Unfortunately, as the rocket continued to accelerate, everything went wrong around mach 3 or so. As I mentioned earlier, the nose fit left a bit to be desired at the pad, and it's possible that it wiggled a bit, or cocked over at a slight angle - it all happened too fast and high to know precisely what happened, but one thing is certain: when a rocket comes apart at mach 3, the re-kitting is pretty thorough:
Now, it's true that in that picture, everything is pretty much disassembled. However, two things pop out if you look closely (especially at the full resolution version):
1) The fins are near the
top of the debris cloud
2) The parachute is near the
bottom of the debris cloud
This indicates that the nose cone came off first, with the fins coming off only after the nose cone had departed and caused the rocket to go unstable at high speed. This is also backed up by the fact that one fin has a substantial chunk taken out of the leading edge, as if something hit it with extreme force. I believe that the nose cone actually impacted this fin's leading edge as it was leaving the rocket. Here's another picture of the debris cloud, but with annotations:
Another thing to note is that the piece of fiberglass tubing just below the actual nosecone is in approximately a million pieces. I found a couple of fragments of this after flight. I will admit, I have to eat my words on this one a bit, as this was one of the areas where I did not go to any special trouble to use extra strong or reinforced materials, and I paid for it in this case. The tubing should have been more than strong enough in a purely axial load however, which is what leads to my assumption that there was a very small amount of wobble or angle on the nose cone attachment, which caused very high sideloads which destroyed the tube. Most of the rest of the rocket was found after flight, but all that was found of this tube was a few small, red-painted pieces that were clearly from the fiberglass tube.
Interestingly, when the fins came off, they delaminated the performance rocketry carbon tube. This indicates that care must be taken in future attempts in how the fins are attached to the tube, and it may even be worth looking into other tubing options. Even though I did cut a moderately deep crosshatch, the delamination consistently occurred at approximately the depth of the crosshatch. Interestingly, one fin came back half-thickness - the forces when the fin was ripped off of the tube were sufficient to delaminate the autoclaved prepreg along the entire area of the fin. Given the forces required to do this, I think it is safe to say that no conceivable fin construction that is even remotely reasonable can survive after a shred at these speeds. However, due to the picture evidence as well as the substantial damage to the leading edge of one fin, I believe I can safely say that the fins were not the failure point.