Originally Posted by
cyccommute
First, there is no such thing as “negative tension”. Tension only works on way. People often confuse compression as the opposite of tension but the only “opposite” is the direction of the force vector. Something in tension has to go to zero before it can be compressed. The chain on the bottom side of the crank is still under a bit of tension because that’s what the arm on the rear derailer is supposed to do…keep it under tension so that it doesn’t flop around. Yes, it is just along for the ride but it is held in a little bit of tension so that it doesn’t hang slack.
Yes, I agree that the front derailer is meant to “derail” the chain from one ring to another. However there are lots of situations where there is too much tension on the system for the derailer to do that job especially when depending on a spring to do the job. When the spring eventually gets to the point where it can cause the chain to deflect off of the ring…either because of the teeth profile on modern chainrings or because the rider eases up enough on pedal pressure to allow it to drop…the chain can drop too rapidly causing that rear derailer arm to snap back resulting in the free chain coming in contact and entangling with the chain running through the jockey pulleys. That entanglement is what can result in the derailer snapping off.
The directionality of the derailer does make a difference. When downshifting on the rear, the cable is forcing the derailer to move the chain, assuming, of course, a high normal derailer. Rear derailers don’t tend to lag like the front can during downshifts unless the rear derailer is a low normal (reversed) derailer. The failure of Rapid Rise was due to relying on a spring to derail the chain under tension…just like the front derailer.
I have no problem with this explanation except that you are drawing the wrong conclusion. Yes, there is more friction and rather than depending on a cable to actively pull the chain off under high tension situations, we depend on a relatively weak spring to knock the chain off. Shimano front derailer springs are actually weaker than SRAM for example. SRAM front derailers tend to work better in my experience because of that strong spring. That said, a high normal front derailer actively pulls the chain off so there is no need for that silly easing of pedaling. Unfortunately high normal front derialers are exceedingly rare.
No argument from me on that other than to point out, again, that there is no such thing as negative tension. Tension can only be there or not. Compression isn’t the opposite of tension.
You're just baiting me with semantics about negative tension. The chainring pushes the chain toward the tension pulley on the rear derailleur. The spring on the tension arm takes it up, hopefully with enough force that the chain won't go slack. I have never seen the chain go slack from "shifting too hard," just like I've never seen it happen from the other end as I described, but I won't say this would never happen. But the only way to apply more tension on the bottom span of the chain is to shorten the chain, get a stiffer spring, i.e., a clutch, or pedal backwards. And the relationship between the chain and the cogs doesn't care whether the shift is initiated by cable tension or a spring. The shift will occur when there's sufficient deflection in the chain to move it from one cog to the other, and that would be facilitated by applying more force from the derailleur or slackening the chain.