A few years back I attended a short sheet metal course to actually experience how to bend up sheet metal, install rivets, hi-loks, trim out damage, etc.... During that class the instructor emphasized to bend the sheet 45 deg to the grain. When I asked him why he emphasized this 45 deg bend, he told me that you get the best of both worlds, higher strength in the axial direction and less likely to crack across the radii. He told me that when he was taught years and years ago, they were taught to bend perpendicular to the grain or 90 deg to the grain and this was always shop practice, but an Engineer later critized him for this because the Engineer was basing his design on the "L" direction and the load was now going to be in the "LT" direction. This got me thinking that I did base all my calculations on the allowables for the "L" direction in the MMPDS (back then Mil-Hdbk-5) and I did agree that good shop practice would be to bend it at a 90 deg. I did a check and found for most aluminums that the L and LT direction allowables were very close, however, some materials (301 1/2H comes to mind) is very dependent on grain direction, especially for Fcy due to the rolling process effect on the material. Anyway, I have since left that company and I think that instructor has retired, so I don't know what they are teaching in that particular class now. For your issue, I feel that the bend radii is your issue, in the H32 condition, your min bend radii should be around 0.12", you could probably get away with a 1t bend radii if you bend it in the "O" condition.
