You want great handling? Look at race cars!

[jeffreylyon]

Here you go: https://www.racingaspirations.com/mods/n0so4nkj/. You'll see that a 25mm drop in ride height results in a 67mm drop in the roll-center.

 

[Dialcaliper]

Here you go: https://www.racingaspirations.com/mods/n0so4nkj/. You'll see that a 25mm drop in ride height results in a 67mm drop in the roll-center.

Nice calculator. I’ve done the same calculation in a big honking spreadsheet and come up with similar numbers based on my own measurements off my car.

For the Mountune springs up front, which have claimed 1.1” drop front, 0.8” rear - my car ride height measured close enough), for the 28mm drop in ride height, I get a 74mm drop in front roll center, so not too far off. The actual CG to roll axis also depends on the rear geometry (for a torsion beam, roll center is effectively at the beam neutral axis (approximately center). Net moment arm (ch to roll axis) increases by 5%, meaning increased kinematic load transfer. Also, inclination of roll axis increases (lower in front), which can produce slightly delayed turn-in response.

I have since removed the Mountune in favor of some DIY custom 225F/200R spring rates at near stock ride height. It’s produced pretty decent results, at least until my longer term custom Bilsteins are finished with custom valving.

Basically, lowering the rear axle *is* beneficial for a torsion beam, it’s the front that sucks struts. Ford in fact knew what they were doing - from base model Fiesta, ride heights for the ST were lowered a measly 0.18” up front and and a full 1.0” in the rear.

But what’s far worse in Macpherson strut cars is that until now we’ve been just talking about static roll center (car at rest). Once a degree or so of roll occurs, the instantaneous roll center actually migrates downward, especially if your control arms go near or below horizontal. Stock ST, they sit about 6 degree downward slope, which is not yet terrible. Lowered 1.1” on the Mountune springs, the control arm angle is reduced to 1 degree slope, meaning the effective roll center drops rapidly as soon as the car rolls even slightly (dropping the inner control arm below horizontal)

 

[Dialcaliper]

For everybody that made it that far I tried to find some numbers for non aero cup cars in old homologation papers (2012-2016) that have spring rates, ofc they also have lowered CG and weight reduction. The stock spring rates I quickly googled so not sure if they are 100% correct

Swift: 400 lbs/inch, 900kg (Öhlins Cup), Stock: 140lbs/inch
M2: 800lbs/inch, 1400kg (KW Comp), Stock: ?
Clio: 450-620lbs/inch, 1100kg (Öhlins Cup), Stock: 180lbs/inch

We are currently building another 99 civic for 3h amateur races ~950 kg using ~450 lbs/inch front+rear eibach springs (stock 160/80 so ~3x front, ~6x rear) no rear roll bar, lowered as much as I can get away with, but this one has double wishbone suspension.

FYI, Spring rates mean nothing without suspension geometry (specifically motion ratios). BMW’s especially tend to run inboard springs which requires a much stiffer spring to hit the same wheel rate.

To be clear, I wasn’t saying that lowering any car is never ever beneficial. What I’ve been getting at is that the Fiesta in particular has fairly compromised front strut geometry due to engine bay packaging directly over the control arm pivots, and lowering the front axle more than about 0.5” can begin to be detrimental. Rather lowering in the rear is limited by other considerations, and is limited by the front axle due to a desire for slight forward rake in ride heights

The rear torsion beam does behave as you describe (roll center barely moves with lower rear ride height). Double wishbone cars can get away with more, but from roll center behavior and camber curves.

Heck, even other strut cars aren’t as bad as the Fiesta and can tolerate running a bit lower The M2 being a prime example as long as you don’t overdo it.

 

[Rutschpartie]

Here you go: https://www.racingaspirations.com/mods/n0so4nkj/. You'll see that a 25mm drop in ride height results in a 67mm drop in the roll-center.

Your example already starts with negative LCA's and pretty low roll center, this car is already lowered a lot to begin with (which is exactly what you are arguing against) so not sure why you choose this scenario?

But OK let's look at your example:

assume static camber/caster/toe is dialed in.
1 and 2 stand for before and after lowering another 2.5cm, or your close to stock vs my stiff spring rate
assume CG 40cm (if you measured yours to be different please correct me) and 2x the spring rate of what @Dialcaliper recommends:

CG1 = 40cm
CG2 = 37.5cm

R1 = 300 lbs/inch
R2 = 600 lbs/inch

Moment arms (distance CG to RC)
D1 = 40cm
D2 = 37.5 + 6.7 = 44.2 cm

Roll will be greatly decreased:
Moment arm and therefore roll moment increases by only 10% BUT spring stiffness increases 2x!!!! So the effective roll acceleration will decrease by almost that factor ~2, and this is not even considering the increased damping force you need to handle the springs.
This is no suprise (for me at least)
Yes the dynamic roll center will decrease even more, but the moment arm length will stay below this factor 2 before it bottoms out in the applet you linked.
(wheels + motion ratio i will ignore here, very likely, like for all mcpherson small cars it will be >0.9)

The dynamic camber change here we ofc also have to look at, because you argue it's influence on grip will be detrimental on grip levels, let's assume a high lateral force turn with maximum static compression mid turn (so you are on/close to your bump stop). Stock static camber is rly not what you want if you want grip but ok...

Suprise again! You can also see in the calculator you provided that it is essentially the same dynamic camber change you traverse (we did not change more than springs) BUT with 2.5cm lowering and stiffer springs you only traversed the smaller portion of it, and not even hit the bump stop (see above roll ratio).
You have even more negative camber at maximum compression for that turn compared to setup 1

-> you have the full advantage of your lowered CG and the decreased total load transfer

Now some of you will argue: but since the RC is lower and the moment arm is longer, total load transfer will increase a lot! Not so fast, this is the mistake that many make.
You can adapt RC heights to balance load transfer between axles, same with stiffness, but not total load transfer.
I am still trying to find a good/simple explanation. A detailed one can be found in the book "RACE CAR HANDLING OPTIMIZATION, Ralph Pürtz".
Maybe citing this article is good enough: "In other words, by moving the roll center we can significantly change the rolling behavior of the car, but we cannot affect the total amount of weight transfer." (https://www.beyondseattime.com/weight-transfer-vs-body-roll-part-1/).

Toe/ackermann I will not look at, I don't have the tire data to decide how much is needed but it should be a similar picture to above.
Maybe it is enough to know that there are people who love static duck stance for fast steering response and accept decreased mid corner grip, others prefer the other way round and this can dominate ackermann% effect mid corner. In reality it will also dominate camber for the drivers feeling.

But now you could argue again for uneven asphalt this 600 lbs/in spring rate will f*** up the tire contact patch, because of this magic 2-2.5kHz i found on the Internet! Yes, maybe on Nordschleife curbs, but not for smooth asphalt tracks/roads. But again, if your roads are bumpy and broken, stock is what you want.

@M-Sport fan: I was able to dig up some number for the front spring rate of the Rally1 Yaris WRC (1260kg) used for a dry CER asphalt stage, it is 120N/mm so around 700 lbs/inch. Stiffer than i expected, but I am sure MR is lower than for the stock Fiesta and the yaris has some aero.

Now this was definitely my last post trying to get my point across, if this and the real world spring rates I provided do not convince you (heck, you even doubt the fiesta cup cars I posted with low CG), stay close to stock and enjoy the comfort

 

[Rutschpartie]

FYI, Spring rates mean nothing without suspension geometry (specifically motion ratios)

I am painfully aware of that (nonlinear complicated MR's do not make it easier to look at dynamic properties), but also pretty sure that for 90% of the small mcpherson cars including the Fiesta you can assume a MR >0.9, so if I talk about >300% spring rate increase I think you will understand that I do not consider it. Also so far I was only talking about static mid corner grip ignoring the dampers and compliance completely, ofc you need to consider them for everything dynamic.
For the BMW you could be right, 800 lbs/in looks like a high number.