quote Originally posted by Zac88GT: I believe you are thinking of instantaneous centers. There are indeed two of these. The roll center is located by the intersection of lines drawn from each instantaneous center through the center of the tire contact patch.

Ah your absolutely right there, I was thinking of it as IC's, but there are also two methods used by people for determining RC's, one I feel more correct than others.
1. As you described, Find the IC's of the suspension linkage, and the axis between the center of traction and the corresponing IC. The intersection of both of these axis will located the RC. (most correct) this method shows at what point the body will "pivot" load transfers in lateral acceleration.
2. Another way is the intersection of those IC's to Tire axis with the vertical axis that passes through CofG. for which there are two seperate points. This way can be more revealing of Jacking forces and how the RC's interact with the cars body. But this, I think, shouldnt be called a roll center becuase it isnt the true datum for zero roll moment.

http://www.neohio-scca.org/...e%20Dynamics2007.pdf
this shows more of what Im talking about, per authors terminology he calls these points Force Application Points.

Vehicle Dynamics isnt an exact science, but applied kinematics and FBD can be vary revealing.
Afterall this is all done, we are still depeding on a street tire for which we dont know much about except overall characteristics, not exacting data.

quote Originally posted by Bloozberry: Thanks for your interest FieroWannaBe. I realize how complicated this subject is, and by no means do I pretend to understand a 10th of it. My ultimate goal here was to see if there was a not-too-complicated way to portray the effects of the changes I plan to make. I knew I needed to map out the suspension coordinates (done), map out some basic kinematics (mostly done, just not all posted yet), and hopefully out of the mess, a few simple conclusions about the effects of the longer wheel base, longer control arms, and a lowered CofG (supported by graphs) would fall in my lap. Theoretically each of these changes should improve performance, but so far, the "falling in my lap" part is still "up in the air". If you or anyone else believes they know how portray these modifications in the best light, then by all means send me a PM with your ideas. My goal with this thread however is to wrap up an abbreviated suspension analysis within the next three or four posts knowing full well there is much, much more to explore... perhaps in another thread. What I'd like to cover in the next three or four posts is the following: 1. the same graphs as above except with my final front configuration using the longer control arms and longer wheelbase with front drop spindles (rather than drop springs) vs stock; 2. the same graphs as above for the rear suspensio but also include a graph depicting the change in roll axis vs bump, and finally; 3. the effects of the redesigned rear cradle. After that, I really want to concentrate on building the rear cradle and emphasize construction instead of planning.

I understand and appreciate your approach, and the work you have done so far is more than can be said for 99% of all modified road cars and 95% of all race cars. So I appluad you. I wont attempt to side track your process here, I only wanted to suggest a method of analysis for maybe a slightly bigger picture of modification effects. I know there are other threads were in depth indulgence and analysis would be more appropriate. I truely appreciate all your efforts, and then uncharacteristic willingness to share this hard earned insight and information. Im learning myself, Im no expert, jsut trying to provide my understandings and insights.
Keep up the good work.

[This message has been edited by FieroWannaBe (edited 01-24-2012).]

quote Originally posted by FieroWannaBe: I won't attempt to side track your process here, I only wanted to suggest a method of analysis for maybe a slightly bigger picture of modification effects.

No need to apologize, in fact your feedback made me realize I was missing an important graph. Although it may not be the specific graph you requested (RC-CGH vs Roll), I do believe it's the one you may have meant. When Zac initally posted some graphs before we realized there were some errors, he included one like the one I think you recommended. The graph below shows how the roll center migrates vertically and laterally across the car for a given roll angle.



There's lots of information on the graph but with the help of the drawing below, it's actually quite intuitive. In bump, the roll center only moves vertically up and down the centerline of the car, but in roll the roll center also moves across the car as well as up and down since the location is a function of what both sets of control arms are doing. Transposing the curves of the above graph onto the rear view of the front suspension, we get this:



Interpreting the curves, it's quite clear that the stock geometry (semi-circular blue line) does an excellent job of keeping the roll center from migrating in both dimensions. That minimizes the change in handling the driver feels as the car rolls (at least for the front suspension... we'll soon see it's not the case for the rear!) The 3" longer control arm configuration (purple line) shows that these arms impact the roll center location in three ways: they drop the roll center, cause it to shift 130 mm (5") to either side of the centerline when rolling left or right 6 degrees, but they also make vertical movement of the roll center nearly zero.

Finally, the full monty configuration (red line) impacts the roll center even more negatively. Once again, the drop springs are the primary cause of the additionally lowered roll center, but the side to side movement of the roll center is even more severely impacted in this config especially considering that the data for this curve only goes to 4 degrees rather than the 6 degrees left to right that the blue and purple lines depict. (The reason for the lack of data past 4 degrees is due to the suspension hitting the bump stops.) At 4 degrees roll, the roll center moves almost 250 mm (9.8") to either side of the centerline, more than a 260% increase over the amount the roll center moves with just the longer conrtol arms (at 4 degrees roll)! This appears to be the most significant impact that the drop springs incur upon performance.

Potential Roll Center Movement Solutions: My understanding is that better lateral control of the roll center with the longer-arms-only-option could be achieved by restoring the stock angles and proportions between the upper and lower arms. Remember that the longer control arms use the same inner mounting points and the same knuckle height as stock, so the effect of the longer arms is to lessen the angle between them by several degrees. To restore the angles, that would mean either raising the lower control arm mounting point by several millimeters or lowering the upper mount several millimeters, or some combination of both. Additionally, the upper control arm would have to be shortened to restore the OEM ratio between the upper and lower arms, which would also necessitate moving the upper control arm mount outboard by an equal amount to maintain the upper ball joint in the extended location. As for improving the drop spring performance (red line), I believe the swap to the drop spindles would bring it in line with the purple line, and then the same modifications to the control arm mounts and ratios would be needed to bring it closer to stock.

[This message has been edited by Bloozberry (edited 02-09-2012).]

------------------
New owner of Silver 88 Fiero GT and second time owner of 1985 Fiero GT. Bought my first fully loaded Fiero GT new in 1985. Fiero's are Fabulous, Fix'em and have Fun! Note, Avatar picture is Mr. Bean (not me, ha ha).

For anyone keeping any copies of the graphs I've been posting, you should be aware that I made changes to the graph and text concerning the front toe vs bump on page 12 of this thread. I needed to make these changes when I discovered that the Lotus software program does not follow SAE convention for the polarity of toe measurements, which screwed things up.

With that out of the way, now it's time for some rear suspension analysis. I'm going to show the results of the same three configurations I used in the front including: OEM stock; a 6" track width increase; and a 6" track width increase combined with a 3" wheelbase stretch and 2" spring drop. To help visualize the differences, here's a before and after drawing showing the stock configuration in blue, and the extended and lowered configuration in red:



Since most of the same concepts apply equally well for the rear as for the front, I won't go into the lengthy explanations about toe, camber, and roll centers unless there are noteworthy differences. For ease in comparing the different graphs front to rear, the graph backdrops are color matched i.e. the two toe graphs are light green, the camber graphs are light blue, etc.

I'll start with the rear toe versus bump graph.



As with the front toe, the stock '88 Fiero rear knuckle is set statically during alignment with some toe-in, 0.15 degrees to be specific. Again, I've chosen to factor this out of the curves above since the toe will neutralize to zero once the car is underway. According to the service manual straightline driving forces will causes manufacturing tolerances to be taken up and reduce the toe to zero. In the stock configuration, it's clear that raising or lowering the suspension causes the wheels to toe-in considerably. This explains the Fiero's tendency to understeer despite the front wheels initially wanting to oversteer slightly. In roll, the outside rear tire toes-in causing the front end of the car to turn less than is commanded by the steering wheel input.

The same happens with the 6" track width increase, only to a much lesser extent so this would tend to neutralize the understeer characteristic compared to the stock car. To increase the rate of toe-in, the front lateral link would need to be shortened and the cradle mount location for it would have to be pushed outboard by an equal amount. This would shorten the arc of the front link and cause it to pull inwards on the front of the knuckle as the jounce increased.

The 2" dropped suspension increases the rate of toe-in considerably causing even greater understeer in roll. In this case, the front lateral link would need to be lengthened somewhat and it's mounting point on the cradle would need to be moved further inboard to reduce the rate.

Rear Camber: There's a lot of discussion on the forum about the poor rear camber characteristics of the Fiero. So just how bad is it? Here are the two most revealing graphs about the rear's performance... one under bump, the other under roll.





Both graphs above show camber after subtracting 1.0 degree of camber as per the stock alignment specs. The effect is to bias the camber in the negative direction to compensate for the poor negative camber gain. The most revealing is the camber vs roll graph which shows that even after the compensating -1.0 degree has been added, there still exists a 3.76 degree difference between the rear wheel's tire contact patch and the road surface at 6 degrees of body roll. This is 0.85 degrees worse than the front tire contact patch.



Were it not for the static alignment spec, these numbers would be 4.76 degrees between the road surface and tire contact patch at 6 degrees body roll, a full 1.85 degrees worse than the front.

The graphs show that the longer control arms improve the negative camber gain slightly but the addition of the drop springs worsen the situation significantly. This is keeping with a commonly known characteristic of strut type suspensions: provided the angle between the lower control arms and the strut as viewed from the rear remains less than 90 degrees, camber gain in jounce will be negative. Shortening the rear springs has the effect of angling the control arms upwards, bringing the angle between them and the strut closer to 90 degrees. The closer to 90 degrees, the slower the negative camber gain, and past 90 degrees the camber gain becomes positive.

Camber Gain Solutions: For a Chapman strut rear suspension, there are several solutions to gaining more negative camber:

a. add even more negative camber to the static alignment, pushing the curves above further to the left, though leaving the rate of change (ie the slope of the curves) unaltered. This would have an overall negative effect by causing uneven wear on the rear tires since most straight line driving would be on the inner edge of the tires. It would also result in a loss of straight line rear traction since the tire contact patches would not be in full contact with the road. Lastly, because the shape of the curves would be unaltered, the new static camber value would correspond to maximum grip at only one specific angle of body roll and be compromised at every other angle.

b. lower the attachment point of the lateral links on the knuckle in a manner similar to what Fieroguru (PFF member) has done here in this thread: www.fiero.nl/forum/Forum2/HTML/121052.html which would change the slope of the curves as shown here: www.fiero.nl/forum/Forum2/HTML/117227-7.html . The benefits shown are significantly improved roll center control, and about 1 additional degree of negative camber gain over 75mm of jounce. The disadvantages are a gain in toe-out (negative toe) of about 0.8 degrees over 75 mm jounce which increases oversteer, and the complexity of the necessary metal fabrication.

c. decouple the fixed link between the strut and the upper knuckle making it a pivoting link, and replace the upper knuckle control with an upper control arm. Depending on the design, the designer could theoretically create the ideal amount of camber change throughout the entire range of bump and roll. The difficulties with this approach are the actual design process which must optimize the control arm location and dimensions given the physical constraints of packaging the arm within the available space.

Next up: rear anti-squat and roll center movement.

(Edited to update link to Fieroguru's knuckle adapter)

[This message has been edited by Bloozberry (edited 02-11-2012).]

[This message has been edited by FieroWannaBe (edited 02-13-2012).]