C4 rear suspension: Changing trailing arm geometry

69427

The Artist formerly known as Turbo84
Joined
Mar 30, 2008
Messages
2,972
Location
Clinging to my guns and religion in KCMO.
As seems to be my habit, as soon as I finish something I keep thinking about ways I perhaps could have done it differently. To save myself time and headache with the C4 suspension transplant, I basically copied the stock geometry in the rear.

IM001059.jpg

The stock system uses a long and a short trailing link. I think I understand the basic reasons why GM did that. However, I'm curious. What would be the result if one was to use the long arm for both links, but arrange them parallel (so the knuckle doesn't cause bind up with the camber rod at full droop). Also, perhaps angle them both up to preserve some anti-squat. These changes would definitely modify the shape of the virtual swing arm, but I'm still trying to gather up all the pros and cons of this configuration versus the stock geometry.
I welcome thoughts on this.

ps: Note to Turtle. The black marks/scratches on the door and fender leading edge are tire marble debris, as I'm sure you are familiar with. This car has spent a lot of miles on road courses, and will return after a few (relatively lower speed) autocrosses to wring out any weaknesses in my design and welds. Just wanted to assure you I did not do all this work just for an engineering exercise. :idea:
 
I'm interested in this too. Could the C4 trailing arms be modified to mount to a single point on the frame, like C3 TAs? How much would that affect geometry? Would it cause the C4 toe or camber adjustments to bind?
 
If you mount them on a single point then that single point to your axle centerline is your virtual swing arm, this virtual arm is the beauty of that system (trailing arm is a fixed swing arm). Now, with these 2 you can create virtually every virtual swing arm you want (although as susp. travels the swing arm's length changes and it's pivot position does also move about).

if you make the 2 rods parallel you loose the virtual swing arm, it will be infinitely long. You also loose the limited anti squat characterstics you can have. The stock C4 arms are placed pretty good, I don't see a need to cahnge them. Keep in mind that messing about with these also changes the wheels position in the opening during suspension travel and that the hub will "rotate" and this affects the other geometry members too.
 
.......................

if you make the 2 rods parallel you loose the virtual swing arm, it will be infinitely long. Yes it's infinitely long, but it's still a virtual arm, though. Why would this be different from the stock virtual arm concept? You also loose the limited anti squat characterstics you can have. Why is that? If they're angled upward, the knuckle thrust vector should still provide anti-squat, shouldn't it? The stock C4 arms are placed pretty good, I don't see a need to cahnge them. Keep in mind that messing about with these also changes the wheels position in the opening during suspension travel and that the hub will "rotate" and this affects the other geometry members too.

I agree on the rotation aspect, as this puts extra bind in the stock (rubber bushed) camber link. By making the two trailing links parallel, the knuckle rotation should be eliminated. My main curiosity/interest in this type of modification is the substitution (where possible) of longer links in place of short links. Longer lengths generally give you less track or wheelbase changes with suspension movement (bump or jounce).
 
Yes it's infinitely long, but it's still a virtual arm, though. Why would this be different from the stock virtual arm concept?

Because of anti squat

Why is that? If they're angled upward, the knuckle thrust vector should still provide anti-squat, shouldn't it?

This one is kind of difficult to explain in words 9for me), anti squat effectiveness in percentages is the result of the swing arm (in your case virtual swing arm) in relation to the Cg location and the front wheel axis plane (vertical), an infinite long swing arm has no vector and as such loses these characteristics. Think about a pole vaulter where the virtual swing arm is the pole, as with the pole bauter, the higher he keeps his end of the pole, the easier it is for him to lift up, exactly similar forces are at work for anti squat.

I agree on the rotation aspect, as this puts extra bind in the stock (rubber bushed) camber link. By making the two trailing links parallel, the knuckle rotation should be eliminated. My main curiosity/interest in this type of modification is the substitution (where possible) of longer links in place of short links. Longer lengths generally give you less track or wheelbase changes with suspension movement (bump or jounce).

Short links exxagerate the hub mvoement, the position is much more dependant on the angle that the rods make, shorter lengths have a greater horizontal and vertical vector component change during suspension travel.

I think what you want to do is something like this:

2492dd47f7af2c.jpg
 
Yes it's infinitely long, but it's still a virtual arm, though. Why would this be different from the stock virtual arm concept?

Because of anti squat

Why is that? If they're angled upward, the knuckle thrust vector should still provide anti-squat, shouldn't it?

This one is kind of difficult to explain in words 9for me), anti squat effectiveness in percentages is the result of the swing arm (in your case virtual swing arm) in relation to the Cg location and the front wheel axis plane (vertical), an infinite long swing arm has no vector and as such loses these characteristics. Think about a pole vaulter where the virtual swing arm is the pole, as with the pole bauter, the higher he keeps his end of the pole, the easier it is for him to lift up, exactly similar forces are at work for anti squat.
Well, as difficult as that was to explain, I'll confess it was pretty similarly difficult to understand for me. The only way anti-squat seems to function in IRS applications, in my mind, is that the propulsive force from the tires gets inputted to the knuckle centerline, and this force from the knuckle then gets placed into the trailing links. If the links point upward (toward the C/G), then anti-squat should occur, as the propulsive force has both forward and upward components.
I agree on the rotation aspect, as this puts extra bind in the stock (rubber bushed) camber link. By making the two trailing links parallel, the knuckle rotation should be eliminated. My main curiosity/interest in this type of modification is the substitution (where possible) of longer links in place of short links. Longer lengths generally give you less track or wheelbase changes with suspension movement (bump or jounce).

Short links exxagerate the hub mvoement, the position is much more dependant on the angle that the rods make, shorter lengths have a greater horizontal and vertical vector component change during suspension travel. We are in agreement. That pretty much echoes what I said above.
I think what you want to do is something like this:

2492dd47f7af2c.jpg

That is exactly what I have been describing. If I'm reading the picture correctly, the two links are the same approximate length, and they would angle up slightly at normal ride height.
 
OK, a novice here trying to figure this stuff out.
How does anti squat differ between a long and short arm (virtual or not) if they're both in the same orientation to the Cg? Is it only that the geometry changes more with the shorter arm during suspension travel?
 
I did a quick comparison of stock C4 control arms and another with two short C4 control arms. These show how the virtual center migrates with 1 inch of wheel travel in both directions. It looks like the difference is that the center moves slightly forward (on jounce) with the 2 short arms and opposite with the stock setup.

449ca69dbda08b.jpg
 
Top