What Happens to Roll Centers When You Lower a Toyota MR2 Turbo | Project Toyota MR2

J.G.
Update by J.G. Pasterjak to the Toyota MR2 Turbo project car
Jan 15, 2021

Shoutout to the SCCA for adding a major national-level Solo event at our official test track, the Florida International Rally & Motorsport Park in Keystone Heights, Florida. Wait, what? It’s in three weeks? Man the wrenches!

Our original plan was to install and sort a new suspension in our 1991 Toyota MR2 Turbo before the mid-March event that usually kicks off the national Solo season, the annual Tire Rack Dixie Tour at South Georgia Motorsports Park. But just last week, the SCCA announced that Pro Solo 1 event—which brings the drag-race starts of Pro Solo to a single-course autocross format—at the FIRM for February 12-13. So our build schedule got compressed a bit. Hopefully we can have a somewhat sorted ride ready for XS-A competition in just a few weeks.

Any suspension upgrade should start with a basic look at what you're trying to accomplish and what problems you're trying to solve. With an SW20-chassis Toyota MR2, those problems are manifold, and the path to fixing them can be a little more complicated than just applying traditional methods.

We know we want to increase spring rate, adjust corner weights and control body motion. So a set of coil-over shocks is the easy solution there, right? Well, yes, but it’s only part of the equation.

For coil-overs, we chose the Variant 3 kit from KW Suspension. All V3 shocks are hand-built at the KW factory in Germany and are independently adjustable for both rebound and compression. The threaded shock bodies on the kit, which retails for $2289, also provide easy adjustment of ride heights for proper corner balance.

The KW V3 kit for the MR2 is designed to work with the OEM upper mounts as well, meaning you have fewer potential points of rattling and clunking than with some coil-overs. We’ll be using a bit more aggressive upper mounting solution for our application, and we’ll get into the details of that in future updates.

Once the KWs are in place, we’ll have better control over body roll due to the increased spring rates (good), better control of those springs through superior damping technology (good), independent adjustment of that damping for specific balance fine-tuning (good), and reduced ride height to lower the center of mass of the car (not necessarily good). Wait, what? Lowering is bad? Do explain, Mr. Car Magazine Smart Guy.

Actually, we’ll defer to an MR2 suspension expert, Alex Wilhelm of Wilhelm Raceworks, to explain the finer aspects of the SW20’s fairly sketchy suspension dynamics.

Let’s look at a few charts that Alex has produced through physical and virtual modeling of the MR2’s suspension.

Chart 1 shows the toe change in the rear of our MR2 when the suspension extends or compresses. With 2 inches of motion, the inside-rear wheel has toed out—turned into the corner—almost 0.2 inch. Meanwhile, the outside compressed wheel has toed in—also pointing toward the corner—a similar amount.

This toe change can produce a “roll understeer” effect where the rear tries too hard to follow the front—right up to the point where all the tires are overwhelmed, physics takes over, and the rear breaks loose unexpectedly.

Toyota addressed this weirdness in the 1993-and-up MR2s, as we can see in Chart 2. Dynamic toe change is still present in compression, but the extension toe change has been mitigated somewhat. Still, it’s not great.

Now let’s look at the front. In Chart 3, we can see that the front also suffers from some dynamic toe changes, with the inside wheel going to toe-in and the outside wheel going to toe-out during cornering. With just 2 inches of compression, total toe-out increases by nearly a quarter of an inch.

Next, let's take a look at what happens to these dynamic tendencies when the car is lowered 1.5 inches, which is the minimum amount that the KW kit affects the car. In Chart 4, that the toe-in tendency in the rear under compression has been mitigated by the slight lowering, but the toe-out under droop is almost unchanged.

Chart 5 shows similar tendencies. Dynamic toe change under compression has been slightly mitigated, while droop response is largely the same.

Can we conclude that lowering the car has little—or maybe a slightly positive—effect on dynamic toe? Well, not so fast, because there’s another secondary effect that changes everything.

In stock form, the roll centers of both suspensions—the points in space that the suspension allows the car to rotate around—sit a few inches above ground and a few inches below the center of mass for the front and rear ends.

When the car is lowered just 1.5 inches, the front roll center dives fully below ground, while the rear one moves down to just above ground level. This greatly increases the distance between the roll centers and the centers of mass. The additional distance creates more leverage, which greatly increases body roll. As a result, any minor gains in dynamic toe change made through lowering alone are instantly erased by increased body roll that aggravates the same dynamic toe situation.

Bottom line: We’re making a big improvement in potential by upgrading to a quality set of double-adjustable KW Variant 3 coil-overs, but to fully realize that upgrade, we’ll need to make some other changes to mitigate the negative second-order effects of lowering.

But that’s a discussion for the next update.

Like what you're reading? We rely on your financial support. For as little as $3, you can support Grassroots Motorsports by becoming a Patron today. 

Become a Patron!

Join Free Join our community to easily find more project updates.
Comments
View comments on the GRM forums
Matt B (fs)
Matt B (fs) UltraDork
1/13/21 11:33 a.m.

I'm going to go out on a limb here and guess that a following installment will involve the Wilhelm geometry kit.  It looks pretty nice. I wish someone made something as comprehensive for the Mk1, but all I can find are the T3 RCAs.

z31maniac
z31maniac MegaDork
1/13/21 11:46 a.m.

So part of the reason for the increased rates is to offset the increased leverage for the roll center change?

JG Pasterjak
JG Pasterjak Production/Art Director
1/13/21 1:08 p.m.
Matt B (fs) said:

I'm going to go out on a limb here and guess that a following installment will involve the Wilhelm geometry kit.  It looks pretty nice. I wish someone made something as comprehensive for the Mk1, but all I can find are the T3 RCAs.

Yeah, did we foreshadow that well enough?

JG Pasterjak
JG Pasterjak Production/Art Director
1/13/21 1:29 p.m.

In reply to z31maniac :

Yeah, if you;re just building a fun street car, you can get awa with lowering and just bumping spring rates a bit to compensate. For something you want to be able to properly tune, though, correcting the roll centers will be necessary. UPS guy is on the way with those solutions :)

Matt B (fs)
Matt B (fs) UltraDork
1/13/21 1:47 p.m.

In reply to JG Pasterjak :

As soon as I saw their name it was clear. smiley

I wasn't aware of that they did that level of modeling though. Cool stuff.  Looking forward to seeing the same graphs with their kit installed.

This kind of thing makes me think if I were a smarter man I'd switch to a Mk2, but I just can't seem to quit my doorstop.

SkinnyG (Forum Supporter)
SkinnyG (Forum Supporter) UberDork
1/13/21 7:58 p.m.

That looks a lot like SusProg3D!

Vajingo
Vajingo Reader
1/13/21 10:51 p.m.

Subbed. Keenly interested. 

Vigo (Forum Supporter)
Vigo (Forum Supporter) MegaDork
1/14/21 9:22 a.m.

These charts make me happy. I've been meaning to 'manually cycle' the rear suspension on one of my cars on the alignment rack to try and diagnose some weirdness. 

It's pretty funny that the net effect of the front and rear toe changes is the car doesn't want to go where you point it. How wonderful lol.

Typ85
Typ85 Reader
1/14/21 5:47 p.m.

Audi and VW's have the same issue with their strut suspension, the cure is installing "ball joint extenders" 

JG Pasterjak
JG Pasterjak Production/Art Director
1/15/21 9:31 p.m.
Typ85 said:

Audi and VW's have the same issue with their strut suspension, the cure is installing "ball joint extenders" 

Yep. We have a similar solution on teh way from Wilhelm. Hopefully his powder coater gets around to them soon so he can get them to us in time for the Pro Solo.

 

Alex_W
Alex_W New Reader
1/18/21 7:59 a.m.

In reply to Matt B (fs) :

At the risk of stealing some of the thunder from one of JG's future articles, the rest of my analysis is on my website: https://wilhelmraceworks.com/suspension-analysis

I can't comment on the toe curves without modeling it, but the camber curves are likely very similar on the MKI, and all of the same improvements will apply.

JG Pasterjak
JG Pasterjak Production/Art Director
1/18/21 12:51 p.m.

In reply to Alex_W :

At least go ahead and make the link work :)

https://wilhelmraceworks.com/suspension-analysis

But, yeah, we'll be leveraging a lot of the hard work Alex has already put in during our build and sharing a lot of his findings. I'm really excited to drive on the stuff soon. I'm hoping to get at least a quick test day in before the Pro Solo, but more likely than that I'll be experiencing everything for the first time at the February Pro. Which, honestly, is fine. I'm obviously really familiar with the FIRM and how the surface behaves, and every autocross there is basically a variation of the same theme, so of all the places to sample a new car that's probably the least risky for me. 

Ugh, now I have to go finish installing that rear swaybar, because I just realized like five minutes ago that the entire rear subframe has to come down a bit to get it in place. Y TOYOTA?!?!

Matt B (fs)
Matt B (fs) UltraDork
1/20/21 4:01 p.m.

In reply to Alex_W :

Pretty impressive improvements! I started to wonder why someone would choose the standard top mounts over the extreme until I saw what they did to the toe curves.

Question - I assume you want the tie-rod spacers to match the height of your RCAs to keep the geometry as close as possible to stock or did you make adjustments?  On a related note, I have to wonder if changing the relationship between the ball joints and tie-rods could actually help the front and rear-ends to act the say way rather than have mirrored toe curves.  Even stock it seems like they're fighting each other. You might not want to mess with the rear for stability under acceleration, but if there was a way to make the front toe behave more like the rear (toe-in under compression) without the wild swing we see when tie-rod spacers are absent.  This would lead to a toe-out condition in the front under straight line acceleration, but seems like a benefit under at least steady-state cornering.  I could be imagining this thought-exercise incorrectly though. Is there something I'm missing?

Matt B (fs)
Matt B (fs) UltraDork
2/1/21 8:46 a.m.

Aleeeeeeeeex, Y-U-No pander to the hare-brained ideas of someone who's mk2 suspension geometry correction research entails reading your site? cheeky

Alex_W
Alex_W New Reader
2/16/21 8:10 a.m.

In reply to Matt B (fs) :

Sorry, I didn't see this till you mentioned in in the comment on the other article!

You want the tie rod spacing to *mostly* match the RCA, but you can certainly play with this a little to change the toe curve.  I somewhat do this already, with the rear tie rod bracket for example.  By rotating the bracket in and up the tie rod is shortened, but the pivot is also raised by maybe 1/4".  The front is easier, you can buy 5/8 dia bump steer spacers for less than a dollar each and make adjustments that way.

They may be pointing opposite directions, but they are both working towards the same goal (roll understeer).  On the front toe out in compression and toe in on rebound causes the wheels to point towards the outside of the corner, so the car turns a little bit less than you would expect for the amount of steering input applied.  On the rear toe in on compression and toe out on rebound points the wheel towards the inside of the corner, so the rear doesn't try so hard to pass you on the outside.  

On my car I have always just tried to set the front to minimze bump steer and have been pretty happy with the result.  The rear is tougher to fine tune, but less toe change vs stock has certainly been a big improvement in my experience.  On your MKI you have the advantage that Toyota gave you a vertical tie rod stud in the rear, so you can replace it with a stud and rod end just like the front and dial in bump steer exactly as you want.  I'm having to create a complete custom knuckle for the MKII to accomplish that (and some other stuff of course).

Oh, and on the top mounts, the main reason to choose standard over extreme is if you don't want to drill holes in the chassis to access a shock adjuster, or don't want to / aren't able to potentially modify the coilovers to get rid of excess camber (slot the brackets).  The extreme top mounts do effect the toe curve a bit, but as I mentioned above that's easy to adjust out provided you are running them with my geometry kit, and even if you don't bother, the resulting toe curve is still not unreasonable.

Matt B (fs)
Matt B (fs) UltraDork
2/18/21 10:10 a.m.

In reply to Alex_W :

No worries - it's not like you have an engineering and fabrication business to run or anything! 

First of all, thanks for the explanation.

Gotcha on the extreme top mounts. I didn't realize there were chassis mods needed for installation. Sounds like the benefits outweigh the negative effects on the toe curve when used with the geometry kit.

As far as the opposite toe curves - to your point I imagine the effect of the front toe-out under compression is mitigated because the wheels are already turned many more degrees than whatever the bump-steer is doing.  If I can get a setup like this cobbled together for my Mk1 I'll probably experiment anyway.  It would be interesting to see what toe-in under compression does to tire temps and turn-in. 

Tom1200
Tom1200 SuperDork
2/18/21 11:05 a.m.

Cool article; I once had an MR2 and while it was fun I didn't particularly like the way the car turned in, now I know why.

Matt B (fs)
Matt B (fs) UltraDork
2/18/21 4:09 p.m.

In reply to Tom1200 :

Curious - was it a Mk2 like the one in the article?  Never had that complaint with my Mk1.

Tom1200
Tom1200 SuperDork
2/18/21 7:30 p.m.

In reply to Matt B (fs) :

Mine was a Mk1.  I can't remember which one of the posts explained the difference between the Mk2 and the Mk1 and after reading it it clicked as to why I prefer Miatas.

 

Matt B (fs)
Matt B (fs) UltraDork
2/19/21 8:07 a.m.

Yeah I could see that. The Miata has an objectively better double wishbone setup, plus they're lighter.  I've just never been excited about them like the MR2 for whatever reason.  I blame it on being a child of the eighties.

dean1484
dean1484 GRM+ Memberand MegaDork
2/19/21 8:21 a.m.

We raced MR2s and the best suspension setup we found was lowering springs and bilstein or koni struts.  Never really cared about corner weight. The other MUCH more important suspension mod was to reinforce the trailing rods in the suspension. We would box them and stitch weld it using either two pieces of angle or a piece of square tube steel with one side cut out so it slip over the rod. We found that these rods bent (and would stay bent) under continual heavy loads constantly changing the suspension geometry. They were also the first failure point with any contact of things like curbs or even the smallest wheel to wheel or wheel to wall contact. Once fixed it totally changes the cars suspension. 

Matt B (fs)
Matt B (fs) UltraDork
2/19/21 1:48 p.m.

In reply to dean1484 :

Interesting.  I assume you discovered this by the perma-bent rods?  What model year(s)?

Pete. (l33t FS)
Pete. (l33t FS) MegaDork
2/19/21 2:10 p.m.
Typ85 said:

Audi and VW's have the same issue with their strut suspension, the cure is installing "ball joint extenders" 

Somehow, the companies that used to make these no longer make them.

 

Curious, isn't it?

Stefan (Forum Supporter)
Stefan (Forum Supporter) MegaDork
2/19/21 2:30 p.m.
Pete. (l33t FS) said:
Typ85 said:

Audi and VW's have the same issue with their strut suspension, the cure is installing "ball joint extenders" 

Somehow, the companies that used to make these no longer make them.

 

Curious, isn't it?

because the demand went away as the market shrank to those that have them and use them and those that no longer own/drive those cars.

Pete. (l33t FS)
Pete. (l33t FS) MegaDork
2/19/21 4:00 p.m.
Stefan (Forum Supporter) said:
Pete. (l33t FS) said:
Typ85 said:

Audi and VW's have the same issue with their strut suspension, the cure is installing "ball joint extenders" 

Somehow, the companies that used to make these no longer make them.

 

Curious, isn't it?

because the demand went away as the market shrank to those that have them and use them and those that no longer own/drive those cars.

They also had a tendency to break off.

 

The fascinating part, is that when they would break, people complained... that they couldn't buy replacements.

Benswen
Benswen Reader
2/19/21 5:15 p.m.
Tom1200 said:

In reply to Matt B (fs) :

Mine was a Mk1.  I can't remember which one of the posts explained the difference between the Mk2 and the Mk1 and after reading it it clicked as to why I prefer Miatas.

RE: Miatas vs. MR2's.  I've had long term experience in a '97 Miata and  a '91 Mr2 non-turbo.  The Miata is a better car, the MR2 is a cooler car.  Miata was better on the road, MR2 is better on the track.   I sold the Miata, will never sell the MR2.

 

JG Pasterjak
JG Pasterjak Production/Art Director
2/19/21 9:54 p.m.
Benswen said:
  The Miata is a better car, the MR2 is a cooler car. 

 

Nailed it.

BrianC72gt (Forum Supporter)
BrianC72gt (Forum Supporter) New Reader
4/2/21 11:33 p.m.

I don't suppose there is space for a custom cross-member to raise the inner mounting points and get the roll center out of the basement on a lowered setup?  Chassis rails in the way?  Who am I kidding, I'm clueless.  I'm grateful for the work you and Wilhelm are doing.

Sponsored by

GRM Ad Dept

Our Preferred Partners
8NvYwzyhMldKZ16PFRJ96vd3CLGVvRhrEb40atyxQv9HsAdeplE2r67YPtDajDR6