Roll Axis Inclination

[ajb]

Should the front or rear RC be higher? Miliken seems to say the front should be higher in RCVD: "If the car is forward weight biased, a rear roll center higher than the front will tend to make it neutral." (p.394) Thus our car (being rear weight biased) should have a front roll center higher than the rear. This makes sense to me, because reducing the roll stiffness on the rear (by lowering the RC) should reduce the load transfer, which would allow the rear tires to grip better.

However, I believe a design judge I spoke with at competition told me that the rear RC should be higher. He said something about how a downward inclined roll axis (from rear to front) would make the car feel more stable to the driver. Are there other pros and cons involved in the inclination of the roll axis?

I'm on the suspension team this year, and have no experience other than reading RCVD and Tune to Win. If there are any serious errors in my understanding, please point them outhttp://fsae.com/groupee_common/emoticons/icon_smile.gif Sorry if this question is covered in another thread. If it is I couldn't find it...

[Jersey Tom]

Neither. You pick your RAI (not the Italian TV broadcaster) based on your requirements.

[STRETCH]

I remember Tune to Win actually indicating that a lower front RC promotes understeer, so is a safe bet. But this is not necessarily true. For a given car, lets assume 50/50 WD and equal roll stiffness all round, imagine what happens when you LIFT the front RC... You will get more geometric weight transfer, and less sprung weight transfer. But because the rear is unchanged, the average roll moment will now be more than the front requires alone. You've therefore increased front WT, an US effect. Obviously the effect is mixed up when there is a WD and roll stiffness distribution, but Paul Haney correctly points out in his tyre book, that balance can be adjusted with the RCs. Lower at the front theoretically adds oversteer, as it biases the total LT distribution to the rear.

Hope this helps?

[ajb]

I understand that this decision needs to be based upon my requirements. I just want to hopefully get a better idea of how the RAI affects the performance of the car. How can I determine, for example, whether a downward inclined roll axis would make the car feel more stable to the driver unless I’ve already designed and built a car…

[ajb]

Thanks Lee, if I understand Miliken correctly, he essentially says what you wrote: that lifting either RC increases the WT at that end. Thus lifting the front RC would contribute to US. From my understanding of what I’ve read, US is more stable than OS, and so it is better to err on the side of US. Thus what would be the advantage of having a downward inclined roll axis (which would contribute to OS)?

[Mikey Antonakakis]

I'm no suspension expert, but why is an understeer car better? What exactly, or even approximately, are you designing your suspension to do? Think of the purpose of this competition, to build an autocross car. What is the most important factor in building a car that goes fast in the autocross? I think generally we are looking to change direction as quickly as possible. Setting up a car to understeer will make it more stable, but will it make it faster? Does a driver NEED the car to be more stable? I do not pretend to have the answer, but you definitely need figure those things out at some point. I can think of at least one vehicle dynamics book that is more focused on passenger cars than race cars. There can be a pretty big difference in a car designed to be stable because most people using the cars have no idea how to react to an oversteer situation (a passenger car) or a car designed to be slightly less than stable if it will make it go faster around corners because the driver has the skill to handle it. I think depending on the skill of your drivers, your first question should be "will ____ make the car go around an autocross track faster?" And the second question should be (if yes to the first one) "Do the drivers of our car have the skill required to control the car if _____ makes it unstable?"

[exFSAE]

Couple things...

1. It is indeed written that raising a RC on one axle will "stiffen" that end geometrically.. ie RAI down at front -> less understeer.

2. Your RAI is not static. It changes based on chassis movement. This may be a significant point for your car depending on how stiffly sprung it is.

3. RAI is not the only thing that determines vehicle balance. Springs and bars are a big contributor. What fraction of the total load transfer is taken up by springs?

Is it generally better to initially start with understeer? IMO yes.

There's the thought that the fraction of load transfer taken up by the RC's occurs more or less instantly with change in tire forces, whereas the sprung fraction requires the sprung mass to move.

If that's the case... maybe you want RAI tuned to oversteer so the car pivots quickly into a rapid steering motion.. whereas the springs lean it out toward understeer mid corner. Could do that with dampers as well.

Maybe your driver likes the car to generally be a little more free (stiffer rear springs) and have the RAI tune to understeer so the handling is a little more numb on steering inputs (also can be done with dampers).

Maybe you want the RAI inclined so it's tuned to the same balance as your springs and bars so the handling is predictable.

Maybe it's all a load of crap.

[HenningO]

There is one more factor which one can take into account (I haven't done any calcs on it, so I can't say anything thing about it's influence).

With an inclined roll axle, the roll motion is going to induce a yaw motion of the sprung mass in transient. With a lower front RCH in corner entry, the yaw motion is going to be of the opposite direction to where the car is steering, thus making the car feeling more stable. The same but reversed is valid for corner exit situation.

Following this reasoning, a higher front RCH then rear will produce a more responsive edgy behaviour.

Without having a very large RAI, in race car applications, where the roll angle is fairly small, I have a hard time seeing how this could have a big effect on handling.

[Ezio_Auditore]

milliken states - " higher the roll center, smaller is the rolling moment about the roll center, and lower the roll center, larger is the moment" doesn't this essentially mean that a larger moment would induce more lateral weight transfer at that end? and as the load transfer increases, the slip angle increases leading to under steer characteristics. please point if im going wrong anywhere...

[Z]

Originally posted by Ezio_Auditore:
... doesn't this essentially mean that a larger moment would induce more lateral weight transfer at that end?
Ezio,

You are going wrong with your thinking in the above quote.

Total Lateral Load Transfer (TLLT) is a function of lateral acceleration, CG height, and track width (to first approximation). Type of suspension (soft, stiff, whatever) makes no difference to TLLT (tfa).

However, the type of suspension does determine which end of the car carries more or less percentage of the TLLT. This is called LLT Distribution.

LLTD is determined by front and rear;
1. geometric/kinematic/n-line/RC-height LLT (these are different names for the same effects from control arm geometry),
2. elastic LLT (from springs),
3. viscous LLT (from dampers).

Very important! Since the car should be reasonably stiff torsionally, FRONT ROLL ANGLE = REAR ROLL ANGLE.

There is no separate "rolling moment about the [front and rear] roll centers". Think of the car as a single 3-D object, not as two independent 2-D pictures.

There are countless threads covering this subject. Check the one from a few days ago, half way down first page.

Z

[BillCobb]

A considerable influence on the placement of the roll axis is the powertrain's inertia ellipsoid. Its principle axis orientation is the subject. This means N/S vs. E/W crank, crank case depth, head component masses, transmission type, and even accessory drive configuration. The concept essentially tries to minimise an additional z*r^2 spring mass roll inertia term that can make a mess of transient handling feel. Thus the roll dynamics are factored into its placement.

But, that's not the only consideration. Weight distribution defines the ride rate of each axle and these rates are usually ratioed to produce a 'flat ride' quality at some designated speed and frequencies. This yields the ride rate components of the total roll moment fraction. Given a roll gradient target for the vehicle, the residual moment components (roll bars and roll axis location) are allocated based on favorable bar sizes and bar package real estate, mass of the proposed bar(s) and vehicle purpose: delta payload, number of passengers, frame dimensions, tire load transfer sensitivity and other things (such as the number of extra parts (cost and mass) needed to provide a rear bar for example). The payload consideration regulates the TLLTD of the max GVW capability of the vehicle, which can't be compromised.

Given a FSAE car package, things are a bit easier, but still ought to be synthesized.

Lastly, the dampers chosen to accompany the ride springs have their own (usually tri or quadlinear) traits which usually adds considerable disturbance to the dynamic roll and single bump feel characteristics of your vehicle.