All:
On the "pod" theory, I have always reinforced the floors going over the frame. Do not trust the floor alone if you have any real expectation of seeing a roll. In my truggy I have 2x3x0.188 square from one side to the other in back from the bottom of one main hoop to the bottom of the other. On the "a-hoop" I have 2x2x0.188 mounted in a sort of cantilever for similar effect but without having to go across the trans (which at the time would have been too much trouble).
And to all those "do it right" folks coming from desert racing and hard core rock crawling, remember the majority focus of this site. Most of these guys are still worried about body panels and are not going to go where they are likely to roll, but wisely want a roll cage "just in case". These trucks do 99% of their time on the street and need to remain relatively comfortable in the winter, rain, hauling the family, etc. They also don't want to "get out their purse" (Brandon and co.

) to go "all the way". So for 99%+ (and I feel VERY safe in that number) of the people on this site who are interested in cages, the frame tie-in story is completely over the top and around the corner. They also have put a lot of money into their trucks and fully expect to use them for "mulit-use vehicles" for years and many thousands of miles to come (whether this is realistic is another debate). The drawbacks and costs of "to the frame" DRAMATICALLY overshadow any benefit they are realistically likely to get from it. So the "pod" style supplies the "just in case" safety as well as the comfort and long term viability requirements. Why are so many people so utterly contemptuous of this? Frankly, I think some of you would insist we all drive a Lamborghini Diablo to and from work in rush hour traffic for 5 miles through downtown Manhattan just because "it's the best" for some imagined reality. Frankly, I would rather drive something more comfortable and suitable to the real needs at hand.
In fact, this strikes me as very similar to the "can't wheel without a Dana 60 front" tendency that runs so rampant on the web. A small subset of us who long-ago (or not so long ago) gave up body and looks as well as realistically "streetable" probably shouldn't even consider "pod" style. In fact, the "pod" is probably MORE work for "us" than "doing it right" to the frame. I did my "pod" long ago before I got nearly where I wound up, and while that cage holds up, I lack the energy, time, money, or inclination to change it now. But I would never rebuild it using the body mounts if I did it now. My son's K5 (which is still not done, he lost interest) is a perfect example of the other side. His truck was to be 99% road truck with "fun days" running the fire roads and moderate rock crawling. It needed to be VERY streetable and he wanted to keep the body mostly stock in over-all appearance. The answer for us was obvious to me. We reinforced his floor (as part of the boat side) much like mine was done and prepared to build a "pod" with seats and belts mounted into the cage.
Bottom line, neither approach is "the right approach" across the board. Each person needs to evaluate their needs and with the pros/cons for *their* usage. I'm very vocal on the side of the "pod" or body mounted cages ONLY because everyone else seems to jump so energetically onto the "TIE IT TO THE FRAME" band wagon. I'm really completely indifferent to either approach and choose based on the intended use and needs for each application.
Jason:
The problem I have with the dog legs is that most form what amounts to a lever that is working one way or the other. I've seen the welded/mounted completely solid with no bushing. This works against the down tube "in bending" and against the frame just like the "bad design" steering box that so many people get up in arms about. If you put a bushing in there, depending on how the bushing is mounted, you still get these forces to varying degrees. And imagine exactly what is going to happen to that "mount" with a very hard roll. Depending on exactly what/where/how it's likely to fold, rip bend anyway, especially if it has become brittle over time from fatigue.
And yes, I believe what Stephen is advocating *for hard core use* is to eliminate all the isolators and mount things solid. So you have a cage welded/mounted to the frame in order to form that rigid "space frame" I mentioned. This cage design is intended to provide protection and rigidity without much concern for the body, so major body mods are generally required. Then the "body" is remounted somehow (depends) to the cage/frame and no longer needs isolators because the flex is mostly gone. In fact, isolators just complicate things at that point.
rdn2blazer:
That's not ideal because you've got the bar fully "in bending". It would work, but I would suggest something like I used in a similar case, 2x3x0.188 square. You know which direction most all the force is coming from, so square is far stronger pound for pound, inch for inch.
Baja Blazer:
Again, "all the serious professionals" don't have to drive their vehicles every day on the street, or too and from the trails in another state, or in the rain and snow… All the "serious professionals" also use 5 point harnesses, wrap around seats, expensive coil overs and high-zoot hand fabricated works of art for tuned suspensions, and so on. Does that make all that the "right choice" for everyone here?
ntsqd:
Agreed 100%. Only thing to add is that when you build the bars that mount the seats to the cage, it almost becomes a "pod" at that point since it will/should keep from punching through the floor (assuming the seat supports are strong enough to distribute the load).
Brandon:
For folks with trail focused trucks, I agree 100%. Trucks where we've already pretty much given up on "body" and "stock looks". What I'm trying to offer here is a voice of reason for the 99% or so of this site who are not in that category where we find ourselves. These guys ARE expecting their "investments" to last for years and thousands of miles, and those concerns (cracked frames, leaks, squeaks, expense, fatigue life, etc.) are valid for them. Your also right about the body mount fatigue, that type of stress always does move to the next weakest point, but between the bushing flex and the mount design, those can handle the stresses a LOT better and longer than the body or frame can in the typical cage/frame tie-in.
All:
My opinion is this, don't do it half way. Either build a "pod" or something similar and call it "good for my needs". Or if, for whatever reason, you feel a frame mounted cage is the "right way for you", then go the full distance and live with the consequences. Most of what *I* consider problems with the "frame mounted" approach come in ONLY because it didn't go all the way to building a rigid chassis. And that requires a (almost) fully rigid triangulated structure in 3 dimensions so that most all forces are transferred into tension/compression. If you decide to go somewhere in between, that's fine too, but you really need to understand the consequences and have judged them acceptable. If so, then good luck and have fun.