RE: CAP 232
Sounds normal to me. If you try to loop or turn any model harder than it is able you will get a stall and at high speeds that generally results in the wing snap rolling the model.
It comes from a few causes. The first is the model being too heavy and with a high wing loading means that it's easy to push the airfoil way beyond it's maximum lift capability and it ends up stalling. The second is that you try to push the wing with too much up elevator and it rotates the model fast enough that the wing goes past it's maximum allowable angle of attack for the airfoil and again stalls the airfoil. The two are actually related in that any model of any wing loading has a given maximum allowable angle of attack for the wing and weight that will result in a snap roll. LIghter models will loop or turn tighter before it happens and heavy ones will have it occur with larger loops or more open turns.
This is why you see the fun fly and other very lightweight 3D models doing stupidly tight loops without falling out into a snap roll. They are so light that even at that size of loop they can generate enough lift to provide the rotation without exceeding the angle of attack where the wing stalls. Of course they also get into the flips and tumbles where they intentionally induce a stall and then maneuver in a post stall mode but that's a different story.
The issue can also be compounded by the selection of the airfoil or building inaccuracies. For example using a diamond shaped stock for the leading edge and not shaping it to the proper rounded shape will tend to induce a stall earlier than a properly rounded profile done to an accruate shape for the airfoil.