MX-3.com

That's an interesting article. I have read that before. The thing that makes me say gimmick is the fact that they are talking about a true electronically controlled 4WS system, whereas we are forced to live with an "Active" rear suspension which converts forces into 4WS "affects".

Maybe the key for the MX-3 is to enhance the 4WS "Affects" to bring it closer to behaving like a true 4WS system. I'm not sure. But I want to know.

wytbishop wrote:Maybe the key for the MX-3 is to enhance the 4WS "Affects" to bring it closer to behaving like a true 4WS system. I'm not sure. But I want to know.

Well, that essentially something totally different that it would be difficult to impliment. The 4WS system is a computer controlled electric-hydraulic system that's been programmed to turn, in responce to speed and steering input, a certain way and a certain angle. The problem with these systems is their reliability and failure because of their complexity.

The Mx-3 system is a reactive system, using the lateral forces to turn the rear wheels. You can't really make it much better than it is. You can give it more play to help for tight maneuvres, but they'll enduce high-speed instability. Basically you'd need to have a system more responsive at low speeds, and less responsive at high speeds.

http://www3.ntu.edu.sg/home/edwwang/con ... icar01.pdf" onclick="window.open(this.href);return false;

Yep, that's exactly it. As usual with mechanics, everything is a compromise. In this case, stability vs. maneuverability.
I think this idea would be great, for example for cars that are used for straight line racing, because it'd make them more stable on the track, but on the street, I'd stick with the active suspension. I have driven a lot of different cars, including several low budget sport cars, some of them great (like the Renault 11 turbo, Renault Fuego, Fiat Uno, to name a few) and some others not so good, but NONE of them had the maneuverability of the MX3. I think a set of your "pretty purple bushings" ( ) would be a great idea to get a crisper steering, but loosing the active suspension would most likely make the MX3 just as dull as any other car out there...

However, if you were to design an LCA like that one WITHOUT getting rid of the 4WS effect...

I wonder if you can made a design that uses a passive hydraulic cylinder for the steering effect, kinda of like a shock. It'll resist small movement, making it stable at higher speeds, but at lower speeds where there would be more lateral forces, it would allow more movement.

Nd4SpdSe wrote:You can give it more play to help for tight maneuvres, but they'll enduce high-speed instability. Basically you'd need to have a system more responsive at low speeds, and less responsive at high speeds.

http://www3.ntu.edu.sg/home/edwwang/con ... icar01.pdf" onclick="window.open(this.href);return false;

That's just an engineering problem. That's the best part!!! My contention is that the suspension on the MX-3, like most other systems on any given production vehicle, is a compromise. As with the exhaust system, pure performance is compromised for ease and cost of production. I can't believe that the active rear suspension as it sits in the production car is optimum, but I admit that I don't fully understand it. The reason I posted the thread was to figure out if other people are as interested in this as I am and to determine what the level of interest would be if I could come up with an affordable solution that would actually improve the car's handling.

As for the link yo uposted. It's very interesting but it doesn't really apply to this argument. The thesis of that paper is to develope an algorithm by which the most efficient path can be planned for an Autonomous Guided Vehicle with 4 articulating steering wheels. The math of that paper is being used to calculate the maximum speed and turning radius of the AGV with the added considerations afforded by the additional steering wheels.

I saw a great movie about a similar thing when I was in school about the automation of a factory in Japan where Lathe beds are made. All the little robotic trucks driving around delivering materials to the different work stations. Pretty cool stuff.

Nd4SpdSe wrote:I wonder if you can made a design that uses a passive hydraulic cylinder for the steering effect, kinda of like a shock. It'll resist small movement, making it stable at higher speeds, but at lower speeds where there would be more lateral forces, it would allow more movement.

I was thinking about something like that actually...you sure you're not an engineer?

Inodoro Pereyra wrote:Yep, that's exactly it. As usual with mechanics, everything is a compromise. In this case, stability vs. maneuverability.
I think this idea would be great, for example for cars that are used for straight line racing, because it'd make them more stable on the track, but on the street, I'd stick with the active suspension. I have driven a lot of different cars, including several low budget sport cars, some of them great (like the Renault 11 turbo, Renault Fuego, Fiat Uno, to name a few) and some others not so good, but NONE of them had the maneuverability of the MX3. I think a set of your "pretty purple bushings" ( ) would be a great idea to get a crisper steering, but loosing the active suspension would most likely make the MX3 just as dull as any other car out there...

However, if you were to design an LCA like that one WITHOUT getting rid of the 4WS effect...

I think you're probably right. Preserving the 4WS "Effect" is necessary to give the car the most maneuverability. My initial drawing was based on the pictures I posted which appear to be totally rigid. Although in fact, I don't know that they are. As I said, I don't fully understand how the system achieves this effect.

I ordered a set of the correct Outer Tie Rod ends so I will know the full affect of the bushings in another week or so. But even with the tape measure alignment I have on it now, it's really really nice.

Here's an idea:

The rear active suspension is nothing but a suspension system purposefully miscalculated to provide bump steering when turning.
A fairly "easy" ( ) solution could be to adapt the front spindle assembly to the rear, make a rear A arm pretty much like the one you drew up, but with a single ball joint to attach the spindle, and craft a link to attach the steering arm on the spindle to a fixed point in the body, calculated to provide the same bump steering. If done properly, that should have the same effect as the stock suspension (or even better), without the binding, since you're using a ball joint to anchor the spindle.

Anyways, it's just an idea...

I really think that the rear steering effect should be linear. The "bump steer" phenomenon is more of an on/off sort of a thing...although I see where you're going by trying to control it with geometry alone. Keeping it very simple.

This has all given me a lot of ideas to think about.

interesting idea, really wonder if it would benefit mx-3 suspension & how it would handle.

Um, I'll be the first one to say, duhhhhh. This is over my head, but it's pretty, and homer want bad. "Mr. Simpson, you have a crayon lodged in your brain." Hahaha, the crayola oblongata.

wytbishop wrote:I was thinking about something like that actually...you sure you're not an engineer?

Haha, no. Actually I'm glad I'm not, I actually don't get along too well them them, they seem to thing only 2-dimensionally, totaly math dependant when it comes to figuring stuff out, and trying to discuss (argue) with them is like talking to a 2D brick wall...I just wish I knew the maths so I could argue better on why I'm right, lol.

I liked the the document showing the calculation for computer controlled 4WS, but I think letting the natural forces determining the steering rather than complex calculations would be better suited. You can see it working in the Prelude video, but you can also see it in some instances try to anticipate and correct itself when it's wrong.
http://www.youtube.com/watch?v=4JCks2Ps53s" onclick="window.open(this.href);return false;

On that, I was also thinking if a speed-sensitive valve could be implimented. The valve limiting or allowing movement of the fluid, thus limiting steering at higher speeds. The only thing you'd have to make sure would be to allow the system to return to center.

The Prelude system was I believe pure mechanical, and was the most reliable of all the systems.

There's a lot of videos on "Quadrasteer" on youtube:
http://www.youtube.com/watch?v=O9g8oqE7Ghc" onclick="window.open(this.href);return false;
http://www.youtube.com/watch?v=Vb3SJeoE ... re=related" onclick="window.open(this.href);return false;

An electrical/powered hydraulic system would probably be too expensive and complex for our cars. In fact, moving away from anything passive towards a more dynamic setup would most likely involve restructuring the complete rear subframe and welding in new pieces. If any sort of active/dynamic setup would be used, the prelude or mx6 "4ws" system would be good to get ahold of, and try to adapt parts, or at least get a better idea of how it works. We could then tweak this to try to minimize complexity on the mx3. Since none of us have any experience in this particular field, we would be starting from scratch - while that is not necessarily a bad thing, it makes things more complicated.


I really do like this idea. I don't want to eliminate the "4ws effect" but would rather improve upon it like everyone is saying. As far as high speed stability, we would probably want to test whatever modifications we do at a track, somewhere that allows speeds of 130 or higher. While most of us will never speeds that great, the car is capable of it. I don't think a speed sensing lock would work to improve high speed stability, because then you would lose the benefits of the 4ws on hard, fast cornering...there should be different functions for an ideal setup though. At parking lot speeds (not that this is really that necessary) it would be nice if the wheel moved just like they did in hard cornering...but you wouldn't want the wheels to turn the same way when you are driving on the highway, you'd want them to turn the opposite.

I believe someone else suggested this or something similar, but the few systems I can visualize are 1) Purely mechanical, 2) Hydraulic, 3) Electronic or a combination of them. Using a mechanical system will have its drawbacks, but it will probably be the longest lasting and cheapest from what we can work with, whether it is cable operated or similar to our current setup. I think that a very strong damper or even a spring/damper coupled with redesigned lower arms to allow wheel movement could work as a basic setup. The damper would prevent "jerky" movement of the suspension. It would have to allow positive and negative movement along its axis, so that it could accommodate both highway driving (where wheels would turn one way due to forces being applied, although w/ this setup I doubt it would move much at all) and hard cornering.

More on my thoughts later, I have to go to breakfast!

Ahh John. I remember being an Engineering student.

Anything that I do, unless it's purely for my personal interest and use, has to be something that would be feasible and useful to a reasonably wide audience. That means that it either has to use the existing system or bolt on to it. My goal is not to redevelope the 4WS system, but to figure out what, if anything, can be done with relative ease to improve the car's handling. I want to know how the car would handle without the active steering.

I think that the active rear steering is actually designed to induce understeer. Understeer is necessary for high speed stability but detrimental to low speed cornering. This is the compromise I was talking about earlier. Mazda has to design their car to be driven safety on the highway by the everyday driver. I want my car to be razor sharp so I'm willing to sacrifice some high speed stability for more neutral low speed handling. I think that I can make the rear of the car far more rigid and then compensate to reduce the oversteer that it would induce. The goal is to make the car neutral handling and stiff as a go cart.

onlytrueromeo wrote:I don't think a speed sensing lock would work to improve high speed stability, because then you would lose the benefits of the 4ws on hard, fast cornering...there should be different functions for an ideal setup though. At parking lot speeds (not that this is really that necessary) it would be nice if the wheel moved just like they did in hard cornering...but you wouldn't want the wheels to turn the same way when you are driving on the highway, you'd want them to turn the opposite.

wytbishop wrote:I think that the active rear steering is actually designed to induce understeer. Understeer is necessary for high speed stability but detrimental to low speed cornering. This is the compromise I was talking about earlier. Mazda has to design their car to be driven safety on the highway by the everyday driver. I want my car to be razor sharp so I'm willing to sacrifice some high speed stability for more neutral low speed handling. I think that I can make the rear of the car far more rigid and then compensate to reduce the oversteer that it would induce. The goal is to make the car neutral handling and stiff as a go cart.

Well to have the rear wheels turn opposite at high speeds would be baddddd. Even all 4WS systems go from opposite to parallel above 35. A stable rear would be much better than an active rear in that case. There's a good example:
http://www.youtube.com/watch?v=X7JgHoeeU-I" onclick="window.open(this.href);return false;

Actually, all cars are designed to understeer because it's a safer setup. Understeer and oversteer are circumstances that happen when limites are reached. Inducing understeer woulld make things worse, what 4WS does is actually allow a tighter turning circle before the front tire are overloaded and understeer occurs.

In our cars, we want to eliminate understeer as much as possible.