Gluing Wood to Unfinished Kevlar

I have a Kevlar canoe which is unfinished on the inside. Recently the metal fastenings which held the central thwart to the gunwhales corroded out (lots of saltwater use with no rinsing afterward). I decided to fasten the thwart back on by screwing onto wooden blocks placed underneath which, in turn, would be glued onto the inside of the hull. I used West System Epoxy (with which I have considerable successful experience in fiberglass and wood) and I cleaned the hull with acetone; but, because the unfinished Kevlar already seemed so rough, I didn't bother to sand.
Well, the thwart did not stay on. I now know (from reading accounts) that sanding Kevlar is a dicey matter. So what is the best way to go about sticking a block of wood onto unfinished Kevlar?

 

Hi, how about saturating the 'unfinished 'kevlar with a few layers of west , letting it dry and then keying it up and bonding with more west mixed with colloidal silca to make a wet paste.
cheers
J

 

Itchy has it, except don't let it dry or you'll be back in the same boat working the Kevlar. Do a wet on wet bond with the wood, then drill for your fasteners.

 

Aha. now Ive learnt something extra too. Thanks PAR

Were all winners here.
J

 

Responding to PAR

As a way of responding to PAR, I should probably be more detailed about the actual gluing job. Namely, I should add that:
I mixed up West System into a wet paste using the #404 High-Density Filler.
This wet paste was applied separately to the wood surface and the Kevlar surface before I brought the two together. I was also very careful to only apply enough pressure to get a little squeeze-out around the edges.
Where did I go wrong?
Is there a way that I could have sanded to get some more "micro-tooth" on the Kevlar surface without raising excessive "fuzziness"?

 

Did you wet out the Kevlar with neat epoxy, prior to applying the thickened epoxy? If not, your joint likely "sucked out" and now you have a whole different set of issues to over come.

 

Second response to PAR

Thank you for reminding me about "two-step bonding". In my memory it was mainly a consideration when using wood (which is obviously porous and therefore has "suction").
So, if I use "two-step bonding" on yellow birch* to clean, unfinished Kevlar, how do you rate my chances of getting a reliable bond for this thwart-support usage?

*each block has a 2inx4in bonding surface

 

Wet out everything, not just the wood. The fabric can leach epoxy away from the bonding surface too. Too many variables to speculate about your "odds" of success. You've done this before, so they're pretty good. No short cuts this time and they're better then pretty good.