Advantages of not using rear corner springs

[Crazy Bored]

Why would you want to do this? Specifically, the Williams FW31 as modeled by iRacing does not have any corner springs (or torsion bars, whatever you want to call them) on the rear suspension. It does have corner dampers, third spring and damper, and an inerter. Everything as I expected, other than the lack of torsion bars.

The front suspension DOES have torsion bars.

The exact quote from iRacing staff:

Eric Hudec wrote:
The iRacing FW31 has no rear corner springs, as per Williams. Roll stiffness is controlled exclusively by an anti-roll bar, and ride stiffness by a heave spring.

I just don't know what factors influence whether or not you could sufficiently control ride and roll without corner springs. Maybe there is very little travel in the rear suspension, it seems to travel a lot less than the front does. I have no idea, I was hoping someone here would.

Also, aren't these torsion bars in this picture of the FW31? I guess it would be easy enough to try running the car with and without them anyway.

[DaveW]

Why would you want to do this?

Ignoring motion ratios, non-linearities, & the odd factor of 2, the symmetric (heave & pitch) stiffness of a three spring & arb layout is equal to the 3rd spring + the corner springs. The antisymmetric (roll & warp) stiffness is equal to the corner springs + the arb. If the arb design & installation is adequate, then the corner springs can be removed with no loss in control - in fact there is a minor advantage because the symmetric & anti-symmetric components of stiffness become decoupled. The advantage is minor, however, because the dampers remain coupled. A 3rd damper helps, but that often means an inerter cannot be fitted. The spring arrangement you describe (no corner springs) is not uncommon, although I have not seen a layout where corner springs cannot be fitted (hence your photo does not necessarily prove the presence of corner springs).

[DaveW]

It does have corner dampers, third spring and damper, and an inerter.

Out of idle curiosity, have you compared performance with a front inerter, rear inerter, both, neither. Did they have any effect & what values of mass did you use?

[Giblet]

OMG OMG

I can't contain my enthusiasm. I am downloading the season 4 update from iRacing which includes the FW-31

I will post soon on my impressions.

[Jersey Tom]

If the heave spring is stiff enough, well... there ya go.

Or hell, they may be trying to get some rear end movement. Increase chassis rake under the brakes? IME, front is much more ride height sensitive than the rear.

Just some thoughts.

[DaveW]

If the heave spring is stiff enough, well... there ya go.

A 3rd spring is fairly efficient, JT. With unity motion ratios everywhere (specified in the conventional way), a 50 N/mm 3rd, for example, is equivalent to 100 N/mm corner springs.

[autogyro]

I would love to see a comparison between the 'hard' McLaren and the 'soft' rb6.
That would give an up to date take on suspension component choice and the reasons behind such choices.

[Crazy Bored]

IME, front is much more ride height sensitive than the rear.

Yea, so could this possibly be one reason for still using torsion bars in the front? In the sim they range from 50-150 N/mm, while the front heave spring is 20-60 N/mm. As far as I know, these are realistic numbers. For comparison, the allowable rear rates are 200-400 N/mm. I just wonder, what would be the problem with increasing the front heave spring enough so that the corner springs weren't needed?

As far as inerters go, the default setup started with no mass in the front and quite a bit for the rear. It isn't terribly fast though, and since I added mass to the front inerter it's been more stable everywhere, especially over flat, but bumpy curbs. It's pretty much impossible to handle high curbs, which makes sense. I have 50 kg in the front and 20 in the rear, but I haven't tested inerters very much. It's taken a while to get comfortable pushing the car. You can use anywhere from 0-100 kg in both inerters.

[DaveW]

A couple more questions, if I may. Do you know motion ratios, & does the model allow bump rubbers & packers?

On the face of it the front 3rd spring range looks to be low, & the rear high. The car should be fun to drive if you remove front "primes" (mechanical o/s, although the diff should be all-powerful). Interesting comments about inerters. I would expect you to have to decrease damping with inerters fitted.

[Crazy Bored]

Unfortunately I don't know the motion ratios, I was wishing that had them so we could really compare the front and rear spring rates.

To my surprise there are no bump rubbers of any kind on the car. It's not a limitation of the sim either, since other cars have them. So I don't know what the story is there.

Some other things to think about, depending on the track the static weight has been around 48% front, and the downforce is also around 47-48% front. Seems to make sense with the wide front tires from 2009. Something else I found interesting, on very low downforce setups the L:D ratio is very close to 4:1.

I'm not sure what you mean by front "primes?" Or mechanical o/s. I've never heard those terms before.

I'll definitely have to try different combinations of damper stiffness and inerter mass.

[riff_raff]

With regards to suspension structures, every component is a "spring", since each has a specific elastic stiffness. Whether tires, a-arms, pushrods, rockers, chassis, etc. If the combined deflections of these elements is more than is required by the chassis, then having an additional "torsional spring" in the system is of no real benefit, right?

Top fuel dragsters get all of the suspension movement, weight transfer, and gear ratio they need to go 320mph without suspension springs or dampeners or a transmission.

riff_raff

[DaveW]

To my surprise there are no bump rubbers of any kind on the car. It's not a limitation of the sim either, since other cars have them. So I don't know what the story is there.

Mmm... No front bump rubbers & low rate springs... Perhaps they have an alternative fixed spring path(s), &/or an extreme "rising rate" geometry. Simple calcs & track video clips suggest that the effective front "heave" spring rate must approach infinity at high airspeeds.

Some other things to think about, depending on the track the static weight has been around 48% front, and the downforce is also around 47-48% front. Seems to make sense with the wide front tires from 2009.

I like your logic. C.g. position will be a tyre-driven compromise in any case. I would guess that nobody runs less than 45%.

I'm not sure what you mean by front "primes?" Or mechanical o/s. I've never heard those terms before.

"Primes" = US shorthand for corner springs - or torsion bars. O/s = over-steer. Removing the front primes (with no other changes) implies (usually) that more of the cornering roll moment will be reacted at the rear axle, which will move "mechanical" (or low-speed) lateral balance towards over-steer. Drivers often like that, but it is likely to reduce corner exit "traction" & hence increase lap times.

I'm interested in your conclusions about inerters. Thus far your comments suggest that the simulation is representative. One characteristic that I believe some teams use is that inerters increase tyre heat rate due to surface irregularities. I find it difficult to believe that would be simulated, however.

[EAKMotorsports]

very interesting topic. Anyone have more info about it?

[Richard]

Apparently Mclaren tried omitting rear springs at Abu Dhabi. Is that the same as the Williams arrangement noted above?

It would be interesting to hear more about this approach.

[DaveW]

It would be interesting to hear more about this approach.

I'll have an attempt.

Basic spring elements on the rear suspension of most F1 vehicles comprise "corner" torsion springs (Kc), a "3rd" coil spring (Kt), an arb (Ka). In addition, non-linear elements such as bump rubbers &/or bellevilles and packers can be added to the "corner" dampers and the "3rd" element (slider, damper or inerter). Just to make life interesting, suspension geometry can be non-linear (usually "rising rate") & the degree of non-linearity can be different for corner elements, the 3rd element & the arb.

To keep the discussion simple, ignore non-linear elements, assume geometry is linear, and consider the overall axle stiffness to have a "heave/pitch" component (Kh) & a "roll" component (Kr). Then

Kh = G1*Kc + G2*Kt (i.e. is a linear combination of the corner & 3rd spring stiffness).

Kr = G3*Kc + G4*Ka (a linear combination of the corner & arb spring stiffness).

It follows that, when all elements are present, Kh cannot be less than the larger of G1*Kc and G2*Kt, whilst Kr cannot be less than the larger of G3*Kc and G4*Ka. The common element is Kc. Minimum values for both Kh & Kr are prescribed once a value of Kc has been selected. Further, any change to the value of Kc will change both Kh & Kr.

Normally it is convenient to be able to change Kh & Kr independently when "tuning" a suspension set-up. This can be achieved by restricting changes to the 3rd and arb elements, which is fine provided than neither is required to be < 0. The most flexible arrangement would be to remove the corner springs entirely, & this will work provided that sufficient ranges of values are available for both the 3rd and arb elements, and the "installation stiffness" of both elements is adequate. From the latter viewpoint, the most efficient structural arrangement is often to share loads between all elements.