Whats the point of Front Wheel drive?

[Moppie]

The more I read this, the more I agree with Nereth on one thing:
For ultimate performance, RWD is better than FWD.
But that was estrablished right at the start of this thread.

From there things get a little lost.
Nereth you seem to be always arguing from the point of ultimate high performance.
Where it is valid and accepted that RWD is better.

What the rest of us are pointing out, is that it is not always better under less demanding circumstances, and in the average road car, FWD can be preform just as well as RWD, and sometimes better.
You don't need a degree in engineering to understand that, you simply need to go out and drive a few cars, or even read a few magazine reviews to see real world examples.

[Moppie]

Originally Posted by slideways...

.....i commented on my opinion of his car.

You did nothing of the sort.
You insulted him, and used his choice of car as your reasoning.
Do it again, or call anyone else here ignorant and I will ban you for a week.

[Nereth]

Originally Posted by GreyGoose006

see, your problem is moderation.
lets say we take your example of a fwd car with a 60:40 f:r weight distribution.
now, lets say that instead of using spring rates as a comparison, we use spring frequency, or hertz. wheel hertz more specifically.
now.

your typical 4dr import sedan has spring frequencies in the 1.3-1.7 hz range.
typically, 1 hz is known as the "brown frequency" and makes people get sick, so it is avoided.

anyway, lets say that for fun, this car has a front spring frequency of 1.5 hz and a rear frequency of 1.6 hz.

this also makes sense as the normal practice for suspension tuning is to make the rear frequency slightly higher than the front frequency so that the cars oscilation front to back will not be un-checked.
if they were both 1.5 hz, the car would tend to rock front to back endlessly. the back is made higher than the front because it helps the rear end "catch up" to the front end after hitting a bump.

now lets assume that since the rear has a higher frequency, it must have a slightly higher roll stiffness.
knowing that understeer is considered safe, an anti-roll bar is fitted to the front of the cars suspension.
lets say for arguments sake that the arb contributes 75% of the roll stiffness of the springs.

Why are you using wheel frequency? To have the rear end 'catch up', you would want to be talking about sprung mass frequency, no? Assuming that is what you meant:

60/40 front to rear and a front natural frequency of 1.5 and rear of 1.6, given that natural frequency=sqrt(spring rate/mass) yields a front spring rate of 135 and a rear spring rate of 102.4. Roll hardness from the springs, assuming equal track front to rear, will be directly proportional to that. This means the rear end takes less of the lateral load transfer, and also is lighter.

Originally Posted by GreyGoose006

this means that if the springs contribute 100 lbft of roll stiffness, the front arb contributes 75 lbft.
i dont know if this is exact, but following your gross estimations, i would not think that this is too far off.

this means that the front now has a much higher roll stiffness than the rear of the car and the car has a natural understeering tendancy built in.

if a rear anti roll bar is desired to add a bit of "sport" to the feel of the car, one may be added.
lets say that the bar on the rear also has a rate equal to 75% of the roll stiffness of the springs.

this means that the roll stiffness of the front is still higher than that of the rear, but the rear is higher, as a percent, with respect to weight.

These assumptions are all wrong assuming my calculations showing front spring rate to be higher are correct. Yes understeer is safer, yes they do want the car to understeer, and yes that means this whole thing doesn't really matter to the average consumer car. Again, I am talking about high performance/race cars, where the goal is to be quite close to drifting. If that is the case, according to my calculations, are large rear swaybar is needed to try to remove the understeering tendency.

Originally Posted by GreyGoose006

again, yes, but you are taking it to the extreme.

http://www.sfxperformance.com/parts/HOT804121.htm

this is a link to a sway bar kit for a toyota scion xb.
it is a relatively well mannered car.

notice that the front sway bar is 1.125" and the rear bar is .5"
i am willing to bet that you could twist that rear bar with your bare hands.
(well not necessarily you, but i surely could.)
if you go and look at kits for other popular import cars, you will see a similar trend.

in the case of the xb, i am willing to bet that the rear bar is nowhere near 75% of the roll stiffness of the springs. it is most likely nearer to 20%

Yes I am taking it to the extreme, but every bit counts. I have a model of a car from a project a while ago. The tyre model comes from some graphs in RCVD (it's not exactly something you can google, so you take what you get). Don't be surprised at the extremely high gees the car can pull, the tyres are basically race tyres and there are no bumps or road contaminants to throw anything off.

Lets take our earlier example car. 900kg front 600kg rear (60:40), with 1.5hz body frequency front, and 1.6hz rear. Toe and camber will be set to 0, and we assume the car doesn't roll, for simplicities sake. Wheelbase will be 100 inches, track will be 60. CGH is 20 inches. We will assume a perfectly rigid chassis. Using the springrates calculated earlier (front 135, rear 102.4), the TLLTD is 56.7% front, 43.3 rear, without swaybars.

In this configuration, the car understeers at 1.093 gees.

If the TLLTD=50:50, the car understeers at 1.103

Trial and error reveals neutral lies at approximately TLLTD=45.74% front, where the car understeers at 1.1071 gees. For a bit of spare drive traction for turn exit, lets pretend that equals 0.46 (it pretty much does, for the accuracy of this program :P). Anyway, to get the 56.7:43.4 to 46:54, the minimum rear swaybar hardness is 23.16 units. Considering that is more than half of the current rear spring rate, I think that is significant.

Anyway, I don't know if you will accept that analysis, since you don't have any reason to trust the program or numbers, but at least it makes me feel a bit more confident.

Originally Posted by GreyGoose006

another thing you are overlooking, especially in the case of a fwd vehicle, is the differential.
the differential, in my opinion, is responsible for more of the way a fwd car handles than the suspension its self.
a lsd with the right tuning (high torque under power, low torque under coast) can give the car LOTS of lift off oversteer.
if a car is understeering, all the driver has to do is give it a shot of gas and lift off while increasing the steering angle and voila, oversteer out whe wazoo.

Lift off oversteer, to the best of my knowledge, will at best temporarily (for a fraction of a second), fix the problem. You cannot expect to be able to get through a long corner, by continuously adding and removing drive torque. If only for the penalty you get every time you add it.

Originally Posted by GreyGoose006

i recommend this program to you
http://www.vehicle-analyser.com/

go ahead and download the lfs version because it includes more car types.

play around with it and you will begin to understand what i, and others here are saying.

Thanks for the program. When I get time I will have a play around with it.

[UncleBob]

Jesus, I hate this kind of "my $500 car is better than your $500 car" BS

With skill, tools and knowledge, I can make ANY $500 car absolutely kick ass in any category. I am so NOT impressed if you prefer one over another for some arbitrary reason

You really think an Iroc can't handle? Do you know where the "iroc" name came from?

thats to slideways, btw

[Nereth]

Originally Posted by KiwiBacon

Lets start with the errors in this paragraph.

Firstly, the roll axis slopes up towards the rear on every vehicle I've looked at. That is, the rear suspension roll centre is higher than the front.
The roll axis is the result of the roll centre, not the other way around.
This is another reason why many production vehicles have sway bars fitted to the front and not the rear.

I had always thought that it was the other way around, but, if you say so I will strike that from my list of arguments.

Originally Posted by KiwiBacon

Secondly, talking spring rates doesn't mean much without including the sprung weight at that end. You probably don't realise but the natural frequency of the suspension matters most and rear suspension is often tuned to have a higher frequency than the front. One of the goals is to prevent see-saw pitching.

I am aware of that. See my post above, even with the rear natural frequency slightly harder, the lack of weight is enough to make the rear springs rate softer in terms of force per distance.

Originally Posted by KiwiBacon

You do not need a "stupidly hard bar" in the rear, I have no idea where you got this idea from, but it possibly stems from you getting the roll axis wrong above.
As I've said, normally you need no sway bar in the rear, if you do need to add one to tip the balance, it's rate does not need to be high at all. One of the purposes of a sway bar is to cause faster load transfer to the outside wheels.
The higher roll centre of the rear wheel already causes faster load transfer without needing additional roll-stiffness to acheive that.

If the roll center is indeed higher in the rear, then admittedly you could soften the sway bar somewhat. However, how much? It would probably go beyond my interest in this discussion to try to make a similar analysis to the above while including suspension geometry.

Originally Posted by KiwiBacon

There's a catch with applying theories, the catch is you need to understand them well enough to apply them correctly. I'm convinced that on this subject you don't.

And I'm convinced I do. Without going through and checking everything, the only new peice of information I ahve been introduced to here is that the roll axis usually slopes upwards rather than downwards. Other than that, all the disagreements have either been 'I have driven a car that drove better than another car', or corrections that winded up not correcting anything (saying the rear wheel needs a higher natural frequency, while ignoring the fact that the rears reduced mass makes that happen at a lower spring rate). Oh, and questions of magnitude, which really have nothing to do with the understanding of the theory.

[Moppie]

Originally Posted by Nereth

Lift off oversteer, to the best of my knowledge, will at best temporarily (for a fraction of a second), fix the problem. You cannot expect to be able to get through a long corner, by continuously adding and removing drive torque. If only for the penalty you get every time you add it.

Ever tried it?
I can tell you from experiance it works, and it works very well.
That experiance both driving a FWD car, and chasing one on a race track.

Your quoting a lot of theory, and lovely results from computer models, but theory can only ever take you so far.
At the end of the day there are always variables involved that a computer model can not test for, all allow for in a realistic way.

I spent the other weekend at the Taupo round of the A1GP, and got to spend a bit of time talking to some of the teams engineers.
They have all sorts of data logging, computer modeling, degrees and experiance to work with.
Yet most of their chassis work was done the same way its been done since people started racing cars.
Tweak, drive it, and see what happens.

[Moppie]

Originally Posted by Nereth

Oh, and questions of magnitude, which really have nothing to do with the understanding of the theory.

Nothing wrong with your theory, just your application of it


If you want to be a really great engineer, then the first thing you need to understand is when and where to apply your theories, and how much to apply them.
You also need to understand that a theory is not a fact, and so is open to being proven false at any time.
Then you need to understand how those theory work out in real life, and just how realistic they are.


An alien with no knowledge of automotive engineering could easily conclude that FWD was better than RWD because there are more FWD cars produced that RWD one's, and be correct in his theory.

[Nereth]

Originally Posted by Moppie

What the rest of us are pointing out, is that it is not always better under less demanding circumstances, and in the average road car, FWD can be preform just as well as RWD, and sometimes better.

I'm aware of that. I am also aware that in adverse conditions (Icy roads, mud, etc), FWD can routinely get more drive traction than RWD.

Allow me to quote myself a few times, in chornological order, and with a few bolded parts to help make my point:

Originally Posted by Nereth

From a performance perspective, it's useless in almost all cases.

Originally Posted by Nereth

And yes, they do handle fine for everyday road use. Obviously we are talking about beyond that.

Originally Posted by Nereth

Sorry if it sounds like I'm making a blanket statement. I know there are FWD cars that handle well. The point I am making is that being FWD isn't helping any.

Originally Posted by Nereth

I am talking about the limits - thats where you are supposed to be for high performance driving, and I thought that was what we were talking about

Originally Posted by Nereth

Drive traction is an issue any time you can spin the wheels. If you are a professional racer, or someone racing a higher end car, it matters a lot of the time. If you are someone racing a small, FWD economy car, or any car without much grunt, not so much.

Originally Posted by Nereth

I have said twice explicitly that I'm not making blanket statements, and many other times implicitly. I know that there are FWD cars that can outhandle RWD cars. The point I am making, again, is that them being FWD is not helping. I think it is very rare when someone decides 'making this car FWD will help lap times'. The only time I can see that happening is low friction surfaces where load transfer doesn't have much chance to take effect, so you can end up with more load over the drive wheels.

[Nereth]

Originally Posted by Moppie

Your quoting a lot of theory, and lovely results from computer models, but theory can only ever take you so far.
At the end of the day there are always variables involved that a computer model can not test for, all allow for in a realistic way.

I spent the other weekend at the Taupo round of the A1GP, and got to spend a bit of time talking to some of the teams engineers.
They have all sorts of data logging, computer modeling, degrees and experiance to work with.
Yet most of their chassis work was done the same way its been done since people started racing cars.
Tweak, drive it, and see what happens.

That is absolutely true. The problem is, those variables should have nothing to do with our discussion. Yes you can make FWD handle very well, but it probably won't be as easy as doing the same with RWD (remember I am talking about the limits), and quite possibly the potential of a RWD car will be greater. There isn't really much advantage to be gained by switching your RWD car to FWD, on the other hand, I think we can all name a handful of advantages for switching your FWD to RWD (assuming, in both cases, you can do some relevant tuning afterwards).

Originally Posted by Moppie

If you want to be a really great engineer, then the first thing you need to understand is when and where to apply your theories, and how much to apply them.

The theories have been proven accurate at a far higher level than what I am using them at now - while it obviously will be innacurate because I am using such simplified models, it is good enough for qualitative discussion. As long as I put enough weasel words in (generally, usually, in most cases), the theories are accurate enough.

The chief advantage of them here is that they simplify the model down so that you can look at FWD versus RWD without including a bunch of irrelevant extras. By using the theories, I got to remove the effects of different suspension geometry, aerodynamics, wheelbase, track, tyre compound, road quality, etc etc.

If the opening question had been 'is a [FWD car] better than a [RWD car]', then the complicating factors become relevant, and the theories at my level become a lot more useless.

[GreyGoose006]

at this point all i really have to say is...
does it really matter that much?

the original question was asked by a NOOB and we are now all sitting here aruguing with more NOOBS who enjoy knocking heads...
lol

[GreyGoose006]

but as for lift off oversteer, it IS a big factor.
any idiot knows that once your tires are sliding as opposed to rolling, they have less grip.
once you get your rear tires sliding slightly a larger radius than the optimal line of the front tires, you can begin to "drift" although you arent applying any opposite lock.
btw, drifting in a powerful FWD car seems fun.
smack a "stupidly stiff" sway bar on the rear, disconnect the front one, and go.
if you get out of control, floor it and it will tighten up your line.
much easier than RWD drifting possibly

[KiwiBacon]

Originally Posted by Nereth

And I'm convinced I do. Without going through and checking everything, the only new peice of information I ahve been introduced to here is that the roll axis usually slopes upwards rather than downwards.

That "new" piece of information is something that could be deduced with 5 minutes of crawling underneath a car to check the suspension geometry.
It's a major mistake and one that shoots down completely every other assumption you've made on the topic.

This quote sums it up quite succinctly.

Originally Posted by Nereth

These assumptions are all wrong assuming

Too much assumption, absolutely no checking of the facts before posting assumptions and misplacing otherwise valid theories.

[Nereth]

Originally Posted by KiwiBacon

That "new" piece of information is something that could be deduced with 5 minutes of crawling underneath a car to check the suspension geometry.

Not really, no. I for one don't intend to put a car on a lift just so I can check that kind of fact.

It's something I picked up somewhere along the way. I had no reason to doubt it and no feasible way to check it.

Originally Posted by KiwiBacon

It's a major mistake and one that shoots down completely every other assumption you've made on the topic.

It only partially reduces the need for a swaybar. Other than that, nothing. Remember I did my calculations assuming no roll axis inclination, and there was still plenty large enough need for a rear roll bar.

Originally Posted by KiwiBacon

Too much assumption, absolutely no checking of the facts before posting assumptions and misplacing otherwise valid theories.

You are the one who assumed higher rear natural frequency equals higher rear spring rate. Should I now pretend that all of your posts are based on faulty assumptions?

You are also assuming that because one FWD car can outperform one RWD car, FWD is better.

I have been posting disadvantages of FWD cars compared to RWD... most of the replies comes down to 'its not that bad', which is really quite a weak argument. I would like it if you people could then tell me some of the advantages it has?

Moppie seems to be the only one who understands I am talking in terms of high performance/racing situations. It would be nice if everyone else could understand that before replying as well.

[KiwiBacon]

Originally Posted by Nereth

Not really, no. I for one don't intend to put a car on a lift just so I can check that kind of fact.

It's something I picked up somewhere along the way. I had no reason to doubt it and no feasible way to check it.

It can be checked by sticking your head underneath. No lift needed. Checking such things avoids looking like an idiot on public forums.

BTW all calculations based on a faulty assumption quickly become worthless. I suggest you reread my posts, your interpretation of my comments is not accurate.

[Nereth]

Originally Posted by KiwiBacon

It can be checked by sticking your head underneath. No lift needed. Checking such things avoids looking like an idiot on public forums.

I have stuck my head under cars before. It isn't as easy as you say to eyeball RCH, especially considerin the difference front to rear of the RCHs is only a few inches. What about with strut type suspensions? Are you claiming you can eyeball a line out a couple of meters perpendicular to a strut? And then intersect it and a couple of other lines and points? And do all this within a few inches of accuracy? And then, moreover, while lying upside down, and with a bunch of other components in the way?

I can barely fit my head under the couple of stock-ride-height cars we have over here, and even if I do, there are fuel tanks and the like in the way.

Originally Posted by KiwiBacon

BTW all calculations based on a faulty assumption quickly become worthless. I suggest you reread my posts, your interpretation of my comments is not accurate.

I didn't base any calculations on a RCH that rises as it moves forward. I made assumptions, yes, but being an engineer, you would know that you have to make assumptions if you want to get anything done in a reasonable amount of time. If you want to remake the only calculation I made and include roll axis inclination with it, be my guest. I for one don't have the time.

I would rather you start naming the performance advantages of FWD though, as I asked in my last post.