I'm about to try a technique for just this kind of purpose: waterproof (important in a wind instrument, for sure), stronger, and dimensionally accurate/consistant.
The idea is to fill the voids in the WSF with a 2-part epoxy resin. Completely.
I did one part previously, though with an incorrect procedure. The results were technically a failure, but pointed to what should be a correct solution.
The basic concept is to immerse the part in the resin, apply a strong vacuum to remove all the air from the WSF (and the resin, as an added bonus), then release the vacuum and drain off the excess resin. The liquid resin should get forced into the gaps by simple atmospheric pressure.
"Failure" came by immersing the part first, then applying vacuum. I used too small a vacuum pump, though. It took so long to pull enough vacuum that the resin was still "foaming" from the WSF venting its air by the time the resin was starting to cure. (Made a heck of a mess!) The new (untested as yet) method will involve pulling the vacuum first, isolating the vacuum source, then releasing the resin into the "tank" in one shot. Air won't have access until the resin has filled the tank entirely - drowning the part, if you will - so only the resin has access to filling the voids.
Bear with me. I'll be running the experiment soon and will update upon completion.