why should rear roll centre higher than the front?

[NR-Cruelty]

hello everyone,
I'm from a young Chinese FSAE team. I have not enough experience to solve some problems. Anyone could help me?
(I'm not good at English, hope you can understand me and point out my errors, thank you! )

Question: some experienced designers give me advice about ensuring the rear suspension roll center higher than the front one, why is that?

My guess: the higher roll centre, the faster lateral weight transfer, this would make a car sharp in tight corner.

I always doubt myself about design. :-P

[exFSAE]

There is no hard fast rule that your rear roll center should be higher than the front, just as much as there is no rule of what color your car should be.

The distribution of your lateral jacking force ratios is entirely up to you based on what YOU want the car to do.

[acedeuce802]

Originally posted by NR-Cruelty:
My guess: the higher roll centre, the faster lateral weight transfer
You are correct about the higher roll center causing a quicker load transfer. So, why would you want the rear to transfer load quicker than the front? Well, you're approaching a corner, and you turn the wheel, the front tires get a slip angle and produce a lateral force. The rear tires haven't yet, until they get some slip angle. Therefore, the rear is lagging behind the front. Having the rear roll center higher helps the rear to "catch up" with the front.

[NR-Cruelty]

acedeuce802,
Thank you very much.
But why we need the rear catch up with the front earlier?would a car be easy to spin if the rear catch up with the front in lag?

[mech5496]

Yes, if you over-do it, it will promote a oversteering attitude during turn in; it's up to you to decide if this is good or bad ant to what extend. I personally prefer a slightly oversteering vehicle during transients but with a slight understeering attitude on steady state cornering.

[NR-Cruelty]

Thank you.
Is the faster lateral weight transfer means faster slip angle forming?
What are the factors which relative to the rate of forming rear lateral force?

[acedeuce802]

Originally posted by NR-Cruelty:
Is the faster lateral weight transfer means faster slip angle forming?

Yes, the slip angle will form faster. Keep in mind that you are still going to be seeing the same slip angles, thus generating the same lateral force. A quicker load transfer does produce a faster more agile car but at the expense of a harder car to drive. The driver must have a faster reaction time to keep up with the quicker load transfers. There is a balance to be found between how agile the car is, and how difficult to drive it is.

Edit: I'm very new to this too, and I'm using knowledge just from past research, and OptimumG, so please correct me if I'm wrong!

[exFSAE]

Load transfer means you build rear slip angle faster? Don't think I believe that, particularly given that it should be compromising both cornering stiffness and ultimate traction.

Who cares what a book or seminar says about X, Y, or Z. Show me your sim work. Prove it.

[Warpspeed]

There are two completely separate issues here.

Steady state understeer/oversteer determined by roll stifness and relative lateral load transfer at each end.

Transient response, or how fast load transfers at each end.

The two are quite independent and are tuned to suit the particular vehicle's mass distribution.

The most common solution usually ends up with the roll centre higher at the rear, and a higher roll stiffness at the front. But not always...

[Z]

NR-C,

Your English is good.

The belief that "higher-rear-roll-centre = good" is, IMO, purely a historical artefact. The development of cars throughout most of the 1900s had most of them with rear beam-axles (high RC), or rear swing-axles (high RC), and independent front suspensions (low RC). This was simply for reasons of convenience of packaging, low cost, robustness, etc., and had little to do with good handling.

Unfortunately, the commonness of this higher-rear-RC has lead many people to think that it must be important for good performance. It is not. Fact is, there are many cars with "lower-rear-RCs" that perform very well.

For a beginning team I suggest you have both RCs about same height, and both about 0-5 cm above ground. Concentrate on a simple, robust design that can be built quickly, so that you can start testing early. Design in adjustability for toe angles (accuracy +/- 0.1deg), camber (+/- 1 deg), and different spring rates (+/- 10%). Don't bother with adjustable kinematics, or ARBs.

Z

[Warpspeed]

Z,
Can you name even one modern well developed production vehicle that has the rear roll centre lower than the front?

Not asking for a dozen examples, just one.

[Charlie]

I have worked on a variety of motorsport formulas and I cannot recall encountering a rear roll center that was lower than the front yet. There are a variety of reasons for that I think, and I would not say it has to be so. But I would disagree that 'it is not' good for performance, or that the idea is dead.

[Trevor]

Originally posted by Z:
Don't bother with adjustable kinematics, or ARBs.

Z

Z, what is your reasoning for skipping anti roll bars? If you can package them in the design phase, they are a much more convenient way to adjust balance than springs which will usually require resetting corner weights or at a minimum getting the car off the ground.

[BillCobb]

Originally posted by Warpspeed:
Z,
Can you name even one modern well developed production vehicle that has the rear roll centre lower than the front?

Not asking for a dozen examples, just one.

I can. And, its from routine measurements made on a commercial MTS K&C machine. Not a one of a kind CAR either. Lots of them were made. Also, in case you are still in disbelief, there are lots of different ones on various models built by the Bite My Wallet Groupen, too. (Just in case the Groupies feel there is some kind of 'philosophy' behind a manufacturer's panache.

You ladies are all hung up on the notion that the roll centers are set and the car designed around them. It bees the other way around, chillens. Roll centers are established by a synthesis process that solves in closed form sets of equations involving TLLTD, D(ynamic)TLLTD, ride balance, and the powertrain inertial axis. Putting it another way, cross products are not good for that ever so popular 'safe and secure' feeling.

[Warpspeed]

I know of only one well known mass produced example, the Hillman Imp. That had an engine well behind the rear axle centreline and the weight distribution that implies.

No doubt plenty of back yard specials and freaks of nature have existed in the past, but can you find something from any of the big well known Multinational Corporations that do serious research into vehicle dynamics and employ real engineers to build world cars ?

Putting the odd ball freaks to one side, the vast majority of vehicles do have a lower front roll centre and higher front roll stiffness.

While testing and experimentation may prove otherwise, starting out with something fairly conventional that has a long history of working may not be a bad staring point for a totally fresh design.

[Claude Rouelle]

there are many cars with "lower-rear-RCs" that perform very well.

Z... you have an example?

Claude

[Z]

Originally posted by Warpspeed:
Can you name even one modern well developed production vehicle that has the rear roll centre lower than the front?
Tony (and Charlie and Claude),

Well... how about... just about every production car using pure trailing-arm rear suspension?

Trailing-arms have nominally ground level RCs. Strictly speaking, we should say they have "lateral n-lines that are always horizontal wrt car body". Combine this with the usual wishbone or McP-strut front suspensions that typically have RC roughly 5-10cm above ground, and... "lower rear RC"!

Historical note: Trailing-arms are cheap, simple, robust, package extremely well, have excellent rough road capability, and, despite the fact that their "RCs" zoom off past Pluto with the slightest body roll, they give very predictable and benign handling. Yet they were the last suspension type to appear on cars! Or, for that matter, on motorbikes (almost universal on bikes nowadays)!

French cars of the 1950s+ were the most common with wishbone front and trailing-arm rear. All the ones I have owned had great handling, in a French sort of way. http://fsae.com/groupee_common/emoticons/icon_smile.gif
~~~o0o~~~


Originally posted by Trevor:
Z, what is your reasoning for skipping anti roll bars?
Trevor,

They are a newbie team.

I have seen cars win FSAE with no suspension movement whatsoever. Yep, win outright! They only needed the "suspension" to get through scrutineering. They would have been faster with some damped suspension movement, but apparently they had "issues" trying to figure out how to do that so it worked.

If the OP tries to design a car with adjustable everything, then chances are they will never finish it in time, and they will perform poorly. So I reckon design a very simple car, get it built early, do lots of testing, and if it doesn't handle 100% perfectly, then that doesn't really matter.

There are still lots of adjustments they have to learn. Tyre types, sizes, and pressures, and then toe and camber angles. That gets you most of the way (still with no suspension). Then spring rates, and maybe damping rates. And then how to drive fast...

Z

[DougMilliken]

Originally posted by Z:
Well... how about... just about every production car using pure trailing-arm rear suspension?

The original Mini (UK):
http://en.wikipedia.org/wiki/Mini
40+ years in production.

[NR-Cruelty]

Thanks all of you.
We are a new team but I had study about adjusting, racing and designing for years.
I can understand most of what you are taking about, and I learnt from your replies.

[Warpspeed]

Ah yes the Mini !
Isn't automotive engineering history a rich and fascinating topic.


Originally posted by Z:

Well... how about... just about every production car using pure trailing-arm rear suspension?...

Z

Nobody uses pure trailing arms anymore at either end, and we all know the reasons why.

I am sure the judges would be really impressed if you presented a pure rear trailing arm FSAE car.

[ZAMR]

This is all well and good, but when you talk about a traditional FSAE car why on earth would you put your rear roll center lower than your front? By traditional I just mean anywhere from 50% to 55% rear weight distribution, rear wheel drive, sla etc. Some people don't make the distinction that in FSAE our steering angles are outrageously high at ~40 degrees for the tightest corners whereas in F1 or traditional racing the steering angles are much much lower (22 degrees in MONACO for F1, much lower for other tracks). To compensate you need to have a bit of extra built in transient oversteer, IMHO.

-Zach

[NR-Cruelty]

In my guess, the higher rear RC is to retain the independence of rear suspension in a same roll stiffness.

Racecars always jump between the curb,once the rear suspension is not independent enough,the inner rear tyre would lift off while the two front tyres climb on the curb, causing much oversteering.
For the equal roll stiffness between front and rear, there should be a strong ARB in front and a high RC in rear.
It's no need to consider curbs in FSAE, so there are some FSAE cars without higher rear RCs.

It's a guess. Please tell me if I was wrong.

[NR-Cruelty]

ZAMR,
maybe you are right. If one fistly make a decision of same distances between RC to MC both in front and rear , then his design will be higher RC in rear because of commonly higher MC in rear.

[Gardong]

Originally posted by NR-Cruelty:
ZAMR,
maybe you are right. If one fistly make a decision of same distances between RC to MC both in front and rear , then his design will be higher RC in rear because of commonly higher MC in rear.
?????RC??MC?????????????????? http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[NR-Cruelty]

?????RC??MC??????????????????
??RC???MC?- - ???????

[Gardong]

Originally posted by NR-Cruelty:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">?????RC??MC??????????????????
??RC???MC?- - ??????? </div></BLOCKQUOTE>


????????????? http://fsae.com/groupee_common/emoticons/icon_frown.gif

[ZAMR]

God help us.

[MCoach]

The de facto language that is used on this message board is English.

[Flight909]

Originally posted by Z:
Well... how about... just about every production car using pure trailing-arm rear suspension?


This is cars which not has performance as it number 1 priority, correct? Sure you can say its a car that handles well, but are they not handling well because the engineers have worked around the cheap suspension design?

In Formula Student, cost maybe is not the important factor for suspension design?

[Warpspeed]

I guess that in another fifty years time, engineers will be rolling around on the carpet, wetting themselves with mirth, about the ignorance of suspension design back in 2012.

Those guys actually used rubber tires inflated with air, hahahahah.

[Z]

Originally posted by NR-Cruelty:
If one firstly make a decision of same distances between RC to MC both in front and rear ...
NR-Cruelty,

I am guessing by "MC" you refer to the "Mass Centre", aka the "Centre of Mass", or "Centre of Gravity", or "CoG", or "CG"?

If so, I hope you don't think that the car body has TWO CGs, one each "in front and rear"!

A reasonably rigid, 3-dimensional car body should be thought of as having only ONE CG. (Is that what Gardong was saying?)
~~~o0o~~~

ZAMR, MCoach,

I am a great admirer of China and some of its ancient writers, but I won't bother learning their langauge. I'm too old, too lazy...

However, if you guys are still youngish, you might have to! Interesting times... http://fsae.com/groupee_common/emoticons/icon_smile.gif
~~~o0o~~~

Flight909,

If FSAE was run on bumpy tracks, then I have no doubt that a well designed trailing-arm rear (and leading-arm front!) suspension would be fastest. However, that would also benefit greatly by "interconnected suspension" principles, which FSAE is still extremely reluctant to embrace (with exception of Uni. Western Aust., due to historical reasons of sponsorship, etc.).

As I have said many times before, the smooth FSAE tracks mean that it can be won with no suspension at all. If a team wants to fit their car with leading and trailing-arms, together with fairly stiff springs, and statically adjustable toe and camber, then they will be at NO DISADVANTAGE to the more common wishbone cars.

BTW, many desert racers (~$100k++s) still use trailing-arms at rear. There is a fashion trend underway towards wishbones "because that's what real racecars have", even though many of the best drivers hate them (constantly varying jacking forces on bumpy corners).

Z

[ZAMR]

http://www.videogamesblogger.com/wp-content/uploads/2007/08/halo-warthog.jpg

You guys might be on to something. If you notice, this is from the future, and features both trailing and leading suspension. Although it tends to slide around A LOT.

-Zach

[ZAMR]

Sorry, my poor attempt at humor. In all seriousness though how can you add a touch of transient oversteer to your car for sudden and tight corner entry if you have pure trailing arm and leading arm suspension? Assuming it is tightly sprung do you just use go-kart logic at that point? Or do you think that it is not necessary? I just wonder.

-Zach

[Warpspeed]

Originally posted by ZAMR:
How can you add a touch of transient oversteer to your car for sudden and tight corner entry .

-Zach
Ding Ding.... we have a winner !!!!

You need to have an increasing understeer gradient for stability.
Spinning out or going backwards at speed can be disorientating, as well as losing lap time.

But stability, combined with just a a bit of transient oversteer will make it fast, responsive and STABLE.

The fastest way from one stable condition to another always involves a bit of overshoot.
Over damping is not fast, and instability is not fast either.

[Z]

Originally posted by ZAMR:
... how can you add a touch of transient oversteer to your car for sudden and tight corner entry ...
Zach,

Huge pro-Ackermann. (There is another thread running on just that subject ...)

Or torgue-vectoring diff. (I'll post on that thread tonight ...)

Or very stiff rear ARB. (Boring ...)

Or lots of rear static toe-out. (Wow! Not boring! http://fsae.com/groupee_common/emoticons/icon_eek.gif)

(Many others (handbrake, driving style...) , but gotta do some work... http://fsae.com/groupee_common/emoticons/icon_smile.gif)

Z

[ZAMR]

Originally posted by Z:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ZAMR:
... how can you add a touch of transient oversteer to your car for sudden and tight corner entry ...
Zach,

Huge pro-Ackermann. (There is another thread running on just that subject ...)

Or torgue-vectoring diff. (I'll post on that thread tonight ...)

Or very stiff rear ARB. (Boring ...)

Or lots of rear static toe-out. (Wow! Not boring! http://fsae.com/groupee_common/emoticons/icon_eek.gif)

(Many others (handbrake, driving style...) , but gotta do some work... http://fsae.com/groupee_common/emoticons/icon_smile.gif)

Z </div></BLOCKQUOTE>

I must not have read the Ackermann thread. Makes sense though.

Not sure how an ARB has a transient effect since presumably any load transfer you apply at corner entry will increase proportionally as you roll.

Static toe out is very interesting and I know GFR plays around with it. I can see how it can achieve the effect we desire, but to what degree I am uncertain (also depends on how much toe-out we are talking about). It obviously works if they do it haha.

I'm not on a team anymore or else I would be working some of this stuff out...

-Zach

[Moop]

What do we mean by transient oversteer? I thought that a linear oversteer car was overdamped in yaw, and that a linear understeer car was underdamped in yaw, resulting in the linear under steer car overshooting the steady state yaw rate in a step steer maneuver

[Warpspeed]

Transient oversteer is when CHANGES of lateral load transfer are made to occur faster at the back.

The car can still be tuned to give a certain basic minimal stable understeer (in steady state cornering) by giving it a higher relative front roll stiffness.

But you can speed up the response to SUDDEN changes in direction by raising the rear roll centre relative to the front roll centre.