I was unfamiliar with this mode of instability, I would bet the rather small fins and the fact that the fins are near a transition could have caused this. My question then is how will this instability mode be effected by larger motors which would reduce my stability margin. It seem that since this condition is caused by the fins needing a high angle of attack to do the job that more speed of the rail may not help, is that correct? Some of the motors I planned to fly this on will get it off the rail faster, but my ultimate plan was to fly an L265 which won't actually get of the pad any quicker despite high top speed and altitude.
As to the ejection issue, the charge space in my rocket is approximately 2" dia x 1.2" long. This gives a piston area of 3.142 in^2 and a volume of 3.77 in^3. I used three 3/32" ABS rods for the shear pins which has a shear strength of 3000 psi or 62 pounds to shear all 3 pins. This equated to just under 20 psi needed to shear the pins. Using the equation found here
https://www.info-central.org/?article=303 I calculated that I would need .038 grams of black powder, from the information found here
https://www.alaska.net/~aleckson/rockets/smokless.html this would suggest I only need 0.013 grams of the smokeless powder we were using. I did ground testing with .025 grams and the shear pins sheared but the rocket separated weakly. We then increased the charge to 0.35 grams and the rocket separated about 4-5ft. For the actual flight the charge was increased to nearly .05 grams. Another thought that occurs to me, is that these spaces aren't vented and the main charges can be heard to go off very close to each other and even combined they didn't do anything, not even shear the pins, not sure how that happened.