Shahram,
I’m not sure there is any rule-of-thumb answer to this one. Usually, we give the loads to the structural engineers and let them figure out an appropriate beam. And so far I’ve had no complaints! The only time I’ve ever had a structural engineer ask for anchor loads to be reduced was in a concrete sleeperway for obvious reasons.
I assume that you are thinking of a new rack rather than an existing one. In the latter case, a thorough structural analysis may be appropriate and the rack may need some beefing up to accept higher-than-normal loads. (I might mention that in California and other earthquake-prone areas of the U.S., structural analyses on existing racks using the latest codes often shows that the racks won’t support the loads in the originally-designed case, let alone adding extra forces.) Also, I assume that you using the common usage of the term “anchor” meaning what might more properly be termed an axial stop. A true anchor—something that prevents all degrees of freedom—imposes moment loads on a beam which can send a structural engineer into fits of anguish and severe depression.
Most of the high-load anchors will be located near the side of the rack (so the lines can loop completely across the structure). Anchors located near the edge, of course, produce a different result on the structure than one in the center of the rack. One “trick” I sometimes use is to locate the anchors at alternate bents (assuming it doesn’t cause a multiple loop system—loops inside loops—to expand into each other; a very unlikely and easily remedied situation anyway). Another consideration is that anchors thoughtfully located can have near-zero loads. To take a simple example, if I have a line with two loops I might start by putting anchors at the ends of the system and guides at appropriate places. I then run an analysis and determine on which bent between the loops the line has minimal axial movement (usually, the location is obvious). Putting a third anchor at that location will present the lowest possible force on the bent and also create a well-balanced system. And adding gaps to the axial stop may result in an actual zero load. Incidentally, I would not recommend removing such a non-acting anchor altogether. The anchor can have a very real benefit if one of the end anchors were to fail.
Another thing to consider (as undoubtedly you are aware) is that no anchor will ever do what the computer says it does: gaps are not exact, beams bend, temperatures vary during the life of the system, yield occurs, etc.
As to your question, “which location of the rack is better to be considered an anchor bay?” again, I don’t think it matters. Some might prefer to use braced bays, but since all the bents are probably connected longitudinally to the braced bays it makes little difference which bay is used.
Regards,
_________________________
Ricardo
Previous Topic
Index
