From my inspection of a removed wing window, it appears there are four possible leak paths. The frame seen edge-on looks something like this:
Three of the leak paths are shown in light blue, which assumes for two of the paths that the vinyl gasket has some hidden rips inside the frame. It also assumes that the "goop" between the panes has failed in one or more spots.
The fourth path is through the corner joint. The frame is in four pieces--the curved piece (which is one continuous extrusion for each part), which has an outside and inside part, and the straight edge, which also has an outside and inside. The two corner joints could possibly allow water to get in between the inside and outside frame pieces.
The above photos show the outside and inside of a corner joint. There is still a lot of crud on the frame from when it was removed.
You can see that the outside part is definitely a separate piece of metal, but even with the screws removed from the back, it wouldn't budge.
Thhis makes me think that it's possible the drawing above is slight in error, that the "outside" piece actually wraps around the two panes, and then is just retained in the "inside" frame by the screws. This would allow the window to be removed without removing the inside piece that's riveted to the shell.
If anyone has disassembled a wing window, can you confirm which drawing is correct?
These windows leak --- like a sieve. I've had up to three inches of water inside them. The problem is that three of the leak paths are improbable--they require at least three failures: the gasket contact with the frame, the goop, and the gasket inside the frame has to be ripped. I've sealed the outside edge of the glass and still had major leaks. Where the heck is it coming in?