demonfire wrote:can you just send me the carbon bar, in a specific length, with threaded inserts?
Ray, the short answer is "no" but the long answer is "shoot me an email."
CV12Steve wrote:Has anyone ever measured the amount of flex at the shock mounts over different surfaces? Inches or degrees of movement and the loading ± in pounds from static?
Yes but. Allan Slocum did this several years ago and posted the results on the Merkur Encyclopedia. It's difficult to take his results and pull meaningful data out of it for this discussion. From the way I read his test he put loads on opposite ends of the rear bumper and measured the resultant deflection between the shock towers. Of course, this isn't how the chassis is loaded in real life. Additionally, there should be no force being applied to the chassis by the shocks when there is no motion (ideally). The shocks only generate force when there is motion. (We know this is "mostly true" because, once compressed, a shock will tend to move very slowly back to an equilibrium position.) So, assuming Allan's load on the chassis puts everything into (a new) equilibrium, the only forces deflecting the shock towers should be related to the load put on the bumper. This is just chassis flex, not reaction to the dampers.
Allan's test effectively proved that there is movement at the shock towers due to loadings elsewhere on the chassis but doesn't show deflection from the forces put out by moving shocks. Is this a problem? No, unless we want to know the magnitude of deflection from the loads generated by the moving shocks. Then his test only tells us that there will be motion, not how much motion there will be.
Here is a link to the Encyclopedia and then a direct link to the article:
http://www.merkurencyclopedia.com/
http://www.merkurencyclopedia.com/Suspe ... _data.html
He concluded that the rear chassis spring rate is 15,400 lb/in. Again, this is going to include things like the stiffness of the bumper in the equivalent spring rate of the car. It's difficult to tell how this impacts test data because the car isn't typically loaded through the bumper. How much of an impact does the stiffness of the bumper have on the chassis when the car is normally loaded at the springs in the STAs?
Here's an additional teaser to translating Allan's data. When sketching the free-body diagram of the loads applied in his test load, it seems as though the loads on the shock tower from the damper and the loads on the shock tower from the weight on the bumper would be opposing each other. The bumper loads would tend to pull the shock towers apart and the damper loads would tend to push them together. This further confounds the data he provided.
I really appreciate Allan for posting this test data. It would be VERY difficult to get better test data. A chassis stiffness test rig usually involves the chassis rigidly fixed at one end and being torqued about the axis of the chassis, not bent along the front-to-rear midline. It gets worse, since one 22 year old chassis will not give the same result as a different 22 year old chassis, given rust and other factors. Yay.
So, that was a short question and a long answer.
The reason I ask is because c/f does relatively poorly in compression (as well as abrasion). The failure modes I‘d be worried about are buckling under compression and the adhesive joints at the ends under repeated compression/tension shock loads.
The cross sectional surface area of the carbon fiber tube is nearly 0.5 in². There is a lot of material to react the forces. The OD is 1.25" so it's got plenty of diameter to prevent buckling (diameter is primary dimension for calculating the resistence to bending). The wall is very thick (0.125") so delamination isn't likely going to be a problem.
The shock tower brace is an "ideal pinned column" so the hardest part of calculating its buckling load is finding the right values of the elastic modulus of the carbon fiber. The following web page covers critical forces in ideal pinned columns.
http://www.engineersedge.com/column_buc ... _ideal.htm
In this case, the variables are as follows:
E = 18,000,000 lb/in² (mimimum value of Young's Modulus for carbon fiber loaded on the axis of the fibers)
I = .07076 in^4 (second moment of area/moment of inertia)
L = 40 in (length of column)
I comes from the following formula:
Where D_O is the outside diameter and D_I is the inside diameter.
The calculation gives a critical force of 7850 lbs for the carbon brace. That should be well over the maximum loading that the brace will see by about a factor of 10.
The failure modes I‘d be worried about are buckling under compression and the adhesive joints at the ends under repeated compression/tension shock loads.
I agree. The adhesive joints are going to be the most highly stressed components of this design. I designed the inserts to have a very large area to react these forces and I chose a structural adhesive appropriate for highly loaded applications. This ain't the crap you get at Wal-Mart.
(Similar concern with the 2011-T3 bars as it’s not the ideals alloy for threaded applications; experiences anyone?).
If the loads are low enough they should be easy to design around.
The inserts are made of 304 stainless.
I hope that I've convinced you that I've planned this out fairly well. This is quite an expensive project (certainly the highest $/unit to produce mc²racing has done to date) and I've been exceptionally careful to avoid problems. The only problem that I've run into is rod end selection. That's why I bothered to post pictures of the recently-purchased rod ends in my previous post in this thread.
Graphics have been snagged from Wikipedia and "Engineer's Edge" (even though Engineer's Edge tried to keep me from snagging their image).
Another one of "Anglin's Retarded Engineering Posts" has now come to an end.
Edit: I proofread the whole thing twice for technical errors but somehow managed to miss "most highest." The error has now been removed.
