is 4 wheel steering a better option????

[Krish1108]

hi guys,

what do u think about 4 wheel steering in our cars????? obviously, the handling will be better but what about the packaging issues.....????

[Krish1108]

hi guys,

what do u think about 4 wheel steering in our cars????? obviously, the handling will be better but what about the packaging issues.....????

[NickFavazzo]

Packaging issues can be resolved by using the Johnson rod
link to a distributor http://kalecoauto.com/index.ph..._info&products_id=18 (http://kalecoauto.com/index.php?main_page=product_info&products_id=18)

[AxelRipper]

If you can make it work within the rules and it works for you, sure. If not, no.

Far more other better things that can make the car better. How about making your rear tires small enough that your rearend rotates just as well?

[exFSAE]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Krish1108:
obviously, the handling will be better </div></BLOCKQUOTE>

BS.

[JDS]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">y views or opinions expressed by me may in no way reflect those of Kettering University, it's stu </div></BLOCKQUOTE>

I think most FSAE teams don't even extract the full potential of a 2-wheel steer vehicle from a vehicle dynamics perspective. The complexity of a 4-wheel steer system is not just twice as complex, it is exponentially more than that. I think saying a blanket statement such as 4-wheel steer will obviously handle better is not accurate at all.

[Kirk Feldkamp]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by NickFavazzo:
Packaging issues can be resolved by using the Johnson rod </div></BLOCKQUOTE>

Why not use the Johnson Pump (http://www.johnson-pump.com/) instead? For power steering, of course! Haha.

[Dr Tron]

[QUOTE]Originally posted by Krish1108:

obviously, the handling will be better QUOTE]

Clearly... I see no downside... bet the farm

[buggaero]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Krish1108:
hi guys,

what do u think about 4 wheel steering in our cars????? obviously, the handling will be better but what about the packaging issues.....???? </div></BLOCKQUOTE>


It changes after slip sets in. Now can you design it accordingly is highly improbable. I would still say...

Beat yourself first, to beat the rest!

[Giova2k11]

Interestingly enough, we are also looking into the possibility of doing a 4-wheel steering car for next season. I'm not entirely convinced about that solution, but the guys who work on the suspension for our car have been working on it for the past couple of weeks...

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Giova2k11:
Interestingly enough, we are also looking into the possibility of doing a 4-wheel steering car for next season... </div></BLOCKQUOTE>
Which way would you steer the rear wheels?

Same way as fronts, or opposite?

And by how much?

Z

[Krish1108]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Giova2k11:
Interestingly enough, we are also looking into the possibility of doing a 4-wheel steering car for next season. I'm not entirely convinced about that solution, but the guys who work on the suspension for our car have been working on it for the past couple of weeks... </div></BLOCKQUOTE>

hi mate...can u tell me more about it as in how are you guys exactly managing it....

[Scott Wordley]

Its been done in the past.
A bit of counter steer makes the car ultra pointy in transients and dramatically shortens your turn radius, but also more unstable and twitchy at the limit during steady state cornering.
Getting enough stiffness into the system whilst minimizing steer effort and compliance is also tricky. As is figuring out what rear geometry you want to be running.

PS: Good to see you back on the boards Z!

[shark.ashwa]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Scott Wordley:

PS: Good to see you back on the boards Z! </div></BLOCKQUOTE>

+1

[Giova2k11]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
Which way would you steer the rear wheels?

Same way as fronts, or opposite?

And by how much?

Z </div></BLOCKQUOTE>Well, you have to steer the rear wheel opposite than the fronts to achieve a narrower turn radius. You could steer it the same way and that would stabilize the rear end and avoid spinning, but at the speeds we go we would be better off with the countersteering.

By how much, I don't know an exact figure but the turn of the rear wheels whould be significantly less than the fronts otherwiese you end spinning; as far as I know we are atalking about 4° max.

@Krish: I don't know the details as it's not the group I'm working with, I work on the engine so I haven't seen the designs yet. From what I know about 4-wheel steering though, the system can be all emchanical, or electrically/hydraulically activated. You can search for Renault, Nissan and Toyota's efforts in the last ten years or so and see what they've done there.

[Big Bird]

OK, consider this. My understanding of the lateral force buildup and cornering response at speed goes something like this (remembering that to build up a lateral force a tyre needs a slip angle):
1. Steer front wheels into corner to build up slip angles on front tyres
2. Slip angles on front tyres generate lateral force on front axle. Vehicle begins to turn in / yaw.
3. As vehicle turns in and lateral inertial force builds up, the chassis becomes angled "inward" relative to the direction of travel, building up a slip angle at the rear wheels.
4. Once slip angle is generated, we can get some lateral grip out of rear wheels, and their contribution to lateral cornering force begins.

Now, with rear wheels turning OUT of the corner, the following happens:

1. Steer front wheels into corner to build up slip angles on front tyres
2a. Slip angles on front tyres generate lateral force on front axle. Vehicle begins to turn in / yaw.
2b. At the same time, the rear wheels turning outwards builds up a slip angle and therefore lateral force in the opposite direct, i.e out of the corner. This speeds up yaw response.
3. As vehicle turns in and lateral inertial force builds up, the chassis becomes angled "inward" relative to the direction of travel. However, for car to actually follow the corner path, and not just spin around its central axis, we need inward lateral force at both front and rear axles. So the rear wheels need to assume an "inward" facing slip angle. But we've turned them "outward" relative to the road. The rear of the car needs to "wag" from it's outward slip angle, past its zero slip angle / zero lateral force point, to an appropriate inward slip angle orientation. And the driver is going to have to catch this tail-happy swing with appropriate countersteering at the front.
4. Once slip angle is generated, we can get some lateral grip out of rear wheels, and their contribution to lateral cornering force begins.

So it can be done, but you'd need a pretty handy driver to be able to catch the "spinning top" tendency and convert it into actual lateral cornering force.

How much countersteering at the rear do you want? Well, what does your tyre data tell you? What is your vehicle's yaw inertia and how quickly does the car "rotate" on turn in. What is your driver telling you? Is your data telling you that the one last thing holding you back from making up that last couple of points to Stuttgart is that you need to crank up your yaw acceleration on turn-in at the expense of lateral force? And I'm trusting by your willingness to go that extra step, with all the complexity and potential grief that it would create, that your team is already performing at Stuttgart level...

[Mervyn]

We have been using 4 wheel steering on plant for quite a while. After some experimenting, much to our bosses annoyance, we have found that if the steering gearing is reduced to give less movement per turn, wwe can move pretty quick & turn without to much problem. We also tried altering the amount of rear axle steer compared to front axle steer. That also helped. Not sure how this would work on highspeed vehicles but am sure there will be a program available somewhere to test this on a pc.

[Big Bird]

That is exactly the problem. At low speeds, when you are relying on the geometric orientation of the wheels for steering, rear outwards is good. I recently spent an hour in a stretch Hummer limo, and the 4 wheel steer on that (rear out) was brilliant. It turned corners like an almost normal car. But at high speeds, when you are relying on slip angle building grip, rear out hurts you - or at least delays the time it takes for the rear to build grip. So a lot of trouble to build something that might end up causing a lot of trouble.

Where are you losing time on track? Slow hairpin corners? And are you willing to trade off some time on the slaloms or high speed stuff to buy a gain in the slow stuff?

[Z]

Krish and Giova2k11,

Please read everything Geoff wrote above (both posts). Then re-read it.

I think it was about 20 years ago when every Japanese car with sporting pretensions came with "4-wheel-steer". Some were active (computerised), some all mechanical. I think most of them gave up on the idea about 10 years ago (?).

IMO the most sensible version (Honda???) was all mechanical and initially steered the rear wheels the SAME direction as fronts. So at small steer angles, as in high speed corners, the car was more stable. As the hand-wheel was turned further for sharper and slower (parking speed) corners, the rear wheels went back the other way and "counter-steered" giving a tighter turning circle. Other systems did something similar, but with electronics and hydraulics (ie. "counter" at slow parking speeds, "same way" at high speed).

If you watch nature documentaries you see the big cats using "active rear steer" when they chase their dinner. The rear briefly steers out on corner entry (all in one big step), then immediately steers (or pushes) inward. (The tail also swings around for "active yaw and roll control".) Downhill ski racers also "hop" the rear of the skis outward on corner entry. But both of these have extremely capable active control systems!

If your team has won every FSAE comp for ten years, and the Design judges are pushing you to "come with something different next year", then active 4WS might be worth considering. http://fsae.com/groupee_common/emoticons/icon_cool.gif

If your team is not a regular winner, then 4WS is almost as likely as 4WD to get you to comp with an overweight and unfinished car. http://fsae.com/groupee_common/emoticons/icon_frown.gif

If you want agile handling I would recommend minimising yaw inertia (centralise all heavy masses), and design your steering with as much "positive ackermann" as you can (ie. inner wheel 10++degrees more that outer). Then if you want even faster turn-in add some static rear wheel toe-out (but only +0.1 degree at a time! http://fsae.com/groupee_common/emoticons/icon_eek.gif ).

Z

[Big Bird]

Just a further argument that I didn't think of earlier. This one not based in geometry and slip angles but rather people.

Z above speaks of the Japanese factories playing around with 4WS around 10-20 years ago. If I've got my facts straight, the chief designer of the Honda 4WS system as used on the Honda Prelude was a very nice man by the name of Dr Shoichi Sano. Those with any knowledge of F1 history will know that the designer of Honda's first race-winning F1 car, way back in 1965 (?) was also a very nice man by the name of Dr Shoichi Sano. And those with any knowledge of FSAE history will know that the team patriarch of the Tokyo Denki FSAE team is a very dear man by the name of, you guessed it, Dr. Shoichi Sano.

So TDU have as their team mentor a gentleman who has designed F1 winning race cars, and who is an expert in 4WS. Do TDU run with 4WS? No.

If you ever have the good fortune to discuss FSAE design with Dr Sano, he will advise you very gently that the way to success in this competition is to keep it simple. I think this is very wise advice, and I for one would not claim to be better informed than someone as humble, wise and well respected as Dr Sano.

[Krish1108]

Thank U guys....I got a lot of valuable information...and some additional stuff too.... http://fsae.com/groupee_common/emoticons/icon_smile.gif i guess i would come to a conclusion that we would not be going for the 4WS.....thanks guys.....!!!! http://fsae.com/groupee_common/emoticons/icon_smile.gif

[JDS]

I think for a 4-wheel steer system to work well with a Formula SAE car, the system would have be active, which is pretty complex for a student team, not to say it can't be done though. The Mazda RX-7 from 1986-88 had a passive system, which was basically a bushing that flexed under lateral loading in the rear causing the rear wheels to steer out increasingly with load. It was not a popular system because for average road drivers as they took a corner fast, it felt like the car was oversteering and going into a spin when it reality it was just turning in more due to the rear-steer. Most drivers would then counter-steer when it wasn't really necessary and this caused a snap snip in the opposite direction. After 88 they abandoned the passive rear-steer.

For rear-steer to work well the rear-wheels would need to steer with intelligence, which virtually eliminates any sort of passive system. I would think a computer sensing velocity, yaw moment, front/rear slip angles, coupled with tire data would be necessary to properly implement rear-steer. I don't think it is out of the question for a Formula team to achieve this, but with the complexity, added weight, and time necessary to achieve such a feat, you would need to carefully analyze the performance benefits and see if it is worth it or not.

[Buckingham]

I would always suggest to master 2 wheel steer before attempting 4 wheel steer, but lets throw this variation on the idea up for some friendly debate:

What about dynamic rear toe (both rears toe-in or both rears toe-out) as a function of steering input angle?

It could conceivably be done with some rather simple linkages (or cables).

[JDS]

It could conceivably be done, but should it be done? Without taking into consideration factors such as vehicle velocity, yaw, and current slip angles I think such a system would be far from ideal.

[Lorenzo Pessa]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by JDS:
I think for a 4-wheel steer system to work well with a Formula SAE car, the system would have be active, which is pretty complex for a student team, not to say it can't be done though. The Mazda RX-7 from 1986-88 had a passive system, which was basically a bushing that flexed under lateral loading in the rear causing the rear wheels to steer out increasingly with load. It was not a popular system because for average road drivers as they took a corner fast, it felt like the car was oversteering and going into a spin when it reality it was just turning in more due to the rear-steer. Most drivers would then counter-steer when it wasn't really necessary and this caused a snap snip in the opposite direction. After 88 they abandoned the passive rear-steer.

For rear-steer to work well the rear-wheels would need to steer with intelligence, which virtually eliminates any sort of passive system. I would think a computer sensing velocity, yaw moment, front/rear slip angles, coupled with tire data would be necessary to properly implement rear-steer. I don't think it is out of the question for a Formula team to achieve this, but with the complexity, added weight, and time necessary to achieve such a feat, you would need to carefully analyze the performance benefits and see if it is worth it or not. </div></BLOCKQUOTE>

A passive system similar to the described one was studied last year by my team. It was not used in the 2011 car because the testing time was too short.
I don't know what will happen this year.

[Giova2k11]

Z, I see what you're talking about. And I agree. I also agree on the way the rear wheels should be turned, countersteering at low speeds and the same than the front wheels at high speeds to make the car more stable; which is exactly what Renault did on its Laguna and Laguna Coupè just a couple years ago. It was all controlled by an ECU and actuated through an electric motor, so it was an active system and not a passive one like an all mechanical system would be.

[Buckingham]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">It could conceivably be done, but should it be done? Without taking into consideration factors such as vehicle velocity, yaw, and current slip angles I think such a system would be far from ideal. </div></BLOCKQUOTE>

But doesn't the system already effectively exist on every car, but with a gain set to zero?

How can you be sure that a gain of +1 and a gain of -1 are both worse?

Just like your static rear toe. You can go through all the tire data in the world, but you should still physically test rear toe +,0,- and just go with whatever the stopwatch says, right?

It doesn't have to be ideal to be better. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[JDS]

A toe setting is static. A rear-steer mechanism that changes based on steering wheel angle is far from static. You wouldn't want the rear wheels steering the same at all velocities, and to be based completely on steering-wheel angle. That is a sure way to have a poorly designed 4-wheel steering system.

[Jan_Dressler]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by JDS:
I think for a 4-wheel steer system to work well with a Formula SAE car, the system would have be active, which is pretty complex for a student team, not to say it can't be done though. </div></BLOCKQUOTE>
TU Graz did it in 2009:
http://racing.tugraz.at/filead...ingGraz/DSC_0023.JPG (http://racing.tugraz.at/fileadmin/galleries/2009/2009_04_25_FotoshootingGraz/DSC_0023.JPG)
http://racing.tugraz.at/filead...Silverstone00060.jpg (http://racing.tugraz.at/fileadmin/galleries/2009/2009_07_17_Silverstone/Silverstone00060.jpg)
http://www.formulastudent.de/f...-20_3522_Almonat.jpg (http://www.formulastudent.de/fileadmin/user_upload/all/2009/pics/FSG09//20090809_Sunday/20090809_13-09-20_3522_Almonat.jpg)
http://www.formulastudent.de/f...-58_3530_Almonat.jpg (http://www.formulastudent.de/fileadmin/user_upload/all/2009/pics/FSG09//20090809_Sunday/20090809_13-11-58_3530_Almonat.jpg)

[PatClarke]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">TU Graz did it in 2009: </div></BLOCKQUOTE>

But, was it actually active in the comp?

Pat

PS, remember, 'Steer by wire' and similar, bans apply to the rear axle too!

[StevenWebb]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by PatClarke:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">TU Graz did it in 2009: </div></BLOCKQUOTE>

PS, remember, 'Steer by wire' and similar, bans apply to the rear axle too! </div></BLOCKQUOTE>

Graz could still have run it back then, steer by wire of the rear wheel was only banned since 2011 (possibly because of graz?)

the old rule was
6.5.4 Rear wheel steering is permitted only if mechanical stops limit the turn angle of the rear
wheels to ± 3 degrees from the straight ahead position.
6.5.5 The steering wheel must be mechanically connected to the front wheels, i.e. “steer-by-wire”
of the front wheels is prohibited.

new rule
B6.5.1 The steering wheel must be mechanically connected to the wheels, i.e. “steer-by-wire” is prohibited.

[Jan_Dressler]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by PatClarke:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">TU Graz did it in 2009: </div></BLOCKQUOTE>

But, was it actually active in the comp?

Pat

PS, remember, 'Steer by wire' and similar, bans apply to the rear axle too! </div></BLOCKQUOTE>
At least in Hockenheim it was not active, according to the TU Graz guys I spoke to back then.
But it would have been legal in 2009.

[Scott Wordley]

I understand and agree with the comments you have made Geoff and Z.

It is definietly not something that I would recommend teams bother looking at until you have a complete and reliable car and a lot of development in other more important areas.

But still I do think that 4WS is a feature worth considering for FSAE, particularly (maybe only?) for very tight and twisty tracks like the Australian comp.

At the end of the day many teams are chasing transient response, and speed through hairpins and slaloms. 4WS could improve those segments of the track dramatically perhaps at the expense of some ultimate cornering potential and some very scary high speed lane changes.

Of course, as JDS recommends, a completely active system would be ideal so that you could control the rear rates based on front steer, speed, yaw etc but such a system appears illegal under the current rules, and besides it sounds like a lot of work and upfront investment in something unproven.

I would prefer to start simple, and test a system with basic adjustability (i.e. linear but adjustable (not dynamically) front:rear steer ratio). Combined with statice rear toe changes that should be plenty to play with and get a feel for how the system might work and what the trade offs might be. I would prefer to do this experimentally, on the car, because I think that the limits of the drivers ability to react to and control the system will be very important. Also important is how well you can eliminate the free play in the system, and keep the steering efforts low. Also important is the quality of steering feel and limit feel the driver is getting back. These aspects that I wouldnt expect to show up clearly in simulations, but could have a dramatic effect on the system's performance and potential viability.

Personally, I think that there is potential for a simple 4WS system to provide increased performance for FSAE on some tracks. It's unproven but also a pretty cheap and easy modification, so it could potentially provide a good return (points/$).

[Big Bird]

Funny, since posting above I've been thinking of a driver actuated rear toe-out on the outside rear - defaults back to conventional when driver lets go of actuation lever. Leave it alone for most of the track, actuate when you get to one of those three-point turn hairpins we love here in Oz. It would be "active" and "intelligent", to the level of how "active" and "intelligent" your driver is. And you could "program" it with a cricket bat. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Scott Wordley]

I think you call that "firmware"!

Jokes aside, thats about the level of 4WS you want, just alternating between parallel / slight toe in for the straights and the maximum amount of rear toe out you like for the twisty bits.

Running that amount of static toe out all the time is not great for tyre wear...

[Big Bird]

Geez Scott, we are agreeing on something. A special occasion like this deserves a beer...

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Big Bird:
... I've been thinking of a driver actuated rear toe-out on the outside rear - defaults back to conventional when driver lets go of actuation lever... </div></BLOCKQUOTE>
Geoff, somewhat "similar" systems are common in rallying and off-road racing. They are "fiddle" or "steer" brakes. The lever is at your left side (assuming you're sitting on the "right" side of the car http://fsae.com/groupee_common/emoticons/icon_smile.gif). Push forward and it applies right rear brake. Pull back and it applies left rear brake. So the car turns in the direction of the push-pull. These are a simple hook-up into the hydraulic brake system, and give awesome yaw acceleration. The yawing force comes from the longitudinal rather than lateral tyre force, so is quicker because no need for slip-angle build up (ie. "relaxation" length).

For those still considering 4WS, I reckon (just speculation...) you want countersteer proportional to steering-wheel velocity. After the steering-wheel stops turning the rear wheels should steer slightly into the corner (same-steer) as lateral G builds. So when the driver gives the steering-wheel a quick turn the rear wheels jump outwards. As soon as the steering-wheel stops turning (or below a threshold dAngle/dt) the rear wheels go to straight ahead. As lateral G builds the rear wheels steer slightly into the corner.

The above could also be done with rear toe-in/out, which might be a safer way to start. This might be possible with an all hydro-mechanical system. A hydraulic damper type mechanism (possibly a gear pump+) senses steering-wheel angular velocity. When velocity is high the rear wheels toe-out. When Vel~0 the wheels toe-in.

Personally, I'd concentrate on a neat and robust front steering arrangement with, say, 50+deg angle for the inner wheel (not a trivial task).

Z

[Chapo]

Has anyone investigated the use of moving the rear toe link so that it does not intersect with the IC (so that you get bump steer) so that as the car rolls the toe is changed slightly giving a passive rear steer?

I think that it could work with a little tuning...

[JDS]

The problem with that is instability under acceleration and braking. Also, as the car rolls, one side is in compression and the other is in rebound, meaning the wheels are toe-ing in opposite direction. Either that or you would only be getting the desired toe change in one direction of turn. Maybe for oval racing! :-)

[Z]

JDS,

Err, half right. "As the car rolls" one wheel toes in, the other toes out, so both steer in the same direction. Hence the term "roll steer" for this behaviour. You are right in that it is a disadvantage during rear squat or lift.
~~~~~o0o~~~~~

Chapo,

This sort of "tuning" is covered in one (or more) of Carroll Smith's books. The gist of it is "We've been given a suspension that always seems to have some sort of bump steer, so how do we adjust it so that it is LEAST BAD."

Z

[Tom Wettenhall]

I believe you'll find that roll steer will perform in a largely similar fashion to 4WS in high speed corners, i.e. it's evil; and in any case where you're not generating maximum lateral g, you'll have a car which doesn't really turn at all.
You could perhaps reverse the direction of the roll steer, start with some toe out which makes your turn-in plenty good, but then recovers when you get some body roll going. I may test that on our car, actually...
The RX-7 system mentioned earlier on strikes me as being much the same thing, a g-sensitive toe control.

[Tom Wettenhall]

I should say, our 2010 car ran like that. Once we got it turning a bit better after comp, it just got nasty, difficult to recover from slides. But that car had other handling issues.
It's really easy to build the possibility of roll steer in to a car, just space your tabs a bit further apart than you usually would, and have different sets of spacers which you can swap to move the pickup up or down.

[JDS]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
JDS,

Err, half right. "As the car rolls" one wheel toes in, the other toes out, so both steer in the same direction. Hence the term "roll steer" for this behaviour. You are right in that it is a disadvantage during rear squat or lift.

Z </div></BLOCKQUOTE>

Whoops. Where was my mind yesterday! http://fsae.com/groupee_common/emoticons/icon_redface.gif Righto!

[Adambomb]

My $0.02 after examining both roll steer and rear "active" steering on fairly comprehensive whole car vehicle dynamics model for the vehicle dynamics class they used to offer here. Sort of a re-hash of what many smarter people have said already, perhaps I can add a little something http://fsae.com/groupee_common/emoticons/icon_wink.gif

For roll steer, since it is on the "reactive" axle and is a reactive effect driven by roll angle, it will not help turn in. Instead, it is simply another way to add steady-state under or over steer. In fact, I prefer to think of it as no more than a fancy application of bump steer (kinematically at least). OK, now here's some fun food for thought: Solid axle cars often have roll steer! For a coil-spring car with trailing arms, think about the "steady state" inclination from a side view of the trailing arms, and how their effective length from a top view changes as the car rolls. Now with leaf spring cars, imagine which end is fixed and which end pivots, and how the spring mounting location moves fore and aft in roll. OK, enough for the random tangent of the day...

Active steer! For our assignment we were to examine the behavior of a system that mimics the control system used in a late '80s Honda Prelude, which sounds quite similar to the one in the RX-7 (and probably most of them in those days). The rear steer angle was determined solely from front steer angle. At low angles it "crabwalked," ie rear wheels turned the same direction. I guess the thought was this would work well for fast lane changes or dodging things on the highway, as you could alter your trajectory and just skip the whole yawing the car business, all without waiting for the rear tires to build up lateral forces. On a steady state corner it would add understeer...because yet again you're avoiding the whole yaw thing. But hey, if one were so inclined, I suppose they could also build in some roll steer to re-cancel that out, right?

Then as steering angle increased the rear steer gain function would change directions and go opposite, for excellent mall parking lot maneuverability and tighter parallel parking. I suppose this may happen at the steering angles you would see on a tight autocross corner; I personally have no idea how well this setup worked in autocross. It would add a steady state oversteer component, but again at steer angles high enough that you would really only see in a driveway or a parking lot (whether at the mall or autocross!). Not only that, but to get to this level of gain the rear steer has to change directions; don't imagine that would do good things for transients.

OK, so with that system, the biggest things you gain are the abilities to dodge things on the highway and make the car easier to park. Sounds like a system well suited to a minivan to me! Not to say that a different control scenario couldn't be devised for racing.

[JDS]

Just to be clear, the RX-7 system was not an active rear-steer. It used flexible bushings, so the amount of rear steer was proportional to the lateral force acting on the rear tires, and not related to steering angle.