Originally Posted by _cy_
turns out the low 18ft lb torque value was NOT the problem for R1200GS. BMW by leaving off requirement of Loctite 643 with such low torque values was the problem.
seems KTM 690 caliper bolts are spec'd at 25 Nm or 18ft lb .. but KTM requires Loctite 643 be used at those low torque values.
It's 243...not 643.
Having slept on this issue overnight, some of the problems very well have been the rider in the linked thread torqued to 18ft/lbs on dry threads. 18ft/lbs on dry threads is going to give a super-low clamp load that will allow the assembled part to side slip pretty easily.
This observation backs up an drives home the importance of filling in the airspace found in between the male and female threads with a liquid that turns to a thermoset plastic with thread lubricating qualities to allow a sufficient clampload to be introduced into an assembly to keep it from falling apart.
I went back and pulled a few Power Point slides directly related to the friction loss on lubricated threads vs. non-lubricated threads. Here's the talking points:
Approximately 85 to 90% of the effort used to tighten a threaded fastener is lost to friction, which leaves only about 10 to 15% of that effort is used to generate clamp load.
Another slide shows clampload (lbs.) generated on different base metals torqued to 10ft/lbs along with a calculated K Factor (Friction Factor)
The following examples are listed as:
Bolt Nut Washer K Factor Clampload
Plain Cadmium Plain 0.15 2,805
Zinc Plain Zinc 0.35 1,546
Zinc Zinc Zinc 0.55 960
The table listed above shoes that Zinc is a very sticky coating with a very high K factor. The low clampload backs up the claim that zinc plated fasteners are sticky.