Originally Posted by
FBinNY
The point is well taken, Yes, it can make a difference when steering changes the curvature of the housing, which is why I preferred to explain how it might make a difference, vs. stating that it did, or didn't.
The question is whether it makes a material difference or not, which given that Shimano now uses spring housing in some kits, might be in doubt. Considering the small amounts that the handlebars are turned while riding, and therefore the small change in neutral axis length, I think that whatever difference there is will be probably be within the working tolerance of the system. But it might make the position of the handlebar when the derailleur trim is adjusted more significant, so I'd do so with the front wheel straight.
Also make note that this seems to be 4mm housing so any change in neutral axis length would be 25% or so less than with 5mm housing.
BTW- let me remind folks that while it might be OK to use spring (brake style) housing for the derailleurs, it definitely is not OK to use index housing for brakes. Index housing lacks sufficient compression strength for the brake lever forces and may burst in an emergency braking situation.
I disagree.
Shimano 9 speed sytstem pulls about 2.5-7mm of cable per rear gear shift. You're telling me that you're not going to have anywhere close to 2mm (less? see below) of effective housing length change as it flexes? With non sis housing? Not to mention the fact that you're constantly countersteering to keep the bike upright. And that people don't only shift gears while riding in a straight line?
Sram 10 speed pulls around 1.6mm of cable.
Shimano 10 speed around 2.3mm (not 6700/7900 series)
Shimano 10 speed around 1.3mm. (6700/7900)
That's for a FULL shift, you and I know it would take much less than that for the gear to not be engaging correctly in it's cog. Unreliable shifting with tighter cog spacing is why shimano introduced SIS housing. It's made because it makes a difference.