[Carisma]

Is it possible to do something for camber and front and rear pressures???

[stretch]

Possible: yes.

Within my capabilities: no.

I think I'm approaching the limits of what I know how to do both mathematically and in Excel. I'm trying to add more stuff, but it's getting very challenging (in a "I'm not having success, this is discouraging" way).

[4banger]

Stretch - If you need any help with Excel, just let me know. If I don't know how to do it, I can probably figure it out.

[kwlsti]

can you do anything for a car with a cage, b/c my body has no flex.

[stretch]

Originally Posted by 4banger

Stretch - If you need any help with Excel, just let me know. If I don't know how to do it, I can probably figure it out.

That'd be awesome.

My to-do list:

Plot a more accurate sway bar rate. As a sway bar pivots, the direction of force changes and the sway bar rate drops. For a short-radius rotation (up front in the STI is just over four inches when using the 29mm Whiteline), this may be significant. The geometry here is a little complex, any ideas?

Then, use solver to impliment this new, dynamic sway bar rate into the weight transfer calculator. If that works, we can also do dynamic rates from bump stops and progressive springs, too. I just need to take the time to figure out solver to make this work (I think).

Predict understeer and oversteer. Graph a tire's grip versus load, then use solver to pull values from that graph based on what we calculate weight transfer to be. Again, I need to learn solver- doesn't look bad, just haven't done it!

Originally Posted by kwlsti

can you do anything for a car with a cage, b/c my body has no flex.

Sorry, I have no idea what you're asking.

[4banger]

To answer the first question about changing sway bar rate as the bar twists, I am going to have to go hit up the book store to look at some books in their "library." I think I need to see how the swaybar rate equation is derived to understand this problem better.

Since you understand this way better than I do, let me throw a few things out there so you can get me down the right path.

Initially, the blade of the swaybar is close to parallel to the ground and the endlink is perpendicular to the blade. Once the blade begins to twist up (or down) the effective lenght of lever arm is shortened and therefore less torque is applied to twist the rest of the swaybar.

I think this is what you talking about because for a long blade, the change would not be as significant as a shorter blade. We may be able to come up with a crude approximation just using cosine of the angle that blade can be twisted to, but I really need to take a look at how the total equation is derived before I can even make a guess.

Do any of the typical suspension books adress this? This sounds like such a cool problem I'm going to head out to the bookstore right now and see what I can find.

Originally Posted by stretch

That'd be awesome.

My to-do list:

Plot a more accurate sway bar rate. As a sway bar pivots, the direction of force changes and the sway bar rate drops. For a short-radius rotation (up front in the STI is just over four inches when using the 29mm Whiteline), this may be significant. The geometry here is a little complex, any ideas?

Then, use solver to impliment this new, dynamic sway bar rate into the weight transfer calculator. If that works, we can also do dynamic rates from bump stops and progressive springs, too. I just need to take the time to figure out solver to make this work (I think).

Predict understeer and oversteer. Graph a tire's grip versus load, then use solver to pull values from that graph based on what we calculate weight transfer to be. Again, I need to learn solver- doesn't look bad, just haven't done it!



Sorry, I have no idea what you're asking.

[joe13]

thank you very much for sharing.

[stretch]

Originally Posted by 4banger

To answer the first question about changing sway bar rate as the bar twists, I am going to have to go hit up the book store to look at some books in their "library." I think I need to see how the swaybar rate equation is derived to understand this problem better.

Since you understand this way better than I do, let me throw a few things out there so you can get me down the right path.

Initially, the blade of the swaybar is close to parallel to the ground and the endlink is perpendicular to the blade. Once the blade begins to twist up (or down) the effective lenght of lever arm is shortened and therefore less torque is applied to twist the rest of the swaybar.

I think this is what you talking about because for a long blade, the change would not be as significant as a shorter blade. We may be able to come up with a crude approximation just using cosine of the angle that blade can be twisted to, but I really need to take a look at how the total equation is derived before I can even make a guess.

Do any of the typical suspension books adress this? This sounds like such a cool problem I'm going to head out to the bookstore right now and see what I can find.

I'm not aware of any books that address this, no.

The "endlink angle problem" as I'll call it is really a dynamic change to the motion ratio. Think of it this way: as the endlink becomes less vertical and more horizontal, the sway bar moves at a lower rate than the suspension arm it is connected to. This is bound to happen since the suspension arm moves vertically while the sway bar arm moves in an arc.

Once the endlink is parallel to the sway bar arm in droop (which must be happening for a stock STI's rear sway bar to flip, as has been reported) or horizontal, the motion ratio is zero because the endlink cannot push any further on the sway bar. That's an extreme scenario, but regardless there is a motion ratio change in smaller movements.

As you know, motion ratios are critical. Even a small change to the motion ratio can make a big change in wheel rate!

[4banger]

stretch,
I hadn't considered the MR problem, but I'm sure it makes a huge difference. I was thinking more that as the bar twists, your "A" is growing on one side and shrinking on the other. This makes me wonder if torsion bars are really variable rate springs. I have a friend who is a mechanical engineer that I'll run this idea by. The point you bring up about MRs is probably much more significant, but I would have never thought of it because it doesn't seem that there is much room for the endlinks to move. If it moves to much to the rear, wouldn't it run into the control arm? Plus I would think the articulation of the bolt/bearing (in the stock links) would help keep the endlink from having to actually move. I wish I had one of those bullet camers so that I could tape it to the underside of my car to find out how significant this is.

Originally Posted by stretch

Once the endlink is parallel to the sway bar arm in droop (which must be happening for a stock STI's rear sway bar to flip, as has been reported)

I can help get a picture of this. Next time I change my oil on the lift I can get pics at full droop all the way around.

I also found this quote from the Whiteline webpage, which makes me think that when Whiteline rates their swaybars, that they have already accounted for change in MR.

The same applies to sway bars but is further complicated by the presence of the swaybars lever arms which also determine the effective bar rate at the wheel. That’s why a 20mm diameter sway bar with short arms can have the same bar rate as a 30mm with long arms. Our Swaybar range often uses changes in motion ratio to design better and more appropriate sway bar shapes to modify the roll characteristics rather than just a straight increase. Our Whiteline "Blade" adjustable bars use these principles to achieve much higher roll resistance with a smaller diameter bar.

As far as adding this to the spreadsheet, I've thought about it and I'm not sure there is a way to accurately know the exact change in MR. One thing that I thought of was to do an experiment where you just jack one side of the car up and take pictures/make measurements of the swaybar/endlinks/control arm. I'm not really sure how you found the motion ratios, so you would have to help me out with that. Take the MR's that we find make a plot of them, and then fit a trendline so that we can make an equation for it. The only problem that I haven't quite figured out yet, is what to relate changes in MR to. I think the right choice is degrees of body roll. I haven't had time to go through all of the calcs in your spreadsheet to find out if how the swaybar rate is apart of the bodyroll calc. I would be a bad hack, but you could have a dummy cell for body roll that the MR cell references and then another cell that calculates body roll as before and then use solver to change the dummy cell to match the calculated bodyroll. Hopefully everything would converge.

[4banger]

For anyone still following - has anyone tried the Cobb rear sway adjustable mounts? I think they were designed to fight this exact problem. They let you adjust the mounting position of the bar so that the endlink is always straight up and down when not under load. They have a pic on their website that shows for a bar set on 'stiff' that the motion ratio goes to crap and that there is a signifcant decrease in MR.

Any thoughts?

[antonch]

Stretch-
Can we add tire spring rate to the resonant frequency calculator?

[Turninconcepts.com]

Originally Posted by antonch

Stretch-
Can we add tire spring rate to the resonant frequency calculator?

We've been talking about that for some time. There are a couple of problems we run into with that. 1) which tires? they're all a bit different. 2) we keep hearing that tires are self damped, but we have yet to find the formulas to prove that or get them to key into the equations.

[antonch]

Originally Posted by Turninconcepts.com

We've been talking about that for some time. There are a couple of problems we run into with that. 1) which tires? they're all a bit different. 2) we keep hearing that tires are self damped, but we have yet to find the formulas to prove that or get them to key into the equations.

I am really impressed how sophisticated the spreadsheet became. It looks modular enough that it can be used on other similar types of vehicles.

If there is a tire selector builtin it would be even more powerful. Given that on Subaru we don't stagger tires it won't make as much of a difference on balance but it would be interesting never the less.

I've found some published specs on some tires, although they don't look like they are for cars:
Hoosier Tire

[4banger]

Tires are not very self damping. I have read that they are about 7% damped.

Antonch - if you wanted to do a very, very rough guess you could measure how much your tires are compressed while parked to come up with a spring rate. Then add that to your coil spring rate as if they are in series.

[antonch]

Originally Posted by 4banger

Antonch - if you wanted to do a very, very rough guess you could measure how much your tires are compressed while parked to come up with a spring rate. Then add that to your coil spring rate as if they are in series.

Overall spring rate reduces when you put 2 springs in series right?