A fantastic book in a fantastic series!
Mr. Hendrickson, on page 102, wonders why the class LG-3 car didn't work out. I note that the previous class LG-2's (and LG-4's) were designed for crane unloading. Thus, cables or chains were most likely passed from one side of the car to the other under the load and then were all tied up, pretty much, for the crane. On the LG-3, also with rigid log "bunks", if workers did that on the ground, the cables would capture the upper parts of the car side truss. It looks like, to unload this car, you'd need to feed the cables inside the truss--a bit time consuming. It also doesn't look like they allowed a lot of clearance for the pass-through of the cables down near the floor; though that could probably have been easily fixed. Also, once in a great while, the logs might have been loaded such that getting that necessary clearance on the sides might have ranged from difficult to impossible.
That said, it's also possible the logs just overpowered the sides with their weight. Loaded up near the top of the sides, the leverage on the side to floor joint is about double that of an un-modified gon.