A few thoughts/comments on this subject-
Early-on this season, I discussed a) how critical it is to get base/chimney leg joints right, and b) how difficult /darn near impossible that is to do if you’re building base and chimney separately, and then mating them up to complete the tower. To review and elaborate:
There are multiple aspects to getting this joint right:
One, as noted by Random Human, is to have top (of base legs) and bottom (of chimney legs) connection areas ‘fully contacting.’ That means when the two leg segments are in….design position/alignment, the faces of both the leg ends are in the same plane, and they’re flat from edge to edge.
The next aspect is that the shape and surface area of the cut ends needs to be the same. If, as many folk try at the start, you cut the top of the base legs so the ends are all in a plane parallel to the test base- to create a “flat” top of base section, because of the lean-in angle of the base legs, the cut ends at the top of base section end up in the shape of a rhomboid – a diamond shape. It has a ‘long axis’ joining the diagonal corners that face toward and away from the vertical centerline of the tower, and a ‘short axis’ joining the other two diagonal corners. Now if you cut the bottom ends of the chimney legs to ‘fully contact’ with the tops of the base legs, because the chimney legs have only a very slight lean-in angle, the (bottom) cut ends end up being very close to being square- the long axis is only a very tiny bit longer than the short axis.
When you mate the two ends, the face of the bottom of the chimney leg is significantly smaller in area than the area of the base leg top end. You can easily see how the long axis of the base leg ends sticks out beyond that of the chimney leg ends. If you manage (on all 4 legs) to perfectly align the short axis of the chimney leg faces with the short axis of the base leg faces (the short axes of both being the same length), then the two leg segments will be ‘axially aligned’- the short axis of both segments will meet in the joint. The force coming down the axis of the chimney leg will go/flow down the axis of the base leg. However, if that ‘axial alignment’ is not perfect, then the base leg segment will not be “axially loaded – the load coming down the axis of the chimney leg will be “off axis” going onto the base leg. Having spent a lot of time (back in 2011/2012- last time we had rules driving a ‘2 part/base/chimney sections’ design requirement) trying to create base and chimney jigging that would perfectly align all 4 legs so that the base legs would be seeing axial loading, I can tell you it is very unlikely you will be able to pull that off. First, the positioning of bottom of chimney legs and top of base legs has to be perfectly matched- in the exact same square (or if not perfect square, the exact same….quadralateral), and along with that the short axes of the end cross sections have to line up on all 4 legs.
Our buckling strength measurements for figuring out how strong and how braced a base leg segment needs to be to carry full/design load assume axial loading- that’s one of the base assumptions in Euler’s buckling equation. If the loading a column/a leg segment sees is off axis, its buckling strength will be significantly less than if its axially loaded. Off-axis loading will immediately, as load starts to develop, induce a bending moment into the column being loaded. Very little load (far short of the load required to buckle an axially loaded column) will start bending- i.e., buckling. The math to calculate the reduction in buckling strength as a function of the amount/degree of non-axial loading, is very complicated. But the important part to understand is that a very small amount of off-axis loading – like less than you can see just looking at the joint – will result in a significant bending moment being produced. Non-axially loaded base legs will fail at significantly lower loading than axially loaded ones.
More importantly, in terms of overall tower performance, is what this means in terms of the …. vertical stability of the chimney section. Because of its narrow width, compared to its height/length, the chimney, as a whole, is very sensitive to off-axis loading. In the perfect case, the legs are oriented in a perfect square, and the load is evenly distributed among the 4 legs – i.e., it is being applied, vertically down the vertical centerline between the 4 legs. If, looking just at the chimney, you get the leg bottoms sanded just right, so the chimney is sitting absolutely vertical, and all 4 leg bottoms are in full/even contact, let’s say if you were to test just this hypothetical chimney, and find it carries 15kg. What would happen if a) the load block was not quite centered, or b) it wasn’t quite vertical, or both a) and b)? With very small mis-alignment of loading or very small mis-alignment from vertical, it would fail (or fall over) at significantly less than 15kg loading.
If our perfect/hypothetical chimney is sitting on a tower base section (let’s say a base that has been independently built and tested to hold 15kg loading), and the leg tops in that base section are perfectly ‘in-plane’, so, as in the test of the chimney by itself on a testing base, it is really sitting vertically, and the load is symmetrically distributed to the 4 legs, what happens if, because of non-axial loading onto one or two of the base leg segments, just a little bit of bending of one/two of the base legs starts? One corner/one side of the level platform the chimney is sitting on is lowered- the chimney will immediately start to lean toward the lowered leg/side. All it takes to start this is… like one or two thousandths of an inch. That lean causes a quick and significant transfer of additional loading to leg(s) its leaning toward- that increases the loading onto the non-axially loaded base leg(s), increasing chimney lean, and in an instant chimney is leaning enough we see the…’classic’ failure mode- chimney falling to one side, chimney/base leg joint failing- the exact situation being discussed.
It is because of these factors- and the way they work together to turn/cascade a set of very minor mis-alignments into tower failure – that I have been recommending a) a single jig approach - a jig that carries/aligns “assembled legs”- where chimney and base legs, glued together at the joint are mounted on the jig, and then the bracing is put on, and b) cutting of the leg ends (both bottom of chimney legs, and top of base legs) so that they are cut at the same angle – that angle being ½ of the angle formed between the base legs and chimney legs. With both leg ends cut at this same angle, the faces that get joined are the exact same cross sectional shape and area – when you mate them up, all corners and faces of the sticks line up perfectly – you have full contact, and loading onto the base legs is axial. In addition to getting the cut at the top of base legs correct, it is important to use some sort of tool/jig to make the bottom of base leg cuts so a) that angle is correct – the bottoms of the base legs sit flat on/in full contact with the test base, and the length is exactly the same. It is also important to carefully fit ladders at top of base- that are strong enough, and while it is not necessary to ....slather the joint with lots of glue, it is not a joint to push the limits on how little glue can you get away with....
There may well be…. clever folk out there
who have figured out other ways to get the precise alignment needed for high performance of the base/chimney leg joints. The way I’ve described is one way, and the best I know of at this point. Hope the discussion above helps to understand what’s going on, and what needs to be controlled/managed to get to high performance.
