Imagine a hose clamp around a 1010 rail tightened. It shears off the buttons, the buttons slide under the hose clamp.
As you hopefully see, I'm going about trying to test and verify this idea before just throwing it on a rail. This is a lot more than I see when people propose new ideas and their first test is when they fly their rocket.
I've tested the shearing in a drop test that simulates flight. I'm hanging weight off the side of three buttons tonight to quantify how much force before something snaps. I've quantified how much it takes a single button to shear in single shear. I agree it ultimately is up to the RSO on the flight.
Hard chuff - I agree that is an issue. That issue can be mitigated by proper motor testing and igniter design. This is a hybrid motor. I have 10 successful static tests of this motor. All of them come up to full thrust in less than a fifth of a second. The rocket is supported by a L-Bracket that the weight of the rocket rests on that instead of buttons.
Quick calculation on crosswind. If I take my maximum crosswind to be 10 mph (which is very high, I'll likely want less than 5mph for this flight) the force generated on the rocket is ~12 pounds. I assumed a cross sectional area of 3.17 square feet (which is about 20% more than there is). If this load is spread equally on all guides that is 4 pounds per guide. This is less than what I observed the individual shear of each button.
I'm doing this research to see if it can be an effective alternative to a tower. I agree towers are a good solution, but rails are much more ubiquitous at launches. Fly away guides are good, but I'm much more concerned with one of those taking off a fin because I've seen it happen, and in videos I've seen them come very close. I feel much less confident in those because the way it releases is less controlled. How far it flies away depends on the speed of the rocket, the moment it releases, the grip it has on the rocket, etc. And what if the grip of the fly away guides slip and are impaled on the front edge of the fins? You could what if every single scenario to death. Engineering is about quantifying the risks and seeing if they can be minimized.
Edward