Ok, Dew, let's look at this as pure engineering, okay? If you really want the FEA done (Finite Element Analysis, a method of simulating engineering structures, for the rest of you, since I refuse to hide behind a language barrier), you can do them yourself; I don't have the time at the moment.
The engineering itself is rather simple, and it makes me doubt the program over at UMA that you can't seem to do it yourself. Both locks are based on putting a pin in quadruple-shear.
The Axis lock uses a cylidrical pin, moved into and out of place by a sliding motion. The Rolling lock uses a semi-cylidrical pin (REKAT's website has a drawing), rotated to either stop the tang, or allow it to pass.
The Axis lock places the entire pin in shear. The Rolling lock places the entire pin in double-shear, and the part where the tang contacts the pin in an additional double-shear. However, since the Rolling lock must use a cam action to account for wear, the area which the tang places in double-shear is, by definition, smaller than the whole pin (and, in practice, must be smaller by a good margin to allow significant amouts of wear to be accounted for). This is an inherently weaker design, as any engineer should know.
As to why the post-tested knife intemittently failed the notorious 'spine-whack' test, this is also a bit of simple engineering. As the tester stated, the pivot pin was bent significantly, and the lock pin was dented significantly. While the pivot pin may or may not be the result of an isolated manufacturing defect, the lock pin comes as no surprise. The Axis pin is a complete cylinder, the Rolling lock pin is a partial cylinder; an inherently weaker structure. The cumulative play of both the bent pivot pin and the dented lock pin add up to enough play to allow the blade tang to pass the lock pin on occasion, when both are flexed slightly more due to the impact force. The intermittent nature of the failure might be due to slightly different amounts of impact energy on different tests (humans can't maintain rigorous force controls, despite Cliff's statements to the contrary), or it might be that the lock was still engaging when it was at the full extreme of it's wear-adjustment cam cycle, but the mechanism of the lock was damaged and did not always hold the lock in that position. I would need to observe it more closely before I would hazard a more definitive answer.
Re Cliff: I'm not shooting the messenger. If you read my posts in the Talonite thread, you will see me say that I think his tests are valid as real-world tests. I question his right to call them science, which they are quite clearly
not. I would be happy to go and get some examples if I had the time, but he has made a lot of posts, and it would take a while to sort through and find examples which would still make sense out-of-context. Make no mistake, Cliff is quite the master of verbal trickery, taking care that he nearly never says outright what he obviously implies so often.
rockspyder, thank you for bringing up reliability. This is another reason I strongly favor the Axis lock. Aside from the frame lock, the Axis lock is the most resistant to failure due contamination, and the Axis is more wear-resistant than the frame lock (thought that margin is quite small for a well-made frame lock).
How's that, y'all? I hope I was able to keep the wording on a fairly readable level, although Dew's request made some engineering terminology unavoidable. If anyone would like further explanations of any of the terms, perhaps emailing me would be a good idea; I could then post all the explanations at once (assuming any requests are actually made), rather than crowding this thread with fifty more posts explaining terminology.
--JB
------------------
e_utopia@hotmail.com