General Suspension Questions

[Norm Peterson]

Those Phoenix guys showed what they were doing in Indy. They're moderately stiff on the front and chewing gum soft on the rear. The Ecoboost car was a little stiffer on the rear, but it didn't put the power down as well, either, which would be consistent logically. Both produced Runoffs championships so that choice was probably well justified. Whoever is screwing those things together must not be a complete jackwagon. They hauled ass all weekend and collected hardware at the end of it.

The EB car has IRS while the Boss and 4.6L Phoenix Mustangs have stick axles. Point being that the rear geometric roll centers for the IRS cars is going to be lower than the geo-RCs on the Mustangs with stick-axles, which implies a somewhat different approach to rear suspension calibration. While I'm not familiar with what the various rulesets involved permit, I'm sure that they have other things going on back there as well.


Norm

Sponsored

 

[TeeLew]

The short story here is that when you've been paid to do dynamic analysis of structures, thinking in either the time domain or the frequency domain and about the effects of damping actually comes in preference to considering only the spring part of a spring-mass-damper system.

I agree with this, and I'm perfectly comfortable conversing in these terms, but I just think that most people's eyes will glaze over while the boffins blather on. Maybe I'm wrong? It happens all the time.

 

[TeeLew]

Right about here is where I think you need to be doing your thinking in the time domain and including the effects of damping. Spring rates or ride frequencies alone really aren't enough for more than establishing whether the ride quality of any given car would be considered "soft", "firm", or somewhere in between.

Flat ride is only "flat" at one speed, and how close it might be to remaining "flat" at any other speed varies (and is not symmetrical for speeds above the theoretical flat ride speed vs speeds below it). You can't "see" those effects when the only data you have is stiffnesses or even frequencies. But they do show up in the driving. The classic example would be an older pickup truck with rear springs chosen for extra payload in the bed when there's no load back there.
..............
That, indeed, is the problem with using simplistic "guidelines". They don't give you any basis for determining when the results they suggest may not be appropriate. Or even if they may be inappropriate. There just isn't enough to work with, so you end up with a 'trial and error' approach from data that isn't itself giving you any real direction.


Norm

1. The funny part about this is, dampers are something you have to measure in the time domain and apply in the one of frequencies. I have a Roehrig crank dyno. I can measure 'full loop' forces at various peak speeds, which will give me a glimpse of the frequency dependence of a given damper, but I lack the ability to do any real frequency sweeps or frequency dependence quantification. In the end, I have to make that qualitative assessment based on experience.

I don't really approach the genesis of a damping curve from a 'soft' or 'stiff' point of view. I try to characterize the frequency content of the road and then match the car to its intended application. The good news is that racetracks tend to give you a lot of the same types of inputs...or, at the very least, the same family of inputs. Each has it's own peculiarities, but in general, there's a lot of low frequency content with annoying-placed high frequency additions.

2. Yes, these things (flat ride) can show up in driving. They can also show up on the data if the car is so equipped. These are all of interest, but they're not what I consider a determining factor for use. That honor is reserved for the stopwatch. If the watch can tell the difference, now I'm interested. I've never had the stopwatch argue in favor of a flat-ride spring arrangement and I've it given plenty of opportunities.

3. No matter what you do, you're going to end up with some amount of trial-and-error. Even Formula 1 falls back to mortar fire from time to time.

I'm not a scientist. I feel no need to derive my choices from first principle origins. I'm an engineer. If I can define a trend, then that will give me places in the future to investigate, but, for the time-being, it narrows the potential choices, which can be very helpful when starting from a clean sheet. I have no issue accepting a theorem and should someone provide anecdotal advice, I'm interested in listening.

I'm not presenting my various guidelines as gospel or a way that *has* to be followed. I'm presenting them for exactly what they are, observations of some random dude on the internet. It could very well be that this guideline doesn't work for this (or any!) car. Even finding this would be of interest as it would help to update the guideline a little further. I'm certainly not presenting my thoughts as automotive law. Having said this, I'm all for discussion. I can justify my choices logically and provide rational arguments in support. There is no mathematical equation to tell you which springs to put on your car, but that doesn't suggest the choice is arbitrary.

 

[bnightstar]

Eibach ProKit for a 1” drop, Steeda Minimum Drop for less at 0.5”. Both are 200/800 with front rates in the GT350 range, but gracefully offer less rate in the rear.

Eibach ProKit is actually progressive 200-314 / 800-914 (1.1" F / 1.0" R). A friend of mine have them on the stock shocks and this is actually very close to how a non HP Mach 1 rides. Very impressive for such a cheap kit considering it actually comes with bump stops and Shock boots.

 

[Roadway 5.0]

Eibach ProKit is actually progressive 200-314 / 800-914 (1.1" F / 1.0" R). A friend of mine have them on the stock shocks and this is actually very close to how a non HP Mach 1 rides. Very impressive for such a cheap kit considering it actually comes with bump stops and Shock boots.

It’s semantics of sorts. The ProKit factors-in the progressive rate of their bump stops. The springs themselves are linear.

Per Eibach: “PRO-KIT spring rates are measured as a system of the front or rear suspension. 10mm of bump-stop deflection is included in the final rate of the system.”

 

[Norm Peterson]

1. The funny part about this is, dampers are something you have to measure in the time domain and apply in the one of frequencies. I have a Roehrig crank dyno. I can measure 'full loop' forces at various peak speeds, which will give me a glimpse of the frequency dependence of a given damper, but I lack the ability to do any real frequency sweeps or frequency dependence quantification. In the end, I have to make that qualitative assessment based on experience.

I don't really approach the genesis of a damping curve from a 'soft' or 'stiff' point of view. I try to characterize the frequency content of the road and then match the car to its intended application. The good news is that racetracks tend to give you a lot of the same types of inputs...or, at the very least, the same family of inputs. Each has it's own peculiarities, but in general, there's a lot of low frequency content with annoying-placed high frequency additions.

Sometimes I wish I'd had similar opportunities. But life somehow didn't let that happen.

2. Yes, these things (flat ride) can show up in driving. They can also show up on the data if the car is so equipped. These are all of interest, but they're not what I consider a determining factor for use. That honor is reserved for the stopwatch. If the watch can tell the difference, now I'm interested. I've never had the stopwatch argue in favor of a flat-ride spring arrangement and I've it given plenty of opportunities.

No surprises there.

The biggest difference in our outlooks here probably boils down to the fact that I'm always going to be tuning for a dual-purpose street/occasional track day car. Have to, as I can't justify the expense or have a place to put a trailer and tow vehicle. Hell, the Mustang won't fit in the garage anyway, so it needs to be registered and insured. A tow vehicle that I'd almost never use for anything but getting a track car to the track and back home again is a non-starter by itself.

I'm not a scientist. I feel no need to derive my choices from first principle origins. I'm an engineer. If I can define a trend, then that will give me places in the future to investigate, but, for the time-being, it narrows the potential choices, which can be very helpful when starting from a clean sheet. I have no issue accepting a theorem and should someone provide anecdotal advice, I'm interested in listening.

I was an engineer as well, for most of my working life. In my case, working from first principles has been my primary source of direction, so I've been working with the available tool. If I can model something, I end up having to understand it a little better than if I only read it somewhere or simply take somebody's word for it. Turned out that learning how to write little spreadsheet applications on the "car side" of my life ended up helping me on the professional side (not automotive), so it's been a win-win.

There is no mathematical equation to tell you which springs to put on your car, but that doesn't suggest the choice is arbitrary.

Of course not. But it's nice to have some idea of what's going on to help you point yourself in what's hopefully the right direction for your wants or needs.

You'll still need to have a little aptitude for evaluating the changes and where to go from there. A little bit test driver.


Norm

 

[TeeLew]

Sometimes I wish I'd had similar opportunities. But life somehow didn't let that happen.

The biggest difference in our outlooks here probably boils down to the fact that I'm always going to be tuning for a dual-purpose street/occasional track day car. Have to, as I can't justify the expense or have a place to put a trailer and tow vehicle. Hell, the Mustang won't fit in the garage anyway, so it needs to be registered and insured. A tow vehicle that I'd almost never use for anything but getting a track car to the track and back home again is a non-starter by itself.

I was an engineer as well, for most of my working life. In my case, working from first principles has been my primary source of direction, so I've been working with the available tool. If I can model something, I end up having to understand it a little better than if I only read it somewhere or simply take somebody's word for it. Turned out that learning how to write little spreadsheet applications on the "car side" of my life ended up helping me on the professional side (not automotive), so it's been a win-win.

Of course not. But it's nice to have some idea of what's going on to help you point yourself in what's hopefully the right direction for your wants or needs.

You'll still need to have a little aptitude for evaluating the changes and where to go from there. A little bit test driver.

Norm

1. Considering the opportunities I've sidestepped, you likely made the better choice.

2. My car has to drive on the street as well. I regularly take it 500 miles in a day, so I have to keep it driveable. For my personal preferences, it's better to reduce the overall heave rate via dropping rear spring as opposed to raising the rate to achieve simmering along the lines of a flat-ride combo. Part of this is personal preference, but much of it is I just don't think 'flat-ride' it's a very powerful tool in a modern car.

3. I get what you're saying on the modeling end. It does help to have something quantifiable, but I guess my biggest suggestion would be to start with the feedback from the driver/car (& a good test driver makes this much easier) and then correlate that to the model rather than starting with something like RCVD & then making your car fit *their* model.

Being able to quantify the difference from step A to step B is very helpful and not overly difficult. It's very, very difficult to fully quantify step A and produce a setup from a clean sheet.

 

[Norm Peterson]

3. I get what you're saying on the modeling end. It does help to have something quantifiable, but I guess my biggest suggestion would be to start with the feedback from the driver/car (& a good test driver makes this much easier) and then correlate that to the model rather than starting with something like RCVD & then making your car fit *their* model.

Precisely why I took my flat ride thinking past such simplified approximations as the 10% rule of thumb and even the one about making the undamped front displacement curve match the undamped rear displacement curve at zero at one cycle. Any good model really needs to be able to show - or at least hint at - the things that you feel, not just the simplified introductory illustrations to the concept.

It's not about fitting my car to some rigidly constrained model. It's really about comparing a rough estimate of how it feels relative to what various combinations of springs and damping suggest in the model. And then looking at speeds on either side of the theoretical (damped) flat ride speed.


Of course, that's only part of it. You may still end up playing around with spring rates for LLTD and rear traction reasons, where you'd probably end up relying more on damping to keep the pitch motions/accelerations down.


Norm

 

[TeeLew]

So, what are your spring rates and where does that put you with respect to the F/R ride frequency ratio (undamped or damped)? Has it been an evolution or a single step? Which criteria are you using to measure relative success?

 

[Norm Peterson]

260 lb/in front (about 1.55 Hz undamped)
220 lb/in rear (about 1.62 Hz undamped)

FWIW, those are roughly where the GT350 wheel rates and undamped ride frequencies fall.

At about 25% critical damping, frequencies drop to about 1.5 Hz Front and 1.57 Hz rear.

Here's a plot for the soft OE springs (roughly 130/140) at 25% and 21% critical damping, respectively, at 50 mph (typical street speed). Undamped ride frequencies roughly 1.14 and 1.33 Hz. At the above dampings, 1.10 and 1.30 Hz.

And a plot for the current setup at 85 mph (a more typical track speed).

Norm

 

[Norm Peterson]

Just so you know, some years ago I was sent some Koni yellow shock force data for the front and rear S197 dampers, with force measurement data taken at every quarter turn of adjustment. Easy enough from there to get a good estimate of Cc for the known spring rates and car weight data, and from there, C/Cc.

I've made a couple of simplifying assumptions in arriving at those plots, but I feel that they're representative enough for the intended comparison purposes. I did go through that spreadsheet and made a couple of minor revisions that didn't change the plots enough to matter here.

Norm

 

[Norm Peterson]

One thing I found out from the Koni shock plots is that Koni' apparently chose linear damping rather than digressive at least for the S197. This made life easier for me with my simplified model (parameter C in C/Cc being the slopes of the various force-velocity traces) but it may also explain in part why there have been complaints of Konis being a bit on the 'harsh' side.

I'm pretty sure that pitch motion has some (probably smallish) potential for causing front and rear wheel loads to vary - after all, the sprung mass is developing some amount of rotational inertia as it rotates in pitch mode.


Norm

 

[TeeLew]

One thing I found out from the Koni shock plots is that Koni' apparently chose linear damping rather than digressive at least for the S197. This made life easier for me with my simplified model (parameter C in C/Cc being the slopes of the various force-velocity traces) but it may also explain in part why there have been complaints of Konis being a bit on the 'harsh' side.


Norm

I'll reply for real when I get free, but that sounds about right. Koni generally runs linear with a pretty heavy rebound bias. The whole Koni tuning method is ancient. They really haven't reassessed their philosophy since the 70's and when you consider that dampers are probably the single biggest chassis change to cars since then, it's pretty disappointing. Linear damping has its place, but, in general, you end up with not enough low speed damping and too much in the high speed.

 

[shogun32]

rebound is frequently (always?) linear.

 

[TeeLew]

rebound is frequently (always?) linear.

It can be whatever you want. Generally speaking, if you have a shaft adjuster on a damper, it adjusts rebound with a needle/seat bleed.

Sponsored