Suspension parameter for chassis design

[nucleardisaster]

I've been doing lots of research on suspension design for our teams first car. I've gone through Carroll Smith's book, and I've started in on Racecar Vehicle Dynamics..its alot, and i dont believe i will have the time to finish the whole book before i design the suspension, so i am in need of some advice on what sections i should be really focusing on for the car.
One of the urgently required data is on difference between the pivot points of the Lower and the Upper A Arms on the chassis as well as the length between the mountings of the two mounting points of the same A Arm on the chassis . Please suggest and would like to learn and work it out as much as possible from here.
Thank You.

[nucleardisaster]

I've been doing lots of research on suspension design for our teams first car. I've gone through Carroll Smith's book, and I've started in on Racecar Vehicle Dynamics..its alot, and i dont believe i will have the time to finish the whole book before i design the suspension, so i am in need of some advice on what sections i should be really focusing on for the car.
One of the urgently required data is on difference between the pivot points of the Lower and the Upper A Arms on the chassis as well as the length between the mountings of the two mounting points of the same A Arm on the chassis . Please suggest and would like to learn and work it out as much as possible from here.
Thank You.

[BillCobb]

Its a four bar linkage.
It reacts forces and moments subject to displacement constraints.
Synthesized axis locations for upper and lower arm pivots produces the desired handling and braking kinematics of the spindle trajectory.
Not a good solution if they break off.
Not a good solution if brakes, bearings, outer tierod ends, and hoses don't fit and or you hit the frame with a wheel in an unfriendly manner.

There's more, but I don't have time to list them all...

[OspreysGoSWOOP]

If you really are reading those books thank you for actually putting in effort before asking a question.

Topics you should concentrate on are:
roll center heights and how the affect handling
roll camber coefficient
anti-features (dive, squat)
chpt 17 in RCVD

Hope this helps.

[nucleardisaster]

So far as per the study we have decided of using Unequal and Parallel A Arms, that means the IC's will be located at infinity and the Roll Centre on the ground Plane. Unfortunately, negative Camber of the unladen wheel we be increased, but since we are using a knuckle of a very high KPI around 9 degrees would this help in reducing the undesirable excess negative camber?

[Buckingham]

It sounds like you are on a good track. Your proposed package is simple, understandable, and you seem aware of some potential trade-offs that come with your decision.

To answer your question, is it possible to counteract some of the camber loss of the unladen wheel by building steer-camber into your steering geometry? Yes.

How much roll do you actually anticipate? How much camber gain/loss does this translate to? How much lateral thrust does this translate to in your tire data? Does your camber gain improve thrust on the laden wheel more than the thrust lost on the unladen wheel?

How much tire steer angle do you anticipate? How much camber gain can you comfortably build into your steering system?

You are the only one who can decide if "how much" makes sense for your team and goals.

It is possible to build a top-10 car with excessive camber gain/loss and it is also possible to build a top-10 car with nonexistent camber gain/loss.

Remember... slip angle generates most of your cornering thrust, and that parameter is quite literally in the hands of the driver. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[nucleardisaster]

After some more research and study, me and my team members have decided to go with Unequal and UnParallel Arms. Making a Front Suspension Geometry on Solidworks we are getting a Roll Centre at a height of 25.4mm above the ground plane.
Track Width=1200mm
KPI=10
SAL=900mm
Static Camber=-1
We are now working for the Rear Geometry and are searching for the Inclination of the Roll Axis(Front Lower or Rear Lower), however we went through all the related Forums about Roll Axis inclination but it didnt help in our case.
Please suggest and would like to learn and work it out as much as possible from here.
Thank You.

[Warpspeed]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by nucleardisaster:
We are now working for the Rear Geometry and are searching for the Inclination of the Roll Axis(Front Lower or Rear Lower), however we went through all the related Forums about Roll Axis inclination but it didnt help in our case.
Please suggest and would like to learn and work it out as much as possible from here.
Thank You. </div></BLOCKQUOTE>
This is where it becomes a bit tricky.

Under steady state static cornering, (as on a skid pad) you can always balance the relative front and rear weight transfer with suitable front and rear antiroll bar stiffness adjustments.

So within reason, any roll axis inclination can be tuned to produce the desired understeer/oversteer balance with appropriate antiroll bar adjustment.

But that all changes with very sudden changes in vehicle direction.
The lateral forces transmitted through the suspension linkages to the tires act pretty much instantaneously, but the forces transmitted through the antiroll bars and springs are applied only as the sprung mass rolls on the springs, and that takes a finite time.

So the relative roll centre heights front/rear will have a big effect on how the car reacts to sudden steering and directional changes, and the lateral forces that sudden change of direction produces.

Higher roll centres create less body roll, and transmit more of the lateral force directly through the suspension links, and do it faster. But this also introduces jacking forces.

Low roll centres create more body roll, and there will be a short delay as that increasing body roll transfers weight off some tires and onto others through the springs and bars.

So it it is transient response phenomena, or how it feels to drive that will be mostly affected by changing the slope of the roll axis, and then compensating for that change with the antiroll bars.

You can always balance the car with the antiroll bars in steady state, but it may not be very nice to dive fast around a chicane if the roll axis is totally inappropriate to the the main vehicle masses.

As a general guide, the roll axis usually slopes down towards the front, with higher roll stiffness at the front on most conventional cars.

This hopefully produces a slightly understeering gradient under steady state, with very slight momentary transient oversteer during sudden directional change to improve feel and speed of response.
The trick is to make it responsive without becoming twitchy or unstable.
As the late Carol Smith says, "a race car should be nimble, it should dance".

First get it balanced on the skid bad with the antiroll bars, then test it on the twisties, and adjust the roll axis for best stability/response.
Rinse and repeat......

This is what I believe to be true, but I am still trying to learn more about all this myself.

[nucleardisaster]

Warpspeed,

Thank You for your valuable contribution, it was of great help. But I want to know HOW the Roll Axis inclining down towards the front produces understeer?

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by nucleardisaster:
I want to know HOW the Roll Axis inclining down towards the front produces understeer? </div></BLOCKQUOTE>
Nuke,

It doesn't. http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif

Read Warpspeed's post again. Then read the other threads that explain LLTD, TLS, and so on.

The answers are all there.

Z

[Warpspeed]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by nucleardisaster:
Warpspeed,

I want to know HOW the Roll Axis inclining down towards the front produces understeer? </div></BLOCKQUOTE>
The understeer comes from making the front take a higher proportion of the total lateral weight transfer in steady state.
That is achieved by fitting a much stiffer front antiroll bar.

The roll axis inclination can then be adjusted to be higher at the rear, to make the rear respond faster to fast and sudden CHANGES in lateral weight transfer through the suspension links.
You would expect this to create transient oversteer, and it does, but only momentarily while body roll is actually changing. What it does is speed up response and make the car feel more "lively" especially during turn in.

You need to separate steady state cornering from transient changes here.
You want steady state slight understeer for stability, but by introducing a "bit" of momentary transient oversteer, it can and will speed up the response to changes.

In summary, tune the bars for steady state balance, and tune the roll axis inclination for fast and stable response to sudden directional changes.

There is no simple magic formula to solve for this, it is the combination that counts, and you are going to have to find it by track testing and making trial adjustments. When it feels right to the driver, it IS right.

While computer simulation would in theory be possible, the problem is that many of the inputs required to truly accurately model the dynamic behavior of the whole car are not going to be available at our level.

[BillCobb]

Somebody needs to create, write, beg, borrow or steal a handling simulation with these concepts added, built-in or swagged. Then you would not be making such incredible statements. Did I say incrediible? I meant uncredible.

These are not the major reasons a roll axis is inclined or declined. Too bad I can't directly show you a histogram of front vs. rear roll center heights (hence roll axis inclination) for a few thousand pass.cars, trucks and race cars. If I can figure that out, I will post it.

Try to remember that you have a rear weight biased car, a rear powertrain, with some cost and mass considerations tossed in.

[Warpspeed]

That you are far more knowledgeable than I am, is entirely possible and very likely true.
Please illuminate us all with your wisdom and experience.

It is all too easy to put a humble altruistic trier down.

Far more difficult to raise the bar of knowledge and experience higher.

Feel very free to share the fruits your superior wisdom with the rest of us.

This is an open Forum for discussion.

[nucleardisaster]

Warpspeed,

First of all i'm very grateful for your elaboration on my priority. But i would like to add some more.
What if we arent using Anti-Roll bars? We are very less on resources, so Anti-Roll bars seems to be out of question. Does having a large roll moment at the rear help in creating a large lateral load transfer so that it exceeds the frictional force, thus causing slip and understeer?
Please illuminate us all with your wisdom and experience!

[Claude Rouelle]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Too bad I can't directly show you a histogram of front vs. rear roll center heights (hence roll axis inclination) for a few thousand pass.cars, trucks and race cars. If I can figure that out, I will post it. </div></BLOCKQUOTE>

Try this as a beginning

Not the best paper ever written on roll centers (not an easy topic anyway). However it shows that the majority of passengers cars measured in this paper study (front or rear wheel drive, SUV or sedan) have higher rear than front roll centers.

http://www.vehicledynamics-exp...1/10.15_lukianov.pdf (http://www.vehicledynamics-expousa.com/08_conf/pdfs/day%201/10.15_lukianov.pdf)

Race cars are not different.

To understand the need of roll axis inclination FS / FSAE students need a to look at the car as a whole (not front or rear separately) and also need to understand the car weight transfer in transient.

Claude Rouelle

[Warpspeed]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by nucleardisaster:
Warpspeed,
What if we aren't using Anti-Roll bars? We are very less on resources, so Anti-Roll bars seems to be out of question.
</div></BLOCKQUOTE>
You can certainly adjust roll stiffness at both ends with just the springs, but you will need to run some pretty fearsome front springs to get a significantly rearward weight biased car to transfer enough weight at the front to reduce the inherent oversteer.
It is certainly possible, but it is really doing it the hard way.
Antiroll bars are about the single most useful tuning tool, and deciding not having any is going to put you at a huge disadvantage.

[Zac]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Claude Rouelle:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Too bad I can't directly show you a histogram of front vs. rear roll center heights (hence roll axis inclination) for a few thousand pass.cars, trucks and race cars. If I can figure that out, I will post it. </div></BLOCKQUOTE>

Try this as a beginning

Not the best paper ever written on roll centers (not an easy topic anyway). However it shows that the majority of passengers cars measured in this paper study (front or rear wheel drive, SUV or sedan) have higher rear than front roll centers.

http://www.vehicledynamics-exp...1/10.15_lukianov.pdf (http://www.vehicledynamics-expousa.com/08_conf/pdfs/day%201/10.15_lukianov.pdf)

Race cars are not different.

To understand the need of roll axis inclination FS / FSAE students need a to look at the car as a whole (not front or rear separately) and also need to understand the car weight transfer in transient.

Claude Rouelle </div></BLOCKQUOTE>

Seems like most passenger cars, trucks, and even race cars are front engine or have a weight bias around 60/40 to 50/50. Farming design trends (as opposed to response trends) from that population of vehicles might not be the best justification.

[Warpspeed]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Zac:
Seems like most passenger cars, trucks, and even race cars are front engine or have a weight bias around 60/40 to 50/50. Farming design trends (as opposed to response trends) from that population of vehicles might not be the best justification. </div></BLOCKQUOTE>
Zac, you are quite right about that.
Circumstances and requirements change, but the basic rules governing steady state and transient load transfer do not.
We still want a very slightly understeering vehicle with good dynamic response no matter what strange mass configuration the vehicle actually has.
And although the final solution may be rather different, the path towards finding that solution does not change.

[Claude Rouelle]

I can't publish our racing customer kinematics design but I can tell you that the trends is the same than for passenger cars. If you can make a winning FS / FSAE car with a kinematics rear roll center much higher than the front one (unless it is a patch on a patch to compensate for compliance) I will be ready to revise my theories; never say never.

[Z]

Just a brief point.

When a car's body "rolls" about an "inclined axis" (ie. rotation about a generally longitudinal, but not horizontal, axis in the car's central plane), then there is necessarily also a certain amount of "yaw" motion (wrt car xyz coords). The magnitudes of roll and yaw are the vector decomposition of the rotation about the sloping axis.

If the axis is down-at-front, then the yaw is in the understeer direction. If down-at-back, then yaw is oversteerish.

Importantly, this is a kinematic effect of the body's motion that is only felt during the "roll" motion, and is in NO WAY CONNECTED TO RELATIVE F/R WHEEL GRIPS.

Just as importantly, it is felt very much by the driver as true under/oversteer. So a down-at-front axis feels like understeer (nose points slightly out from the corner tangent), even if the front tyres have massive grip. Vice versa for down-at-back.

And importantly for FSAE, or circuit racing in general, none of the above matters much if the car is so stiffly sprung (or ARB'd) that there is negligible body "roll".

(BTW, ARBs = grip reducers = mostly bad! http://fsae.com/groupee_common/emoticons/icon_smile.gif )

Z

[Warpspeed]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
And importantly for FSAE, or circuit racing in general, none of the above matters much if the car is so stiffly sprung (or ARB'd) that there is negligible body "roll".

(BTW, ARBs = grip reducers = mostly bad! http://fsae.com/groupee_common/emoticons/icon_smile.gif )

Z </div></BLOCKQUOTE>
Even a factory sports car might only roll perhaps two degrees, and any serious race car much less than that.
Unless the roll axis tilt is absolutely huge, any actual suspension link induced yaw would have to be be pretty slight compared to the deliberate yaw required to negotiate the corner we are trying to get around.

And yes, ARB's do shift the grip away to the other end, but if the change is appropriate, the result will be a beneficial improvement in overall vehicle balance.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Warpspeed:
Even a factory sports car might only roll perhaps two degrees, and any serious race car much less than that... </div></BLOCKQUOTE>
Tony,

Ah, yes, "serious race cars".

I remember once watching an off-road racing "truggy" having difficulty negotiating a corner. (Truggy = truck/buggy, with V8 front engine, wishbones at front, and live axle at rear. This one, with ~700hp, cost perhaps $200-300K, so I guess not very serious? http://fsae.com/groupee_common/emoticons/icon_smile.gif Hey, they get all covered in mud, so they CANNOT be serious! http://fsae.com/groupee_common/emoticons/icon_smile.gif)

Anyway, the driver was getting around the corner in a series of lurches. Enter the corner at speed, feel massive understeer, off-throttle (maybe some brakes), un-roll, apply throttle, massive understeer again, repeat cycle...

The thing is, from my viewpoint the front wheels had plenty of grip. They were NOT sliding. But the truggy had (notional) ground level front RC, and a rear RC height about 0.6m, due to a high mounted upper A-arm for axle-lateral/axle-torque control.

If the driver kept his right foot down he would have got around the corner just fine.

I think passenger cars have a little of this roll/yaw "understeer" so as not to frighten the drivers. The truggy just had too much of it (and/or a driver who didn't understand it).

Z

[Warpspeed]

At least in years gone by, it seems the main goal of the production car suspension design engineer was to get the wide eyed, white knuckle driver to wet his pants long before the vehicle actually goes totally out of control.

There is a sort of perverse logic in that, that brings a smile to a corporate lawyers face.

[Zac]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Claude Rouelle:
I can't publish our racing customer kinematics design but I can tell you that the trends is the same than for passenger cars. If you can make a winning FS / FSAE car with a kinematics rear roll center much higher than the front one (unless it is a patch on a patch to compensate for compliance) I will be ready to revise my theories; never say never. </div></BLOCKQUOTE>

I can't publish my customer's kinematics designs either, but I am aware of at least one world championship winning FSAE car that has a front roll center higher than the rear, one where it is approx parallel, and several where the rear is higher than the front.

But this has nothing to do with my earlier point. If a student showed up in the design tent with "that's how they do it on real race cars" as their design justification for anything you would tear them apart. That line of thinking is no more valid here.

[DougMilliken]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Z:
...But the truggy had (notional) ground level front RC, and a rear RC height about 0.6m, due to a high mounted upper A-arm for axle-lateral/axle-torque control.

If the driver kept his right foot down he would have got around the corner just fine. ...
</div></BLOCKQUOTE>

Nice story, thanks. For anyone looking for pictures and more discussion of inclined roll axis, Olley discusses it in "Chassis Design" (shameless plug for our book).
-- Doug Milliken

[nucleardisaster]

Thank You all for your help!
Got to learn a lot!

[EPMPaul]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by DougMilliken
For anyone looking for pictures and more discussion of inclined roll axis, Olley discusses it in "Chassis Design"
</div></BLOCKQUOTE>

Another good read for trying to figure out what your roll center should be(assuming you have a model that can deal with transients and steady states as well as some basic roll dynamics) is the 7th chapter of A multibody dynamics approach to vehicle dynamics. It won't give the whole answer but it gives interesting cues about what is felt by the driver and hence what yaw/sideslip/roll metrics you should be looking out for when simulating / acquiring data on the track.

As for roll center locations in themselves, having not really had the time to simulate/try this stuff extensively, putting your front roll center a little lower than your rear one and not too far from the ground (figure 2 -3 inches, which seems to be what you're looking at) is probably a safe bet for getting through first round of design/ having an ok handling car, providing you can give the explanation Warpspeed gave.

Seeing as it's your first car, it'd take a fairly safe bet, go with it and get the car running. That'll have a much bigger effect on your results than moving your roll center 1 inch in any direction you want providing you're in the right ballpark(which you seem to be)

[OspreysGoSWOOP]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">The reduced roll couple at the rear leads to balanced handling and any ‘error’ in calculating the roll axis height is easily tuned out with spring rates or motion ratios. The low roll axis means that jacking forces are negligible, and can effectively be ignored.
</div></BLOCKQUOTE>
Pat Clarke's explanation for an inclined roll axis.

He writes a good article about roll centers.

http://www.formulastudent.de/d...ats-column-february/ (http://www.formulastudent.de/de/academy/pats-corner/advice-details/article/pats-column-february/)

[nucleardisaster]

I realized that there is no Perfect position for the Roll Centre placement so we had our Rear Roll Centre a little higher than the front, so an inclined Roll Axis towards the front. I went through the threads of the Control Arm Load Calculations, Double Wishbone Shape. What should be next step after roughly drafting the suspension geomtries?
Also how to avoid the disadvantages of using the Rod Ends? I know the Design Judges hate them.

[PatClarke]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Also how to avoid the disadvantages of using the Rod Ends? I know the Design Judges hate them. </div></BLOCKQUOTE>
ND, the judges don't hate rod ends but they do hate when they are used incorrectly or inappropriately.

Pat

[Warpspeed]

Just reflecting on Z's anecdote about the off road racing trucks.

Those machines always have a relatively high centre of gravity, because of the requirements for a large ground clearance.
They also have a requirement for very soft long travel suspensions to climb over obstacles while maintaining as much traction as possible.

If you take all of that into consideration, the roll centre placements are going to have to be unusually high at both ends to limit body roll, compared to what we are used to seeing on our own ground hugging cars.

The design requirements are certainly very different for off road, and roll centre placements must be part of an overall plan that best suits the intended purpose.

There is no "strictly correct" answer to roll centre heights, just perhaps an acceptable compromise of a whole bunch of conflicting requirements.
Some time spent thinking the whole thing through, and some experimental changes made during track testing will be amply rewarded.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Warpspeed:
... the roll centre placements are going to have to be unusually high at both ends to limit body roll... </div></BLOCKQUOTE>
Tony,

I forgot to mention that the truggies "apparent" understeer was due to the massive roll angle typical of cornering off-roaders (this then coupling with yaw...).

Large roll-angles, and the consequent adverse tyre camber angles, are NOT a problem off-road. There are so many bumps and ruts that the tyres don't really have any measurable camber angle, at least not for more than a millisecond or so, after which it changes to the opposite value...

However, jacking can be a serious problem, because it can quickly lead to vehicle roll-over. So most off-roaders have (notionally) ground level RCs for their independent suspensions. Beam-axles don't suffer from jacking (unless very badly located!) so can have high roll centres. IMO the truggy would have worked much better if it had a lower rear RC (essentially, swap the axle locating links top for bottom).

Another important consideration off-road is to minimise gyroscopic forces (coming from big, high MoI wheels, going over big bumps, all of the time). This means minimal camber change of the wheel during bump, so very long (infinite) FVSAs, and thus zero "camber compensation".

Practically, this means trailing-arms (like the original Beetle and still very common off-road), or long, equal-length and parallel, lateral wishbones. Both of these have the RC shooting off to infinity, or at least out past Pluto, with the slightest roll motion. But NOTHING bad happens! http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Warpspeed]

Agree, the generally huge aggressively treaded round section tires suffer very little from eye popping camber angles on dirt, mud or gravel.

Traction probably has much more to do with actually keeping the tires planted on the ground rather than any ultra fine tuning of camber angle or scrub.

But I disagree about very low roll centres being an advantage on these types of tall vehicles.

With very soft suspension and very high CG, the vehicle would likely fall over with ultra low roll centres. Can you show us a picture of an off roader with roll centres at near ground level both ends? I have never seen it in off road.

While jacking is the obvious big disadvantage of having higher roll centres, I an told it is the lesser evil to having massive uncontrolled body roll.

Yeah, the gyroscopic effects and steering kickback of those monster wheels must really be something. Most seem to run hydraulic steering dampers and power steering.
It's a whole different world to FSAE, that's for sure.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Warpspeed:
With very soft suspension and very high CG, the vehicle would likely fall over with ultra low roll centres. Can you show us a picture of an off roader with roll centres at near ground level both ends? I have never seen it in off road. </div></BLOCKQUOTE>
Tony,

"Never seen" low RCs off-road??? Wow! You ought to get out more often... http://fsae.com/groupee_common/emoticons/icon_smile.gif

The independent suspensions of off-road buggies, broadly speaking, are split into two classes.
~~~o0o~~~

1. Original "VW Beetle" style. This has twin-trailing arms at front, and pure-trailing arm at rear. (Note that original Beetle had semi-trailing "swing-axle" at rear, which was quickly dropped because too much jacking from the high RC.)

Both these trailing arm setups (pivoting about horizontal-lateral axes) have horizontal n-lines in end view, which always stay horizontal wrt car-coords throughout the range of travel. So notional ground level RCs at both ends.

The main disadvantage of this setup is that for long front suspension travel, the trailing arms must be long, and hence there must be a lot of structure a long way in front of the front wheels (which then bumps into things).
~~~o0o~~~

2. Long Lateral Double-Wishbone style (or "A-Arm" in US). These can be fitted all-round, or only at front with pure-trailing arms at rear.

These are a "fashion" development, because "That's what real racecars use, so, err... they must be better...". Yes, I have heard exactly that reasoning used too many times to justify these. I know that many drivers don't like them because of their unpredictable jacking in corners (jacking dependent on arm angles, so effected by bumps). Nevertheless, they are "the way everyone is going now..., the most expensive buggies use them, so they must be good...". http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif

Anyway, these lateral, parallel, equal-length, wishbones are setup so they are about horizontal at mid-travel. This again gives horizontal n-lines and ground level RCs. (Unfortunately the n-lines change angle with wheel travel, hence the unpredictable jacking...)
~~~o0o~~~

Here is a picture of one at full droop. Note that here the RC would be about half the height of the floor pan, the highest the RC ever gets. This is about 2cm in FSAE terms.

http://image.off-roadweb.com/f/31271974/111010or_8881+lucas_oil_off_road_racing_series+loo rs_buggies.jpg

Finally, off-roaders have been able to get higher while avoiding rollover by simply becoming wider. Most have stopped at 2.4m (8ft) O/A width to allow legal trailering, although some now fit special narrow inboard wheels for transport.

Z

[Warpspeed]

The roll centres in that picture appear to be about ten feet in tha air Z, nowhere near the ground, hehehehe.

But around here, most of the crazy 4WD the off road racers still run beam axles both ends, and that makes ultra low roll centres rather more difficult to accommodate.

But I do know for rock crawling and other crimes against nature, massive suspension articulation, very soft long travel springs and high roll centres are the go.
I guess we have been talking at cross purposes.