Understeer Dilemma

Suspension Mods so far (05 2.5GT daily driver):

Koni Adjustable Shocks

Rallitek Sport Springs

Rallitek 22mm Front Sway Bar

Rallitek 19mm Rear Sway Bar

Rallitek Front HD Endlinks

AVO Steering Rack Bushings

Goodyear Eagle GT tires (crap for autocross, I know)

Stock Alignment Settings

The problem:

I have autocrossed a few times this past year and have noticed a lot of understeer with this current setup. At the last race of the year, I tried to play around with the Koni settings and tire pressures some more to tune out this understeer. I ended up putting the rear shocks to almost full hard and turning the front struts down to only 1/2 turn from full soft. I was able to take about 2 seconds off of my lap time by doing this, but I still had a lot of understeer (and the front tires showed it!!!). Before I went out on my first run, I put chalk on my tires to see how much they were rolling in turns as well. At 45 psi, they still roll a good bit! I have been doing a lot of reading on this forum and also referencing some of the motorsport engineering books from school since then and have narrowed down some fixes to a few ideas.

1. Go to a larger sway bar - I am having problems with this change because I feel that I would be taking away some of the effectiveness of the independent suspension. Having too stiff of a sway bar acts almost as a solid axle suspension. Also, at this point in time, Perrin has stopped making their rear sway bars for our cars. So that leaves me with companies that all make their rear sways smaller than their fronts.

2. Put the stock front sway bar back on and put in poly bushings - I am leaning more towards this route since I can do this for the cost of only the $25 bushings. If I don't like the feeling, I can always go back to the Rallitek front bar. The down side is I will be getting more roll in the turns. I can always try and correct this by making the front struts more stiff, but I don't know how much roll I can tune out by this.

3. Leave the sways the way they are and change my alignment settings - I have been reading that changing our front camber to -1.5 degrees does wonders for the track. Seeing that my front tires have been wearing on the outsides due to the understeer, I would really like to give this option a try. I could also get the alignment changed and swap out front sways at the same time. I am not sure how much just the alignment settings will do alone.

4. Get better tires - I know that better tires are the best suspension mod on the market and that they will improve lap times tremendously. My only problem with this is if I don't fix the understeer before I get new tires, I will just be fighting this issue later on while wearing my new tires badly. Tires are definetely in the future, I just would like to narrow down the problem first.

If any of you experienced autocrossers have any other ideas or input, please let me know.

Drop your front sway back to stock, add an adjustable rear (max of 22mm) but min of 20mm. You're overdriving the rear suspension with your current settings at the front. Set your tire pressures with 2 psi bias, and increase/decrease pressures in 2 psi increments until you get the feeling you want. Also, don't underestimate the impact of good tires.

Set your alignment at -1.25 rear, -.75 front camber (about the most you'll get with the stock setup), 0 toe F/R, 0 cross-camber. To get beyond these settings, you'll need to add lower adjustable control arms at the rear and aftermarket camber bolts at the front, unless you go with adjustable camber plates at the top.

If you're not using the offset LCA bushings, I'd do that as it gives you more positive caster and the car will literally pirouette

Drop your front sway back to stock, add an adjustable rear (max of 22mm) but min of 20mm. You're overdriving the rear suspension with your current settings at the front. Set your tire pressures with 2 psi bias, and increase/decrease pressures in 2 psi increments until you get the feeling you want. Also, don't underestimate the impact of good tires.

 

Set your alignment at -1.25 rear, -.75 front camber (about the most you'll get with the stock setup), 0 toe F/R, 0 cross-camber. To get beyond these settings, you'll need to add lower adjustable control arms at the rear and aftermarket camber bolts at the front, unless you go with adjustable camber plates at the top.

 

If you're not using the offset LCA bushings, I'd do that as it gives you more positive caster and the car will literally pirouette

Awesome reply!!! Thank you very much for your input! This sounds like a solid solution and sort of what I was leaning towards. I have been looking into the LCA bushings as a DIY project, but it doesn't look that way. I have done all of the work on all of my cars before, but without a bushing press, this mod does not look like a job for the at home mechanic. I hate people that I don't know working on my car.... I have seen a lot of friends get screwed by crappy technicians.

SBT is mostly corect (sorry man), it's a combination of 1 through 4 really.

In this order, follow SBT's suggestion for now except please reverse that camber suggestion using aftermarket camber bolts and experiment with 2-4 psi more in front on the pressures.

Next would be the Hotchkis/Cobb sways or Whiteline. Front on min and rear on max. "effectiveness of the independent suspension" ...not really, if Auto-X is your thing. You need larger bars.

Get better tires! The alignment specs I suggest create NO additional wear. Do it in confidence.

Simple...done, now go race the crap out of it!!

Step 1: Get an alignment. It is ridiculous to think that the car will behave well if you don't have a proper alignment. Small changes make HUGE differences.

Step 2: Better tires. You can't auto-x on all seasons.

Step 3: Get bigger bars front and rear. I run the hotchkis 25mmF/22mmR combo from FBP. It is a common misconception that big rear bars reduce understeer. Running a big rear bar simply promotes oversteer, which is entirely the wrong way to go about making a car handle well.

Step 4: Practice.

Fast street cars will remain neutral and understeer at the very limit.

I agree with both Drew888 and BAC5.2 except for your selections on sway bars. You both talk about the Hotchkis 25mmF/22mmR sways. These bars are made out of a hollow tube where the Rallitek's and Whiteline's are made from solid steel. Rallitek offers the 22mmF/19mmR setup, even though these seem to be much smaller than the Hotchkis, they are stronger.

For example:

Read the following document produced by Whiteline http://www.whiteline.com.au/docs/bulletins/Hollow%20vs%20Solid%20Swaybar.pdf. There are two formulas inside to follow when calculating the shear stress of these two different types of bars. If we assume that the inner diameter of the Hotchkis bar is 4mm less than the outer diameter, we end up with the following results.

EDIT: (Removed error in tables, see posts below)

The only way the Hotchkis bars would be stronger than in Ralliteks is if the tube thickness was 5mm on the rear bar and 6mm on the front bar. These are very thick values and be harder to manufacture. The tubular bars have to be bent with mandrel bends to be successful (if you don't know about mandrel bends, look it up, it is pretty neat).

These values are not the spring rates, but the spring rates are directly proportional to the shear stress. Since the Hotchkis rear bars are adjustable, they may see a small increase in the spring rate (may put it equal with the Rallitek rear). If you guys want, I can post the formulas to calculate the spring rates of the sway bars.

The tubular bars will deform or bend easier than the steel bars. The one advantage the tubular bars have over the solid versions are the reduction is weight. Even though this sounds great, the amount of performance seen in street/autocross applications is minimal.

Now, if I were to upgrade bars, my only option at this point would be the Whiteline bars, they are a little larger than the Ralliteks (1mm each). For the amount of money I would have to pay to get the minimal performance gain would be a waste I think.

I know a lot of you will disagree with what I have said in the post, but in the end it all comes down to personal preference. When I started looking into sways, Perrin already had stopped producing their rear bars. To me, my only other two choices were Rallitek and Whiteline. Rallitek gave me a killer deal on the sways and it was very hard to pass up.

I agree with both Drew888 and BAC5.2 except for your selections on sway bars. You both talk about the Hotchkis 25mmF/22mmR sways. These bars are made out of a hollow tube where the Rallitek's and Whiteline's are made from solid steel. Rallitek offers the 22mmF/19mmR setup, even though these seem to be much smaller than the Hotchkis, they are stronger.

 

For example:

Read the following document produced by Whiteline http://www.whiteline.com.au/docs/bulletins/Hollow%20vs%20Solid%20Swaybar.pdf. There are two formulas inside to follow when calculating the shear stress of these two different types of bars. If we assume that the inner diameter of the Hotchkis bar is 4mm less than the outer diameter, we end up with the following results.

 

Hotchkis Max Shear Strength:

Front - 19256 MPa

Rear - 12692 MPa

 

Rallitek Max Shear Strength:

Front - 22998 MPa

Rear - 12794 MPa

 

The only way the Hotchkis bars would be stronger than in Ralliteks is if the tube thickness was 5mm on the rear bar and 6mm on the front bar. These are very thick values and be harder to manufacture. The tubular bars have to be bent with mandrel bends to be successful (if you don't know about mandrel bends, look it up, it is pretty neat).

These values are not the spring rates, but the spring rates are directly proportional to the shear strength. Since the Hotchkis rear bars are adjustable, they may see a small increase in the spring rate (may put it equal with the Rallitek rear). If you guys want, I can post the formulas to calculate the spring rates of the sway bars.

 

The tubular bars will deform or bend easier than the steel bars. The one advantage the tubular bars have over the solid versions are the reduction is weight. Even though this sounds great, the amount of performance seen in street/autocross applications is minimal.

 

Now, if I were to upgrade bars, my only option at this point would be the Whiteline bars, they are a little larger than the Ralliteks (1mm each). For the amount of money I would have to pay to get the minimal performance gain would be a waste I think.

 

I know a lot of you will disagree with what I have said in the post, but in the end it all comes down to personal preference. When I started looking into sways, Perrin already had stopped producing their rear bars. To me, my only other two choices were Rallitek and Whiteline. Rallitek gave me a killer deal on the sways and it was very hard to pass up.

There are a few things you are forgetting and not understanding. First, is you have forgotten that you are calculating shear stresses, not shear strength. The two are completely different. Second, your math is wrong. Third, calculating shear stress is only a part of the design process.

To address the first, it appears you followed Whitelines shear-stress calculation and the calculation of the polar moment of inertia (which I'll get to in a second). This is calculating shear stress, NOT shear strength. Shear strength is a material property. Shear stress is a physical property. The whole point of the Whiteline article is that shear stress is higher in a hollow bar than a solid bar (of a given constant outer diameter). This is expected, nothing new, and no big deal as #3 will go into further.

The second, about your math. You left out a HUGE step. You only calculated the moment of inertia, not the actual shear stress. If you redo your math, you'll find you get much more understandable numbers. Units for moment of inertia are mm^4, MPa is a unit of stress.

Third, shear stress (not strength) is not a critical factor. All that tells us is that, for a given load, the absolute highest stress we will see is what was calculated. In Whiteline's example, they said 108MPa for a hollow bar. For a standard 4130 bar, we have a yield strength of ~460MPA. So for the Whiteline example, the 100Nm load has a factor of safety of 4.6. That's pretty substantial. Especially when considering that the example assumes a completely fixed bar. In reality the bar rotates, and sees much less load overall.

The bottom line of THAT, is that the paint will crack long before the bar will actually fail. In addition to that, any manufacturer worth their weight will have hopefully heat treated the bars appropriately in a way that the bar will yield before fracture. That means you'll bend the bar long before it actually fails, and you'll know that happens when you see lots of flakes of paint on the ground.

That said, the whole point of this process is to determine that the bar as designed will meet and exceed the strength requirements of the task. While a solid bar will be "stronger", it will be heavier without a significant increase in stiffness.

Yeah, I got ahead of myself while posting . The numbers I have listed are the moment of inertia, I posted the wrong values instead of the shear stress. I did not do my math wrong, I calculated the moment of inertia and the units are mm^4. I am not trying to offend your decision in sway bars, but I do feel that the solid bars are a better design and the math does show it. I am not going to keep posting on this, I admit I posted the wrong values and I labeled them as shear strength instead of stress. I do understand the difference.

I do also understand that the equations we worked are for worst case scenario and that we will never see these forces on the road or track.

I never once stated that any of the bars would actually break, its common sence that they will bend before any real damage is done. i just wanted to show that the solid bars are stronger and that the weight difference between the two types are not going to be noticed for everyday and autocross driving. Sorry for my post mistakes, that's what happens when you are in a rush.

Here are the corrected tables that should have been in the earlier post.

Hotchkis Moment of Inertia:

Front - 19256 mm^4

Rear - 12692 mm^4

Rallitek Moment of Inertia:

Front - 22998 mm^4

Rear - 12794 mm^4

Hotchkis Shear Stress at 100Nm Torque:

Front - 64.91 MPa OD, 54.52 MPa ID

Rear - 86.67 MPA OD, 70.91 MPa ID

Rallitek Shear Stress at 100Nm Torque:

Front - 47.83 MPa

Rear - 74.25 MPa

Most think that that these seemingly small differences aren't noticed but I notice them. Now what I have personally compared are two (seemingly, as one was an LGT the other a spec-b, one is slightly heavier also) identical LGT's at stg 2 with the exact same mods except mine had Cobb bars, aluminum arms (Spec-B), rims lighter by 4lbs, and tires by a pound (to the best of my knowledge as this was 50k and 2yrs ago).

Same stretch of canyon road I felt more confident in my car (maybe just more careful in his), but we traded back and forth all afternoon and he also noticed the same.

I believe it's the cumulative differences that added up make it something you can indeed notice/feel.

Not unlike TGV deletes that by themselves add 1-2hp but in the grand scheme of things...it all adds up.

So the question is...what are you after?

Most think that that these seemingly small differences aren't noticed but I notice them. Now what I have personally compared are two (seemingly, as one was an LGT the other a spec-b, one is slightly heavier also) identical LGT's at stg 2 with the exact same mods except mine had Cobb bars, aluminum arms (Spec-B), rims lighter by 4lbs, and tires by a pound (to the best of my knowledge as this was 50k and 2yrs ago).

Same stretch of canyon road I felt more confident in my car (maybe just more careful in his), but we traded back and forth all afternoon and he also noticed the same.

I believe it's the cumulative differences that added up make it something you can indeed notice/feel.

Not unlike TGV deletes that by themselves add 1-2hp but in the grand scheme of things...it all adds up.

 

So the question is...what are you after?

Thanks Drew888, I also believe these small differences are noticed. A good example is the JDM rear sway bar upgrade. The US spec rear sway bar is a 16mm hollow design and the JDM is a 19mm hollow design. The 3mm difference may not show up as much when you do the shear stress calc, but when it is applied to calculate the spring rate of the bar it is noticed. When we are driving, we feel the spring rate. All of the posts of people replacing their US rear sway with the JDM version feel a big difference.

I just want a neutral steering car. If it takes changing my sway bars, then that is fine. I just wanted to make a point that there are differences between the brands and the designs and that the Ralliteks already on my car are already pretty beefy.

You are putting too much emphasis on shear stress. What matters is stiffness. They are not the same thing.

Calculating shear stress is merely a check to ensure the part won't fail. Calculating it gives no other information.

You should still get more neg camber before you mess with the sway bars. Not enough neg camber will give you bad understeer. Camber, tires, then decide on what changes need to be made to the sway bars.

This.

Neutral handling won't be fixed with just sway bars, it's a combination of:

front springrates vs rear spring rates

strut dampening rates

camber & alignment specs

front swaybar stiffness vs rear swaybar stiffness

tire pressures

front & rear ride height

weight balance

And most importantly driving style. Carrying too much speed into a turn is not understeer. Turning the steering wheel a lil too much while braking isn't oversteer. Neutral steering to one person might be understeer to someone else.

Yes the brand of parts do matter, tho alot are pretty similar. Just find the combination of parts that fit a driving style that you're comfortable with, and just go with it.

You should still get more neg camber before you mess with the sway bars. Not enough neg camber will give you bad understeer. Camber, tires, then decide on what changes need to be made to the sway bars.

Sounds like a plan.

Neutral handling won't be fixed with just sway bars, it's a combination of:

 

front springrates vs rear spring rates

strut dampening rates

camber & alignment specs

front swaybar stiffness vs rear swaybar stiffness

tire pressures

front & rear ride height

weight balance

 

And most importantly driving style. Carrying too much speed into a turn is not understeer. Turning the steering wheel a lil too much while braking isn't oversteer. Neutral steering to one person might be understeer to someone else.

 

Yes the brand of parts do matter, tho alot are pretty similar. Just find the combination of parts that fit a driving style that you're comfortable with, and just go with it.

Thank you.

A good example is the JDM rear sway bar upgrade. The US spec rear sway bar is a 16mm hollow design and the JDM is a 19mm hollow design.

JDM Rear sway is 20mm - hollow. That's all I use, and am more than happy with this. Part of that is the rationale that my master mechanic gave me when we were discussing handling and rear LSD on Subarus. His counsel was to stick with a minimal upgrade, no more than a 5-6mm upgrade over stock as the LSD provides the additional grip and loading, unloading is easier/faster if you match the two (diff and sway). On an open diff, add as much rear sway as you can get without upsetting the overall car's balance or breaking something...

Using the settings I described above, my car is absolutely flat neutral. If I want to dial-in oversteer, I increase rear PSI by 4 PSI, or 2 above the fronts (and only for short durations unless under load, i.e., towing). But, like everything else, vehicle-to-vehicle, YMMV.

GL and keep us poste on what's working for you.

Drop your front sway back to stock, add an adjustable rear (max of 22mm) but min of 20mm. You're overdriving the rear suspension with your current settings at the front. Set your tire pressures with 2 psi bias, and increase/decrease pressures in 2 psi increments until you get the feeling you want. Also, don't underestimate the impact of good tires.

 

Set your alignment at -1.25 rear, -.75 front camber (about the most you'll get with the stock setup), 0 toe F/R, 0 cross-camber. To get beyond these settings, you'll need to add lower adjustable control arms at the rear and aftermarket camber bolts at the front, unless you go with adjustable camber plates at the top.

 

If you're not using the offset LCA bushings, I'd do that as it gives you more positive caster and the car will literally pirouette

Agree.

^^Interested more about this...

I suppose my driving style is just much more aggressive, which is ok... just different.

I agree also with everything except camber. So little in the front is doing nothing to help with understeer especially with so much neg camber in the rear (for the sake of this conversation, all things being equal and all about camber and sways). This may feel neutral to you until you push it much harder then you loose front end grip (yes, even if you load the front properly by tapping the brake or easing off the brake as you make the transition).

Drew - Not taking issue with your comments, more a commentary on them. I've driven the SoCal mountains at 8-9/10s (ask edekba about our Aug 09 fun run) and 9/10s at WSR up in Lancaster, and I can tell you that the car is absolutely flat with the settings I'm using.

The one caveat is that it's a LGTW vs an LGT so there are some differences weight-wise over the rear. As I mentioned YMMV car-to-car.

Cliff Notes: The car handles great at the limit, with the settings and tires I've mentioned.

1. Get better tires

2. What part of the corner are you usually understeering in?

3. I change the settings on my BCs the exact opposite you run on your Konis. i.e. I run my fronts stiffer than the back.