Cross tower brace benefits

[p66]

Is our OEM cross tower brace effective? Would like your opinions on the benefits of a brace such as the Mason Engineering brace in the link below. I have the steel version of said brace but removed it because the paint finish was deteriorating from engine heat and was going to get it repainted. Then wondered if it was worth it.

http://www.masonengineering.net/images/E82front.jpg

Thanks

[redux]

Modern chassis systems are a lot more effective than the old wet noodles produced prior to usage FEA during a vehicles development cycle. So I'd like to that say the OEM piece is more than adequate for anything short of a running low tread wear R-Compound (I.E. Hoosiers) or slicks. Of course, I base that estimation on absolutely nothing; but I have put plenty of track time on the stock brace with little issues of chassis flex.

[MDORPHN]

I ran Mason Engineering strut tower brace on my 99 M Coupe. An aftermarket piece is not needed on the 1M.

Neil

[p66]

Thanks. I had this brace on my 08 135i and carried it over with the 1M. It is fully depreciated and I could save 6 lbs. That should improve my lap times by .00001 second. Yesss!

[Pete_vB]

So I agree with leaving it off, lots of other places to stiffen a car up first and I'm not sure I like the design. However I do find what they did to the M235i racing to stiffen up the shock towers interesting:

Two braces, one that triangulates the strut towers to the frame and then a second carbon piece coming around the front like the new M3/4. Looks impressively stiff.

This is in a fully caged race car, so much stiffer than our cars to begin with. The presumably sticky rubber and stiff suspension put a premium on chassis rigidity, and this bracing looks clearly designed to work with the cage. That said, it suggests that for a track car a properly designed brace would probably be an advantage- BMW didn't go to all that trouble to simply look cool.

[MDORPHN]

Wow, that's some beautiful engineering!

Neil

[dmw16]

That looks so sweet, but I'm not really sure what they are going for? Maybe it behaves almost like a torsion bar? Either way, it looks bad a$$. Too bad they stuck an automatic transmission in it.

[MDORPHN]

Originally Posted by dmw16

That looks so sweet, but I'm not really sure what they are going for? Maybe it behaves almost like a torsion bar?

Not unless there's an outrageous amount of chassis flex!

I imagine it simply further stiffens things.

Neil

[Nugget]

I have the Active Autowerke one, didn't notice any difference but it looks good.

[Pete_vB]

Originally Posted by MDORPHN

Not unless there's an outrageous amount of chassis flex!

I imagine it simply further stiffens things.

Neil

Yep. A regular shock tower brace is designed to maintain the distance between the strut tops. This helps with camber compliance- the top of the outside strut doesn't deflect sideways as much under load, which helps reduce camber loss- a good thing. However a traditional strut brace does nothing to reduce motion forward/ aft, nor does it prevent both struts from deflecting sideways.

This setup does both of these, significantly improving on a regular strut brace.

The reduction in forward/ backward motion is probably the bigger benefit, because this motion occurs when the entire chassis twists. The front U effectively triangulates the strut tops, taking this forward load and tying it to the front of the car (note to bolts on the nose) then back to the other shock tower. Now when the car tries to twist in torsion and the strut tops try to move the U tries to pull the entire nose of the car sideways, and by quite a lot. Including all of this additional nose structure into the equation must significantly increase torsional stiffness, while the traditional strut brace continues to maintain the distance between the shock tops and triangulate the structure.

It's a slick design first seen on the new M3. Either this shares a bunch of parts with that car, or they have invested very heavily in production tooling for a racing variant. If it's the latter is makes me suspicious that we could be looking at M2 parts... but that's conjecture.

[scotth944]

Any idea what magnitude the deflection would be in the strut tower when the car is at it's limits?
Given the strut support is mounted in rubber, what would be the advantage of additional reinforcing unless you had already replaced the mounts with camber plates using hard mounted spherical bearings?

[Pete_vB]

Originally Posted by scotth944

Any idea what magnitude the deflection would be in the strut tower when the car is at it's limits?
Given the strut support is mounted in rubber, what would be the advantage of additional reinforcing unless you had already replaced the mounts with camber plates using hard mounted spherical bearings?

I don't have data for our cars, but I can give you a wag:

Something between .1 to .15 degrees per KN (or .45 to .66 degree per thousand pounds of side load) is a decent guess for camber compliance based on other modern sports cars. That the total of everything- chassis flex, bushing deflection, springiness of the suspension arms, bearing play, etc. Applied to our cars these values would mean a wheel would "lose" ~.8 degrees of camber while cornering, which sounds bad until you realize that the average car can be 4 times higher:

Clearly bushings are going to be much of this, I've often seen roughly half half, with the various metal bits adding up to the other half. If the shock tower was responsible for 25% of that remaining bit it'd need to deflect ~.7 mm at max cornering, so you're talking very small deflections.

Now consider chassis twist- if we're 20,000 Nm per degree then over a bump we could see .5 degrees of chassis twist on an impact. This in turn probably moves the shock tops ~3mm forward/ aft relative to one another, so that's what the U brace is trying to reduce, probably quite effectively.

I do think there is a big advantage to reducing these chassis movements even if you have rubber shock tops. It cuts total deflection, but importantly it also increases resonant frequency and hence grip as well as feel/ feedback. Consider that when just the wheel and shock move you have just those components as the sprung mass, and given that they are stiff they resonate at a high frequency, which makes it easy for them to react quickly to the road surface, and the rubber also probably has some internal damping.

When you begin twisting and moving the car body, on the other hand, you suddenly have very large masses that you're moving around, and these react and settle much more slowly both because they have less damping and because they "bounce" at much lower frequencies. Go over a bump in a flexy car and the whole body shakes for quite a noticeable amount of time before it calms down. Hence reducing chassis flex is still a good thing, even with rubber bushings.

[scotth944]

Thanks for the details so I can appreciate the dynamics involved Pete.

I think one detail to also consider is how much camber has been added in by the suspension geometry design as the wheel is loaded down so .8 might not be as dramatic in comparison, unfortunately I don't have details on how much to expect but I bet you do?

I imagine a full roll cage could also help considerably with the chassis flex between the suspension mounting point, not that most of us are willing to go down that road with these cars.

[Pete_vB]

Kinematic camber gain... Again, not sure on these cars, but it's often in the range of .015 to .03 degrees per mm on modern cars, usually towards the higher end of this range at the back and the low end at the front. So if you're compressing the outside suspension 25mm in a corner you gain nearly as much camber as you lost due to compliance.

Then you take static camber into account, maybe -1.7 degrees, and lean angle of the car, maybe 1.8 degrees, and you get -1.7 (static) + .8 (camber loss, compliance) -.7 (camber gain, kinematic) + 1.8 (lean angle) = +.2 degree camber tire camber to the road, nearly flat for maximum cornering grip but slightly less to improve straight line traction. Usually the front will be noticeably more angle to the road for increasing stability in the turns.

Run these numbers with Hoosiers, where the higher Gs create more vehicle lean and camber compliance loss, and you see why it's important to add more camber with sticky tires.

[scotth944]

Sorry for hijacking the thread but two more questions:
1- Is the only way to evaluate you are running the appropriate static camber for the tire setup your using is to measure the tire temp between inside-mid-outside zones?
2- Is there any metric available to evaluate a specific tire's potential lateral traction capabilities in an ideal condition?

[Pete_vB]

Originally Posted by scotth944

1- Is the only way to evaluate you are running the appropriate static camber for the tire setup your using is to measure the tire temp between inside-mid-outside zones?

Practically speaking I don't think so. Even tire temperature isn't ideal; it's a guide, but it's really the clock that decides. If you have nearly all the data for a car you can also use a vehicle simulator, but in practice only manufactures really do that.

Originally Posted by scotth944

2- Is there any metric available to evaluate a specific tire's potential lateral traction capabilities in an ideal condition?

For mortals, you mean a cheater method, like durometer? Don't think so. If you can get data from one of these, then yes:

But you can't, or it's very difficult. These machines come in various sizes/ flavors up to stories tall, and manufactures use data from them to design their suspensions, etc. This data is extremely closely held, and you'll rarely find any current applicable data in the wild. You can however often find generic models and perhaps tweak it with specific testing, such as: http://www.optimumg.com/software/optimumtire/

What are you trying to do, chose between different tires, or?

[scotth944]

Wow, thanks for the follow up.

My main goal is to understand what the fundamentals are that give the 1M it's handling personality it has and what directions I might be able to go to modify it. Thanks again for all the details.