Originally Posted by VincentVega
Hi. I have yet to purchase my bus, but when I do, I’d like to not chop and raise the entire roof, but to cut a mostly full length, rectangular hole in the ceiling, and add a hightop, much like what is used in van conversions. Any issues or tips for preventing the side walls from caving in?
What you're considering would be similar to trolley-top buses like this one: http://soderholmbus.com/images/buses-big/carriage.jpg
... which are unfortunately notorious for leaking badly.
Really, your main problem with this approach is that in most buses the walls and ceiling are formed from continuous ribs ("channels"), and to open up a skybox of any sort like this down the middle of the bus, you have to cut the ceiling portion of these ribs out and you lose their structural integrity (as well as the strength and stiffness provided by the sheet metal removed from the hole).
In a conventional roof raise these channels are also cut (usually at window height) but their strength is preserved by the welded-in extender bars (which are often even stronger than the channels). You can't do this for a skybox opening, obviously, so you have to restore the lost structural integrity in a different way.
I was originally planning to do the same kind of thing on my bus, and I priced out the materials that would be required. Basically, I was going to cut out about a 4' section from the middle of each channel (only about 7 ribs total, starting from the third from the back and going forward), with the sheet metal also cut out but 2" to the inside of the channel cuts (this would allow me to bend the existing sheet roof up and create a 2" flange).
Next I would weld together a set of essentially S-brackets, with a 6" piece of channel extender at the bottom at the same angle as the roof, a 1' vertical piece, then another 6" piece of bar extender at the roof angle. The bottom bit of extender would go into the bit of channel still there after the center piece was removed (and be bolted in place since the roof sheet is still in place blocking where you would normally weld it), then the vertical piece goes up a foot, and then the top extender on the S-bracket goes into the cut piece of channel (which is now held a foot above where it was originally). Since I'd be cutting 7 channels, I'd need to fabricate 14 of these S-brackets.
So essentially you're just raising the center 4-foot section of each channel by a foot. These raised ribs would be open, and you would then sheet them over with new 4'x8' sheets (which are pretty much the cheapest size) laid side-to-side (the use of 8' sheets would constrain how wide + how high your raised section would be). The edges would be bent and then riveted to the existing roof sheeting (which would also have the bent-up flange underneath the new sheets) to make it watertight.
These raised channels might also require longitudinal stiffening, and an additional reason to add beams that run along the cut ends of the channels is that weight on the roof would tend to make the cut channels "pinch" inwards, putting stress on the welds on these S-brackets. Over a long enough run, even very strong beams would have a tendency to bend slightly and allow this pinching to happen somewhat, but this could be mitigated by a cross-beam in the center that ran from beam-to-beam (this would be a potential head-smasher, but my thought was to disguise it as a pole to hang pots and pans from).
Since I wouldn't be doing this over the entire roof length, I would have had to figure out how to do the transition from the raised part back down to the unraised part over the span between two channels. Getting this watertight would be tricky, but my plan was to make a test piece out of cardboard and get that fitted right, and then cut the metal piece. Lots and lots of riveting with this approach, but I don't think all that much more than a conventional raise.
Cost-wise, I've found a place near me where I can get 4'x8' sheets of 16 ga steel for $40 and I would have needed 5 sheets (I think) for the length I imagined. Altogether I think the materials might be $500 to $600 plus more (possibly a lot more) for the rivets + renting or buying a rivet gun setup. It could possibly even require no welding, since the extenders would be bolted to the channels and you could (and possibly should) engage a fabrication shop to weld together the S-brackets.
There would be difficulties in making this watertight, but I don't think it would be any worse than with a normal roof raise. Also note that there is no way this construction would hold up in a rollover accident like an unmodified (or a conventionally-raised) bus body would, so under no circumstances would it be a good idea to ride in the back of this thing. Unless, of course, you were willing to accept the same risks as anybody who rides in an RV.
You should take this exposition with a grain of salt: the only reason I'm (probably) not going to be raising my roof like this is that I'm going to be lowering the floor instead.