I'm not aware that a swaybar does anything useful when traveling straight ahead.

ISTR that when the highway up Pikes Peak was paved only with gravel, the fast racers would remove swaybars, or at least use thinner bars. I think their operating theory/guess was that increased body roll would put more weight on the outside tires, causing them to dig deeper in the gravel, forming a dynamic/transient berm and allowing increased corner speed.

If they were right, your answer depends on the surface you're racing on.
If you have a mix of surfaces, maybe you need a dynamically disconnectable swaybar, which you can clutch/declutch with whatever time you can free up from other driving tasks...

Mike Halloran
Pembroke Pines, FL, USA

Mike,
The track is 100 percent asphalt
I had in mind without sway bar, for compensating roll use stiffer springs

I mean:
If we can achieve an acceptable body roll and good corner balance without swaybars, there is no reasons for swaybar use?

Radek

Right, I think. Stiffen the springs enough to limit body roll, and remove the swaybars. Ride quality will suffer, but you don't care about that unless the track is bumpy, and maybe not then, depending on how old you are.

Mike Halloran
Pembroke Pines, FL, USA

Thanks Mike

Quote (sierra)

is a good idea?
with low center of gravity 4WD racing car without swaybars? (maybe only small rear bar for balance)

Adding a sta-bar increases wheel rate - making that end of the car less compliant, resulting in greater variation of wheel loads over whatever suspension movement occurs. It also reduces the "independence" of wheels on a nominally independent suspension. I think years ago it was Porsche whose opinion was that avoiding the need for a sta-bar was a goal to be sought, reasonable perhaps in the days of low, 1500 lb sports roadsters with horizontally opposed engines and no other single item of substantial mass other than the driver.

Powerful RWD solid axle cars may not use a rear sta-bar at all; this is normal at least in oval track racing whether on asphalt or on dirt.

Quote (Mike)

I'm not aware that a swaybar does anything useful when traveling straight ahead.

There is a clear advantage to having a rear stabilizer bar in the very narrow situation of drag racing a car with a solid rear axle. What you're trying to do there is use roll stiffness distribution to oppose driveline torque and equalize the rear tire loading for maximum total longitudinal grip. Rear sta-bars used in this manner tend to be far stiffer than would be sensible for street driving, and may even be coupled with the removal of the front sta-bar.


Norm

OK Norm,
I have now a stiff front spring for travel limitations and 0.7" front sway bar for body roll limitations
Also very stiff rear springs and big bar for understeer compensation - this causing inner rear wheel traction loss during acceleration out of slow corners

I mean ,when swaybars stiffness decrease the same values front and rear - I get better acceleration out of slow corners

BUT - what will be the impact on the corner balance - when I have the front rollcenter low than the rear?
will to more oversteer?
is correctly?

Are you talking about load transfer distribution?

You need to keep load on the inside rear, this means reducing rear roll stiffness or increasing front roll stiffness. Unfortunately this means an increase in understeer.

I have a load transfer calculator that may be of use to you for this purpose?

Yes Komodo86,

This is my load distribution dilemma

How can I get this calculator?

It is here: http://blackartdynamics.com/Chassis/LoadTransferMX.... Best viewed on a desktop/laptop with Chrome or Firefox. It is not Internet Explorer friendly.

Default inputs are Mazda MX5, you'll need to know a fair amount of detail to fill it in for a Sierra, but most of it should be available if you don't already know it.

At the moment the ARB calculation is only a basic straight, right angled arm one, I've not yet got round to angled arms or anything else. Should still be useful as a comparison between two different setups on the same car, just not between cars with different suspension designs.

Ideally you want the LTD being within a few percent of the weight distribution for a RWD car.

Roll angle is defined as that generated by suspension movement alone. Tyre deflection is not accounted for.

Erm, any more questions, give me a shout!

"Ideally you want the LTD being within a few percent of the weight distribution for a RWD car."

This thread is mostly about AWD, but I think a slightly more robust rule of thumb would be LLTD should be biased towards the undriven axle, and towards the front of the car. A good rule of thumb for RWD road cars would be LLTD on front axle = WD+5%

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Quote (Komodo)

It is here: http://blackartdynamics.com/Chassis/LoadTransferMX.... Best viewed on a desktop/laptop with Chrome or Firefox. It is not Internet Explorer friendly.

Seems to work OK in IE11, which was released only very recently . . . and I'm guessing that the unsprung mass entries represent the entire unsprung mass at each "axle" rather than for just one corner.


Norm

I'm struggling with meaning of LTD and LLTD

(Lateral) load transfer distribution.

That is when cornering what percentage of the load transfer takes place on which axle. If the RCH is zero at both ends then it is the same as the roll stiffness distribution.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Greg, yes you are correct on the LLTD being WD+5% for RWD.

On FWD Honda's, they worked best at around 46% LLTD when the WD was around 62%.

I have to disagree on the LTD being the roll stiffness distribution. As you change the lateral acceleration, the LTD changes, the RSD does not.

Norm, good to hear it works on the latest version. I'm not all that bothered about re-scripting it for the older versions, people should be well away from them now but I still need to mention it as there are a few dinosaurs lurking that are still using IE6-9!

The unsprung mass is that in each corner.

How do you quote on this horrible forum software?

Thanks Komodo86 ,
nice toy I will testing it

sorry , but what it means : LTD,LLTD,RSD, etc.

Thank You,
Radek

LTD / LLTD / FLLTD = Front Lateral Load Transfer Distribution. The percentage of the total load transfer which is taken across the front axle. A higher number typically means more understeer bias. Changes with lateral acceleration as the stiffer axle transfers load at a faster rate.

RSD = Roll stiffness distribution. The percentage of the total roll stiffness taken by each axle. This is fixed.

I only know the syntax, but there is an icon for inserting quotes about 6th from the right that I've never tried.

[quot name]paste text here[/quot] . . . misspellings of the word "quote" are intentional else you won't see it. Including the name is optional.


Norm

Thank You,
how to estimate CoG height for calculation?

NO. The TLLTD does NOT change with lateral g unless you have non-linear springs. Its usually VERY constant unless you have some weird camber changes or a direct acting front bar and high steer angles or very high tire MX/FY action.

sierra - fantastically good question. everything else is moot until you know this. (a) wag 600mm (b) measure it (c) work it out from your vehicle roll gradient and measured roll stiffness etc (d) build a spreadsheet model.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Sorry ,I asked the wrong
What is the typical value of the CoG height for example:
WRC car (asphalt - low version)
WTCC car
DTM car
NASCAR
I know that my calculation will be incorrect,but help understand the connection.

I know that an MX5 is 18". The rest is just a guess tbh. I would estimate most saloon road cars to be 20-22".

Quote:

NO. The TLLTD does NOT change with lateral g unless you have non-linear springs. Its usually VERY constant unless you have some weird camber changes or a direct acting front bar and high steer angles or very high tire MX/FY action.

You and I have a very different understanding of FLLTD then. How can it NOT change?

At 0g the FLLTD is 50%. At the point of both wheels lifting, it is also 50%.

At any point between those two extremes, it is dependant on the RSD and the SWD.

In a 50/50 WD vehicle, with a non equal RSD, the stiffer axle will transfer load faster, up until the inside wheel is completely unloaded and lifts off the road, at which point the LT across that axle is 100%. The softer axle will be at less than 100% load transfer as the inside wheel is still loaded, and so the FLTTD at that point is not 50%. Past this point, the softer axle continues to transfer load until it to has transferred 100% of it's static load, at which point the FLLTD is once again 50%.

Please explain how you think it remains constant.

Thanks, I thought that probably is higher than Mazda

Yes, RX8 is also in the 18" region, Porsche Cayman and Toyota GT86 are also very low in the 17-18" range. These are all very low CG production vehicles.

Work from those examples and have a think about where the weight is in your car. A saloon car with a high roofline, large glasshouse, upright engine with heavy DOHC head and perhaps high mounted turbocharging equipement is all going to add to that height significantly.

An upright alloy block with heavy DOHC head is going to have a higher CG than a cast iron block V8 with pushrod alloy heads, a lot of sound deadending removed from the floor of the car will move it up, poly windows will move it down, so on so on.

I understand,
look ,is nothing superfluous

Rollcage is heavy and high mounted though... ;)

Really, the only way to be sure is to determine it with a tilt test, but it's a bit time consuming as you need to replace the front suspension with rigid links to do it accurately.

Quote:

Quote:

NO. The TLLTD does NOT change with lateral g unless you have non-linear springs. Its usually VERY constant unless you have some weird camber changes or a direct acting front bar and high steer angles or very high tire MX/FY action.

You and I have a very different understanding of FLLTD then. How can it NOT change?

At 0g the FLLTD is 50%. At the point of both wheels lifting, it is also 50%.

At any point between those two extremes, it is dependant on the RSD and the SWD.

In a 50/50 WD vehicle, with a non equal RSD, the stiffer axle will transfer load faster, up until the inside wheel is completely unloaded and lifts off the road, at which point the LT across that axle is 100%. The softer axle will be at less than 100% load transfer as the inside wheel is still loaded, and so the FLTTD at that point is not 50%. Past this point, the softer axle continues to transfer load until it to has transferred 100% of it's static load, at which point the FLLTD is once again 50%.

Please explain how you think it remains constant.

Just to expand on this some more.

At 0g, '0% LT', FLLTD is 50%. At point of both wheels become unloaded, FLLTD is also 50%. Lets call this '100% LT'.

Let's look at 10%, 20% and 30% LT.

2000lb 50/50 SWD, with a 60% front RSD.

At 10% LT there is 200lbs LT, of which 60% is taken by the front.

Front LT is 200 x .6 = 120lbs. The outside front tyre has 500+120 = 620lbs load.
Rear LT is 200 x .4 = 80lbs. The inside fear tyre has 500-80 = 420lbs load.

FLLTD is (620+420)/2000 = 52%

At 20% LT there is 400lbs LT, of which 60% is taken by the front.

Front LT is 400 x .6 = 240lbs. The outside front tyre has 500 + 240 = 740lbs load.
Rear LT is 400 x .4 = 160lbs. The inside fear tyre has 500-160 = 340lbs load.

FLLTD is (740+340)/2000 = 54%

At 30% LT there is 600lbs LT, of which 60% is taken by the front.

Front LT is 600 x .6 = 360lbs. The outside front tyre has 500 + 360 = 860lbs load.
Rear LT is 600 x .4 = 240lbs. The inside fear tyre has 500-240 = 260lbs load.

FLLTD is (860+260)/2000 = 56%.

Thus, FLLTD varies with acceleration.

not a lot of added weight with rollcage,
but a dramatic increase torsional stiffness!!really

It might not be a lot in absolute terms, but in relation to the sheet sheets of ally and perspex windows it shares it's space with up there, it will be heavy.

It won't affect the CG greatly, but it will certainly raise it slightly vs having no cage at all.

Given the spartan interior and I'm guessing an iron block DOHC engine, as a total guess I'd probably quote your CG being in the 19-20" region.

Yes ,
but engine is 4.0 V6 (60 °) SOHC mustang cast-iron block+alu heads

Er, god knows then! Tilt test would be best for you, shouldn't be too much hassle looking at the work gone in to the build!

Crude approximations of vehicle CG height have been given as the height to the camshaft for cars with pushrod (cam in block) V-type engines.

More reliably, there is this NHTSA database (the CG information starts on page 10).

http://www.nhtsa.gov/DOT/NHTSA/NRD/Multimedia/PDFs...


Norm

Quote (Komodo)

Front LT is 200 x .6 = 120lbs. The outside front tyre has 500+120 = 620lbs load.
Rear LT is 200 x .4 = 80lbs. The inside fear tyre has 500-80 = 420lbs load.

I think you may be confusing lateral load transfer with the relationships among the resulting wheel loads after the load has transferred.

LLT = Lateral Load Transfer, which is only the 120 and 80 numbers in that particular case.

The only way the distribution of LLT can vary is if there is some nonlinearity involved. The only way that would amount to much would be if said nonlinearity is significant - such as in the case where an inside wheel lifts, or where significantly nonlinear-rate springs are fitted at one end and linear rate springs at the other.


Norm

Quote:

I think you may be confusing lateral load transfer with the relationships among the resulting wheel loads after the load has transferred.

FLLTD = Front Lateral Load Transfer Distribution i.e. how much load transfer has happened at the front. It changes with lateral acceleration.

Perhaps it should be called 'Front Diagonal'.

Whatever it is called, the differential between front and rear load transfer is not constant.

Of course the load transfer itself changes with lateral acceleration. But the static snapshot of how it is distributed front vs rear is not, assuming that there are no nonlinearities in the geometry or stiffness terms.

You'll note that your 120/80, 240/160, and 360/240 load transfers are all in a 60/40 distribution.

Adding outside front and inside rear cornerweights and expressing that as a percent of total car weight is crossweight or "wedge" in circle track slang, and this will of course vary.


Norm

Ok, I had the wrong terminology there then, thanks!

Thank You all for responses!!

Radek