MZ3 vs. MS3 Eibach Pro kit measurements

Here are some photos showing the difference between the two types of springs. You can really see the difference between linear and progressive styles here, and the thickness between the two.

This first photo shows the MS3 spring on the right, MZ3 on the left.

Rear springs. Progressive (MS3) on the left, linear on the right (MZ3).

Thickness difference of the front springs. MS3 left, MZ3 right.

Here are the specs from the alignment that I had done on 4/5/08:

Front: Left

Camber:
Actual: -1.4*
Before: -1.5*

Toe:
Actual: .01*
Before: .16*


Front: Right

Camber:
Actual: -1.5*
Before: -1.4*

Toe:
Actual: .08*
Before: 2.6*


Rear: Left

Camber:
Actual: -1.5*
Before: -1.5*

Toe:
Actual: .21*
Before: .21*


Rear: Right

Camber:
Actual: -1.5*
Before: -1.6*

Toe:
Actual: .14*
Before: .13*

It hasn't been mentioned yet so I thought I'd chime in.

Neutral toe static settings will result in a toe out condition at roads speeds, thanks to tire drag, and bushing deflection. Even with FWD cars, most high performance tires will usually try to pull themselves outward at higher speeds. For a daily driver I would stay pretty close to the factory toe-in settings because they take that into account.

If all you care about is track performance, all bets are off. But if you don't like things like jumpy steering response, harsh ride, and excessive tire wear on your drive to work, stay with the stock settings.

I've been very consistent with my utter disdain for the stock dampers that come with the MS3. They suck, so I'm changing them. But alignment specs are very well considered - the MS3 has a different spec for the MZ3, for example. I'd like to think that was the result of extensive testing that OEM cars go through, but I'm willing at accept that I'm wrong, especially considering a spring change to MS3-specific Pro Kits. Or especially Mazdaspeed coilovers, which come with links to keep the rear end geometry correct. There's a reason for all that.

IIRC, the factory specified toe-in settings are very mild.

kwsmithphoto, what you say about static neutral toe settings is generally true on typical MacPherson strut-type rear suspensions on FWD cars, but is not nearly as much of an issue with the SLA "Control Blade" setup used on the Mazda3. The long trailing-arm-esque "control blade" keeps longitudinal deflection to a minimum. There is very little deflection resulting in toe changes at road speeds, and dynamic rear toe is way more dependent on suspension compression than anything else.

With the type of suspension used on these cars, there is actually a form of passive rear steering at work. When a rear wheel's suspension is under compression, the two upper lateral suspension links fight each others' movement through their respective arcs, and the longitudinal "control blade" link flexes, as the frontmost upper lateral link pulls the front of the wheel in (creating a toe-in effect on that corner - probably just enough to overcome the tire's slip angle). This effectively makes the car rotate better in the turn. Having too much static rear toe-in eggagerates this effect, and can actually make the car feel a bit squirrelly (in addition to the accelerated tire wear it can cause) and create unsettling rear-end "bumpsteer".

I think it's good to use a very small amount of static rear toe-in, as it can keep things from approaching a toe-out situation under heavy braking that can cause stability problems if coupled with an abrupt steering input, but I prefer to run closer to zero rear static toe. I've found it gives me the best combination of predictability/stability and rotation, while not adversely affecting tire wear.

This is after having tracked cars with the same type of rear suspension setup (I had 2 Ford Focuses before the MS3, which had the same basic type of rear suspension) for roughly 7 or 8 years, and from trying different alignment variations. For the street, I don't think there's anything wrong with the factory recommended settings, but one of the problems with that method is that the "factory recommended settings" vary from alignment machine to alignment machine, and you pretty much have to know the numbers you want going in (and can't always rely on what the machine tells you is the correct value).

Just my $.02