andrus said:
^Interesting! Is that "150% reference force" according to the new euro codes rules? Or it's just a generalization from your side?
To be fair, I encountered that example in the Eurocode for concrete structures, but I would think the same applies to steel structures as the logic is generally the same everywhere.
Basically, what you do when designing a structure, is:
1) Find reference force, which would be the strongest force to plausibly to be inflicted on the structure during normal operation/weather event.
2) Multiply reference force by a factor. I'm a little unsure when to apply the factor 1.2 or 1.5, both are used and I've seen up to 1.8 too, but it has something to do with the load configuration (basically, where the forces act on the structure). For instance on a house roof, you'd have to consider situations with both half the roof and the entire roof are covered with snow. The former is actually more dangerous, as there is nothing to compensate for the force on the snow-covered half (think of a scale weight). Yet since the house is weighed down heavier when snow covers the entire roof, you have to consider both situations, and pick the most dangerous situation when you proceed.
3) Find the yield strength of the materials in question. Multiply with various factors to account for the possibility that the materials might perform worse than specified. For concrete, you multiply with 0,85/1.5 (I'm not entirely sure why the two numbers aren't combined to a single factor, as our course didn't go that deeply into things). For steel rebar, the factor is 1/1.15. If I'm not entirely mistaken, the original material specifications are often also estimated to be rather low, for instance concrete strength is specified as the pressure strength of the concrete 28 days after pouring, yet concrete will continue to harden and achive up to 30 % more strength after that stage.
4) Choose the dimensions of the structural element necessary to withstand the re-calculated reference force, given the re-calculated material parameters. For concrete elements (again, concrete was the course I had) beam thickness is usually limited by architectural limits (ceiling-to-floor-above distance, usually), so what is crucial is the amount of rebar you put into the element.
5) Round up the answer from 4) to the closest practical industrial standard. If you get the necessary amount of rebar to be 2000 mm2 over the cross-section of the beam, for instance, you have no practical way to get that exact number using industry standard rebar diameters. Using 4x 25 mm rods, for instance, will land you at 1963 mm, which is too little according to your previous assumptions (without applying the correction factors, it would be plenty). Five rods, however, would bring you above the treshold, so that's what you use (for simplicity's sake, you try to avoid mixing rod diametres).
Now your structure will have a higher rebar content than is necessary in order to withstand forces way larger than it is planned to need to, under the worst possible load configuration, assuming that the materials are weaker than they are specified to be. You might call that "over-supported", yet it's actually the minimum of what the code requires. I'm sure Hixee could explain this better than me, since structural engineering is not my speciality and it's been a few months since last I had a structural engineering course.
As for the supports, also keep in mind that the concrete tower and the coaster are separate structures. The tower doesn't appear to carry any load from the coaster, which would make sense considering that the two are subject to way different dynamic loads (mainly wind for the tower, and train movements for the coaster), and interference between them would be a nightmare to design for.
This separation of the intertwined structures requires some wonky and unusual structure geometry, and the chosen solutions might be less than optimal from a structural standpoint, resulting in higher stresses than would be the case for a simpler, free-standing structure. This is solved by making the structure more robust, leading to more supports per track length than is usual for less complex coaster structures.
