Mazda 2 Front Suspension Geometry

[Zach@DDMWorks]

OK so here is the main thing I was working on this holiday season. I took measurements of the front suspension and put them into CAD so I could put the suspension through it's paces and see how the camber curves are. I also changed a few things to see what I would get out. So this is all based on my car which in the front is 23.375" from the ground to the top of the fender. I also have -3 degrees of camber. Here is the plot and I will explain it below.

OK, so the Y axis is the suspension travel. 0 is my ride height and I put the suspension through +2 inches of compression and -2 inches of droop. For those changes in movement I have the camber change plotted. Not absolute camber but the camber change. The first data set is the stock data. As you can see when in compression the camber gain starts to drop off as it compressed, but even with 2" of compression it never goes negative. Also you can see in droop it is pretty linear which is good because it should help to keep the inside flat to help with wheelspin. OK so the next couple of data sets I changes a number of things. First I changed the upper mount with the top hat inwards 0.5". Basically what camber plates will do. This helped to get more camber gain even though it was a small amount. The potential down side is that the scrub radius changed from about 0.25" to 0.04" which will give less steering feel. The next set was if the lower control arm was 0.5" longer. This gave more camber gain, but it also had the same effect on the scrub radius. The next set after that is I moved the lower control arm ball joint down 0.5" like a roll center adjuster would do. This gave the most additional camber gain, but again it wasn't anything huge. This gave a slight increase in scrub radius as well to 0.32". I then did all 3 together for the last set of data. This gave a noticeable amount of additional camber gain. Scrub radius was down to 0.13".

So the way I see all of this is that they all make improvements, but nothing that huge. As an example my car, just in roll, my suspension should compress and droop about 1" so having all 3 done I would have 0.15 degrees more camber in roll. So it would be better, but nothing major really especially for the work it would take. The main thing that this did show me is that the car does have some camber gain to it even with a bunch of compression. Right now when my car compresses about 1" I am on the bump stops. Well any extra roll from there just lifts the inside and not compress the loaded corner since it is on the bump stop. So the car rolls more, but the loaded tire doesn't compress which doesn't gain camber. So at least I am going to swap out to my shorter bumps stops and possibly even cut them a bit to take advantage of the extra travel to gain a little more camber on that loaded tire. After the first track day of this year I am also going to be upping the spring rates even more. Thinking 700F 600R to further reduce roll and keep the tires happy......you know because racecar.

Any questions let me know. I just wanted to post this up for the other racers out there like Andres, Tony, and Jeff.

Zach

 

[Ar5h3d]

I'll be honest, its this kind of thing that got me to going for a 2. Would it be okay to ask for a copy of the CAD? i'm kind of curious to see how it translates to my current setup. Or maybe just the measurements you took?

 

[eyeballs]

Good info, and easy to interpret graphic. It's kind a bummer the camber gains are all so minimal. Would it be correct to assume extra travel due to bumpstop compression is insignificant? I ask because I don't think I've ever seen it included in suspension calculations.

 

[Zach@DDMWorks]

yea, there would be some travel there, but not much depending on how hard they are.

When I get home tonight if I remember I will post up the dimensions I used to model it up.

As a point of reference the Miata camber curves are easy to find online. The front is about -2 degrees for 2" of compression and the rear is -1.5 degrees for 2" of compression.

 

[Ar5h3d]

^actually isn't the lack of camber change better considering nobody makes rollcenter correctors for the 2, So lowering the car isn't going to put you in a weird place with dynamic camber? Then just brute force compensate with static camber?

 

[vdsracing]

^actually isn't the lack of camber change better considering nobody makes rollcenter correctors for the 2, So lowering the car isn't going to put you in a weird place with dynamic camber? Then just brute force compensate with static camber?

Yes this is true. This is maybe why the japanese mz2 racecars for super taikyu are lowered so much.

 

[Zach@DDMWorks]

The more gain of negative camber you can get as the suspension compresses the better. Well there is a limit, but with a mcpherson suspension your never really going to get there. Right now with a Mazda 2 you basically just compensate with a lot of static camber like you mention because the downsides don't really show up. The downsides being putting the power down and braking. Well, we don't have much power so that isn't an issue, and since we don't have much power we don't need to brake much either.

You also have to remember that the number I posted are all changed based on whatever your static alignment is. If I lower my car it would gain negative camber, but I could always re align it back to what I had. That is what I did in my model as well. I changed the camber back to -3 degrees after making the suspension adjustments.

Someone has mentioned roll center to me on facebook and I am going to be getting information as well since it is easy, I actually already updated my model to include it. The data is easy. I am just trying to come up with the best way to explain what it means accurately without getting too technical with equations and that is going to take some time. having the roll center heights is one thing, but actually knowing what it means and how it effects things is a different story and I want to explain that now just give the data.

What I will say at this point is that for a racecar low CG is always a good thing which is why the super taikyu cars are low. On a street car there are issues which arrise, but I will get into those later once I work on the best way to explain it all so everyone can understand without getting too technical.

 

[Ar5h3d]

Oh no, technical is fine with me. I honestly want the measurements so I can try modelling it myself as measuring myself isn't really possible in my case. In anycase the increased roll moment from lowering isn't that much of a deal anyway as roll has a relatively small effect on weight transfer compared to the effect of the lowered CoG. What roll really does affect though is the position of the suspension arms but since the change is relatively small in this case the advantages do outweight the disadvantages. On a road car the main thing that would stop you is bottoming out and the springrate needed to prevent that from happening is too high to be tolerable on normal roads.

 

[Zach@DDMWorks]

What roll really does affect though is the position of the suspension arms but since the change is relatively small in this case the advantages do outweight the disadvantages.

The changes are bigger than you think for a car more set up for normal roads. That is one of the things which I am going to make a point about in my write up.

The fact that we have small amount of camber gain is the issue. When you say that the change is relatively small what are you referencing to, as to what is changing? Roll? Just want to make sure I am reading the right thing.

 

[okvol]

With my autocross experience, this explains the "bite" you get with this car when cornering hard. You can really push hard, way past your normal expectations, until the tires finally give up. I've autocrossed a Scion Tc and a 1Gen Neon too, and neither had the bite of this car.

 

[Ar5h3d]

I'm trying to affirm that for dedicated track cars it's understandable that low as possible is indeed still better. From what i've observed on my lowered 2, even though there is some camber loss as the car corners, i've been able to counter it with around 35psi of tire pressure combined with the 1 deg negative static camber. I'm currently guessing that around 2.5 degrees static would be enough that i can go back to stock tire pressures.

Yeah, i kind of meant that roll itself doesn't seem to affect weight transfer all that much compared to the height of the CoG. Also, is it right that lowering the car in general increases roll because the roll center is lowered a lot more compared to the CoG?

 

[Zach@DDMWorks]

Ok so I am getting back to this. To first reply to your questions. yea you need a good bit of static camber. I have -3 and it still isn't enough for my R888's. The roll center height has a bigger effect then you would think when it comes to handling.

So time to list the dimensions I used. Remember for the height numbers this is based on my car which is 23.375 from the ground to the top of the fender.

1. The LCA mounts to the chassis are 29.125" apart and 5.56" from the ground.
2. The top hats for the struts are 43.56" across and 29.375" from the ground.
3. The LCA is 13" long.
4. The center of the wheel was 10.875" from the ground.
5. I modeled it with 3 degrees of negative camber.
6. The wheel mounting face was 3.75" away from the axis from the lower ball joint to the top hat.
7. the height of the wheel mounting face was 5.125" up from the lower control arm ball joint.

That is enough to get it all modeled up.

RC information coming soon. Going to type it up tonight then read it again tomorrow before I post it. Time to play Forza.

 

[jasyatz]

The one thing you have to keep in mind, and it's a trade off for sure, is that the more static negative ca,her you put into the car, the more it reduces the a certain contact patch. You'd think that wheelspin wouldn't be an issue on a 100hp car. Whence you get the car cracking on course though, hard turn in and getting on the power early lights up the inside front whence you get above -2.2 static camber.

I found the sweet spot right at -2.2 but I'm also running 700lb front springs and have the car lowered to 23.25" up front. There's not much camber change under compression because there's not much compression to begin with.

 

[Zach@DDMWorks]

True, but even in 2nd I have yet to have wheelspin issues unless the car is sideways from too much lift off on turn in, which you are going slow at that point anyway. Also for track stuff I am mainly in 3rd or higher. I get into 2nd for a few turns, but not many. Also with the smoother lines you do on track I don't think that you get that harder and quicker change which really lightens the front. We will see this first trackday though because I am planning on running RE71's this yreat since I am doing time attack events.

 

[Zach@DDMWorks]

Firstly I am going to say that this whole conversation is all based on the geometric roll center. Now a days force based roll centers are used since they are more accurate. The geometric roll center is pretty much accurate for just as the car is sitting there and how it changes with ride height is accurate as well. Forces based are usually only done by computers because of the complexity. Also I am going to try to avoid just typing a bunch of equations and the numbers involved. I will mention some numbers in a casual method, but I want to keep this as casual as I can.
So here is the data for the front suspension.

Now time to put it all together for what it all means. Firstly if you find this interesting and want to know more read How to make your car handle by Fred Puhn. Well written and easy to read. I have read the whole thingmore than once.

OK so here is where it all comes together. When you go into a turn, weight transfers from the inside to the outside tires. The less weight which transfers the more grip you will have because the tires are being more efficient. When you adjust your suspension with spring rates and damper changes you basically are changing how the weight shifts on all 4 tires which adjust the handling at that given moment in time.

Firstly physics is physics, so in the end for a certain height of CG and track width there has to be a certain total amount of weight transfer. Nothing can change this. Basically imagine a metal box with 4 small pins on each corner. As that sees a side load from cornering, there is a certain amount of weight transfer. This is why the lower the car, the lower the CG the less overall weight transfer. Same with going wider it leads to less weight transfer.

Now when you look at weight transfer there are 3 parts which add up to the total weight transfer for a given end of a car. Firstly there is the weight transfer caused by the roll of the car. This is the main part which looks at the track width and the distance between the CG and the roll center. The second part is the weight transfer caused by the height of the roll center compared to the ground. This uses the track width and roll center height. The third part is from the CG of the unsprung mass. I am going to ignore explaining this third part because it is pretty much constant and has the least effect on weight transfer.

So this is where it gets interesting. As you lower the car the CG goes down, but the roll center lowers more. Roughly 0.5 of lowering lowers the RC 1.3. The distance between the CG and roll center increases 0.8, which means there will be more weight transfer caused by the roll of the car since the length of the lever arm causing the car to roll (CG->RC) increased that 0.8. Now, when you look at the weight transfer caused by the roll center height, the roll center is lower so there is less weight transfer caused by the roll center height. The roll center compared to the ground dropped 1.3 so the lever arm is shorter leading to less weight transfer. Now going back to what I said easier, the CG is lower so the overall weight transfer will be less which is great, but more of the weight transfer will be from roll instead of roll center height. That is where the extra roll comes from.

Now to put some numbers to it, my front suspension uses 525lbs springs and no sway bars. If I lower my car 0.5 in a 1g turn I will have ~8lbs less weight transfer overall because of the lower CG which is good, but there is an increase of ~20lbs of weight transfer due to roll. So, even though overall there is less weight transfer there is more roll. Now with my car that equates to ~1mm (0.040) more compression on the outside suspension and the same amount of extra droop on the inside. With how wide the car is that is 0.08 degrees more roll. So I would need that amount more camber. Now lets keep in mind that this is a relatively small amount and when you look at changes this small there are a lot of other factors like bushing deflection and chassis flex which also have an effect in both + and ways. Now if you have a soft suspension like, lowering springs, this will cause a lot more body roll, but with just lowering springs your main problem is going to be hitting the bumps stops so the increase in roll is going to be a more minor issue.

So in the end, lowering the car will cause less weight transfer because of the lower CG, but there will be more roll of the chassis which can cause the tires to be off camber with the road more so you just need to adjust your alignment accordingly. Also unless you are out to win money and are checking your tires temperatures with a probe pyrometer, you never really know if your camber settings are optimal anyway. Oh yea I will be getting a good pyrometer probably before my first trackday this year.

Feel free to ask questions, I tried my best to make it easy to understand. In the end the effect was more than I originally though. Mainly because the roll center moved more than I thought.

Zach

 

[eyeballs]

I think I'm going to have to get that book.

 

[Ar5h3d]

Any luck asking for that geometry? I'd like to check the values on my car too . Don't really have the tools to measure things myself, as our apartment parking is sloped.

 

[Zach@DDMWorks]

Post 12 I give the numbers for what all I measured.