Stand by my statement, Carole *LINK*
Eddie Pacheco; San Jose, Ca
>The glue layer, in a vertical or thickness direction is very thin. A few thousands, in some areas, in this case. Remember the lathe bed and headstock mounting surface are not super flat. They're flat but not super flat. Grainy surfaces as well. So it stands to reason, especially in this case, that a CA glue bond it would be weaker in tension. In a shear direction you have to fight the whole "glue area" to break it free. In this case that could be a half inch or more of glue patch depending on how much was spilled, and how rapidly it was able to creep or wick underneath the headstock.
Now if the headstock is able to be pushed up from a corner or opposite end of the glued surface then you can create a peel situation, and then it may take even less force to break the joint.
We (IBM) used CA glue (Black Max), applied in a microscopic drop via syringe, to bond the small read write heads to their metal suspensions on hard drives. These heads sometimes wring themselves onto the super smooth surface of the disk during temp and humidity changes within a harddrive. When the disk starts to spin the dab of glue would hold so well that the metal suspensions would get bent all to heck before a head would let go from the suspension or disk for that matter.
Sometime Wikipedia info is not correct. Or it could be that the data was stated as "Tensile Shear Strength". The test was done in shear
Here are some links to look at.
Even "our" Russ Fairfield has stated that CA glue joints are weak in tension
http://209.85.173.132/search?q=cache:GoLMu7GxWJUJ:www.bemrc.org/bemrc-build-class/ca-glue-article.pdf+shear+strength+of+CA+glue&cd=10&hl=en&ct=clnk&gl=us&client=firefox-a
http://www.woodcentral.com/russ/russ6.shtml
http://www.roymech.co.uk/Useful_Tables/Adhesives/Adhesives.html