Bent Axles/Camber Issues - Event Date: 08 May 2019 - 15 May 2019

All 3 depending on what kind of bump load you have.  The way the bending forces work on the axle tube which is a cantilever, you can have two cases. 

Case 1) If you hit say a speed bump with both wheels at the same time, then the stress on the axle is the same at and everywhere between the two frame rail support points. So the axle can bend anywhere or everywhere along that length. Outward from the supports the bending stress drops off, so bending at the very ends of the tube is unlikely. 

Case 2) If you hit a bump with one wheel only, the stress on the tube is max at the support nearest that wheel and tapers off both outbound and inbound from that point, so you are most likely to bend the axle right at the support.

Case 2 is probably more likely than 1 but both appear to be pretty common. 

One thing to point out is that the tube is really stiff and so does not deflect much at all before it yields under load. The point of yield is considered a bending failure because after that it won't spring back to its original shape. The points of maximum deflection are the ends and at the center, but it only deflects about a couple of tenths of an inch or so at either location before it fails. So its going to be hard to see the bend especially outward from the supports. Its easier to see if it has occurred inboard because you can run a string line and measure up in the center. At the ends its easier to see the bend as negative camber than it is to see it in the axle itself. 

Yes, you can push the axle up in the center to try to remove some camber, but that won't bend the axle ends back where they were it they were bent. Better (my opinion) to take it to an axle shop which has the proper equipment  to do it correctly. 

It doesn't help to only support the axle tube in the center. I think one thing that's maybe confusing folks is that the points of maximum deflection are not the points of maximum stress in this case. In a cantilever the points of maximum stress (the support points) are actually the points of zero deflection. The center is also one of the points of max stress in case 1, but all the other points between the two supports get loaded the same so supporting the center is only going to move the bending failure somewhere else. In case 2 the stress at the center of the axle is pretty low. 

It does help to reinforce the axle tube along its entire length of course, which is what "my" u bolted angle does. If I didn't support it continuously I would be creating a weak point at the center which would then become the point of failure in case 1.  

It does help to add support to the tube ends like both marwayne's and toyanvil's designs do. In those designs the supports can apply torque to the rpod frame rails, which potentially creates other problems. Marwayne and toyanvil have considered and dealt with that issue in different ways.