Suspension Optimization and Tuning Help

[Ron Sutton]

Tuning the split in a panhard bar for neutral handling

For most road course & autocross applications, a watt's link works well and is the preferred choice of many. I agree & support this ... but because I'm a tuner, I prefer a panhard bar ... but only if it is height adjustable on both sides & centered in the chassis. This is not a thread to steer someone towards or away from either. Just an understanding of how they work & how they can & can not be tuned.

Pros & cons of Watt’s links:
• Pro: Does not cause different roll centers, roll resistance or tire loads on LH & RH corners.
• Con: Can not be tuned to provide different roll resistance or tire loads on LH & RH corners.

Pros & cons of panhard bars:
• Con: If simply set level, causes different roll centers, roll resistance & tire loads on LH & RH corners.
• Pro: Can be tuned with simple angle change to achieve neutral & equal roll resistance & tire loads on LH & RH corners.
• Pro: Can be tuned to provide different roll resistance & tire loads on LH & RH corners.

The natural tendency for a panhard bar (mounted to the housing on the left & the chassis on the right):
• In LH turns, the left side rod end raises only a small amount with tire extension & the right side rod ends lowers significantly with suspension compression. This moves the roll center down dynamically on LH corners.
• If the panhard bar had been level at 10”, the math could look like this: LH 10.25” & RH 9” = DRC 9.625” & 1.25” split
• In RH turns, the left side rod end lowers only a small amount with tire compression & the right side rod end raises significantly with suspension extension. This moves the roll center up dynamically on RH corners.
• If the panhard bar had been level at 10”, the math could look like this: LH 9.75” & RH 11” = DRC 10.375” & 1.25” split

The split is the same both directions … at 1.25” in this example:
• But the DRC (dynamic roll center) is ¾” lower on LH corners than on RH corners.
• The car rolls more on LH corners … making the car more free or looser.
• The car rolls less on RH corners … making the car more tight or pushy.
• If the panhard bar mounting is reversed, everything above is reversed too.
• This can be corrected easily. More on the “how to” later.

The irony is … the torque steer from centered 3-link, 4-link, torque arm & truck arm suspensions … behave the opposite under power on corner exit. So under power the combination of torque steer & panhard effects are neutralizing each other to some degree … but just on corner exit. You would still have the car behaving more freely or looser in the entry & middle of LH corners … and more tight or pushy in the entry & middle of RH corners. So it’s not the ideal solution by any means.

But with an offset 3-link … if the top link is offset to the right the correct amount to counter the torque … you end up with zero torque steer under power. Combine this with a Watt’s link or “Balanced Panhard Bar” and you will have even loading throughout the corner … on both left & right hand corners.

What is a Balanced Panhard Bar?
We already know that Watt’s links do not cause different roll centers or loads on LH & RH corners … and they can not be adjusted to do so. If you’re running a Watt’s link you will have even loading. If you’re running a panhard bar … with an offset 3-link … you need to tune the panhard bar angle to neutralize the effects discussed above and have a balanced panhard bar set-up. It is easy to do, but needs to be done.



To counter the typical panhard bar effects, you simply run angle … also known as split … in the panhard bar … with the left side lower & right side higher. (Assuming the panhard bar is mounted to the housing on the left & the chassis on the right.) For discussion purposes only, let’s say you run the same 10” roll center but with a 1/2” split. The left side would be 9.75” & right side 10.25”. The dynamic results during cornering would look like this:
• In LH turns: LH 10” & RH 9.25” = DRC 9.625” & 0.75” split
• In RH turns LH 9.5” & RH 11.25” = DRC 10.375” & 1.75” split

This next part throws people off as it is a little challenging to digest:
• The dynamic roll centers do not change with panhard bar split.
• But the differences in dynamic panhard bar angles … change the chassis roll resistance & tire loading when cornering.
• The lower levels of split on LH corners & the higher levels of split on RH corners … balance the handling left & right.
• Of course the numbers above are just an example & the split needs to be fine tuned on any race or track car.

So Ron … why don’t you run a Watt’s link?
Simple answer really. There are some variables that sometimes call for the race car to load the tires more one direction of corner & less the other direction. An adjustable panhard bar lets me fine tune the balance of tire loading.

Situations to tune balance with panhard bar split:
1. Car not 50/50 balanced side to side. Some cars are left heavy due to driver placement.
2. Some tracks have more rights. Some tracks more lefts.
3. Some tracks have high speed lefts & low speed rights … or vice versa.
4. Some tracks have more banked lefts & flatter rights … or vice versa.
5. Some track have off-camber corners, combined with flat and/or banked corners.

At the end of the day, winning races is about overcoming the challenges of the track better than your competitors. If I have a tuning tool that lets me tune for any of these 5 conditions better than my competitors that don’t have this tool … that’s an advantage I’ll take.

We ran a course with 7 right hand turns and only 2 lefts. One left was banked & the other was tight & flat. We used several tools to “wedge” the opposite direction of oval track racers. This gripped up the RH turns & freed up the LH turns. We had amazing grip on all the right hand turns. The banking providing “ok” grip for the first left turn. And freeing up the other super tight left corner made us faster there too. We were faster in 8 out of the 9 corners … won a lot a of races … and won two championships at that track.

---------------------------------------------------------------------------------

Whew! That’s a lot to take in. Let’s summarize it.

1. Centered 3-link, 4-link, torque arm & truck arm suspensions experience torque steer under acceleration.
2. Offset 3-links, if offset correctly, do not experience torque steer.
3. Watt’s links do not cause different roll centers or loads on LH & RH corners … and cannot be adjusted to do so.
4. Panhard bars do cause different roll centers & loads on LH & RH corners … but can be adjusted back to neutral & balanced.
5. Panhard bars can be run neutral & balanced … or tuned to load the tires differently if needed.

---------------------------------------------------------------------------------

With a centered 3-link, 4-link, Torque Arm or Truck Arms & a Watt’s link:
• You will have even loading on entry & middle for both left & right hand corners.
• You are stuck with torque steer on corner exits.

With a centered 3-link, 4-link, Torque Arm or Truck Arms & a adjustable panhard bar:
You can use the panhard bar angle to effectively counteract "torque steer" … to a degree. The keys to doing this are simple:
• Keep the center height where you want the roll center.
• Lower the side you want to load the tire more on & raise the other side the same amount.

This strategy is always a compromise. But the suspension already had a compromise.
You’re just shifting where the compromise is.

Think of it this way:
• If you eliminate 100% of the torque steer, you will have 100% of the effects provided by an unbalanced panhard bar set-up.
• If you balanced panhard bar set-up 100%, you will have 100% of the of the torque steer.
• Most racers shoot for a 50/50 balance and then tune for track conditions.
• You can always shift the balance to gain here & give up there.

With an offset 3-link & a Watt’s link:
• You have zero torque steer and can achieve a “Balanced Panhard Bar” for even loading throughout both left & right hand corners.

With an offset 3-link & an adjustable panhard bar:
• You have zero torque steer and can achieve a “Balanced Panhard Bar” for even loading throughout both left & right hand corners.
• Plus, you can tune tire load balance with panhard bar split if the situation calls for it.

Make sense?

[AutoX_a_Truck?]

After you read all of this over ... digest it, answer my few questions ... and let's figure out what we need to do make your truck an even better handling machine.

Ron, thanks again for your assistance. I almost always end up reading your posts multiple times, but you certainly are very good at providing concise explanations that break down the concepts. It was a lot of information to digest, but I think I have some grasp of the outlined concepts. Below are my answers to each of your questions, which I hope will clear up any confusion.

4. We may ... or may not ... find you have one big thing "off." And that would be what I call dynamic toe, which is a combination of ackerman, static toe out & "bump out" (toe out from bump steer). I can't tell for sure, because we haven't mapped out the steering ... yet. But that spindle has anti-ackerman in it ... so I have concerns.

My Dad started experimenting with modifying the spindle steering arms. That project is finally to a mock up stage. Here is a picture. The spindle itself is the same, but the overall length of the steering arm is approximately 1" shorter than before, and the steering arm is bent as far as brake rotor clearance allows, which is about 2” further inward. The plan is to reset the alignment and then check and adjust bump steer as necessary. The basic goal is to try and at least eliminate the anti-Ackerman steering to achieve some percentage of 100% Ackerman.

5. The two fenderwell camera videos in post #1 ... show me different dynamic tire angles ... which has me a bit perplexed. Are they from the same day? Were there any changes between those runs?

One video was from GG Des Moines July 2012 and the other was from GG Kansas September 2012. I do not believe any chassis or suspension components were changed between the two events. It is possible that there was a slight change in the front end alignment between the two events though.

* The geometry shows plenty of camber, but one of the videos shows the car driving on the outside edge of the outside front tire. Any changes between the date of the video & the date of the geometry capture?

Those suspension view videos were taken with the old truck body. The truck body was slightly lighter in the front and lighter in the rear. The underlying chassis components are the same though.

6. If I understand the order of things ...

a. You had stiffer springs & the 28mm sway bar, then ...

This was the original setup that started as a stock style S10 front suspension with a conventional spring, stock style shock, and drop spindle.

b. Went to softer front springs & the same 28mm bar, then ...

At this point the ridetech Triple Adjustable Coilovers were installed with a 10" long 450 lbs/in spring (softer than the previous conventional spring). The stock style 28 mm S10 sway bar was used, albeit with aftermarket end links. This is the setup used in all of the videos I posted.

c. You switched to the 33mm bar ... but haven't run it yet.
... is this correct ?

Correct.

8. My calcs ... and the software show you have much higher caster than your specs ... and different from side to side. 11.5° on the left & 10.6° on the right. So either your ball joint measurements are off or your caster measuring process is off.

The caster measurements were only recently taken using a newly acquired Fastrax gauge. I hope the error was in the cater measuring process. I will look into verifying both measurements. When my Dad and I mapped out the suspension, we spent the better part of a Saturday and Sunday making sure the measurements were accurate.

9. That is a ton for an AutoX vehicle. Where it hurts the most is in the tightest corners ... and especially if you have to give the truck some additional steering input when you're in the corner. You can band aid this some by running an higher/wider entry line ... turning in later & harder ... and apexing later. This will help, but the solution is longer control arms & deeper back spaced front wheels to get that scrub radius down closer to zero.

I would like to redo the entire front end with better components and ditch the heavy, cast Cragars for a set of 18” light weight forged wheels with deep backspacing, but for the time being I cannot afford to change or upgrade my wheels and all of the components so I am stuck reusing components in a different way or targeting aspects that can be tweaked or fabricated.

10. There are several differences in your geometry settings from side to side. What this does is make the dynamic geometry different on LH turns versus RH turns ... which makes the truck handle differently on LH turns versus RH turns. This is common ... especially in production car ... but never good. My focus is on "how bad is it" and does it merit reworking & TLC'ing the geometry points to correct it.

Yours is off significantly. Your RC migrates to the left quite a bit (21") on LH turns ... and a ton (184") on right hand turns.

11. The front roll center is too low one direction (25" below ground level) ... and way too high (5"+) the other direction. If all other things were neutral, this would make the front end of the truck roll more on LH turns & less on RH turns. With the numbers I'm seeing, the effect won't be small, so you should be feeling it & seeing it.

P.S. All things are not neutral, as you have 3" of split in the panhard bar. I'll address that in the next post.

For a 2.5°-3° high roll angle/Autox set-up I feel around 1.5"± above ground would be closer to optimum. Both directions obviously. For a 1.0°-1.5° low roll angle set-up, you'll want to be around zero (ground level).

Dependent upon your input, I think tweaking the upper and lower control arms pivot points should be able to provide two benefits. First, (if I am picturing the dimensions correctly) I think it should be possible to square up a large amount of the front end geometry by tweaking the horizontal and vertical position of the upper and lower control arm pivots on either side. For instance, I double checked, and the ball joint pivots to control arm pivots are the same on either side. Second, where feasible, it might also be beneficial to improve the geometry rather than simply square it up. That said, I am not the expert here so I will defer to your suggestions.

13. We also need to look at what your bump steer does through your realistic travel. Have you bump steered each side individually yet?

I will add measuring bump steer to the "to do list."

15. Remember earlier when I said ...
"You have a slight mismatch in front & rear roll angles. Ideally we want the front roll angle to be .3° to .4° less roll angle than the rear. This creates a balanced handling competition vehicle that is disengaging the inside rear tire more than the inside front tire. Yours is off a little. It is rolling slightly less in the rear than the front ... I'd say around .2° more in the front ... which is about .5°-.6° from ideal."

What this means is as you work out your spring rate & sway bar rate combo ... you need rear springs & sway bars to achieve about .3°-.4° more rear roll angle than front. As an example, i have a set up that runs the front end at 1.2° & the rear at 1.55° roll angles. This is keep the inside front tire working better & disengage the inside rear tire "to a specific degree" to help the truck turn. Make sense?

Yes, the concept makes sense.

[AutoX_a_Truck?]

To counter the typical panhard bar effects, you simply run angle … also known as split … in the panhard bar … with the left side lower & right side higher. (Assuming the panhard bar is mounted to the housing on the left & the chassis on the right.)

It sounds like the current panhard bar split is the opposite of ideal. The housing mount on the left is a fixed design set at 9 5/8" and the chassis mount on the right is adjustable in 1" increments from 6 5/8" to 9 5/8"

At first, we usually just ran the panhard bar level but at one point it was disconnected and put back in the lowest hole (probably for no good reason). It should not be a huge problem to alter the current design of the panhard bar by either tweaking the mounting points or by fabricating new mounting solutions.