If you've read this thread you'll see that I'm look for this at alternatives for a tower. A tower is a large piece of specialized equipment and rails are ubiquitous. I'm just testing to see if this can be an alternative. I also fully expected the negative reaction of 'THIS IS THE END OF ROCKETRY AS WE KNOW IT'.
For those that haven't read the thread, the rocket is 3' diameter, N1100 motor, 104" tall and the velocity at the end of a 16-foot rail is 125 feet per second. Also, you'd see I'm just not pursuing this blindly, but I'm actually going through quite a bit of testing to quantify when the buttons shear, how they shear and what kind of forces they resist in different planes.
I've seen fly away guides cripple a rocket before and don't want to take the chance with this flight profile. For a regular 4" rocket, 4FNC, J/K motor, I'd use fly away guides hands down, if I didn't want buttons on the rocket.
For the last of the box of set screws I went the opposite way on the next tests. I was able to find a place where I could cobble together almost 30 feet of 1515 rail vertically for drop testing. I used 2.5" PVC that was 6 foot long and made the total weight 30 pounds, what my rocket will most likely come in at. Math says somewhere near 40 ft/s at the end of the drop. The first time I totally blanked on cushioning the fall and it hit the concrete and shattered the PVC and lead balls went all over the place. D'oh!
After that mishap I placed some cardboard boxes with pillows inside and the 'airframe' fared much better. No problems shearing the buttons, came off very cleanly as the buttons stacked up. It does tend to add small flat spots to the buttons. I haven't had a lot of time lately to pursue further testing, but I do have another two boxes of set screws on order. I just need to find the time to give this method a proper write up.
Edward