I feel it's important to share these failures and subsequent repairs because that's part of being a professional builder...standing behind not only your product, but your customer as well, despite the origin of the issue. It the responsibility and the cost of doing business in this field.
The last two weeks, I've had two bikes come back that needed some love. Both were a little over 5 years old, both used the original Zona stays, both were Rohloff builds, both had a failure of the seat stay tubing about an inch up from the dropout termination brazing. One could argue that the failure is in the HAZ zone, that the cantilever design of the Paragon sliders places coupled with the disc brake and the Rohloff torque places too much force on the frame, shoulda used a stay brace, or that when you ride hard off road for a long time stuff wears out. Regardless, they need fixed.
Chris's bike was designed as a single speed, but had the Rohloff installed about 2 years ago to help him climb the hillier terrain of Phoenix when he relocated his family to Arizona.
He noticed a small crack forming in the finish of the left seat stay and sent the frame in for review.
When constructed, the stay is slotted for the tab, then filled with a lot of silver so that the tab and interior of the stay encapsulate each other, forming a solid bond. What I found once the paint was stripped off was a small hairline circumferential crack beginning about 1 inch up from the end of the dropout tab. In this case, the tubing failure would be towards the cooler edge of the heat affected zone. The accumulated force of the cantilevered dropout, constant forward torque from the Rohloff, and the pulsating forward pressure of the disc brake fatigued the stay material, exceeding it's ability to absorb the forces.
Would a stay brace have helped? I've seen failures of this type in both configurations on diverse builders frames, so it's tough to say. What I have concluded through the years is that while the sliding dropouts are excellent for single speed or geared use, they are less than optimum for use with a Rohloff long term.
To fix this, the stay was cut out about 2 inches up, a solid piece of 1018 steel was turned down on the lathe to the stay OD, shouldered for 2 inches with a chamfered edge to slide up into the stay, and scalloped to match the original shape at the dropout. The piece was then bored out to leave a .120" wall, plenty sufficient for the forces anticipated.
The piece was then brazed in place with the stay and welded to the dropout.
Now, to hustle back to show stuff.