here you can see the result wich I draw in the first sketch. It’s a bit different because for me it was to risky with the thin st and the dropped stays.
Next time I will use one with a sleeve or a wall thickness of 1.2mm for dropped stays
4 Likes
I didn’t care enough to go any further than I did. After measuring a slight increase in torsional flex, which doesn’t make any sense, I concluded that it doesn’t matter enough to worry about. This is to say, my testing errors were bigger than the effect. If I saw flex decrease from .020" (0.5mm) to say .010" then that fits with intuition and is measurable so I would conclude that it does have an effect.
Determining and quantifying the flex of a bicycle being used IRL is something ‘we’ have been trying at for a long time (see other post re: columbus testing from the 80s). I’m not going to do better in my shop than many other, much better funded labs are going to do.
To be clear, I always add them unless there is a specific reason not to.
@adamsklar
I suspect confirmation bias is strong on this. The total flex I measured was like 0.020" without bridges (and slightly more with) so even if adding bridges removed or doubled all flex (which it didn’t), I doubt we can feel 0.5mm of flex. 3psi of tire pressure would cause more deflection than a 0.5mm (this is hyperbole, I didn’t do the math on this 
).
I can see where adding bridges would decrease cantilever or disc brake induced flex in a single seatstay or chainstay as it would effectively brace it against the other stay and shorten it. As a complete structure, the rear end is too triangulated to have bridges affect ride though.
Would love to be proved wrong with data.
1 Like
I think Adam is talking about SS to CS bridges for disc brakes:
This is also a known failure mode (brake tab cracking the SS), so it definitely makes a difference!
2 Likes
OOOHHHH!!! Yeah, totally ignore what I said about that then! 
I must have misread that.
I’m talking about bridges up near the seattube or BB.