The risk isn’t just what the spring does but what happens to something caught in it. Springs can release energy a lot more quickly than a counterweight which can only accelerate a 9.8m/s/s, and therefore more risky to disable.
A few hundred points accelerating at 1g is still plenty of energy. It also comes with it's own challenges when it comes to releasing it.
It also provides a constant force, whereas the force required to raise a garage door linearly decreases the higher it gets. With the consequence that a garage door with a failed motor would slam into its stops over your head rather then into the ground. Followed shortly thereafter by the now liberated counterweight slamming into the ground.
True but by adjusting the angle you still end up with a fixed weight throughout the run. A spring force changes throughout the movement of the garage door and that’s chosen to match the unusual fact that the garage door’s weight changes as it’s gets progressively rolled up onto the top rails.
A fixed weight but not angle means the force varies. Cut the cable and acceleration would therefore also vary with that angle along the track.
Springs make a lot of sense from an installation perspective and can be easily tuned to match the specific door design, but if I was working from scratch on a DIY project I would prefer to use a counterweight if there was somewhere it fit.
Plus a counterweight is a big fat weight hanging in midair, so it's pretty obvious what will happen if you let it fall (and where it will land), whereas with a spring it's much harder to tell whether it's under tension or relaxed, and what it might do when the tension is released...
