Do we need suspension?

[WilO]

Dave,

My comment was a poor attempt at humor....apologies. Just thought I'd beat someone to it, that's all.

Wil

[hardingfv32]

"although I recall somebody claiming to have found a solution".... ME ?

Well, of coarse my solution would be to only race on dead smooth tracks.

For my own understanding: Is it correct to say that most of the tire dampening requirements are because of the tires response to road irregularities? Are drive and brake forces going to cause significant dampening issues?

Brian

[DaveW]

"although I recall somebody claiming to have found a solution".... ME ?

Er, no, I'm afraid.

For my own understanding: Is it correct to say that most of the tire dampening requirements are because of the tires response to road irregularities? Are drive and brake forces going to cause significant dampening issues?

It was certainly a requirement that they worked on real circuits, but driver inputs were/are a significant disturbance, I think, & they tend to happen when control over grip is required most.

[DaveW]

My comment was a poor attempt at humor....apologies. Just thought I'd beat someone to it, that's all.

No need to apologise, Wil. I thought I would take the opportunity to defend depleted uranium (or GEC heavy metal, as it was known). I don't think it is any more dangerous than many other heavy metals (lead, for example). Certainly radioactivity was not an issue - we took care to check that - & it was used widely at the time in the aircraft industry for mass balance weights.

[mep]

You only need a suspension when you have a rough track. A dead smooth track does not require a suspension or improve the cornering abilities of the car.... Fact!

The weight transfer formula for an F1 car: weight transfer = weight x cg height / wheel track x g

You will notice there is no variable/input for wheel or spring rates, roll rate or sway bar size, etc.

Brian

I quickly fly over this topic and it seems nobody really went into detail of this statement yet so I will try it. It is correct that there is nothing in this formula representing springs, dampers, antirollbars and tire stiffness but there is a simple reason for that. This formula is just valid for steady state cornering it does not represent the transient condition. So it’s a car driving permanently in a circle with no change in radius or velocity. You probably know that transient conditions require a lot more math but I am sure some guys here can give you formulas for transient conditions.

Now the problem with infinite stiff suspension (or no suspension as you call it even when that is incorrect because you need something to attach your tires to the chassis) you get instant load transfer during cornering. That’s something your tires might not like because friction coefficient drops with load. Your car is faster when your load is equally spread among your tires. Also this instant load transfer might make the car very nervous because it instantly reacts on any input. You ever noticed tail waving of F1 cars when they accelerate? There is some load transferred in an oscillating fashion. Some bad habit because it reduces accelerating performance and has to be controlled trough your suspension.

Then you want tires who don't flex vertically or in any direction therefore infinite stiff but do you know what this causes?
Infinite tire causes the contact patch of a round object to go towards zero, therefore you have no grip.

Also hardingfv32 would could you use the quote function and prevent double postings?
It makes the topic more readable.
This topic somehow reminds me of the "Diffusor Confusion" topic.

[DaveW]

Thanks, mep. Actually, there is rather more. Whilst 4 tyres remain in contact with the road, it is possible to vary the way "weight transfer" (a poor & misleading phrase) is shared between the axles. Carrying more on the front axle will move balance towards understeer, whilst carrying more on the rear axle will move balance towards oversteer. Working out how that can be varied (with "weight jacking" banned in F1) may explain why a suspension is useful, regardless of other benefits.

[hardingfv32]

1) I think weight transfer as measured in lb. or kg can not have a time dimension. Is this not correct?

I am certain that the formula is correct as found in most publications. There is also never any modification made to the formula for transients that I have ever come across.

2) You are absolutely correct, the tires are not going to be happy with any weight transfer (transients?). BUT THERE IS NOTHING YOU CAN DO ABOUT IT outside the parameters in the formula.

I think this has something to do with one of Newton's Laws. When a lateral force is applied (g) to the car I believe it is the tires that must react to it at the same time. There can be no delay.

The fact that there is body roll or shocks resisting the roll has no relevance.

3) I agree the tires bring complexities to the discussion.

Brian

[hardingfv32]

Yes, the load transfer can be reallocate among the 4 wheels, but you can not change the total.

Brian

[Jersey Tom]

1) I think weight transfer as measured in lb. or kg can not have a time dimension. Is this not correct?

I am certain that the formula is correct as found in most publications. There is also never any modification made to the formula for transients that I have ever come across.

Your formula might be valid for some steady state case. Load transfer on a real car DOES have a time response. While the inertial force arises instantly, there is rotational inertia associated with the sprung mass. Load can only be transferred once the force elements (springs, etc) have been displaced.

[mep]

Yes true Dave.
I wanted to write about that too but then left it out or just forgot about it. Maybe we can come back to it later once the transient weight transfer is understood. Stay prepared to correct me in case I do something wrong. Like Ciro always says: We are all here to learn.

1) I think weight transfer as measured in lb. or kg can not have a time dimension. Is this not correct?

Yes but it can change with time. Like your bank deposit. The unit (currency) is Euro, Pound, Dollar or whatever but it does change with time. You can create a graph f(t) where you see how your bank deposit rises or falls with time.

I am certain that the formula is correct as found in most publications. There is also never any modification made to the formula for transients that I have ever come across.

The formula is correct it just does not represent the transition condition. It gives you the load transfer at let's say 4g's. It does not tell you how you reach them. Even when there is a sprung input of your g's (g's go from zero to 4g in infinite short time) you will not get a sprung response of load transfer. Imagine you have a spring standing on your desk and you slam with a hammer on it. The desk will not get the total force immediately (in this case it might never get the full force) because the spring compresses. Now slam your hammer directly on the desk. You will create a much bigger crater on the surface. It is kind of the same as a car cornering just that during steady state cornering the forces stay up for much longer so you will reach your max weight transfer after some time and that is when external forces are in equilibrium with internal forces (springs) and the body roll stops.

The reason why you haven't found the formulas in a book yet might be because transient stuff comes with nasty math which many people don't like. But you can ask DaveW, Jersey Tom or someone else if they can give you better formulas. I would be curious too.

[hardingfv32]

1) Yes, load transfer is measure at a point in time. At that same moment the tire reacts to that same force. Values are equal and can not be delayed.

2) Yes, there is a transitional force. May I assume since it is transitional it is only applicable when the load transfer numbers are changing and not in play when the car is steady state in the middle of the turn when the load transfer become, in a simple case, a constant value.

I am familiar in theory with this "transitional force" as part of my shock tuning efforts. There is nothing in the literature that I have found on this transitional force subject other than it does exist, although I am not sure we are using the correct term.

I am not sure of its magnitude or what might modify its values. I do know that it must be in addition to the normal load transfers values.

It does not do anything to negate my claim that there is nothing that you do about load transfer outside the parameters in the formula.

Yes, someone teach use something

Brian

[ubrben]

I'm midway through a 24 hour race and haven't slept for over 30 hours, but somehow I don't think that's the only reason that post sounds like complete bollocks...

Ben

[DaveW]

... Yes, someone teach use something.

From this thread two pages ago:

I suggest reading up on simple handling models - bicycle model, 4 corner rigid body dynamics models, sprung models - and revisit this thread later once you have a better understanding of what's important and what drives the 1st order effects.

Can I suggest "Race Car Vehicle Dynamics" by the Millikens?

Yes, the load transfer can be reallocate among the 4 wheels...

Thanks. Now, have you thought about how the load distribution can be adjusted without a suspension?

[hardingfv32]

I am game and have the book out.

My question would be: Does load transfer in response to lateral force (g) in a instantaneous fashion.

There is a lot on load transfer but it all seem to be steady state. There are transient stability and control discussions but i do not see it tied to load transfer.

How should I refine the search? What labels or terms would be helpful?

Brian

[hardingfv32]

"Thanks. Now, have you thought about how the load distribution can be adjusted without a suspension?"

I assume that the assumption is that we have an imbalance that needs correcting. While I agree traditional methods are handy to use, they are not the only option. We can change tire size and compounds, camber, toe, weight distribution, and any number of aero tools.

Brian