Long term suspension project

[ShatterPoints]

I was pointed here from mustang6g.

I am finally getting away from AutoX. It is fun but I want to participate in dedicated track days now. I am coming from a 997tt that I ran a custom valved coilover setup among other suspension mods. I have a factory 2017 gt350 and I will be fitting custom valved dampers and 450/750 springs.
Lurking here it looks as if there is more technical information so I will chronical my project here as well. I have also sourced a shock dyno and may soon begin testing shim stack configurations for myself. I will also be getting the rear control arms form cortex racing in order to fit traditional round end mounted dampers.

For my setup I have calculated:
66lbs/in = critical damping for the front
72lbs/in = critical damping for the rear

With 450lbs springs front and 750lbs springs rear, that yields a 2.0hz front frequency and a 2.2hz rear frequency. Based on milliken, milliken RCVD 1.5-2.5hz is the threshold, or break point range for subjectively "stiff" setups.

I will be running a 62% LS damping rate and a 25% HS rate in the front
&
a 54% LS rate with a 22% HS rate in the rear

Because I will have a two way shock I figured a forced match shock would be best as I could have bump and rebound on their own bleed circuits and I could dial in the correct asymmetry if needed. I have run force matched bump: rebound on my 997 and it was great. I think though after additional experience I think the lower damping rate and lower gas pressure helped the soften the feel quite a lot and that, maybe there is some use for the rebound bias on a road car. I plan to data log the OEM setup and then my new custom setup at COTA. I will be logging lap times with force matched settings, rebound biased and bump biased configurations. Is there any interest in this sort of thing here? I am out to quantify data for setups and to objectively prove what works and what doesn't. I know the 350 has terminal understeer, and I may see what ways there are to combat this in the future, although it may ultimately be a OE chassis / linkage limitation. If interested I can post my collection of links for shock data, shock tuning theory, as well as other sources which have inspired this project from me.

 

[Dave_W]

Very interested in your progress using objective data. I have yet to buy RCVD, as I'm still working my way through Smith's books, though I've read Puhn. I did attend an FSAE conference about 10 years ago where a younger Milliken gave a lecture. Please post your reference links when you have a chance.

 

[ShatterPoints]

Very interested in your progress using objective data. I have yet to buy RCVD, as I'm still working my way through Smith's books, though I've read Puhn. I did attend an FSAE conference about 10 years ago where a younger Milliken gave a lecture. Please post your reference links when you have a chance.

Books I have read and recommend:
Milliken, Milliken RCVD
The Shock Absorber Handbook - John C Dixon
Tune To Win - Carol Smith
Engineer to win - Carol Smith
Vehicle Dynamics and Damping- Jan Zuijdijk (JRZ's founder)
Several Jim Kasprzack seminars

Here is the mother load of reading, brew coffee, set aside a weekend, or two.
Shock design and calculation:
https://www.chassissim.com/the-damper-workbook/

http://www.kaztechnologies.com/wp-content/uploads/2014/03/A-Guide-To-Your-Dampers-Chapter-from-FSAE-Book-by-Jim-Kasprzak.pdf

Shock Tuning and theory:
This is from a motorcycle forum, however a shock is a shock and the shim configurations will in principle yield the same results. There is different leverages in forks vs struts so the speeds and force numbers between them will share no real similarities.

Dyno data –And what it tells us about how to tune a shim stack and control the shape of the damping force curve

Dyno Data on TT You'd think dyno data would be all over the place in suspension forums comparing shim stacks, bike setups and year-to-year suspension tuning changes. Turns out dyno data is pretty rare. If you dig around there are a few dyno runs scattered through the TT suspension forum. This thr...

thumpertalk.com

https://thumpertalk.com/forums/topic/677243-shim-deflection-and-piston-design/

http://www.shimrestackor.com/Code/Sample_Applications/Damping_Profile/damping-profile.htm (this is a must imo, I have purchased and used it already to moderate success. I need a shock dyno to really show where I am at)

SAGE Journals: Your gateway to world-class research journals

Subscription and open access journals from SAGE Publishing, the world's leading independent academic publisher.

journals.sagepub.com

https://www.penskeshocks.com/wp-content/uploads/2018/07/ITM-12.1-BLEED-SHIM-V-HOLE-05-01-12.pdf

Off-Road Shock Tuning Guide | How To Tune and Valve Off-Road Racing Shocks

Crawlpedia is an off-road encyclopedia featuring a detailed off-road racing shock tuning guide with details shim stack definitions and explanations.

www.crawlpedia.com

https://www.f1technical.net/forum/viewtopic.php?f=6&t=8937&sid=b880addb9c0bfe25f419754ab850a496 (lots of REALLY good info here)
https://www.clublexus.com/forums/gs...-shock-revalve-parts-1-3-a-2.html#post8869547 (a similar project as mine- I drew a fair amount of inspiration / motivation from this post)

Suspension/Stiffening - Critical Damping Analysis of...

www.iwsti.com

http://farnorthracing.com/autocross_secrets19.html (This whole site has extremely good info)
https://bluetoad.com/publication/?m=&l=1&i=470289&view=articleBrowser&article_id=2991492 (small magazine article which discusses why things are designed to run on bump stops IMO)

 

[xr7]

You should have given me the list back in March

 

[Norm Peterson]

Thanks for the links. I have some of the books but I don't think I've seen any of the other references.


Norm

 

[blacksheep-1]

let's cut to the chase, do you own a shock dyno?

 

[ShatterPoints]

let's cut to the chase, do you own a shock dyno?

I will soon. I finally sourced an affordable one as well as potentiometers so that on top of dyno data I can speak to real world performance as well. I hope to combine shim restackor, motec, chassis sim, and shock dyno data to deliver empirical data and not just "my best guess".

 

[blacksheep-1]

Rorhig..or something like that, grew up in my neighborhood, he invented the first one around 86

 

[ShatterPoints]

Rorhig..or something like that, grew up in my neighborhood, he invented the first one around 86

Yea I'd love to be able to bankroll a Roehrig dnyo... those things are expensive. I think I found a suitable alternative that I can afford anyways.

 

[blacksheep-1]

Penske..the only guys providing customer service at the runoffs.

 

[ShatterPoints]

I wanted to add onto this post for context, in both my own conclusions as well as information made available in a multitude of resources.

Here are some notable quotes from Milliken & Milliken Race Car Vehicle Dynamics:

"A brief discussion of choice of damper characteristics is given in Ref. 51, Chapter 3 (by Richard Hodkin). He points out that in practical dampers some friction exists, due to piston movement and seal-also inertia and hysteresis in the valves. Thus dampers seldom develop forces strictly proportional to velocity. He notes that the work done by the damper per cycle is measured by the area under the force-velocity curve. The bump setting of the damper is generally low to minimize the forces put into the body, whereas the rebound setting is larger and accounts for more of the damper work. This is the case when optimum ride is sought but he suggests that for road-holding, bump and rebound settings may be closer together. "

" Up to this point we have simplified the system by assuming the automobile has only one main spring and one wheel. As soon as we consider the more realistic model of an automobile with a wheel at each end, we should consider the dynamic behavior of one end relative to the other. In general a lightly damped vehicle will exhibit two modes of vibration: a "pitchy" bounce and a "bouncy" pitch, each with its own center of oscillation and frequency. The "pitchy" bounce has its center of oscillation outside the wheelbase and the "bouncy" pitch has its center of oscillation within the wheelbase.

Taking the example shown in the upper part of Figure 22.12, the front wheel will hit the bump first and begin moving at 84 cpm whereas the rear wheel will start moving a little later and move at 68 cpm. The comparative behavior is shown in Figure 22.l3 and will tend to give a pitchy ride. These frequency and mode positions are controlled by the static deflection (mass/stiffness) at each end. The upper part of Figure 22.12 shows a static deflection (referenced to the unloaded height) of 5 in. at the front and 7.5 in. at the rear. By softening the front springs and stiffening the rear springs we can give a front static deflection of 7.5 in. and a rear one of 5 in.

Hence the rear of the automobile will move faster than the front and tend to catch up, with the relative behavior as shown in Figure 22.14. This gives the "flat" ride. The dynamics of a vehicle can be modeled on a computer and run on a typical road where the input to the rear wheel is delayed according to the speed.

The third graph in Figure 22.15 shows how the soft-frontlstiff-rear car exhibits less than half the pitch angle displacement, giving the flat ride. To obtain the flat ride in a lightly damped vehicle it is important to have the static deflection at the front greater than the static deflection at the rear. With heavily damped vehicles, where the main ride resonance is well suppressed, this requirement is less important (race cars, for example)."

^^This basically translates to: A stiffer damper slows the frequency response == More static deflection. Paired with soft springs in the front and stiff springs in the rear causes the phase delay for flat ride to occur. (Flat ride confers comfort to the driver and passengers).


Some take-aways from this is, we know that our cars stock are still an OE passenger car. As such the car has terminal understeer (mainly for safety reasons). But you can also do the math to find that our cars are most likely lightly to moderately dampened as with the spring rates we have ~1.35hz front and ~1.55hz rear ride frequencies; or about 15% flat ride which is significant as roughly 20% is the delta between each end of the vehicle to provide the correct "flat ride".

Ultimately this tells you that the less aero and the more you run street tires the more you should target flat ride as resonances will affect both the comfort and performance of the vehicle.

As you load the car with Aero AND racing tires you will be adding damping (stiffness) everywhere. Both for road holding and to control aero pitch/ balance. So you should target flat ride less.

 

[ShatterPoints]

One more thing;

Reading on here about spring rates and such, I feel as if there should be more scrutiny regarding spring rates.

You want to have higher Motion Ratios (MR) as that lets you run a lower spring rate AND you get more useful stroke/ sweep from your dampers.

The Factory front on the GT350 is a coilover//strut so the damper and spring MR is roughly 1.0. The rear of the car has a factory spring MR of 0.50 (which is how it can be 900+ lbs vs the front's 194/240) and the rear shock MR is 0.70 which is not awful compared to most factory rear geometries. Cortex sells a rear control arm that improves the MR to 0.77 and allows you to fit a conventional round end shock to the control arm, allowing you to go rear coilover if you so chose. Here is an example of why that is so impactful:

GT 350 Rear spring MR of 0.50 uses a 914lbs spring to provide a ~1.55hz rear ride rate

GT 350 Rear spring MR of 0.77 uses a ~400lbs spring to provide a 1.55hz rear ride rate; nearly HALF the original spring rate!

 

[Norm Peterson]

Some take-aways from this is, we know that our cars stock are still an OE passenger car. As such the car has terminal understeer (mainly for safety reasons). But you can also do the math to find that our cars are most likely lightly to moderately dampened as with the spring rates we have ~1.35hz front and ~1.55hz rear ride frequencies; or about 15% flat ride which is significant as roughly 20% is the delta between each end of the vehicle to provide the correct "flat ride".

Ultimately this tells you that the less aero and the more you run street tires the more you should target flat ride as resonances will affect both the comfort and performance of the vehicle.

As you load the car with Aero AND racing tires you will be adding damping (stiffness) everywhere. Both for road holding and to control aero pitch/ balance. So you should target flat ride less.

Lots of damping quiets ride motions down pretty rapidly, which is justification all by itself for paying less attention to "flat ride" as your car more and more becomes a race car.

Sometimes I think that enthusiast-level introductions to flat ride do as much harm as good. There's a tendency to describe it in terms of targeting some fixed percentage difference between front and rear ride frequencies, which ignores damping and any notion at all that the range of speeds of interest might matter.


Norm

 

[Norm Peterson]

XenForo strikes again.

 

[ShatterPoints]

Yes... I posted as much here:

"The third graph in Figure 22.15 shows how the soft-frontlstiff-rear car exhibits less than half the pitch angle displacement, giving the flat ride. To obtain the flat ride in a lightly damped vehicle it is important to have the static deflection at the front greater than the static deflection at the rear. With heavily damped vehicles, where the main ride resonance is well suppressed, this requirement is less important (race cars, for example)."

Fixation on ride frequencies is just for practical assistance, as flat ride really occurs at a specific speed given any particular setup. In "general" it is a percentage split front to rear to produce a higher frequency response for one end of the car over another.

Personally I use flat ride as a justification to explore setups that are not 3:1 rebound biased because the internet said so.

 

[Grant 302]

for road-holding, bump and rebound settings may be closer together. "

This is one reason why believe in the way CorteX/JRi DA unit ties low speed compression and rebound in one adjuster. ‘Total’ damping and I suppose relative compression/rebound ratios can be adjusted somewhat via the high speed rebound. I don’t typically look at it that way, and I see high speed rebound as a ‘platform’ setting. And I generally tune it to as little as possible to control what isn’t damped by the low speed end of the curve.

 

[JAJ]

Yes... I posted as much here:

"The third graph in Figure 22.15 shows how the soft-frontlstiff-rear car exhibits less than half the pitch angle displacement, giving the flat ride. To obtain the flat ride in a lightly damped vehicle it is important to have the static deflection at the front greater than the static deflection at the rear. With heavily damped vehicles, where the main ride resonance is well suppressed, this requirement is less important (race cars, for example)."

Fixation on ride frequencies is just for practical assistance, as flat ride really occurs at a specific speed given any particular setup. In "general" it is a percentage split front to rear to produce a higher frequency response for one end of the car over another.

Personally I use flat ride as a justification to explore setups that are not 3:1 rebound biased because the internet said so.

You might want to take a look at this Youtube channel. Lots of interesting stuff about how suspensions work: https://www.youtube.com/channel/UCr_eB4JLRa59_XW1cTgYqlQ

The video I started with is this one:

Something you can do when controlling magride dampers is implement the 6DOF model of the suspension in the magride controller to get coordinated damping by corner, similar to and potentially better than the hydraulic version that McLaren built for the MP4-12C suspension. This kind of full body motion control is something you can't achieve with four separate hydraulic dampers.

In terms of my own experience with this stuff, I've gone down the same road you're on and spent a lot of time and money in pursuit of better handling. Double-adjustable KW and JRZ suspensions on a couple of cars, a handful of single-adjustable setups and now magride. While you can get good handling with normal hydraulic dampers, single or double adjustable or otherwise, the GT350 magride solution by Ford is eerily excellent at what it does. It may not be what you need for your purposes, but don't be surprised if it's remarkably close to your custom built solution.

 

[Norm Peterson]

Fixation on ride frequencies is just for practical assistance, as flat ride really occurs at a specific speed given any particular setup.

Theoretically perfect flat ride, yes. But there's a range of speeds either side of that where it's still close enough to flat that it's not particularly noticeable. I think you might want to more or less match that range up with your range of speeds. Plotting it up may improve visualization (as far as flat ride is concerned, you can't do anything about the first quarter cycle or so of front displacement, so I didn't bother doing anything with pitch in that region).



Norm

 

[Norm Peterson]

You might want to take a look at this Youtube channel. Lots of interesting stuff about how suspensions work: https://www.youtube.com/channel/UCr_eB4JLRa59_XW1cTgYqlQ

Thanks . . . something else to try to understand. Unfortunately for me, matrix operations were never a strong point


Norm

 

[JAJ]

Thanks . . . something else to try to understand. Unfortunately for me, matrix operations were never a strong point

Norm

No sweat - there's a programming language for that. It's called A Programming Language (APL): https://en.wikipedia.org/wiki/APL_(programming_language)

The first desktop computer I ever used was a purpose-built APL machine called an MCM/70 from Micro Computing Machines in Ontario. It could do the math in the 6DOF model in a single command. The problem with APL as a programming tool is that you can perform so many operations in a single sentence that it's almost impossible to read code you didn't write yourself.