GhostPants_ wrote:When choosing sticks for competition towers, I would do a simple "push down" test on a scale and record that number as well as the weight of the stick. I then selected four sticks that had high scale test values while also being in the desired weight range
My first observation has to be, this approach worked, and worked well for you. So the following observations are not saying/suggesting anything wrong with your approach, just looking at what we might learn/understand from what you achieved.
Absent a failure in a bracing member (wood breaking, or glue joint letting go), and assuming symmetrical construction (i.e., each leg is seeing equal force under loading):
a) It is the actual buckling strength of the legs that determines performance, and
b) It is the weakest leg (actually weakest section of the weakest leg) that will fail first.
What these basic… principles say is that the lightest tower (that will hold near full load) you can get from the set of sticks you have to choose from is the 4 lightest sticks that each have a buckling strength at/above ‘design strength’ (the strength that when braced, gives you, in each braced interval, a buckling strength at/above the force the leg sees under near full load). Depending on B or C tower, and meeting 29cm circle bonus or not, with a 4 legged tower, that force is between 3.77-3.84kg.
As I’ve discussed, with a ladders and Xs bracing configuration, we know that buckling strength measured (in a 36” stick) under ‘pinned-pinned end conditions’ [what you have in a single finger push-down test (SFPD)], gives you, in the ‘fixed-fixed end conditions’ of each braced segment/interval you get when bracing with ladders and Xs, 2.3 times the SFPD, times 1/the proportion squared [where the proportion is the proportion of the length (along the leg) of the braced intervals divided by the length of the stick as tested (36”/91.6cm)]. How we get to the 2.3x factor – effective length factor – is discussed in detail in a post in the “Measuring/using buckling strength-new info” thread.
Also, as discussed, we know there is inherent variability (of “E”; the modulus of elasticity; the inherent stiffness), within each stick. If you test a stick, and get a SFPD reading of, say… 40gr, if you were to cut it in half and test each half (which will increase buckling strength by a factor of 4), you might see, say 170gr in one half, and 150gr in the other half. So, as a practical engineering matter, some level of safety factor (SF) is needed- a calculated in-place buckling strength some percentage above the ‘design strength.’ In our experience, a 20% SF is conservative- is almost always sufficient; a 10% SF is usually sufficient.
Last, as I’ve discussed before, while the design approach described above works (its consistent with known engineering principles/theory, and in multiple tower tests, actual performance is consistent with calculated performance), I do not yet have a comparable design calculation basis for an Xs only bracing configuration- specifically the effective length factor; it is obviously a lot less than the 2.3x factor that applies in a ladders and Xs configuration.
The data from your tower provides some useful insight into what that factor might be. Median SFPD buckling strength for a 1.2gr/36 stick is around 30gr. Were the BS test results you saw for your 4 leg sticks close to that? In a ladders and Xs bracing configuration, at 1/5 interval bracing, the in-place buckling strength would be about 3900gr (just a bit over design strength). At 1/10 interval, which if I counted correctly, you were running in your tower, that in-place buckling strength would, with Xs and ladders, be about 15,600gr- 4 times as much. So, just back calculating, this suggests that about ¼ of 2.3 would be an accurate effective length factor for Xs only; about 0.8.
Raleway wrote:Question for BS test: We have been using different forms of BS testing to try to get the optimal one but we all know w/o perfect testing situations the BS test is quite... erroneous. Thus we tried to come up with a solution: use same density pieces of wood (to at least 10ths) and make 2x2 (taped together) of each stick 36 inches long and use that for push down. That way the stick is more parallel and the middle is stuck together for about a week to try to get it straighter then untied right before testing or have only one keeping them together. We then took the average of the test to apply to each stick. Any suggestions or opinions?
That’s an interesting approach.
First, I have to say I don’t agree with your take that without ‘perfect conditions’ the BS test is ‘quite erroneous.’ We’ve been seeing leg performance very consistent with testing (as I’ve described in detail in various posts - single finger push-down at 36” x 2.3 effective length factor (for ladders and Xs) x 1/the proportion the braced interval is of 36”, with a safety factor in the 15-20% range). And, as I’m sure you’ve seen in BS testing, once you have bowing in a stick being tested by….an inch or two, scale reading varies very little if you move your finger around- i.e., you don’t have to have it perfectly vertical. If stick being tested has a slight bow, it will, of course, bow in that direction under pushdown. If you…induce it to bow in the opposite direction, you’ll see a BS…pretty close, sometimes a few grams higher- average the readings. Testing at shorter length (e.g., a piece just a centimeter or two longer than installed leg length) does require greater precision to get consistent/predictive results.
There are a couple of things I don’t understand on your 2x2 bundle testing- you are talking taking 4 sticks; yes? And you’re talking testing the BS of the 4 sticks together? (with them taped together at the midpoint)? I’m very curious a) what sort of readings you are seeing for a bundle, and what that gets you to for a given stick. As I understand what you’re saying, you take ‘bundle BS’, and…..divide by 4 to get individual stick BS?? Don’t know what density sticks you’ve been working with, but, for instance, for 1.2gr/36” sticks should be getting around 30gr; for 1.3gr/36”, getting mid-30s; for 1.4gr/36”, getting high 30s. How does that align with what you come up with?