TLDR: Started nerding out and typed out a monster below. Using properly spec’d shear pins to handle the recoil should be the easy answer. Bent pins means this was not done — I could easily see not multiplying by a Dynamic Amplification Factor as probable cause as I see professional engineers screw that up. Relying on friction is very problematic and should not be done if at all possible.
I should probably sit down and do the math on this at some point as it always interests me. Problem is, I don’t know what material fasteners are used, I could guess, but that’s vital information for the calcs.
In general it is bad practice to have a shear load held in friction especially a shock load like rifle recoil. (Yes most rail mounts do it, but it’s not best) What UM is trying to do is actually a hard problem to solve, and you really can’t make the torque value “dummy proof” if that’s what they are trying to do. Degreased vs greased threads, bases, rails, whether loctite is used. Thread spec tolerances, torque wrench calibration. All these things can compound to easily be +/-50% on how much torque is actually converted to preload in the joint. And is why you shouldn’t do it.
You can’t just specify a “torque value”. It absolutely has to be a process. UM could take a step in the right direction and define that process and that would be awesome to see. It should spec what degreaser to use, where and how to apply loctite, and torque. I’ve never seen a manufacturer do it (for rings or bases).
Design process for coming up with a torque value would look like this :
1. Take max scope weight with max recoil impulse. To get static F = ma load.
2. Critical part I see every single recoil calculator online fail at… you need to double this load as a “Dynamic Amplification Factor”. Theoretical maximum DAF is 2.0 — which rifle recoil would be approaching.
3. Determine frictional factor to use based on material combination in the joint.
4. Determine a “K factor” to use for what ever lubricant is chosen (probably loctite 243 )
5. Now here comes the hard part. To account for thread tolerances, torque wrench calibration, variances, etc you need to be able to handle at least +/- 25% preload uncertainty. That’s what amount of torque actually gets converted into preload. Spec’ing a torque value that will not plastically deform a bolt at the upper end, yet will still provide enough preload at the bottom end will be difficult.
6. Add in properly sized shear pins once you realize this math likely doesn’t close accounting for worst case frictional factors under all conditions. (Unless you start using exotic bolt materials).
The guys who aren’t having problems and haven’t used shear pins, likely have followed a rigorous prep/ install process, know how to use a torque wrench properly, and haven’t had tolerances and inherent variances stack up worst case against them.