[4banger]

The whole 65-70% thing is really just where people find the best compromise between the speed of response and the ammount of overshoot when your spring/damper combo hits a bump. The 65 and 100% cases take about the same ammount of time, it is just that one case has no overshoot at all, and the other overshoots, then undershoots and then settles down. If you go over 100% critically damped, the response time is longer. There is still no overshoot, but the system takes a longer time to reach it's new steady state. I suppose that if you don't find the response too harsh, there may not be that big of a difference between say 65-110%.

Originally Posted by Turninconcepts.com

here's the thing that bothers me on this. Do we know that 65-70% is ideal for this car, and under what conditions?

What gets me thinking this is the math vs. the real world feel. While the numbers (and, Stretch, I do believe you have them correct unless we're both missing something major) show that these should be rather overdamped I have to argue that they feel fantastic when put on the car and driven.

So, where are we seeing the disconnect, and why?

[Turninconcepts.com]

Originally Posted by 4banger

The whole 65-70% thing is really just where people find the best compromise between the speed of response and the ammount of overshoot when your spring/damper combo hits a bump. The 65 and 100% cases take about the same ammount of time, it is just that one case has no overshoot at all, and the other overshoots, then undershoots and then settles down. If you go over 100% critically damped, the response time is longer. There is still no overshoot, but the system takes a longer time to reach it's new steady state. I suppose that if you don't find the response too harsh, there may not be that big of a difference between say 65-110%.

Which is where I'm curious. I'm going to shoot this over to the Ohlins guys along with the AST guys, and perhaps the Koni guys and try and get some feedback from them.

I would LOVE to be able to get the input from the engineer at Subaru who spec'd out the requirements on the stock dampers just to see their input, but I think that one may be darn near impossible.

[DuncanG]

One of the reasons for the disconnect is that the dampers have to be a compromise for all 4 suspension modes (heave, roll, pitch and warp). Ideally these modes would be separated and each have its own spring rate and damping factor. There have been some systems that go a long way towards that (eg Racing for Holland LeMans car).

Conventional systems get some advantage by being velocity sensitive and for track purposes you can make an approximation by using the low damper speeds for roll control and the higher speeds for heave (or bump) control. Remember also that the anti-roll bars (sway-bars) increase the spring rate for roll so that ideally would be accounted for in the damping-ratio calculations.


Millikan is a very old book and doesn't give much insight into modern dampers. Dixon's Shock Absorber Handbook is also rather disappointing from a practical pov.

Also rally dampers need to be rather different from race spec and there is no published info that I can find dealing with that. Remember Subarus are more akin to rally cars than race cars.

[stretch]

Originally Posted by DuncanG

One of the reasons for the disconnect is that the dampers have to be a compromise for all 4 suspension modes (heave, roll, pitch and warp). Ideally these modes would be separated and each have its own spring rate and damping factor. There have been some systems that go a long way towards that (eg Racing for Holland LeMans car).

Conventional systems get some advantage by being velocity sensitive and for track purposes you can make an approximation by using the low damper speeds for roll control and the higher speeds for heave (or bump) control. Remember also that the anti-roll bars (sway-bars) increase the spring rate for roll so that ideally would be accounted for in the damping-ratio calculations.

I agree with this; it could make sense to exceed 65% cd (or even 100% cd) to dampen a car's additional roll resistance. The thing is, with body roll (especially on a stiff car) being 4in/second or less (some say 3in/sec or less), why does % critically damped on the Ohlins peak closer to 6in/second and never fall below 65% critically damped even after that mark? That's what I don't understand. I'd expect the peak to be a little earlier but, more importantly, with a sharper roll-off thereafter. It's only because the Ohlins are always above 65% cd (even in the high-speed bump range) that I consider them overdamped no matter the application. I just don't see how they could be anything but.

Looking through Penske's catalog, the Ohlins dynos seem like they match the Penske equivelant of a "high flow linear" stack design. Perhaps the only way to properly compensate for roll resistance is using the VDP stacks (variable dependent piston) which can do peak cd at very low speeds (100% cd at 2in/sec), then taper down to a constant cd at a piston speed of your choice (say, 50% cd everywhere beyond 4in/sec). Penske's can become progressive at even higher (extreme) piston speeds if needed to prevent bottoming out. Neat stuff, but I'm sure other brands can do the same.

[DuncanG]

Originally Posted by stretch

I agree with this; it could make sense to exceed 65% cd (or even 100% cd) to dampen a car's additional roll resistance. The thing is, with body roll (especially on a stiff car) being 4in/second or less (some say 3in/sec or less), why does % critically damped on the Ohlins peak closer to 6in/second and never fall below 65% critically damped even after that mark? That's what I don't understand. I'd expect the peak to be a little earlier but, more importantly, with a sharper roll-off thereafter. It's only because the Ohlins are always above 65% cd (even in the high-speed bump range) that I consider them overdamped no matter the application. I just don't see how they could be anything but.

I don't think you can draw too sharp a boundary at 4"/sec especially on a compromise suspension. Consider going through a slalom and lets suppose a 2" bump/droop at max-g how long do you want to wait for it to take a set from one side to the other? I would want it to settle in no more than 0.5 sec (and I'm sure many will want faster), so that gives 8"/sec just on roll on average and the peak will be higher. Of course for a dedicated track suspension you can raise the spring rates and reduce cornering deflection and rate.


As for hi-speed damping, I'm not convinced that you can ignore bump damping as seems to be implied in these discussions. According to Millikan and Dixon for optimum performance (tire grip) we're looking for a df of around 60-70% but that is based on experiments with linear equal bump and rebound damping (ie text-book damper). To
get an overall damping factor of 65% with a rebound/bump ratio of 2 leads to df of 87% rebound and 43% bump. For a dedicated track setup you would possibly go for a higher rebound/bump ratio but 2.0-2.5 is close to the mark for a compromise setup.

[4banger]

Originally Posted by Turninconcepts.com

Which is where I'm curious. I'm going to shoot this over to the Ohlins guys along with the AST guys, and perhaps the Koni guys and try and get some feedback from them.

Any replies from the big guns?