Well, everybody seems like they agree on damper valving, so I'm going to go ahead and be the turd in the punchbowl...as usual.
---------------------------------------------------------------------------------------------------------------------------------------
First, my philosophy has been developed through trial & error and I’ve now wrapped a mathematical justification around it (which I’m going to leave in the background for simplicity). Usually, it’s done the other way around. This approach has evolved mostly through trial and error. Keep in mind, I’ve been in the position to make a lot of errors. I’m not making my argument from authority, I’m just being honest that I can’t provide a book and page number to support my approach. That may or may not be a good thing.
My opening given statements:
- Rebound damping controls the unsprung mass while compression damping is more useful in controlling the sprung mass.
- I’m perfectly comfortable with a damping ratio higher than 1 (mathematically ‘overdamped’) in the low speed ranges of the curve.
- I can justify just about any damping curve shape (linear, progressive, digressive, etc.) depending on the road inputs. It just so happens that most tracks are similar in terms of the spectrum of road inputs (within relatively small ranges).
- I tend to run relatively soft springs (which improves mechanical grip, bump/curb riding and tire life) and provide handling response with low-speed damping forces.
My first statement is probably my most controversial. It’s opposite of the Koni tuning advice sheet (which they’ve handed out for about 40 years, same sheet) and most experts. I arrived a justification through these thought experiments.
Thought experiment A: If a car is driving on a bridge and we somehow secure the body of the car and pull away the bridge, what will happen? The wheels and tires will move in the droop direction and the rebound forces will oppose the spring load and the inertia of the unsprung mass. In this we’re scenario, we’d be calling the car chassis the ‘ground’ of our spring/mass/damper system, which I think is valid in most cases because the sprung mass is usually ~7-10:1 greater than the unsprung.
Thought experiment B: If a car is driving down the road and hits some sort of protrusion bump (ordinary speed bump) at speed, it’s the compression side of the damper than will initially be active. In this scenario, we need to provide enough damping to reduce unsprung mass over-shoot, but not so much as to disturb the sprung mass. The difference here is we have the entire sprung mass inertia resisting the road input (as opposed to the much smaller unsprung mass in T.e._A). We have the spring, damper and unsprung inertial forces to absorb the input kinetic energy produced by the road. The forces of interest are on the compression side of the damper.
To extend these thoughts just a bit, on the back side of the bump, we’re back to experiment A. If we assume the tire will momentarily leave the pavement in this scenario (due to overshoot). The more rebound damping we have, the longer the tire will be disconnected from the road. That’s bad. So, given that I favor this view of the motion, it’s really no surprise why I end up with the compression forces being several times higher than the rebound side.
It’s fairly common to calculate forces using a ‘damping ratio’ from a quarter car mathematical model. This is a ratio of the force used compared to the force necessary to limit the system motion to exactly 1 up/down cycle when excited. Several books (RCVD and The Shock Handbook among others) have graphs which leads us to keep this damping ratio in the range of 0.7 (70% of critical damping) for performance applications and closer to 0.3 (30% critical) for ride comfort. The problem with these recommendations is that they seem to emerge from thin air. For me, they never adequately explain how these numbers are derived.
I see damping as a way to control inputs. Broadly speaking, cars see 2 types of suspension input. The first is low frequency, which is the type of input which we want the body of the car to follow. Think CA freeway roller bumps or anything resulting from driver inputs. The other type is high frequency bumps. A pavement transition, pothole or brake zone chatter bumps are all good examples. With these, we want the tire be able to follow the road perfectly (so as not to disturb the contact patch), but we don’t want any of the road input to influence the chassis. A well set up baja truck busting across the desert is a great visual example of this with wheels bouncing everywhere, but the chassis stable.
Generally speaking, at racetracks, we find that low frequency inputs correspond to low damper shaft speeds and high frequency inputs correspond to high damper shaft speeds. It’s *very* important to understand that these are not the same things, just that we often find this correlation. If a track were to have a large, long duration undulation, then we could easily see high shaft speeds from a low-frequency input. When tuning with driver feedback, I refer to low-frequency as “Roller bumps” and high-frequency as “Impact bumps”. That seems to be terminology to which most people can relate.
I don’t want to get too deep into the weeds, but it turns out that low frequency inputs are better controlled by increasing damping and high frequency inputs are better controlled by decreasing damping. What we want is a damper which provides both. When developing a damper valving package, I’ll pick a transition (or knee) point, 3 in/sec damper velocity gives a good first cut. In the 0-3 range, which is predominately low frequency, I have forces in excess of critical damping. It will depend on many factors, but anywhere from 110% to 175% critical is within reason. At the knee point, I dump force and by the time the damper is moving 10 in/sec the damping ratio is much closer to that ‘ride comfort’ number of 30-50% critical. It depends on the damper design/valving as to how easily this is accomplished, but it can usually be done to some extent. Keep in mind, this is painting with a very wide brush.
There are plenty of outlier racetracks which give different road inputs and require different damping curves. The relative importance of the low/high shaft speed vs. low/high frequency inputs is really the determining factor concerning the shape of the curve. This is why I say I can justify about any curve shape. I don’t believe there is a ‘perfect’ curve shape generically or even any perfect curve shape for a given car. It’s all the factors including car, tire and road input.
I tend to run springs on the soft side to gain mechanical grip. The downside to reducing spring rates is that the car will gain compliance, but lose response to the driver. When the driver reports a lack of response and vague car feedback, that’s a pretty good clue you’re too soft. However, if you run higher damping forces in the range of inputs which relate to driver feedback, you find the spring rate floor become lower. It’s a trade which is usually beneficial to make.
Lastly, I want to calm concerns about too much compression. I’ve often been told that too much compression will cause the tire to skip and lose grip. I think this has been found by people who are already running excessive rebound. What they’re bumping up against is too much damping overall. It’s not that compression per say is too high, it’s that it’s too high for the existing amount of rebound. By reducing the rebound, I’ve found compression forces can be safely increased without causing compliance problems.
I don’t particularly want anyone to believe what I’m saying. I’m not writing a gospel or trying to convince everyone I’m ‘right’. I’m simply explaining my process and providing some food for thought. I’m interested to see its reception.