Matt Romanowski

Has anyone here done an advanced suspension analysis? Calculated the ride frequency, calculated springs rates, dampening curves, etc?

Anyone worked backwards from shock data gathered on the track?

Anyone used ChassisSim?

IcemanG17

When I read "Chassis Engineering" by Herb Adams I tried to do some of the calculations & just figured out that it is beyond my math abilities...... However I did learn quite a bit about how to adjust little things and got a better understanding of how the suspension-tires react as a unit to the road and car...

mobonic

I have looked into this and its not for the faint of heart and I'd say for non pro stuff its not worth it, better to just use a pro's buttmometer and pay him to set up your car for a track.

To do it right you need:
1. Full Data with Shock Pots
2. GOOD engineer with tons of suspension experience
3. Very good or pro driver that can feel and report back on differences
4. Lots of money to repeat this at every track you go to as set up for one track is not good for another.

If your chasing 1 or 2 tenths for a championship and you have the resources then DO IT.

Matt Romanowski

I don't think most of the measurements and math are too bad. I have the shock pots and all the car data.

I remember someone said they had a spreadsheet for figuring their shock valving, but can't remember who. Most of it isn't too hard, but the results are surprising.

DOUGLAP1

Yes, I built a spreadsheet that uses my corner weights together with an estimated sprung vs. unsprung weight to calculate a suspension natural frequency based on a specified spring rate. The calculations are based on wheel rate considering the motion ratio of the suspension in the normal travel range.

The spreadsheet then calculates the critical damping rate at various displacement velocities, and I use this to calculate an ideal range of damping rates, which are geneally about 70% of the critical damping rate.

I then sent these damping rate curves to AST (the guys that bought Moton), and they custom valves a set of coil over struts that are adjustable from about 50% - 100% of the critical damping curve. I had them do a shock dynomometer test and send me the curves with the dampers, so I know exactly which curve I am dialing in for a specific number of clicks.

The result is much better than I could achieve from my numb butt...

Matt Romanowski

I'm working on the same, but what seems to be calculated for spring rates is much softer than I would have expected. Haven't done the damping ranges yet.

Yargk

Keep us posted. By the way, what are the motion ratios? I've heard conflicting numbers in the past.

lordpantsington

I've only nicked the surface of this working on my rf project, and I know that I've made a few assumptions which might not make the results apply to real-world applications. If it helps any, I built up my spreadsheet from these pages:
http://www.optimumg.com/technical/technical-papers/ (Springs and dampers section)

Although, those papers might be a bit too generic for any specific car. When working 997 Cup values backwards to find the damping ratio from known rates to apply to 996, I get something that isn't exactly in line with the rules of thumb being presented. Even the multiplier/split for bump and rebound end up wildly different. I've also talked to a few suspension gurus and been told it is isn't an exact science. Those guys have my utmost respect. There is certainly some artistry going on.

There are two guys I follow on twitter Buddy Fey and Jeff Braun that drop interesting tech bits here and there. Both seem really approachable when it comes to questions and perhaps can show you the way? I know Buddy is a ChassisSim user. For the gurus out there it is their livelihood, so information may not be easy to come by if you DIY. Best of luck!

DOUGLAP1

It depends on the natural frequency you are trying to hit. I am using about 2.2 Hz front and 2.5 Hz rear. Motion ratio on my car (Boxster S) measured 0.9 front and 0.94 rear. Of course, the wheel rate conversion is the square of this motion ratio. I am using GT3 control arms, so I would guess the current 911's have a fairly similar geometrical motion ratio.

stownsen914

Matt,

I've spent many hours doing analysis for a suspension redesign project on my 914 using a program called susprog. It does pretty much everything for static and dynamic suspension analysis, and costs a couple hundred bucks or so. So far I've built a new strut-based front suspension for the car. Using susprog, I was able to move suspension pickup points around to get the roll center and camber change where I wanted them while minimizing toe change and scrub (side to side movement of the tire) as the suspension moves up and down through bump and droop, and also getting the ackerman where I want it. I built all new lower arms and struts and had to fabricate new pickup points on the chassis as well, so obviously a lot of work. I recently drove the car for the first time since the redesign and am extremely happy with the result.

I have the rear suspension redesigned as well using susprog, but haven't built anything yet.

Scott

amso3

Matt

I have some calculations somewhere. I put shock pots on a 997 and was chasing a spring/shock imbalance. I think it is similar to what you are doing. Not sure they survived my computer crash a couple months ago. I'll search and e-mail it to you if I find them

DOUGLAP1

That Susprog program looks quite good, and more economical than Wingeo. It seems that Susprog also has a number of Porsche suspension configurations pre-modeled so you don't have to start from scratch.

If I was looking at moving suspension pick up points, I would go with that for sure.

ProCoach

Matt, the information you have from John Block, the information in Bob Knox's book and the information in Jorge Segers' book should be enough of a foundation to start with the analysis portion using the measurements of the car to build the math channels required to accurately model what is actually happening for further testing and analysis.

The question appears to have morphed into two, separate threads.

The first is the measurement of "what is happening," which is achievable through Race Studio 2, Pi Toolbox, CDS TrackMaster and MoTeC i2 Pro.

The other is simulation and/or modeling of the potential behavior of a given geometry, using WinGeo 3, Simulink, SusProg3D and others, as well as Claude Rouelle's new software.

For the former, there are optimal behaviors measurable with the shock pots, optimal dynamic behaviors desired through analysis of the synthetic outputs from raw data (math channel products) and just the simple validation of proper function can go a long way.

For the latter, well, that is much more involved, but still valuable.

I use the former to capture what happens during "flowing brilliance," that magical performance that becomes the benchmark!

Matt Romanowski

I'm able to see what is happening without trouble. The problem I'm having is in how to take the identified areas of improvement and figuring out how to do the improvement. I know all the areas where my shocks are not as good as they could be, but how do I decide what valving will make things better?

With winter coming, what better way to kill time than run simulations?

ProCoach

Well, the way many top professional driver coaches and car engineers objectively validate setup improvement on the same day, the same track and the same driver is percentage full throttle over the lap... ☺️

Also, the objective way to set shocks is included in both John's advanced AiM course and in Bob Knox's book. There is a particular distribution of the histogram bins that result in better car behavior, universal among platforms.

It pains me that folks who are hard pressed to drive three laps within a tenth and a half together at places like Elkhart, the Glen and VIR are qualitatively validating shock changes in one or two click increments... But that's just the former chief mechanic in me talking, not the driver coach! ��

Do you have Ross' e-paper on shock adjustment?