was on rails around corners. mind you if all speedway cars were like that no one would go to watch, bloody boring as hell!

Of course there's the famous Porsche 928 Weissach rear toe compensation setup ...

http://www.autozine....21.htm#Weissach

When the Nissan GTR "Godzilla's" campained in the Australian touring car championship they had their rear steering disconnected only because "it's just another thing we have to understand and we have enough on our plate already" (Fred Gibson, Team Manager).

Edited by cheapracer, 07 November 2011 - 16:25.

[sharo]

From the typical road cars Citroen Xantia (and I think the other models with hydro-pneumatic suspension) has a "passive RW steering". The rear axle frame is mounted on 4 metal-rubber bushes and in combination with the trailing arm type design, over certain cornering force the frame moves within 3 degrees.

[Greg Locock]

From the typical road cars Citroen Xantia (and I think the other models with hydro-pneumatic suspension) has a "passive RW steering". The rear axle frame is mounted on 4 metal-rubber bushes and in combination with the trailing arm type design, over certain cornering force the frame moves within 3 degrees.

Any car with a rear twist beam, that is to say, most FWDs, will have rear compliance steer. The natural effect on an RTB is compliance oversteer, which is unhelpful. Over the years people have turned that into compliance understeer using angled bushes, bushes with wedges in, weirdo linkages, and a watts link. The last named is an act of desperation, as it eliminates all of the advantages of a twist beam in return for a slight handling advantage that fails to close the gap on an IRS - the difference in handling between the watts and non-watts version is due to the tires, their watts system is so soft as to be meaningless.

Calling it passive rear wheel steer is marketing jargon, in my not very humble opinion.

[desmo]

Is there any reason to believe that 4WS could- even theoretically- significantly reduce lap times in road racing? It seems to me that if a conventional car is set up to use close to all four tires' potential, there's just not a lot of possible upside.

[DaveW]

Greg, you will probably recall the active rear steer research vehicles produced by Lotus Engineering in the late 80's.

These controlled rear steer using a "trajectory demand" algorithm using front steer position as an input. The effect on vehicle handing was profound, improving the steering time constant (one mid engine example was measured at around 6 Hz, I recall), giving mere mortals the capabilities of a rally driver and, incidentally, making it impossible to ride in the back seats of a 4 seater for more than a few minutes (the initial turn acceleration was in the wrong direction). We didn't do a race car, but it is difficult to believe that it would not improve exit speeds from low-speed corners.

When a turn was initiated the rear steer angle immediately turned to pro-steer, decreasing as the turn became established, ending up mildly counter-steer. That would suggest, I suppose, that the correct strategy for a passive rear steer arrangement would be "compliance understeer", but this is likely to have a negative impact on steering time constant (not necessarily bad in a road car).

If I remember correctly, SAAB had a different philosophy.

Edited by DaveW, 08 November 2011 - 09:43.

[cheapracer]

When a turn was initiated the rear steer angle immediately turned to pro-steer, decreasing as the turn became established, ending up mildly counter-steer. That would suggest, I suppose, that the correct strategy for a passive rear steer arrangement would be "compliance understeer", but this is likely to have a negative impact on steering time constant (not necessary bad in a road car).

That's not a long way from what the mass produced Honda and Mazda systems did as you turned the steering wheel - first a little pro-steer, going back to straight through to countersteer nearing full lock. The Honda being pure mechanical and the Mazda electronic which i believe was speed sensative. Many good driving reports from both but Greg may have more insider view on them...?

I don't know what the Nissan GTR's sequence was.

[DaveW]

That's not a long way from what the mass produced Honda and Mazda systems did as you turned the steering wheel - first a little pro-steer, going back to straight through to countersteer nearing full lock. The Honda being pure mechanical and the Mazda electronic which i believe was speed sensative. Many good driving reports from both but Greg may have more insider view on them...?

I don't know what the Nissan GTR's sequence was.

I don't know what Nissan did, but I am fairly sure that the Honda system was "scheduled" (not quite as you describe, perhaps), rather than "active" (my sequence occurred with no change in steer angle, or speed). In that sense the Honda system was deficient, because it responded in an unhelpful way to "opposite lock" events.

Edited by DaveW, 08 November 2011 - 10:16.

[sharo]

Any car with a rear twist beam, that is to say, most FWDs, will have rear compliance steer. The natural effect on an RTB is compliance oversteer, which is unhelpful. Over the years people have turned that into compliance understeer using angled bushes, bushes with wedges in, weirdo linkages, and a watts link. The last named is an act of desperation, as it eliminates all of the advantages of a twist beam in return for a slight handling advantage that fails to close the gap on an IRS - the difference in handling between the watts and non-watts version is due to the tires, their watts system is so soft as to be meaningless.

Calling it passive rear wheel steer is marketing jargon, in my not very humble opinion.

I admit it's somewhat difficult for me to understand everything, because I lack some knowledge in English terms and (maybe) auto slang. For example I now wonder what "compliance steer" or "compliance oversteer" is. Would be much obliged if someone enlightens me.
But as far as I can see, you are talking about driving a touring car in race manner, which is a different matter than normal driving. My only experience is with the Xantia and I've tested its behavior in a sharp turn carrying a bit more speed where when the "passive steering" kicks there's an added oversteer and the feeling of losing the back, which requires a slight correction with the steering lock.

[gruntguru]

For example I now wonder what "compliance steer" or "compliance oversteer" is. Would be much obliged if someone enlightens me.

Compliance steer occurs when the wheels "steer" in response to lateral (cornering) forces. Compliance oversteer (of the rear wheels) occurs when the wheels turn towards the outside of the corner.

This should not be confused with roll steer (a response to suspension travel) which is of course another form of passive rear wheel steer.

[cheapracer]

I don't know what Nissan did, but I am fairly sure that the Honda system was "scheduled" (not quite as you describe, perhaps), rather than "active" (my sequence occurred with no change in steer angle, or speed).

yes, I see what yours did and no the Honda wasn't capable of that being an uninteruptable pure mechanical sequence dictated by the steering wheel only. How complex or not the Mazda was I don't know.

In that sense the Honda system was deficient, because it responded in an unhelpful way to "opposite lock" events.

I have never driven one let alone had one sideways so I don't know, may have even reduced the slide angle back to a slip angle and regained traction...


How did your system know where it was relative to the corner?

Edited by cheapracer, 08 November 2011 - 12:22.

[Magoo]

http://www.qcxy.hb.c...ge/96596856.pdf

[DaveW]

How did your system know where it was relative to the corner?

It simply responded to steering commands & vehicle responses, using the algorithm defined in http://www.freepaten...com/5348111.pdf

Essentially, the driver demanded a trajectory with the steering wheel, & the system attempted to oblige.

[cheapracer]

It simply responded to steering commands & vehicle responses, using the algorithm defined in http://www.freepaten...com/5348111.pdf

Essentially, the driver demanded a trajectory with the steering wheel, & the system attempted to oblige.

Thanks DaveW, will have a look'see tomorrow

[Greg Locock]

I'm guessing the Lotus car was probably'Trilby' , but I left after it had run but before rear wheel steer had been tried, and as a mere NVH engineer I wouldn't have got more than a demo anyway.

The way I remember the Honda system was that for small angles it gave a small positive gain (understeer) then neutral and then for large angles it gave vast amounts of oversteer, to improve the turning circle etc. This was all done inside a little gearbox with a steering column going to the rear rack.

And if it wasn't like that it should have been.

So far as compliance steer goes there are several effects. To start with if you push sideways inwards at the CP it is obvious you'll get positive camber. In plan view you'll get some toe, either in or out depending on where the elastic centre of the suspension is.

But bear in mind in the real world the sideways force is applied at the pneumatic trail point, which can be anything up to 60mm behind the nominal CP. So we either measure the compliance steer at two different trails, or measure the toe/self aligning torque as well as toe/lat force. Then we know how compliance steer changes with pneumatic trail.

If you draw a rear twistbeam in plan view I think it is fairly obvious that unless you do something clever you'll get compliance oversteer. The general rule of thumb with any rear suspension is to make everything behind the axle as stiff as possible, and to make any compromises in front of the axle, so as to give compliance understeer. This applies to subframes, subframe bushes, arm bushes and so on.

With the front suspension one might think the opposite is true, that is you could afford to have soft bushes behind the axle. In practice this may give great linear range understeer, but you end up with horrible steering feel.

In practice compliance steer can only be used in tiny amounts, most of the understeer in a car with nice steering feel comes from the weight transfer/tires, and rollsteer. Lookee here for a fictitious but not completely ridiculous example http://en.wikipedia....undorf_analysis (although 4 deg/g is an awful lot)

Edited by Greg Locock, 08 November 2011 - 22:04.

[Lee Nicolle]

Flicking through this weeks Autosport mag, there is a quite from Matt Neal along the lines of "we had the best rear-wheel steering system on the car this year" (re: the 2011 BTCC Civic).

Does anyone have any insight into what systems the touring cars are running, or was it some kind of in-joke by Neal?

I suspect it may be an in joke, probably about a design fault in the car. Or accident damage.

[jatwarks]

I suspect it may be an in joke, probably about a design fault in the car. Or accident damage.

I took it as a typical, cryptic throwaway line from a driver.

As far as the Honda and Mazda production cars were concerned I believe that one, or both, allowed the rear wheels to turn parallel to the fronts at higher speeds, for smooth lane changing, and opposite to the fronts at slow speed for tighter turning circle.

I would suggest that a true 4WS car, in which the rear wheel movement exactly mirrors the front, would have the advantage of having the tyres always rolling in the direction of wheel travel, without side slip, at least when the car is being driven within certain performance limits.

Perhaps 4WS with 50/50 4 wheel drive would be an interesting exercise - with some good traction control to keep it civil!

[MatsNorway]

I believe the 3000GT from mitchubitiisis something had that.

the 300ZX also had it i believe.

[Fat Boy]

Many moons ago I remember watching a 4WS FSAE car going through it's paces. It was fairly good on the ridiculously tight tracks they had it on. They had all sorts of gears and pulley and crap hanging off the back of the car to make it work. The funny thing was watching it in the drag strip run.

The driver launched and got a little squirrely initially. It was probably a torque steer thing. He almost had it gathered up then hit second gear. Now it got a little more squirrely. Third gear didn't help. He hit fourth just as the car departed from controlled flight and went spinning off through cones and scattering track marshals. No one got hit or hurt, but I think a lesson was learned.

Rear toe can be a super powerful tool. I'm amazed at how really small toe changes can effect a car at times, especially on ovals.

[cheapracer]

As far Mazda production cars were concerned ..... allowed the rear wheels to turn parallel to the fronts at higher speeds, for smooth lane changing, and opposite to the fronts at slow speed for tighter turning circle.

From Wiki ....

Used from 1987-1997 on the majority of MX-6s around the world[citation needed], this was the MX-6s centerpiece.
It was available on both generations, although the North American market only received it for the 1989 model year. South African MX-6s never received 4WS.
According to Mazda, the system provided:
Superior cornering stability
Improved steering responsiveness and precision
High-speed straightline stability
Notable improvement in rapid lane-changing maneuvers
Smaller turning radius and tight-space maneuverability at low vehicle speed range
The system electronically controlled a rear rack that was behind the rear wheels. At low speeds (of up to 35 km/h), the rear wheels would move in the opposite direction to the front wheels, aiding parking and U-turns by lowering the turning circle. Above these speeds, the rear wheels would move in the same direction as the fronts, meaning control during high speed manoeuvres such as lane changes or cornering was improved. Either way, the turning angle of the rear wheels was slight at just five degrees, a measurement Mazda determined to be optimally effective and natural to human sensitivity.
Note:
When the engine is turned off, the rear wheels would straighten up. They would change back to the angle of the front wheels when the engine is restarted. This is caused by the 4WS control unit powering down, and the failsafe system overriding the rear rack.
If the system ever faults, as a failsafe the rear wheels would lock straight to allow the vehicle normal 2WS functionality.

Perhaps 4WS with 50/50 4 wheel drive would be an interesting exercise - with some good traction control to keep it civil!

Earlier Nissan Skyline GTR's came close to all that, R32, R33, not sure about the R34.

Ahh, what would life be without Wiki ... Fatboy, note the highlight below ..

HICAS (High Capacity Actively Controlled Suspension) is Nissan's rear wheel steering system found on cars ranging from the more recent Skyline and Fairlady Z (300ZX) iterations to smaller models like the Nissan Cefiro (A31), 240SX/Silvia (S13 & S15)/180SX and Nissan Serena/Nissan Largo. It is also found on models from Nissan's Infiniti division, such as the Q45, M45/M35 and G37. Unlike other four wheel steering systems, HICAS and Super HICAS are fitted to improve handling rather than as a parking aid.
Earlier HICAS versions used hydraulics to steer the rear wheels. The hydraulic system was powered by the power steering pump and used speed sensors to determine how much and in which direction to steer the rear wheels. Later versions, called Super HICAS, moved to an electric actuator for the rear steering rack, making the system much lighter. The Super HICAS system also used its own computer to control the system instead of speed sensors. HICAS and Super HICAS rear wheel steering is limited to about 1 degree in either direction.
HICAS was introduced on the 1986 Skyline GTS coupes (GTS, GTS-R, and GTS-X). The system was later adapted to work on many models in the Nissan range, beginning with the Passage GT.

Edited by cheapracer, 09 November 2011 - 17:58.

[cheapracer]

geez, forgot how many cars actually had 4WS back then - Mitsibishi VR4



The explanation in Japanese but follow'able ..

Part 1/

Part 2 is very interesting ....

Part 2/

Edited by cheapracer, 09 November 2011 - 18:25.

[BRG]

geez, forgot how many cars actually had 4WS back then - Mitsibishi VR4

I am pretty sure that Ralliart used to disconnect the AWS for rallying.

[cheapracer]

I am pretty sure that Ralliart used to disconnect the AWS for rallying.

[Bloggsworth]

If you want rear wheel steering, get an Austin 7. It had 1/4 elliptic rear springs, as soon as the car started to roll the outside wheel mooved back relative to the inside and turned the car into the corner. You could have done a row of pylons by getting the back seat passengers to rock from side to side while you held the steering wheel in the straight ahead position...

[jatwarks]

Didn't Freddie Dixon, Tony Rolt & Harry Ferguson try to develop "The Crab", a 4WS 4WD special? By all accounts the name must have been an afterthought!

Presumably the steering rack gearing for a 4WS car would have to be adjusted to allow for the fact that the rear is helping to turn? If front and rear wheels were to turn through the same angle then half the steering input would be required.

The sensitivity of driver inputs would be much more critical.

The sense of driving from the seat of the pants would change too!