The issue I have with your recommendation, or calculations, is that a 2000lb CRX sees over 600lbs/front axle at rest.

300lb front springs will give you a wheel rate of about 130lb's on a CRX. That means it only takes 130lbs to compress the spring 1 inch. If at rest the axle sees over 600lb's (pretty much 5 times the weight it takes to compress the spring one inch) that is a hell of a lot of movement, and I shudder to think of dive or hard transitions. In tuning a competition vehicle, we like to stay above a 50% corner weight to wheel rate ratio.

Now lets look at the 14K recommendation. A lot of people get scared when rates like that are tossed around, because 14K or 787lb's according to our handy dandy chart seems like a lot. But when using wheel rates you can see that only works out to 330lbs. Not so crazy now? And hence why I asked if you are calculating wheel rates. (On a side note, this calculation you are using must be for wheel rates. If you intend to apply it to all vehicles, be aware that if you don't use wheel rates you are disregarding motion ratios, and that completely nullifies the data)

Anyways, this 330lbs is going to be roughly 50% of our corner weight, and that means even resting the spring has 2 inches of compression, which is an ample amount of travel to absorb bumps and NOT skip and lose grip.

Like I said, any skipping or loss of grip will be attributed to the over damped shocks you are using and their pack down effect.

I notice you are also a fan of the no swaybar mentality. The debate between swaybar vs no swaybar has been going on for a long time, and its not going to be solved here. I will give my opinion on it anyhow.

I agree with you to a point, however I don't like to be so extreme in my views as to completely disregard the importance of a swaybar.

Yes, its true that a swaybar compromises the independant nature of our suspension systems, and can be a little harder to tune, but it is also quite useful in controlling body roll, creating more unfiform weight distribution, helping to unload not needed weight, and ultimatley resulting in the creation of oversteer that would have been harder to create with springs alone.

I usually recommend to people to run no front swaybar. The front rates we tune with are generally good enough to aid with body roll, and are stiff enough to minimize any effect the swaybar would have had in the first place.

In the rear I recommend a swaybar however. It evens out front tire distribution, which as you know results in more grip. It also aids in reducing body roll on the entire system, and helps remove unnecessary weight from the inside rear. Because of the lower rear spring rates I tend to use when tuning, the swaybars also help with creating more oversteer in addition to alignment and other settings. I'll generally start with just the stock rear swaybar, and only go bigger if the need is there. For road race setups, that is rarely ever the case.

Remember, a lot of these "rules and calculations" were created by people racing RWD platforms. FWD is a slightly different ballgame, and we play by our own rules many times.

Cheers.

 

Great insight here.. !

 

oh sorry, forgot to post that up!

EG hatch

2Join Perf - great info dood, thank you! i agree with you that the damper makes a big difference on the spring rates. i was scared of my rates at first for daily use, but was extremely suprised at how 'supple' it is for daily. the rear will skip a bit but i'm not driving like a racecar on my way to work

 

These rules and calculations can be applied to more than just rwd cars. It applies to all types. Once you have your base numbers, you need to adjust accordingly after that to tailor to your specific type of racing (autocross or whatever) and the dynamics of the type of car you are racing (fr, fwd, awd, rwd, mr...).

Once again, I'm talking spring rates not wheel rates. And yes I understand the difference.

300in/lbs front and 500in/lbs is a perfectly exceptional spring rate for autocrossing a 2000lb crx. Like I said before, you have to tailor to your specific type of racing and the type of tires your using. If you're using really good racing slicks and you're on a race track with longer sweeping corners, it makes sense to go stiffer.

The thing about using REALLY stiff-*** springs on a light car is, no matter how well damped it is, when you hit a bump, and the wheel moves up, it's going to moving the whole corner of the vehicle up with it because there is no give with super stiff springs, therefore your upsetting the balance of the car at that moment. It's like having no suspension.

With a softer spring, the tire remains in contact with the ground at all times, and the vehicle does not become up set by the movement of the wheel over the bump.


Back to sway bars: (I'm thoroughly enjoying this discussion...)

You agree with me front sway bars are not necessary. Thats fine.

Consider this: If we took 2 identical 2000lb crx's. they both have the same suspension. The only difference is one is setup with a rear sway bar and the other is setup without a rear sway bar and they are both setup for the same race track (both with same tires of course).

The crx with no rear sway bar, I would put money on this, would have more mechanical grip, hold a higher G, and be more stable in a corner than the one with a rear sway bar.

 

I don't know that you do. Lets assume you are calculating spring rates for a car weighing 2000lbs with a 62/38 weight distribution. According to you, we need to use 300lb front spring rates for this car. Fine.

But now assume we have two of these cars. Both weigh 2000lbs, both have the same weight distribution. The only difference is the front motion ratio. Car A has a motion ratio of 0.5, Car B a motion ratio of 0.9.

Therefore:

Car A: 300 x .5 x .5 = 75lbs

Car B: 300 x .9 x .9 = 245lbs

So using the SAME spring, Car B has effectively over 3 times the spring rate of Car A. This is going to directly translate into how well both cars handle, and how "stiff" the suspension feels.

So how is it possible that this one formula you have can give me the "perfect" setup for my car if it completely disregards such a vital component of suspension tuning?

300lb springs are only 20% of the corner weight. Sounds like a fine street setup, but good luck keeping the body under control with so little resistance. Especially if you are not using swaybars.

As I have shown in my previous response, even 800lb springs have a good 2 inches of compression at rest. That is more then enough to absorb bumps and dips, so no, it is not like riding with no suspension. Yes its stiffer, but not in a bad way.

If we are building a dune buggy, a rock climber, or a rally car going for gravel, yes, we would go softer.

Again, if the wheel is getting upset, its the valving.

Yes, its nice to get some real discussion in the suspension forum, instead of questions regarding what color to paint wheels.

And that is where I disagree. As I said earlier, a rear swaybar will offer MORE grip to the front wheels, where it counts. As you know, grip is a function that relies on the difference in weight between two tires along the same axle. A rear swaybar will minimize that difference offering MORE grip.

Also, in order for you to get the same amount of roll couple in the rear as the car with the rear swaybar, you would need to up your spring rate. Keeping in mind that we are working with front heavy FWD cars, you will quickly be approaching the limits of the rear corner weight to wheel rate ratio, resulting in over springing the rear.

The car with the swaybar will be able to retain it's composure in the straights and under light turns, whereas your extremely stiff rear springs will have you hopping and skipping.

But since you are not upping the rear spring rate in this hypothetical, I would have the advantage of less understeer. All additional spring rate present during cornering can be controlled with low speed valving.

Cheers.

 

All I posted was a chart

 

And a good chart it was haha

 

LOL, I wasn't expecting it to grow to such an in-depth discussion...

 

Ok yeah, I'm an idiot... it's wheel rates I've been calculating, not actual spring rates.

Now that I've confused everyone...lol.

Anyways, back to the discussion:

Let me recalculate for the actual spring rates.

Do you actually know the specific motion ratio of the front suspension in a crx? If so, what is it? and what type of tires do you run?

I will still say, the results, even though they will be stiffer than 300lbs, will still be nowhere near as stiff as 14k (or 7XXlbs), nor is it necessary to be that stiff for a vehicle this light weight.


This formula came from a guy who won the most autocross championships in North America...ever. So I think he knows what he's doing. I also believe it works for any car for autocross and track racing setups. You just have to adjust and tailor to the dynamics of your vehicle and the type of racing you're doing.


I've seen real life results from using this formula, so I'm not just spouting pure theory.

 

.65 for the front, .75 for the rear.

Usually some sort of R-compounds. Tires can get very complex, very fast.

I think you may be surprised. 14K springs result in roughly 330lb wheel rates, just a hair above the 300lb's you calculated originally. Plus we never accounted for the weight of the driver, so tack on another 150-200lbs to the weight of the car.

I don't doubt it works for the guy, but what I am personally cautious about is how this theory of his applies to FWD cars. Now that we have established it is wheel rates, not spring rates, I can agree with the outcome of the front wheel rate.

However I am still reserved about the rear wheel rate. The reason being is because we now enter into the realm of front to rear spring bias. There are people who tune their FWD cars with massive amounts of rear spring, and I am not one of them. Some of the reasons I have already outlined in previous responses.

The nice thing about suspension tuning is that there is never a correct answer, only degrees of wrong, and more then one way to skin a cat.

 

I'm just trying to prove that there is a way for anyone out there struggling for the right setup (on any vehicle, for autocross or race track) to get a very good baseline setup to start fine tuning from.

I wish I could setup a car for you and let you drive it, before and after, to show you what I mean. And there is a lot more I would love to explain and discuss, but through a forum on the internet just sucks for that. I'd be typing for days...lol.

What I can tell you is this:

We (couple friends and I) have an 87 MKI MR2 sc (the one in my sig) with too many mods to list.

Everything suspension-wise is pretty much race preped. We used the formula and came up with an autocross setup of 600in/lbs fronts and 350in/lbs rear spring rates. We ordered the springs and they just arrived monday. I installed the fronts this morning and will install the rears tomorrow morning. (sway bars removed of course). (running koni yellow dampers, incase you're wondering).

Our 1st race this season is this coming sunday, may 10th.

I'll post up how the car handles. It would probably need 400lb springs in the rear for road race track setup, but this is our baseline setup and we'll go from here. I'm confident it will handle very well none-the-less.

Want to get back into the sway bar discussion??? lol I could go for weeks on topics like this...

 

I look forward to your re-calculations tomorrow.

Keep in mind the rear motion ratio of most Honda's is more effective then the front, meaning it takes less rear spring to be just as stiff as the front. But I can see why you would want a stiffer rear rate, because you don't use rear swaybars. What I am able to achieve with a swaybar, you need to work in with extra spring rate.

If there is a nice simple method that will help people ballpark a suspension setup I'm all for it. Suspension tuning doesn't get enough attention and anything we can do to help set people up properly is valuable.

I just know that when putting together a setup I consider MANY things. Natural frequencies, corner to wheel rate ratios, flat ride oscillations, motion ratios, front rear bias, etc. Hence why I have my doubts about a universal formula that works for everything, but anything that makes life easier is a good thing.

Hah, I know what you mean.

Good luck! Do let us know how you did.

 

So for the recalculation for a 2000lb crx with 62/38 weight distribution, the base numbers come out as:

Front 410in/lbs (7k) spring rate

Rear 620in/lbs (11k) spring rate

Going from there, for an autocross setup, I'd go with 450(8k) front springs and 650(11.5k) rear springs.

For a race track setup, I'd go with 450(8k) front springs and 600(10.5k) rear springs.

(again these setups would be with no sway bars).


Getting back to the sway bar discussion. When you say the rear sway bar aids in front grip, I agree with that, but I think you're forgetting it's changing whats happening in the rear end. It's taking away spring pressure from the rear inside tire and adding it to the outside, basically changing the spring rates by an unknown amount (unless you know the specific spring rate of the sway bar you are using). By adding to the outside spring rate it can overload the outside rear tire, thus reducing overall mechanical grip.

As long as you don't over load the outside tire (which is more likely to happen with the use of a sway bar) then both setups will work. but the setup with no sway bar will allow a higher cornering G and stability over bumps than the setup with sways bars would.

 

Could you please post the calculations? That way we can see how you arrived at the results.

When I use your previous formula I get:

1100lbs x .38 x .8 = ~330lbs
1100lbs x .62 x .8 = ~545lbs

So I'm left wondering what has changed?


In a FWD car, we don't care about unloading the inside rear. Any weight on the inside rear, could have been on the inside front.

As for overloading the outside rear, perhaps with a large enough swaybar. However, the combination of lower rear spring rates with a mild rear bar allows for greater stability in the straights, as opposed to no swaybar and having to compensate with constantly oversprung rear springs.

Don't forget too, that along with the increase in spring rate at the outside rear, I am also experiencing severe weight transfer over that wheel in turn in situations. We don't live in a steady state, and so this extra mass over the outside rear is used to control the natural frequency of the increased spring rate. Therefore retaining stability, and bumps are easily absorbed.

As you noted, both methods achieve the same result, but with different characteristics.

 

Here are the basic calculation I made to find base number for spring rates for autocross:

Vehicle: 1990 honda CRX
Weight: 2000lbs. (approx)
Weight dist: 62/38
Motion ratio: Front = .65
Rear = .75

The calculations:

Half the weight x the inverse weight dist % x .8 for R compound tires = wheel rate.

Then:

Wheel rate x the inverse of the motion ratio %age + the wheel rate = the spring rate.

So to show some numbers:

Front:

1000 x .38 x .8 = 304

motion ratio is .65 so: 304 x .35 + 304 = 410 OR 304 x 1.35 = 410in/lbs

Rear:

1000 x .62 x .8 = 496

Motion ratio is .75 so: 496 x 1.25 = 620in/lbs

That is how I came up with those values.

Now for autocross this would be a very good base line to start with.

For a race track setup, it would obviously require some different adjustments from this point on.

 

Just for kicks I dusted off the ps2 and fired up gt4.

I find this game (although it is a game) is quite realistic and after inputing the exact same data for the setup we had (before race springs) with our 87 mr2 sc, I was very surprised to see it handled pretty much exactly like our real car.

So I chose a 1990 honda crx si. Modded to 186hp and weighs just under 2000lbs. I selected full race tires (medium softness) and fully adjustable race suspension.

I set it up with 14k front and 12k rear, adjusted the dampers to match, turned the front sway bar off and the rear sway bar to a mild thickness and hit tsukuba circuit.

The best lap time I could muster after about 10 laps was 1:05:063.

I must admit, I was surprised how well the car handled. I could get right back on the gas after braking and pretty much floor it all the way through to corner exit. The car felt a bit squirly and the back end felt very loose. It seemed to slide around a bit though not significantly and nothing that couldn't be save with a quick flick of counter steer. But I did notice bit of pushing in the front at mid corner.

So then I set the crx up with the rates I calculated above.

I started with 8k front and 11.6k rear, sway bars turned to zero and dampers set to match.

Mid corner was fantastic but it had a little push at corner entry and exit, so I bumped the rear up to about 12k and it seemed to be better. With this setup the car felt alot more "planted" to the ground and predictable.

Happy with that setup, I did 10 more hot laps and managed to pull off a 1:04:674.

Not too much of a difference. I know tsukuba circuit very well, so figured it would be a good track for this test. Some tight corners, slight elevation change, and a long sweeping corner.

Honestly though, it's splitting hairs really. There are a number of factors that could have effected the out come of these results. Basically I'm not saying that either setup is the better setup. That depends on the individual.

I guess in the end, both our setups work, (not that I doubted yours, you've obviouly done your home work and know what works best for you).

We could go on and on about this for months and not come to a solid conclusion that we would both agree on. The only way to really tell if this works is to test it in real life, with several different types of cars.

 

Ok, so the first part is exactly the same as what I have, except I factored in the weight of the driver.

The second part of your calculation is wrong though. It appears you are trying to convert from wheel rate to spring rate. Problem is, you don't take into account the fact that we square the motion ratio when looking for wheel rates.

Therefore if you wish to go from wheel rates to spring rates, your calculation looks as follows:

1000 x .38 x .8 = 304
304/(.65^2) = ~720lbs

How do we confirm?

720 x .65 x .65 = 304lbs

Lets do the rear:

1000 x .62 x .8 = 496
496/(.75^2) = 880lbs

Again, lets confirm.

880 x .75 x .75 = 496lbs

So you'd need rates in the 13KF/16KR area, according to your formula.

 

Ok, here's where I got screwed up.

The way I was thinking of the motion ratio is this:

The distance from the chassis pivot point of the lower control arm to the point where the force of the spring intersects the arm divided by the distance from the chassis pivot point to the ball joint (full length of control arm).

After some quick research to find out why you calculated it the way you did I now realize it's a leverage ratio. duh. lol.

Basically, for those who want to know: for a motion ratio of .65 ... for every 1" of travel the wheel makes, the spring compresses .65".

So yes, your calculations are correct. I stand corrected.

Thats basically whats been screwing me up this whole time.

MKI MR2`s have a motion ratio of close to 1 so there is no need to include that factor in the formula.

So the formula still works! It was just my own blindness that I thought those rates were way too high.

So with your calculations you got:

13k front and 16k rear.

I then agree, that would be an ideal setup without the use of sway bars. This, calculated from Steve`s formula (guy who invented it). Though 16k, to me, still seems a bit high, but then using 12 or 14k in the rear with a sway bar, will easily become a 16k spring because of the function of the sway bar.

I`m going to try that setup in gt4 and see how that works. (13k front and 16k rear with no sway bars)



I`m glad this discussion went this far, because other wise I`d go on calculating for other types of cars, not including the motion ratio factor. So thanks for the in-depth discussion.

Suspension setup has always been such a mystery to me, and I`m glad in the past few years (since started racing autocross) I`ve been able to crack it.

Thanks for the geek-talk 2join!

Would you agree now, that the formula does give you a good ball-park base number for spring rates for all types of vehicles?

 

Here`s a link to a vid of Steve`s MKI MR2 autocross car. He is running the same springs we have just installed in ours, calculated from the same formula we just used to find the ideal spring rates for a honda crx.

Enjoy:YouTube - SE-Div ACU4 Sunday

 

Glad to see it all come together.

Exactly. As we noted earlier, both methods get you to the same place. I'm still of the opinion that such high rear rates are compromising more then lower rates plus a mild rear bar. It's like I was saying earlier, in order to get the rotation and benefits of the swaybar, you need to up the rear spring to levels that are testing the limits of a front heavy FF layout.

Looking over my own numbers, I would go with something like 14Kf/11Kr and a mild bar. More/less rotation can be dialed in with alignment specs or staggered tires.

Yeah, let us know how it compares to your other trial runs. Though only a game, its still interesting to see the perceived effects.

Thank you for the enjoyable convo.

I think this has turned into one of the best threads on TCC.

It would appear so. The front rates are pretty much exactly what I said since the beginning, so I have no issue with them. The rear rates are a little high, but if it's kept in mind that that rate can be reached either with a spring alone, or a spring and bar combination, then I really have no issue at all.

So yes, looks like the formula gives a pretty good ballpark.

Awesome! Can't wait to hear your impressions from your first track day.

Cheers!