If you've been following my occasional rants in Appendix J on suspension setup, you might have noticed some critical issues. Much of what I talk about is based on theory, math, a lot of assumptions, and rudimentary measurements while the car is sitting still. It's basically all that the average Joe can do when it comes to setup.
If you wondered what the big boys really do to figure out suspension, and more importantly damping setup, the answer lies in the seven post machine. Unfortunately these machines are rare, almost as rare as full size wind tunnels, and only F1, NASCAR teams, and OEMs have the budget to gain access to one. Finding people smart enough to set up and use one isn't easy either. Just last week I finally had the opportunity to see one in action at KW Suspension's global headquarters in Germany (KW managed to buy theirs off of BAR Honda).
The seven post machine is, like the name implies, made up of seven hydraulically actuated posts attached to a complete car. Four main posts have pads which support the wheels and tires, much like some Disneyland virtual reality rides. The other three posts (three points define a plane) are attached to the car's body and push or pull on the chassis to simulate aerodynamic loads and other body motions caused by inertia. This now turns the car into a giant, very expensive driving simulator, which might be great for games, but doesn't do anything if you don't know what to do with the data that it collects.
To understand that, you have to first remember what the suspension on a car is supposed to do. A suspension's primary purpose is to insulate the car chassis from unnecessary motions and vibrations while at the same time keeping the tires on the ground as much as possible. In essence, it has to control the excessive motions of two masses attached to each other, the car itself and the wheels and tires. It does this in two ways.
The springs support the vehicle's weight and also generate the normal force that keeps the tires on the ground to generate friction. Springs typically play a static role in suspension, i.e. when everything is sitting nice and still, whether the car is just parked or is rounding a perfectly smooth, constant radius corner. Figuring out spring rates and roll rates is fairly easy when you know a car's mass.
Damping controls the dynamic motions in the suspension. In other words, springs control how much a suspension moves, and damping controls how fast it moves. More damping means more resistance to the spring compressing or extending in a short time span. Because of this, damping control takes priority over spring rates in suspension design, and the difference between a good and bad suspension comes mostly from how well the shock is tuned and maintains damping control.
The problem is that you can't just go off paper theory and spec a spring rate, stroke, and appropriate damping force based on motion ratios and weights. It just doesn't work like that. There are a lot more interactions in chassis stiffness, bushing compliance, and suspension geometry that we can't calculate. The best example is the difference between a BMW 3 Series and a Mercedes-Benz C Class. Both cars share the same front engine, rear drive layout, weight, weight distribution, and comparable suspension layout. Yet the wheel rates and damping forces to make either car feel right are totally different from each other, and what you would figure just from a rough calculation.
The seven post machine offers probably the most scientific way of testing a real world suspension and how it interacts with the car. By using the machine, there's no need to look at motion ratios, spring rates, damping forces or calculating the exact weight supported by each wheel and partially sprung weight like the suspension and control arms. The machine simply simulates road inputs and sees how the chassis reacts by measuring the difference in acceleration between the wheels and different points on the chassis.
In terms of evaluating damping and how well it controls body motion, the seven poster operates in a heave mode. For this type of testing, only the four posts exciting the wheels are in motion and the chassis is not constrained. The four posts will heave each corner up and down in unison to compress and extend each suspension and do this at increasing speeds from 1 to 25 hertz. (A video of this test is available on www.SportCompactCarWeb.com).
Six accelerometers are placed on the car, one on each wheel and two on the chassis on the front and rear axle lines between the wheels. These accelerometers measure the acceleration forces on each wheel and the front and rear of the chassis as the machine sweeps through each frequency. Since most cars are not perfectly balanced in front to rear weight distribution, the idea is to figure out the delay between chassis motion and wheel motion. If improperly tuned (like in the video example), the chassis will show signs of unnecessary chassis pitching at the natural frequency of the front or rear suspension. All of us have felt this while driving on concrete paved freeways with constant undulations that cause the car to dive and squat unnecessarily because the chassis is moving more than the wheels were moved.
This is due to the spring mass natural harmonic frequency of each axle, which is stipulated by the spring rate and the masses attached to it. Every system has it, including you. Just jump up and down on your bed, you'll notice that the frequency of your bounces is pretty consistent and difficult to change even if you tried. That's the harmonic frequency of your weight and the spring rate of the bed.
Without proper damping control, this pitching will occur as the suspension is excited at the natural frequency of the front and rear suspension. This is the whole point of the heave test, to find the natural frequencies where the car isn't happy and adjusting damping of each axle to eliminate unnecessary motions.
Mathematically speaking, this goes back to my rant about spring mass systems, critical damping forces, and damping ratios (August 2007). By looking at the relative acceleration of the chassis to the wheels, you can figure out the delay in motion. Perform a little FFT (Fast Fourier Transform) and covert the data from the time domain into frequency domain, and you can look at the phase delay in terms of degrees and frequency. Without going into this in detail, ultimately, a delay phase angle of 50 degrees correlates to an ideal (according to some schools of thought) damping ratio of 0.6, or slightly underdamped. A critically damped suspension will have a phase angle of 90 degrees. Try explaining this to your drunk car friends next time you're up late wrenching.
The seven poster also has several other modes of operation. Twisting or warp tests can be used to establish the influence of anti-roll bars. A full track simulation, which is what F1 and NASCAR teams typically use this machine for, can run the car through a known track using previously collected data to ensure that the suspension is near perfect from a mechanical perspective.
While few people will ever see or understand the immense possibilities that a seven post machine offers, it's good to see one that's trickled down to the aftermarket and being used for products we can afford. It might be overkill for a street suspension, but in some cases, even suspension gurus with highly calibrated butts have to resort to the hard data the seven post machine spits out.