mikeschn wrote:
P.S. Can we conclude that the tongue has to handle the highest loads? and that we should pay extra attention to the tongue?
Well, I'm not an engineer, but I did go to engineering school for a couple of years before switching to IT, and I grew up (and still live) on a farm, where we fix almost everything ourselves.

The tongue wasn't handling the most LOAD, but was under a fair amount of STRESS. And from what I've seen, you didn't have any structural failure at all, only welding/fastener failure. Now from the pics there's no way for us to know if the weld was inadequate or if the design wasn't up to the load or stress, or both. Personally, if I had been welding that, I would have wanted some triangular gussets at the intersection of the tongue pieces and the frame pieces, along the sides of the pieces.

(this image is from
http://weldingdesign.com/processes/usin ... -correctly)
This pic only illustrates the basic idea. I know your frame pieces lapped over, didn't butt together.
Not only would those additional welds spread out the load, but the vertical weld would trying to pull the pieces along the axis of the weld rather than perpendicular to it. A properly sized and designed weld should be strong enough regardless of the direction of the stress applied to it, but I'll say I've seen more failed welds due to perpendicular stress than shear stress. This seems to be mostly due to perpendicular stresses causing one or both pieces to flex, and the welds can't flex like the material can. So then you get tiny cracks, which allow moisture in, which weakens it further (and repeats until failure).
I'm an amateur welder at best, so if any of you welder pros can correct me on any of this, please do.