How much is it actually drooping? Have you pulled a string across the length of the wall of the lower box, just above the bottom edge, and gotten some idea of the amount of droop between the ends? And noted the location of maximum droop? How about marking the wall near the front and rear of the box, at a fixed distance down from the edge of the roof, almost to the bottom edge. Pull a string across these two points, and confirm that the shape there matches or doesn't match the shape at the lower edge.
I'm not sure where I am going with this, but it seems easy to confirm that the wall is still solid and square, and the problem is indeed some sort of detachment of the floor from the wall.
Similarly, you might pull a string along the metal cross-members under the floor. Are they straight, or bowed? Can you tell if one of them is straight but is not in the same plane as the others - in other words, has an end of one of them dropped?
I'm just noodling here, and perhaps you have investigated all of this. I'm just trying to get the picture in my mind. Somehow, I'm thinking that you should be able to push the floor (?) back up, and if you can't reattach it to the wall, weld a piece of I-beam front-rear between two adjacent cross beams to support the floor. This is kind of a mini version of the plate Shane suggested. I would like to go out and look at mine to help' me visualize, but it has been captive at a service place for two weeks.
Bill
|