A arms and a 3rd link

[VLADD]

Rick
That jig is FREAKIN AWESOME    It clearly represents that you can find that ''sweet spot'' with pinpoint accuracy.
I like it
I was pondering the jig set up If you were to mount the 3rd link on a rod such as it is in the jig so it rotated on that axis (same as the control arms) you could eliminate all the controversy here, Altho it would add some weight , but I'm sure you have already thought of that since you built the jig

[Rick S.]

Maybe the confusion here is because you guys look at this problem as a link with a pivot (or heim) at each end. This is NOT how I view it. I only look at the geometry. This is nothing more than a geometry problem. (An exercise) This was all worked out using points and lines.

It all starts in the 2D rear view. In this 2D view the link is nothing more than a line. and this line pivots about a point. (the sweet spot.) When we assign a true length to this line (representing the link) it becomes the long side (hypotenuse) of a triangle. The 2D length of the line that we see in the rear view is the short side of this triangle. (Side opposite)
The point (sweet spot) that the line pivots about in the rear view, is the adjacent side of the triangle.
Now try to imagine this as just a triangle in space. When we cycle this model, the triangle rotates about the adjacent side. In other words, the adjacent side is the axis of rotation.
The adjacent side is fixed in space, it never moves. The hypotenuse and opposite side rotate about it. There is only one axis here because I'm only dealing with the geometry. And the geometry don't lie.
 

[fabr]

Neither does the deflection needed to make it work. If it were a true axis the mount was on there would be no change in length of any of the links.  The rear a-arms are a fixed length.The forward mount is a fixed pivot point and the link is a fixed length in the real world. The model will not function without deflection(link length change) and the cars setup will not work either without deflection allowing it to work so therefore there is no single pivot point at the "axis" where the front mount is located.No single point means no axis can exist. Now if you want to take those points and "average " them together and call them an axis feel free.That's what lets this work..So instead of making a model to prove your point(VERY nice model BTW) which allows you to prove it works in the real world that I feel no one is arguing about you need to do a 3D cad model that shows it working without deflection or ANY change in mounting points positions or link lengths. This cad model MUST replicate your cars setup to be valid.

[Engineer]

Neither does the deflection needed to make it work. If it were a true axis the mount was on there would be no change in length of any of the links.  The rear a-arms are a fixed length.The forward mount is a fixed pivot point and the link is a fixed length in the real world. The model will not function without deflection(link length change) and the cars setup will not work either without deflection allowing it to work so therefore there is no single pivot point at the "axis" where the front mount is located.No single point means no axis can exist. Now if you want to take those points and "average " them together and call them an axis feel free.That's what lets this work..So instead of making a model to prove your point(VERY nice model BTW) which allows you to prove it works in the real world that I feel no one is arguing about you need to do a 3D cad model that shows it working without deflection or ANY change in mounting points positions or link lengths. This cad model MUST replicate your cars setup to be valid.

Fabr,  The cad drawing of it working is posted on the "other site". 

[fabr]

Post it up.I'm all for being corrected .But how can that be if the link changes length as it cycles in ricks model?

[Rick S.]

The smallest deflection numbers on the jig were .005 (with the link to the rear) This tells us that the true deviation of the sweet spot is a number less than .005
If this .005 is what you have a problem with, then you are truly splitting hairs my friend.

[Engineer]

Neither does the deflection needed to make it work. If it were a true axis the mount was on there would be no change in length of any of the links.  The rear a-arms are a fixed length.The forward mount is a fixed pivot point and the link is a fixed length in the real world. The model will not function without deflection(link length change) and the cars setup will not work either without deflection allowing it to work so therefore there is no single pivot point at the "axis" where the front mount is located.No single point means no axis can exist. Now if you want to take those points and "average " them together and call them an axis feel free.That's what lets this work..So instead of making a model to prove your point(VERY nice model BTW) which allows you to prove it works in the real world that I feel no one is arguing about you need to do a 3D cad model that shows it working without deflection or ANY change in mounting points positions or link lengths. This cad model MUST replicate your cars setup to be valid.

The deflection is indicative of the real world.  A 3 leg stool sits flat, a 4 legged one deflects to sit flat, and a 16 legged stool would have to deflect to sit flat.

Think of it this way, Theoretically, when a fly lands on the deck of an aircraft carrier, it bends.  When you start welding parts together crap bends.  Then you get binding.  That is why most people prefer only 2 A-arms, no binding.

[fabr]

 Rick you  are correct.I am splitting hairs and it's been fun. Like I said IMO the idea of using that link is a very good one.Kudos to you on it! I do like it!
 Engineer I understand that and I've just been pushing the subject because I love these types of discussions.They are usually educational. This one IMO was great.

[Rick S.]

Masterfabr,
I will give you this one. Yes deflection will always exist. When we develop our link in cad or cardboard we pick the sweet spot based on an arc that is actually a slight ellipse. Always knew that, and never denied it. In fact it was a point I tried to get across on the original mbn thread. Once we pick our point we treat it as an axis. For our geometry model it is a true axis.
For the real world model we see the .005 deflection, caused by the fact that even the sweet spot does not truly exist. The only way an absolute sweet spot can ever exist is when your a arms form a true equilateral parallelogram. Same with the axis. therefore I am wrong.
If any of you are planning to try this link you do so at your own risk. I assume no responsibility if it fails.

[Rick S.]

Do I still need to post a video? :-\

[Engineer]

I stole this from Doug's post at "the other site" hehe.

I assumed that it was a circle.

The only problem that I have with the link is that it doesn't provide that much support to the carrier, because it is mounted with a heim at the rear.  When you hit the gas or the brakes, the wheel carrier tries to pivot around the end of the A-arm.  No help there, because its mounted with a heim.  If you hit a tree with the tire, once again, the carrier is going to twist rearward, and the link isn't going to do a lot to help because it isnt rigid to anything at the rear.

[fabr]

Masterfabr,
I will give you this one. Yes deflection will always exist. When we develop our link in cad or cardboard we pick the sweet spot based on an arc that is actually a slight ellipse. Always knew that, and never denied it. In fact it was a point I tried to get across on the original mbn thread. Once we pick our point we treat it as an axis. For our geometry model it is a true axis.
For the real world model we see the .005 deflection, caused by the fact that even the sweet spot does not truly exist. The only way an absolute sweet spot can ever exist is when your a arms form a true equilateral parallelogram. Same with the axis. therefore I am wrong.
If any of you are planning to try this link you do so at your own risk. I assume no responsibility if it fails.

I knew we really agreed all along.Again thanks for a great topic and your willingness to continue it to this point.

[fabr]

I stole this from Doug's post at "the other site" hehe.

I assumed that it was a circle.

The only problem that I have with the link is that it doesn't provide that much support to the carrier, because it is mounted with a heim at the rear.  When you hit the gas or the brakes, the wheel carrier tries to pivot around the end of the A-arm.  No help there, because its mounted with a heim.  If you hit a tree with the tire, once again, the carrier is going to twist rearward, and the link isn't going to do a lot to help because it isnt rigid to anything at the rear.

Thanks engineer.I'm at home on dialup today.I'll watch it at work tomorrow.

[fabr]

I stole this from Doug's post at "the other site" hehe.

I assumed that it was a circle.

The only problem that I have with the link is that it doesn't provide that much support to the carrier, because it is mounted with a heim at the rear.  When you hit the gas or the brakes, the wheel carrier tries to pivot around the end of the A-arm.  No help there, because its mounted with a heim.  If you hit a tree with the tire, once again, the carrier is going to twist rearward, and the link isn't going to do a lot to help because it isnt rigid to anything at the rear.

Engineer and rick,do you guys see a way to prevent what engineer is describing? Rick ,do you see it as an issue?

[Rick S.]

Yes the carrier is still trying to toe in and out. I'm not sure how you'd correct that. Other than getting the rod's attachment point directly over the tire's contact patch. The outer a arm bushings still see some of these forces, but they're reduced quite a bit because the link attaches outboard of the a arm attachment points on the carrier. If you use this on your car make sure to place the link as far outboard on the carrier as possible. At the same time keep it tight (to the carrier) fore/aft. I shoot for around 1/2" to 3/4" clearance to my wheel. Look at the overhead view that Doug posted earlier. The rod is very close to the wheel. This is what you want. If you can place it halfway between the a arm pivots and the contact patch of the wheel, you've probably cut the forces in half. It also changes the direction of the forces seen on the pivots. I have no experience with The Edge set-up without the link. I've only been told that the bushings keep wearing out. Tony from The Edge suggested I try this when I brought up my concerns. After 3 years, mine (with the links) showed no signs of wear. (So in my mind it's an upgrade.) This summer I did have to replace the upright axle bearings, because the seals went. But the a arms were still tight. A long link is going to offer more crash protection, but the attachment point is forced out away from contact patch of the wheel.
If you go with a shorter link you can move it inside the wheel more, and decrease the toe forces.
The seat is always an issue with a long link. The reason the link in the photo below is mounted low on the carrier, is because the seat drives it's location down. The ideal location would probably be a little higher. At the same height as the axle. You can really tailor this thing the way you want it. I prefer the crash protection of a long link. And the way I've been setting these up seems to work well on the bushings too.