Originally Posted by bostontrevor
Argh. Apparently it was a mistake to use the 100% thing as an example.
It's common in physics to first consider an ideal system and proceed from there to the real world. I wasn't saying a real-world drivetrain is 100% efficient. I was considering the ideal drivetrain which is to first illustrate the state we are approximating with our actual bicycles.
In doing so we see that as efficiency goes up, so too does friction (and thus wear) diminish. That's all.
OK. Sorry. I was a bit too sarcastic. Practically everyone who mountain bikes has personal experience with running a chain dry and feels (as well as hears) the difference after they lubricate it. I can't just ignore what actual experience has taught me because of the results of a lab test. To me, when you describe the noise that needed to be dampened, it's just what happens when there is no film of lubrication between the moving parts and you have metal riding on metal. I may be over-simplifying, and I am not any kind of physics expert, but (until now) I never thought there was any question about whether or not to lube the chain--just a lot of opinions about what works best.