Sub-frame connectors... Please build them correctly!
No need for computer models, a torsional test will separate fact from fiction. This type of fixture with attachment points at the wheels is the proper method:
Blame Sedanman. All his fault for stirring this up again. ;)
Like I said before, until someone does a full stress/strain analysis on a complete chassis and compares the different types of subframe connectors it's all just guessing. Until that happens, saying someone is "doing it wrong" makes assumptions that can't be scientifically justified. All of the "evidence" that anyone has is pretty much anecdotal. The chassis and body structure are too complex to just "wing it".
Heck, even if someone did do a full on computer model with a stress/strain analysis there would still be points to argue, based on the assumptions that have to be made in order to model welded joints in both the original structure and the subframe connectors. It would be much easier to model the tubular subframe connectors, the contoured and welded to the floor type would be a royal PITA to model for analysis. And they way that they get modeled absolutely has an effect on the end results. Or maybe we should compare two cars with different style connectors using dynamic loads in the real world? Consider the lousy quality control when they left the factory, then consider it's 40+ years later. How much hidden damage is there? Who knows. Maybe one of the cars has a couple extra spot welds because it wasn't Friday afternoon. Test 50 cars and maybe you're getting somewhere, but who's going to do that?
All of the different types of subframe connectors out there are better than nothing. Better than each other? Good luck. The Hotchkis Taxi is a full second faster than a 2012 3 series BMW on TireRack's test track with TireRack's driver. Meaning, in a one day test with a 1970 4-door Satellite he beat his own best time in a 2012 3-series BMW that he makes hundreds of laps in. And since I have a set of Hotchkis subframe connectors that I'm currently installing on my Challenger, I can tell you they aren't even the strongest subframe connectors I've seen. The ones I built for my Duster myself at home are stronger, at least comparing subframe connector to subframe connector. But obviously the Hotchkis pieces work, even if they're not the beefiest, or biggest tubes and even if they don't get welded to the floor.
No need for computer models, a torsional test will separate fact from fiction. This type of fixture with attachment points at the wheels is the proper method:
