Spring Rates & Sway Bars (road racing)

[gmx]

Originally Posted by berns

<snip>
1:55 - Oh boy, this is just embarassing. It's really difficult to get this thing stopped in time on street tires with aggressive pads, 3200lbs, 133mph top speed into this super difficult turn. USUALLY, I turn in early, clip the curbing, bounce into a power-on oversteer, lose time and cross the finish. This time, I couldn't get the car slowed and turned in time, which resulted in me hopping off the brakes and turning in, and praying for enough front grip not to go off, and to finish the lap. The car understeered here because of me, I didn't fuck the turn up because of the understeer, if that makes sense.

Will continue playing around with damper settings, going to add more front aero, will try stickier tires soon too (A052) and go from there. If the car is still not doing what I want, I'll see about softening the front rate some.

Double also -- going to swap to E9x M knuckles up front too.

I notice as the car is stiffer, you cannot throw it around like on OEM wheel rates. Got to be precise with inputs or stuff like that happens. You'll love A052, except they wear as fast as PTRs lol.
Please post weights of the E9x knuckle if you get a chance It would help to figure out unsprung/sprung weight then really calculate wheel rate and hence frequency. By my estimations we're all running way over 2.5hz on the front. I have some theories for this and none of it is based on speed

On to my setup:
-3.4F camber, 3mm toe out
-2.5R camber, 2.5mm toe

570/800 Bilstein CS, nothing else changing or modified as per how it ships. Wanted to run it how it is before attacking and changing rates.
In addition to standard M bits:
-Solid subframe mounts
-Spherical upper rear arms. Basically the entire rear end is solid/spherical mounted except the RTABs (do that later), inboard camber arm bush. The upper shock mount is Powerflex Black.

I've found a some springs that'll work in the front (lengthwise) from 450lbs and up and am contemplating swapping those in. Would like to avoid slapping on a swaybar but it might be inevitable.

This lap is pretty bad and running a range of street tyres with way different wear/usage rates.

This one I ran with different rebound/comp settings on the LF shock and didn't realise. I wouldn't be looking at the braking characteristics and exits from right-handers Only mid corner stuff.

[bbnks2]

Originally Posted by gmx

I notice as the car is stiffer, you cannot throw it around like on OEM wheel rates. Got to be precise with inputs or stuff like that happens. You'll love A052, except they wear as fast as PTRs lol.

I wouldn't be looking at the braking characteristics and exits from right-handers Only mid corner stuff.

I am assuming you are talking about the car being loose on exit. All of the places where your car is loose you are on the throttle. Sometimes very early in the turn like @ 1:19

[gmx]

Yeah, my settings were off on the LF damper. It's inverted and I was adjusting it by the number of clicks blindly. I got that wrong and also spun the wrong adjuster trying to grip it without raising the car (shoved my arm in there). It's a bit stubborn to turn. It ended up softer than the rest, you can tell where it's most obvious as the car veers braking for T1 from 140~ at the end.

That lap was not clean, it was the first session I used Sport button where the tip-in throttle is much more sensitive before it flattens out. It's a custom throttle-map, I wasn't used to immediately yet.
As for my feeling, it's catchable but I feel the front is way too stiff. There's not much you can do about this diff once it starts to lock-up except replace it

[tsk94]

Originally Posted by cwlo

He lists pricing on the website. Anyway, from my perspective, there are so many variables with suspension tuning that getting a turn key solution designed exactly to my needs has a lot of value, and will probably be cheaper in the long run.

Its also interesting reading through the threads on this forum, and also FCM's video's, that much of the stuff out there is tuned for understeer, or other characteristics that may feel fast, but are really slower than an optimized neutral setup.

Once I get the spreadsheet, I'll post it, and see if Shaikh can comment on its design. He did one for the e46 chassis, and there's a thread out there with good explanation.

I spent a long time researching suspension options for my E92 race car build, for a while I was strongly considering FCM but ultimately went with KW on a custom competition setup.

Shaikh is great guy, and knowledgeable. After all my research and experience with my own setup, I have to disagree with his 'flat-ride' approach to an extent. The caveat being I'm talking strictly track or race application, on a car that sees driving on the street this may not apply to the same extent.

Firstly, FCM's idea of having a higher rear frequency is balanced out by the fact that Shaikh often is designing these around using a big front sway bar and either a small or no rear sway bar. This creates more understeer then the frequencies themselves represent, as sway bar rates aren't taken into account with frequencies - they however are represented in roll-couple %'s but that's another discussion. So while other aftermarket setups may seem more 'understeer' biased based on their stiffer front spring rates, this must be taken with a grain of salt as an FCM setup, with relatively soft front spring rates to maintain a lower front freq relative to the rear, relies on a lot of spring rate coming from the sway bar to reduce the oversteer characteristics of a rear biased freq ratio.

There is also some good published research examining frequencies for race applications. Basically the conclusion being; while a flat-ride setup (like FCM does) provides better ride characteristics, for race applications a higher front frequency will perform better. In my case, with my current setup, I'm running ~3.0hz front and ~2.1hz rear. This setup initially was very oversteer biased before tweaks to alignment, ride height and damper settings was made. The setup is now very neutral. My point is, frequencies only tell part of the story. The final, and main reason why I didn't end up going with FCM was the lack of adjustability. I could go on about this, but for a race or serious track car, an adjustable damper is necessary - there's no way around it.

If you do some research, you'll find that almost all modern professional GT cars with non-extreme amounts of aero, are running higher front frequencies. If it were faster to do otherwise, professional teams would be doing so.

Lot's of people seem to be happy with FCM products, and I'm sure you will be as well. It's a different approach that can definitely be made to work well, but in my opinion it's not ideal. And again, proper suspension setup goes far beyond just frequencies, but that's primarily what this thread is discussing.

[gmx]

I agree with you to an extent. Frequency primarily depends on what rubber you're using. I presume you're on slicks and nothing less? Oversteer is very vague. I recall you had oversteer in entry/braking, the rest was good?

FCM relies on the big bar soft spring theory. Nothing really wrong with it neither new with it actually.

I have noticed since the E36 generation with "professional" race cars the driver applies such input and the the front end of the car visibly oscillates and even skips along the surface. I see this a lot on American race-prepped E36s but I do not know anything about the condition of rubber, track, ie. surface type etc. The tyre, roll-centre, surface type are key variables regardless of suspension, specifically front/rear spring rate coupling. I've assumed they don't know wtf they're doing and have slammed the front for more static camber while ruining or not correcting the roll-centre. Who knows. Or maybe crap tyres, brash inputs etc - a lot can cause that.
Anyway, I NEVER see such things on raeder motorsport (Manthey Racing), schirmer cars lapping the ring. You can see this slight behavior in my own car at 0:38 in the first video.

I agree with cwlo , much of the stuff is tuned for understeer. These brands have a reputation to uphold. And to add, most of them copy eachother without doing any engineering. More than we think imo.

@tsk94 To get your frequencies, you have unsprung weight for the M3 noted down, right? Would you mind sharing them, specifically the front/rear knuckle and rest of the factory arms.

[tsk94]

Originally Posted by gmx

I agree with you to an extent. Frequency primarily depends on what rubber you're using. I presume you're on slicks and nothing less? Oversteer is very vague. I recall you had oversteer in entry/braking, the rest was good?

FCM relies on the big bar soft spring theory. Nothing really wrong with it neither new with it actually.

I have noticed since the E36 generation with "professional" race cars the driver applies such input and the the front end of the car visibly oscillates and even skips along the surface. I see this a lot on American race-prepped E36s but I do not know anything about the condition of rubber, track, ie. surface type etc. The tyre, roll-centre, surface type are key variables regardless of suspension, specifically front/rear spring rate coupling. I've assumed they don't know wtf they're doing and have slammed the front for more static camber while ruining or not correcting the roll-centre. Who knows. Or maybe crap tyres, brash inputs etc - a lot can cause that.
Anyway, I NEVER see such things on raeder motorsport (Manthey Racing), schirmer cars lapping the ring. You can see this slight behavior in my own car at 0:38 in the first video.

I agree with cwlo , much of the stuff is tuned for understeer. These brands have a reputation to uphold. And to add, most of them copy eachother without doing any engineering. More than we think imo.

@tsk94 To get your frequencies, you have unsprung weight for the M3 noted down, right? Would you mind sharing them, specifically the front/rear knuckle and rest of the factory arms.

Yeah, the oversteer was primarily entry-mid corner. All of that has been sorted since now I'm experimenting with tire setups and tweaking damper settings now that the alignment and ride heights have been sorted. I'm running R-comps and scrub slicks.
Yes I have all of the unsprung weights. With that being said I'm not running any factory arms, all have been replaced with spherical arms. These vary a decent amount depending on my wheel and tire setup but ~110 for the front corners, ~100 for the rear (total). I have the individual weights from when we built the car and I added them up, but that's buried somewhere in a pile of paperwork..

I'm not saying that FCM's approach cannot work, I hope I didn't come across that way. But, from my experience, the vast majority of well-setup track and race cars are running higher front freq. And beyond that, you could make a 'flat-ride' approach setup understeer biased with things like alignment and ride heights.

Also to add onto the soft spring, big bar theory thought. I agree this is nothing new. TC Kline implies basically the same principle, the difference being he doesn't try to maintain a higher rear freq compared to the fronts.

I will agree that a lot of the street or semi-trackable stuff is very likely understeer biased. This thread pertains to track spring rates so that theory doesn't hold as much.

[tsk94]

Those unsprung weights are just off the top of my head estimates. I have the actual weights somewhere..

[cwlo]

Good info Tsk. I'm still in planning with Shaikh, so I will be interested to hear his response. My application is primarily autocross, with the ability to drive to the event without my fillings falling out, so a 3hz front might be a bit much for my application. Nevertheless, I appreciate the constructive criticism.

Gmx, I've got a bunch of suspension components, including a whole rear setup, so I can provide some weights or measurements. I've also got an F80 front spindle and hub I can weigh which is very simiar to e90 m3. Also, fe1rx had measurements of 115 front, 121 rear for unsprung weight on his 135i.

[gmx]

Actual would be good to disregard wheel/tyre as that'll be different across the board. I took these off a French guy (M3/1M) in kg:
Strut Assembly 32 (Not sure what exactly includes)
Fr Brake disc: 10.6
Rear Brake disc: 9.5
Caliper: 7.2
Rear Caliper: 5.6

I realise some will have a BBK etc.

Foolishly while I was replacing all my bushes and had the opportunity to measure arms while also doing my suspension - I forgot...

[tsk94]

Originally Posted by cwlo

Good info Tsk. I'm still in planning with Shaikh, so I will be interested to hear his response. My application is primarily autocross, with the ability to drive to the event without my fillings falling out, so a 3hz front might be a bit much for my application. Nevertheless, I appreciate the constructive criticism.

Gmx, I've got a bunch of suspension components, including a whole rear setup, so I can provide some weights or measurements. I've also got an F80 front spindle and hub I can weigh which is very simiar to e90 m3. Also, fe1rx had measurements of 115 front, 121 rear for unsprung weight on his 135i.

No problem, keep in my mind I was talking from a track perspective. I realize setup for autocross will vary in comparison, certainly a a looser (more oversteer biased) setup can be be made to work better in autox compared to track use.

With that being said, I've driven my car on the road 2-3 times and the ride is surprisingly subtle. This setup rides kerbs at the track better and with more compliance than it did when the car was totally stock with stock spring rates. A true high quality damper can do amazing things.

gmx I believe the weights I have are with my front dampers installed in the hubs, so I don't have the hub weight alone. I will check but I believe when we weighed it it was together. And as I mentioned above I'm not using the factory arms, SPL up front and ground control in the rear.

[ShaikhA]

Greetings, all - and thanks to cwlo for point me to this thread. I'd like to add a few point re: the tuning approach I've come to for optimizing suspensions.

This thread is long but packed with good information, great conversations, and even more importantly, the feedback of E46 M3 owners who tried Flat Ride and uniformly preferred it to the pitch-based setup:

http://www.m3forum.net/m3forum/showthread.php?t=601913

and this thread has some feedback from a few of those driver including a great few posts and on-track videos (by 'bigjae1976') on his M3 track experience with pitch vs. Flat Ride:

http://www.m3forum.net/m3forum/showthread.php?t=611348

Bigjae's post #4:

http://www.m3forum.net/m3forum/showp...95&postcount=4

and post #7

http://www.m3forum.net/m3forum/showp...46&postcount=7

===

From a very simple perspective, Flat Ride is the most efficient way to make a vehicle settle quickly when experiencing a ride disturbance (straight-line). The front suspension experiences a bump or dip first, then after crossing the wheelbase, the rear suspension experiences it. To prevent pitching, you'd want the rear to 'catch up' to the front and in numerical terms, that means making the front ride frequency lower than the rear. This has been done widely on passenger and even sports cars for many decades (many BMWs, all Porsche, Corvettes, etc.). The exceptions prove the rule.

Another benefit of Flat Ride tuning is that a higher rear ride frequency than front creates a more rapid initial yaw moment (rotation / turning) of the chassis. So the rear is PUSHING the front into the turn, instead of a stiffer front DRAGGING the rear through the turn. You can feel this in your body - even my last g/f noted from the pax seat that the car felt like it was rotating from the rear. This makes a car more responsive (when combined with other variables like alignment, bump stops, etc.) to driver input, reducing understeer which the factory also designs into every production car and we end up fighting on most aftermarket / track setups.

Strangely, I do see some OEs tuning for pitch these days (and it seems BMW started doing this a few years ago on some M models) which I believe is to give the driver a 'sportier' experience without actually increase performance as much as a Flat Ride-based setup would have. Pitch makes you THINK you're going faster than you are, because your body is moving around more - but if your body is bouncing, so are the tires ... how is that good for grip, or lap times...?

===

In terms of the contribution of springs vs. sways, when you do the roll stiffness calcs, you'll find that essentially ANY strut-based, RWD sports car will have the front bar contribution >50% of the total roll stiffness. This is even at the higher ride frequencies (> 2 Hz) for 'racing' setups. Even at 3 Hz front frequency you're likely under 25% front spring contribution to total roll stiffness vs. 45-50% for the front bar. So no matter whether your ride frequencies are oriented toward pitch or Flat Ride, your front bar is probably TWICE as important for roll control. These specific calcs are for a track-driven E46 M3 but the percentages would likely scale for a 1 Series. I have done some initial calcs for cwlo's setup and also looked at other setups posted in this thread for helpful reference.

I'm not an expert on the nuances of BMW geometry for anti-squat, anti-dive but it seems they design in a reasonable amount of each (at least on my E46) so once you get into the 2.3-2.5 Hz range you're really not gaining as much front anti-dive with VERY stiff front springs as you might be losing in terms of increasing tire contact patch load variation, reducing total grip and stability at the limit.

The winning E36 and E46 NASA ST cars (throughout the season) I've been working with all eventually (through a bit of resistance!) switched to Flat Ride, with frequencies well above 2 Hz, and using a fair amount of aero on wide sticky R-compounds. I've got about a half-dozen drivers I can point to who've felt the improvements when you optimize the suspension around Flat Ride.

===

If you're getting into LMP territory then aero grip likely becomes a more dominant factor so ground clearance concerns dictate what springs you'll need to run. I fully accept Flat Ride isn't the answer to everything but on a production-based competition car, I've seen more proof that it's a better foundation to build on vs. pitch. And honestly, saying 'others do it' is not an argument for the correctness, or optimization, of an approach. There may be valid reasons in some situations to use pitch vs. Flat Ride, but those would need to be backed by before-after testing, not argument to authority. There were PLENTY of skeptics in the E46 M3 community before I explained my reasons and they went and tested! I always base my conclusions on real-world testing along with applied theory, and I hope you all do the same. That's the true nature of science, so you're making improvements not just changes.

A slightly higher rear frequency doesn't by itself make the car undriveable - the factory tuned it that way on most BMWs to begin with. It's how you take the other variables into account for high-performance / competition situations that helps balance the vehicle. The damper tuning really matters, along with the damper gas pressure (esp. for the monotubes many of us prefer), then sway bars, bump stops, alignment, ride height, tire pressures, etc.

===

This variable doesn't get nearly enough attention: how your dampers are dynamically influencing understeer/oversteer. Many aftermarket (and a lot of OE) dampers use linear rebound vs. digressive compression which allows the suspension to 'jack down' over a series of chassis movements. This usually results in bump stop contact up front and will cause terminal understeer. Or, if you have more rebound bias in the rear, or insufficient droop travel, you could get rear bump stop engagement / inside wheel lift from the jacking down and oversteer. Sometimes you'll get both understeer and oversteer in the same corner - this happens far more often than many realize. I've focused on providing very neutral damping up front and a *slight* rebound bias in the rear. This both helps counter-act the factory-designed methods for understeer (esp. less front camber gain vs. rear in bump) and also lets a competent driver safely and quickly drive a car to its limit when the environment and reason allow for it.

It's very important to realize that how the dampers are tuned will influence the 'zero point' of the suspension, the effective ride height while driving. Rally cars uses damping to generally bias the suspension upward, NASCAR biases downward - where should we be? It depends on your driving style and roads, but an excess of either compression or rebound (or gas pressure) will cause unintended effects that will make a car more unstable and harder to keep on the edge, preventing you from getting the maximum possible grip from the tires.

===

I believe any mechanical engineer will agree that one key to maximum tire adhesion to the road is minimizing the variation of the tire's contact patch load. You want consistent, and predictable, changes in tire loading. The driver's behavior influences this greatly, but a suspension that has pitch (higher front frequency vs. rear) will by its nature induce MORE contact patch variation and reduce grip.

I appreciate this observation by gmx:

I have noticed since the E36 generation with "professional" race cars the driver applies such input and the the front end of the car visibly oscillates and even skips along the surface. I see this a lot on American race-prepped E36s but I do not know anything about the condition of rubber, track, ie. surface type etc. The tyre, roll-centre, surface type are key variables regardless of suspension, specifically front/rear spring rate coupling. I've assumed they don't know wtf they're doing and have slammed the front for more static camber while ruining or not correcting the roll-centre. Who knows. Or maybe crap tyres, brash inputs etc - a lot can cause that.

Anyway, I NEVER see such things on raeder motorsport (Manthey Racing), schirmer cars lapping the ring. You can see this slight behavior in my own car at 0:38 in the first video.

It was this kind of pogoing / oscillating that I observed even with overly-stiff autocross cars (before I got more involved in track driving/tuning) that made me question the wisdom of the higher front vs. rear ride frequency approach. E46 M3 track driver bigjae1976 observed that in his comments above.

===

The M3forum.net thread I started (with the spreadsheet I developed for them, and am willing to also create/share with this community) has well over a dozen instances of E46 M3 owners switching their spring rates to achieve Flat Ride and reporting benefits to grip, composure, stability, and lap times.

There's more than one variable at play - just getting Flat Ride in your spring rate choice won't bring peace on Earth. But I've found and consider it the foundation. I didn't invent 'big bars, soft springs' - but I understand the pros vs. cons pretty well. The approach is use I called Ride Harmony (3 principles) and Race Synergy (3 additional elements) and it seems to work very well for street, track, autocross, etc. I'll hat tip Dennis Grant/Far North Racing and take some credit for providing empirical and analytic tools to show how effective Flat Ride is when applied as part of a whole suspension tuning philosophy.

[ShaikhA]

P.S. A specific comment to the gent who mentioned that 'any serious racer knows you need knobs.' I honestly felt the same way for years. If you're covering a VERY wide temperature range in your racing environment then having an adjustable damper makes sense. But I don't think that's the case for anyone here. If it's below freezing you're likely not running a national championship. If it's 130F you're probably inside with air conditioning. I've worked with single-adjustables, designed my own double-adjustable reservoirs (with the help of ME's of course!), and incorporated technology from rally and F1. When you start to zero-in on the variables and make the reasonable assumption of a liveable temperature range for competition, the knobs becomes more of a psychological crutch than a required tuning aid.

I'm not being mean, just expressing my and many other people's experience. For those who may be familiar with Dennis Grant, he used data acquisition extensively back in the 90s while pursuing SCCA national championships. He found that adjustment was far less necessary than he had originally believed, once he'd made some good design choices (including Flat Ride, although he didn't specifically call it that). My own testing has shown similar conclusions.

I used to always rely on knobs to make setup changes, but looking back I realized that I hadn't optimized multiple other variables and I had NO IDEA what the dampers were really doing and if the change I was making was actually better in terms of lap times. I've found that the people who rely on knobs the most are using pitch-based setups where you NEED more damping to keep the oscillations under control. Using Flat Ride cancels those oscillations out so you focus on things like tire pressure temps that may be unique for a given track.

All the recent winning NASA and SCCA road racers I work with are using setups that have no damping adjustment. The bigger change they make is when it's very wet, they'll disconnect bars - I got that feedback from Anthony Z @ Edge after a Sonoma NASA race. This is why the springs + damping is such a CRUCIAL combination to get 'right' - because when you do, the vehicle works across a variety of track conditions, surface types, and temperatures (from above freezing to sizzling summer).

Yes, there may be a slight fall-off in damping, but a virtue of tuning for Flat Ride is you DON'T need as much damping compared to a pitch-based setup! You can really feel the tires working, and concentrate on your line, braking zones, etc., instead of going to the pits to tweak a knob. I did that dance for a long while and it was very frustrating and ultimately slower. You're not losing anything necessary when there isn't a knob - provided you're approaching suspension tuning from a thorough perspective. You're actually gaining a lot in terms of what I can make a damper (lower high speed compression slope) do when you use an 8mm shim stack vs. 12mm (drilled shaft) shim stack.

Quick summary of video - one damper (with 12mm post) had a high speed compression slope of 10.9 lb / in/sec, while another (with 8mm post) had 6.5 lb / in/sec. That's about 40 lb less high speed compression force every 10 in/sec of damper velocity. From a technical perspective, you end up with a wider high speed compression (and rebound) shim when you use an adjustable shaft, which leads to a higher high speed force and more potential for 'launching up' or 'jacking down.' It's much harder to create the kind of 'curb absorption' with an adjustable shafts.

[ShaikhA]

Originally Posted by bbnks2

The more I mess with suspension the more I learn that spring rates and even sway bar tuning doesn't mean much if you're not considering the many others aspects that affect the cars handling. Springs are least important of all.

Hey bbnks2, I fully agree that all crucial variables need to be considered to have a fast, responsive, predictable car. However, my experience differs re: 'springs being least important of all.' In fact, I'd say they are #1 in importance to set up a 'cancelling / Flat Ride' relationship between front and rear suspensions. From there, all the tuning is MUCH easier and more natural since the springs aren't fighting you where the driver feels a need to increase damping to add more 'control.'

The biggest revelation I ever had was a test ride in an Audi S4 Avant that had DEAD dampers but air springs that created Flat Ride. It absorbed small and medium bumps quite well, and though bounced more on larger bumps it settled more rapidly than I expect. Plus, it was able to 'slalom' quite well. It obviously needed (and received) more damping than zero but as a test, running just Flat Ride was very eye-opening!

Your car has Flat Ride, with the 6k/16k setup (~2.0 Hz front, 2.1 Hz rear ) so perhaps that what you mean by neutral / balanced spring rates? A Flat Ride car isn't fighting itself, vs. one with pitch. It naturally wants to stop oscillating. This means it's easier to run softer damping (and improve grip + comfort) with Flat Ride.

I am a firm believer in running neutral balanced spring rates, that work well within your struts range and damping, and then tuning the car to handle how you like through other means. There are massively more important things to consider than changing the spring rates by 100lb.

Yes .. provided you're staying with Flat Ride vs. going into pitch territory. I'm going to create a public version of the spreadsheet I made for Chris' 128i so users here can put in variables for their car and compare different setups.

I am still running 6k/16k (336/896) with E92 M3 sways front and rear and I don't have any plans to touch that anytime soon. Car handled great last season. Ride height, rake, track width, alignment, bump stops, strut travel, corner balance, rear differential behavior, etc. are all other things I am looking at for 2020. Any of these things being different from someone else who replies to this thread can drastically change how their car "feel," or "handles" vs yours. No one on this forum ever wants to get into actual suspension tuning though it's just always "you have to run x spring rate because it works for me."

Totally agreed - it's hard to compare one setup to the other without knowing many variables, but with at least a few key ones known it's possible to speak reasonably about how a car would handle with one setup vs. another. You also didn't mention damper behavior in there, which I'd put next to springs and sways in terms of determining the car's handling (and ride) behavior.

[bbnks2]

Originally Posted by ShaikhA

Hey bbnks2, I fully agree that all crucial variables need to be considered to have a fast, responsive, predictable car. However, my experience differs re: 'springs being least important of all.' In fact, I'd say they are #1 in importance to set up a 'cancelling / Flat Ride' relationship between front and rear suspensions. From there, all the tuning is MUCH easier and more natural since the springs aren't fighting you where the driver feels a need to increase damping to add more 'control.'

The biggest revelation I ever had was a test ride in an Audi S4 Avant that had DEAD dampers but air springs that created Flat Ride. It absorbed small and medium bumps quite well, and though bounced more on larger bumps it settled more rapidly than I expect. Plus, it was able to 'slalom' quite well. It obviously needed (and received) more damping than zero but as a test, running just Flat Ride was very eye-opening!

Your car has Flat Ride, with the 6k/16k setup (~2.0 Hz front, 2.1 Hz rear ) so perhaps that what you mean by neutral / balanced spring rates? A Flat Ride car isn't fighting itself, vs. one with pitch. It naturally wants to stop oscillating. This means it's easier to run softer damping (and improve grip + comfort) with Flat Ride.

Yes .. provided you're staying with Flat Ride vs. going into pitch territory. I'm going to create a public version of the spreadsheet I made for Chris' 128i so users here can put in variables for their car and compare different setups.

Totally agreed - it's hard to compare one setup to the other without knowing many variables, but with at least a few key ones known it's possible to speak reasonably about how a car would handle with one setup vs. another. You also didn't mention damper behavior in there, which I'd put next to springs and sways in terms of determining the car's handling (and ride) behavior.

I think I meant more "spring rates" being least important, as in, running 350lb/in vs running 450lb/in. That small change is not going to do much for you at all if spring rates are all that you are looking at. Is suspension travel even F:R when motion ratio is accounted for? Bump stop length appropriate for the motion ratio of the struts? Does the car over-steer because you dialed in 1.5" of rake?

I was more trying to say that people end up tweaking things unnecessarily because of purely self-induced handling issues. Like, running stock-like rear spring rates which puts the car down onto the rear bump stops almost instantly in roll or acceleration...

Can't agree with tsk94 post. Not going to reply in-line as Shaik explained things well. Flat ride is not about having balanced roll couple distribution. That is a big misinterpretation. You can have springs tuned for flat ride and still build yourself an under-steer oriented car. That is what is great about starting out with neutral balanced springs lol How would you even maintain, or determine, balanced roll couple distribution if your springs aren't balanced? The springs are a component of roll couple distribution lol...

Also, race cars do indeed use flat-ride. There are tons of technical article out there explaining about how everything from dirt rally to oval racing using spring frequencies as part of their suspension tuning strategy. It's literally what every OEM in existence does as well but on an even more complex scale than what we as enthusiasts try to recreate on a 2D scale. There is a lot of good information in the post but I just think he is looking at it all wrong. Tune a cars springs for flat ride, or close to it, and then use a million other things to fine tune its balance to your liking. Like, an adjustable sway bar. Cheap, quick, easy and can be done at the track to fine tune the car on the spot. Can't swap springs at track side and most race cars can't change springs at all as they are homulgated to run a certain spring rate.

An example I've broken down in great detail in other posts is the M235iR. At first glace it doesn't seem to be tuned for a "flat-ride." But, it actually is pretty close. It is homulgated to run 16K/16-18k springs in race form. But, it has a true real coilover and a 54/46 weight distribution. When you crunch the numbers, the car is very close to having neutral spring rates in the ~3hz range. That is on a purpose built race car with aero running Pirelli slicks.

Since Shaik mentioned the E46, that is another big sticking point for me. Everyone wants to run E46 spring rates on these cars, including TcKline who produced some of the first kits for the E8x/e9x chassis (450/700). But, the rear motion ratio and suspension is much different. So, why would the same spring rates that work well on an E46 work well on an E82? Without looking at any math or frequencies you can call BS on it. Actually crunch the numbers and again you see that to get the same exact ride that 450/700 produces on an E46 you need to run closer to 400/900 on an E82. I didn't actually crunch the numbers... just going off the top of my head from the comparison that I did a long time ago and the E46 having a higher .46? motion ratio. All numbers I crunched for people back in some autocross argument threads lol Complain about under-steer but refuse to believe the front-end is relatively stiffer than the rear. "But all the fast E46's run these rates."

Originally Posted by ShaikhA

Another benefit of Flat Ride tuning is that a higher rear ride frequency than front creates a more rapid initial yaw moment (rotation / turning) of the chassis.

Yes, when you actually read SAE whitepapers on suspension theory it holds true that the rear should be relatively stiffer than the front so that it can respond to inputs at the same time, to put it simply. People get obsessed with big front sway bars and lots of front roll stiffness but I don't see any scientific or suspension theory to support it. Just anecdotal experience and the "feeling" of better transient response.

I am half-way through reading Dennis Grant's book. He "autocrossed and won" in a car no-one thought could be competitive. And, he did so tuning around basic suspension theory and not the "monkey-see-monkey-do" approach.

[bbnks2]

Originally Posted by gmx

I have noticed since the E36 generation with "professional" race cars the driver applies such input and the the front end of the car visibly oscillates and even skips along the surface. I see this a lot on American race-prepped E36s but I do not know anything about the condition of rubber, track, ie. surface type etc. The tyre, roll-centre, surface type are key variables regardless of suspension, specifically front/rear spring rate coupling. I've assumed they don't know wtf they're doing and have slammed the front for more static camber while ruining or not correcting the roll-centre. Who knows. Or maybe crap tyres, brash inputs etc - a lot can cause that.

Anyway, I NEVER see such things on raeder motorsport (Manthey Racing), schirmer cars lapping the ring. You can see this slight behavior in my own car at 0:38 in the first video.

Just to give an example other than the M235iR... since someone said race cars don't run neutral balanced springs, and you mentioned Manthey racing, here are the build specs for the SRO GT4 MR.

https://www.manthey-racing.de/Techni...port_MR_V1.pdf

A car with 45/55 weight distribution, relatively close to 1:1 motion ratio in both the front and the rear (mcpherson struts), and they are homulgated to run several spring packages that are all close to a 3Hz and would produce a relatively flat-ride:

The following spring configurations are available:
130-140 (VA) MTH343530 and 130-170 (HA) MTH333534
120-140 (VA) MTH333533 and 110-170 (HA) MTH333535
150-140 (VA) MTH333531 and 150-170 (HA) MTH333532

They run a front sway with a very wide range of adjustment which is where most of the on-site "balancing" is done. Pretty interesting to see the types of equipment that goes into these cars... KW clubsports with basic 2 way adjustment. Same as the M235iR. People run more pimp suspension on their HPDE cars!

Quite honestly, 99% of the most useful information I have found (practical/applied) has been on RC forums and not "real race car" threads. Want to see someone nerd out about roll centers and suspension frequencies? Go check out an RC car forum lol Using WinGeo to plot their cars and tune suspensions. Not as practical on a real world scale but all the same principles apply.

Start with Episode 1 and continue through episode 10:

[gmx]

Thanks for that will give it a watch.

That's why I put pro in inverted quotes. The work that goes into Schirmer (when revealed), although the price is astronomical or simplicity of Raeder (they are the same dudes as Manthey) is sometimes genuinely jaw dropping, simple, yet precise.

I have just rough spring rates plucked all over the net in my spreadsheet. I might keep these as variables. I initially didn't see much point since they're stripped and I don't know if any lightening went on with the arms/knuckles nor know the weight of the disc let alone the entire car (the AP part numbers are in the PDF below if you feel like hunting). Would be difficult to figure out the frequency and irrelevant really as I will never run such tyre that offers this grip nor the aero. The front/rear wheel rate ratios are good to compare across a range (IIRC 3 main springs for homologation). But then, setup is up to the team, if they use the softest/stiffest front/rear or vice versa, we won't know.

E92:
https://www.bmwusa.com/content/dam/b...rsion_09_2.pdf
You can probably find these on real-oem too.

26.5mm front ARB
Main spring options (helper: 60-60-2):
140-60-160
140-60-180
140-60-200

22.5mm rear ARB
Rear C/O (use the known damper motion ratio, 0.83 IIRC with 80-50-20 helper):
160-50-130
160-50-150
160-50-170

F82:
ARB 30.4mm with triple adjustment
080-005 Helper (no idea what that is, 80mm long & 5nm/mm?)
140-60-180
140-60-200
140-60-220

Not surprised with these spring selection. Stripped weight would be as low as the E9x. The swivel bearing (aka knuckle) is practically the same sans some kingpin inclination improvement. Car is structurally stiffer and looks like new rules allowed some more aero performance. Not sure of tyre advancements either.

F82 rear C/O
ARB 26.5mm with triple adjustment
080-005 Helper
170-60-150
170-60-170
170-60-190

Despite rear C/O, I noticed (via realoem, and that is not 100% always correct) the lower camber arm is still the OEM part. Maybe homologation rule, no idea. Motorsport24/Schirmer are known to build their own lower link with their rear C/O.
However, the rear end is completely rose jointed, no rubber anywhere as you would expect. The rear F8x trailing arm (wishbone by internal references) stands out as do the axial mount on the toe arm, trailing arm (wishbone) which are standard on the road-going version as well.

Also no idea what the rear damper motion ratio is of the F8x.

[gmx]

On the 718 GT4 (SRO), the KW could and are more likely to be competitions rather than clubsport. THey also look like aluminium bodies?

There was 1M here using this. Actually it was the white one posted on this forum:
https://www.1addicts.com/forums/showthread.php?t=684308

They ended up modifying it a lot (esp front top plate) speaking to someone buying parts from their retired car. They've since moved on to racing F80s since the pure (we don't get really get strippers here) comes in under MSRP making it eligible. One of the 135i (same team or a competitor ran it, don't remember) competition dampers ended up on this member's car:
https://www.1addicts.com/forums/show....php?t=1302309

[bbnks2]

Originally Posted by gmx

Not surprised with these spring selection. Stripped weight would be as low as the E9x. The swivel bearing (aka knuckle) is practically the same sans some kingpin inclination improvement. Car is structurally stiffer and looks like new rules allowed some more aero performance. Not sure of tyre advancements either.

F82 rear C/O
ARB 26.5mm with triple adjustment
080-005 Helper
170-60-150
170-60-170
170-60-190

Despite rear C/O, I noticed (via realoem, and that is not 100% always correct) the lower camber arm is still the OEM part. Maybe homologation rule, no idea. Motorsport24/Schirmer are known to build their own lower link with their rear C/O.
However, the rear end is completely rose jointed, no rubber anywhere as you would expect. The rear F8x trailing arm (wishbone by internal references) stands out as do the axial mount on the toe arm, trailing arm (wishbone) which are standard on the road-going version as well.

Also no idea what the rear damper motion ratio is of the F8x.

F series uses the same rear HA-5 multi-link rear-end as the E8x/E9x series cars. G series does as well. Obviously they made small improvements, but, these chassis will all share similar motion ratios unlike E36 and E46 generations which had different designs. You can read about some of the changes in the BMW academy suspension docs floating around the net.

The two cars you're referencing use a true rear coilover. They can do this on race cars because the upper strut mount is tied into the roll cage and reinforced. The rear springs motion ratio is the same as the strut which is going to be around .72.

This is an F8x converted to a true rear coilover. You can see where the spring USED to sit (much more in-board). My oversimplified calculation of motion ratio for an F8x with stock location springs is .59^2 = .35.
https://i.ibb.co/Gsf0BNx/ohlins-f8x-4-1.jpg

The F-series front spindles do have slightly different geometry. I know Hydra Performance could give you more details as he just swapped M4 GTS front spindles into his 135i. He said they produced more camber than E92 M3 spindles. I think the Front control arm is longer which is part of it too? Not really sure but those little nuances of these different cars doesn't matter much within the context of this thread. Point was that for the most part I see relatively neutral balanced springs on most of the actual race cars I can pull up data on... more balanced than many typical setups anyway.

[gmx]

I've got those docs. Although the motion ratio would be similar, I wouldn't suggest making such assumption. There were good examples in the doc itself that look like they will result in different motion ratios due to packaging and pickup points even though they use the same suspension generational design and concept.

Case in point, I have 0.83 written down for the rear damper motion ratio! Did you measure 0.72? The problem with the internet :P

Btw, why do you square .59 ? Isn't that only squared to find / calculate the wheel rate (after multiplying spring rate)

There is some BMW document technical comparison at the launch of the F82, where they compare to previous gen front M3 geometry. The only difference was slightly more SAI (KPI angle) which will affect camber. Apparently the GTS has the same overall geometry except a larger hole for the bespoke damper. You'll see JDM cars usually have a strut / knuckle fastened with an eccentric bolt allowing adjustment here.
I noticed this over the last 2 seasons. Asked a former pro as he has both GTS (E92 and F82). His first comment, much more neutral balance and can get away with less Fr static camber whereas the E92 had inherent understeer. It was also evident by his tyre wear / usage on the front.

Not to mention F8x cars easily get -4 static camber very easily. The strut tower is huge and simply has more space to place the strut inboard. Those 1M/135i racing above ended up with -5 by using an additional offset sandwich plate. One (or maybe more?) other team simply notched the towers. Important to note, you weren't allowed to change the spring location in those championships effectively banning rear C/O. Moving up a category, you could like this car (revalved, rear C/O clubsports):
https://f80.bimmerpost.com/forums/sh....php?t=1674814
I talked to the suspension builder and he explained they use a slightly lower frequency due to the range of tyres used, and inexperience of the drivers (gentlemen racers).

I might have an opportunity to acquire a set of F80 knuckles, I'd need to get new BBK brackets built for it though.

ShaikhA Happy to see you post here. We spoke before mutually deciding that I preferred local support.

If you're local, I think it's a no-brainer for you guys.

[bbnks2]

Originally Posted by gmx

I've got those docs. Although the motion ratio would be similar, I wouldn't suggest making such assumption. There were good examples in the doc itself that look like they will result in different motion ratios due to packaging and pickup points even though they use the same suspension generational design and concept.

Case in point, I have 0.83 written down for the rear damper motion ratio! Did you measure 0.72? The problem with the internet :P

Btw, why do you square .59 ? Isn't that only squared to find / calculate the wheel rate (after multiplying spring rate)

There is some BMW document technical comparison at the launch of the F82, where they compare to previous gen front M3 geometry. The only difference was slightly more SAI (KPI angle) which will affect camber. Apparently the GTS has the same overall geometry except a larger hole for the bespoke damper. You'll see JDM cars usually have a strut / knuckle fastened with an eccentric bolt allowing adjustment here.
I noticed this over the last 2 seasons. Asked a former pro as he has both GTS (E92 and F82). His first comment, much more neutral balance and can get away with less Fr static camber whereas the E92 had inherent understeer. It was also evident by his tyre wear / usage on the front.

Not to mention F8x cars easily get -4 static camber very easily. The strut tower is huge and simply has more space to place the strut inboard. Those 1M/135i racing above ended up with -5 by using an additional offset sandwich plate. One (or maybe more?) other team simply notched the towers. Important to note, you weren't allowed to change the spring location in those championships effectively banning rear C/O. Moving up a category, you could like this car (revalved, rear C/O clubsports):
https://f80.bimmerpost.com/forums/sh....php?t=1674814
I talked to the suspension builder and he explained they use a slightly lower frequency due to the range of tyres used, and inexperience of the drivers (gentlemen racers).

I might have an opportunity to acquire a set of F80 knuckles, I'd need to get new BBK brackets built for it though.

ShaikhA Happy to see you post here. We spoke before mutually deciding that I preferred local support.

If you're local, I think it's a no-brainer for you guys.

Not really sure to what end point youre mentioning any of this stuff to. We are talking about 135i's. The data was only presented to show that race cars run fairly neutral spring rates when looking at things at a macro level, for the most part. At the very least, they are not running springs operating at 3hz up front and 2hz on the rear like many 135i's. People also said all of the same things when moving from E36 and E46 M3's to the E92 M3. They needed less camber.

135i is supposed to be:
Springs:
Front 0.960^2 0.9216
Rear 0.563^2 0.316969

Damper:
Front 0.960^2 0.9216
Rear 0.813^2 0.660969

Motion ratio is further reduce by angle but it's a fairly lower % adjustment so most people just ignore the "ARC" component of the formula. You could throw a camber gauge on the rear spring and strut to get an idea of how "vertical" they are.

And yes I was rattling off numbers based on that pic. not a perfect shot it's taken from an angle and measurements are probably skewed by aspect ratio a bit. The numbers being close enough to actual though. Just to show conceptually that where the spring mounts in the control arms changes how effective it is... not to present factual numbers. Whether or not an M4 has a strut ratio of .72, .69, .68 or whatever doesn't really matter and I don't have an M4 GTS to measure.

Pickup points matter for kinematic geometry and dynamic spring rate. Static motion ratio as we are calculating is a simple formula dependent upon the length of the lower camber arm and the attachment points and angle of the spring and strut. It's a simple proportion, really. Seeing as how you are saying the lower control arm is the same the motion ratio would also be the same or very close to it. Part number is not the same for a regular F8x and a M4 GTS, but, neither is the 135i and 1M part numbers. Yet, the spring and strut sit in the same spot. The strut just mounts differently and the arm is aluminum. Motion ratio between a 135i and a 1M should be almost the same. These two cars having different toe arm pickup points does not change the static strut or spring motion ratio calculation.

[gmx]

Unless rules or cost prohibit it, I don't think there is a problem including the M equivalent cars or factory homologated cars. As you said, the MR is likely the same. Everyone in here runs different camber which will result in different spring angles and hence MR anyway.
F8x front knuckles bolt on and have improved SAI resulting in better dynamic camber curve. Unless you can draw up and mill custom uprights, that's a no-brainer cost/time permitting.
I used those cars to compare wheel rate front / rear ratios since nothing really further than this can be compared, the sprung and unsprung weights can barely be deduced. You can if you try, the BBS, AP part numbers are all there to go down that rabbit hole. The overall suspension frequency range is pointless since (ie. actual spring rate) the tyre they run, weights and kg/points of downforce they make is far different especially BoP SRO GT4 vs old FIA GT4 regs.

They're actually not neutral to me unless you're talking behavior or feel if you've been in the hot seat of one? You get close to neutral if, of the homologated spring selections, you pair the softest front spring and stiffest rear.

fer1x did mention this 57% "magic number", didn't give this as much thought until I threw those factory purpose offered cars into the picture. The OP's option #1 hits this bang on and he has never been faster, nor been happy with the balance in options following this. Let's be real now, unless you know the points in downforce from your aero (and whether moved aero balance forward / rear) is making or car is visibly bottoming out, there's no reason to throw stiffer springs at it just because so and so said. While I make a sample of him, his car's journey, it's not a dig. Nor do I aim to make this thread out to be.

Anyway, nailed down my unsprung / sprung weights. Currently at 2.77 & 2.04hz. Looking at either a 425 or 450 front spring to test (2.40 - 2.46hz) maybe in time for March.

[FrostyDC4]

Originally Posted by chris_flies

I'm not in time trial or racing (yet?), but my car is huge amounts of fun on track. It's my only car, so it's not too extreme, but it's well beyond most people's tolerance for NVH.

Diff: MFactory 1.5-way with 3.73:1 ratio.
Suspension: Ohlins Road and Track with 1M-spec valving from 3DM Motorsport and 7k/16k (F/R) Swift springs.
Sway Bars: No front sway bar. ECS Tuning 12mm solid rear sway bar with Meyle HD end links.
Others: M front control arms, Dinan camber plates. M rear upper control arms and guide rods, Condor Speed Shop UHMW (solid) subframe mounts. 225/45-17 RE71R's all the way 'round.

One thing that made a big difference in how the car handles, was the alignment. My car is currently sitting at -2.1* camber and 0.2* toe out up front, and -0.8* camber and 0.2* toe in for the rear.

Deleting your sway bar may be drastic, but I actually looked at pictures of my car when I had an Eibach 28mm solid bar up front, and I was barely compressing the outside suspension, and was lifting the inside tire about an inch off the ground, which means there was a little more grip to be had. As a stop-gap measure, I removed the front sway bar, drove 7 hours to Mid-Ohio, ended up loving it, and haven't turned back. The deletion of the front sway bar drastically changed the behavior of the car. Even on the highway it felt so much better than before. Presently, it's definitely biased towards oversteer, but it isn't excessive. It really likes to come around at corner entry, which feels great, and if you catch it right you can get a smooth 4-wheel slide going through to corner exit, at which point all 250-odd horsepower stabilizes the rear end and you're off to the next Miata. I think an extra half degree of camber in the rear, or maybe deletion of the rear sway bar will neutralize it to perfection.

Do you have a lot of body lean up front with the sway bar delete? Asking because the fast 350Z/370z drivers either run a big bar up front and/or run without a rear sway bar.