Designs

[Vizard007]

Thanks Balsa! That was extremely helpful.

And, as to whether to use a tower or not more than once... I think it really depends on the design. I've found that it really depends on the quality of the tower you've made. There have been times in the past in which my team, in the past, have made bridges/boomilevers that have worked for both regionals and states. But, generally, imo, it is best to make a new one. (Better safe than sorry?) Because, theoretically, the best tower would break at the 15kg mark, right?

[SLM]

Here's a question I just thought of...
When you build bridges and test them, if it holds all the weight and your totally happy about its weight, do you use it again? Or because you put force onto it, its weaker than before?

We usually compete at tournaments using pre-tested structures. A few years ago, for Elevated Bridge, we pre-tested our competition bridge many times (around 100 times). The pre-tested bridge earned us 1st place at the state tournament and fifth place at the nationals that year. The repeated testing was easy for a short structure (the bridge was only 30 cm in height) using a commercial testing apparatus that uses a force gauge. The apparatus allowed us to load an unload the structure in a matter of seconds; we did the 100 loading cycles in less than one hour.

However, since the testing apparatus cannot accommodate structures taller than 40 cm, we tested our competition tower only a few times (around 10), using sand, last year. That tower earned first place at the state tournament, but the team did not make it to the nationals.

Balsa wood is an elastic material. That basically means, within limits, the strength of the material is not going to be effected by its repeated loading and unloading. As long as the load stays within those limits, repeated loading of a well-built tower does not negatively impact its ability to carry the load in any significant way. However, if you know the tower would fail, say, at 15.1 kg, you don't want to pre-test it up to 15 kg. My rule-of-thumb is to load it up to 85% of its capacity.

[jma]

Hi SLM,

What is the purpose of pre-test the same structure several times?

[mrsteven]

Hi SLM,

What is the purpose of pre-test the same structure several times?

I agree, I can understand 1 or 2 times but any more than that seem redundant

[SLM]

Hi SLM,

What is the purpose of pre-test the same structure several times?

Good question.

If you are certain that the tower is going to perform as intended, then there is no need to pre-test it. However, it is almost impossible to build with such a certainty. So, it does make sense to test the tower to make sure it can hold a decent amount of load (say, 85% of its intended load capacity). But, why is it necessary to test it more than a few times?

Say you have designed a light weight tower to carry the full load, and you have successfully pre-tested it once. But, you wonder if it could hold the full load again. You ask, what if during the first loading cycle a few micro-cracks have developed in the wood weakening its strength? What if some of the members have weakened to the point that they will fail prematurely next time around? You just don't know.

In engineering/material science, fatigue is the ability of the material to loose some of its strength due to cyclic loading and unloading. If the material is stressed excessively, then fatigue could become a problem. Let's call the stress (or load) that causes the development of significant micro-cracks in a tower its critical load. This load is generally unknown to us, So, If during pre-testing the applied load reaches the critical load micro-cracks start to develop. Repeated loading eventually causes the tower to fail. The challenge here is that we cannot predict how many testing cycles it will take before the tower fails. It could be 2, 3 or more.

For elastic materials (like wood) that are not excessively stressed, it takes about 10,000 loading/unloading cycles before fatigue becomes an issue. So, by testing the tower multiple times you tend to develop a higher level of confidence that it is not at the critical stage, that fatal micro-cracks are not being developed, that if it can handle 100 loading cycles, the probability that it would fail in the next loading cycle is not that great.

[jma]

Thanks for your response.

For new builders like us, if we don't know how to calculate the capacity of our tower,the only way to know is to test until our tower breaks or is there a better way?

[Balsa Man]

Thanks for your response.

For new builders like us, if we don't know how to calculate the capacity of our tower,the only way to know is to test until our tower breaks or is there a better way?

Yeah, testing is the only way to know for sure.
Setting up a test rig will take some time. Be sure to get it pretty darn level.
There is one very helpful, let's call it "development tool", to consider- a "safety tower"- search past posts- and see what it looks like in the Image Gallery. A tripod made out of 1x2s, that't taller than the tower, that has a chain, from which hangs an eyebolt, with the eye up, threaded end down. You can get a long coupling nut (at like Home Depot. You put the safety tower over/around your tower, you set up the top so thet the load block is at the height of top-of-tower- with the upward-extending threaded section of the load block eyebolt into the coupling nut from the botton, and the threaded section of the eyebolt hanging from the safety tower coming into the top of the coupling nut. Then adjust the height of the loadblock so its, oh, maybe 1/8th inch below the top pf tower. Load 'er up. What the safety tower does is allows failure to happen, but without destroying the structure. You will know what broke first, how/why.......

[SLM]

Thanks for your response.

For new builders like us, if we don't know how to calculate the capacity of our tower,the only way to know is to test until our tower breaks or is there a better way?

Calculating the load carrying capacity of a tower is a challenging task, even for experienced builders. Using physics/engineering, you can estimate the strength of your tower design with 80% to 85% accuracy. Beyond that, you need to resort to testing to more accurately determine how much load the tower can hold.

The problem with destructive testing (testing until the tower fails), however, is that you end up destroying the tower. Now you have to build another one which, most probably, is not going to have the exact strength as the one you just tested. The point is, you rarely can accurately determine the strength of a competition tower. The best you can do is to estimate it.

A good tower design, based on sound engineering principles, is a good starting point in this process. With a good design, if you can construct the tower accurately using the right pieces of balsa, then the chances are the tower can hold the intended load with 80% to 85% probability. Beyond this point, you can start making incremental changes to the tower, and a lot of testing, in order to improve its efficiency.

[Balsa Man]

SLM's take on 80-85% range matches with our experience. To get there you have to have a) some test data (which could be whole tower, section (lower/base, or upper/chimney, or compression/column testing of individual piece), b) decent jigs (so the shape/geometry is very close to the same for each build). Wood, as a natural material, comes with "built-in" variability. Even if you match density- like a leg set where they all are within 1/100th, or even 1/1000 of a gram, there is going to be variability in strength/stiffness. Our experience suggests that 15% for bass, and 20% for balsa is a workable/comfortable safety factor.

As I noted in my earlier post today, a safety tower gets you around the problem of destroying a tower in testing. That's why I referred to it as a "development tool." For instance, in settling in on leg density/weight, you could build with legs of different weight. Record weights, color code (magic marker) so you know which is which. The lightest (probably the weakest, depending on the range of weights you use) will (likely) fail first. If you're using the safety tower, and have the distance the block can fall set right, so that the failure happens, but things don't progress enough to break anything else, you can "splint" the leg that broke - straighten it, and glue lamination strips on (which will strengthen it way up), then re-test. If the next heavier leg is strong enough, and it holds, there's your leg weight target to go with (for all the legs) on the next build; if its not strong enough, it'll break- splint it up, re-test, and see if the next heavier is enough. This way, in a single test session, with one tower, you can make great headway on knowing what weight/density range (that 15-20% range) you're going to need.... Same approach works for ladders & diagonal bracing.

[SLM]

As I noted in my earlier post today, a safety tower gets you around the problem of destroying a tower in testing. That's why I referred to it as a "development tool." ...

Using a testing platform that limits the amount of failure to one member at a time is a good idea for sizing structural members. As Balsa Man suggested, you can load the tower until a member fails, strengthen the member by increasing its size, then test the tower again. This technique is particularly effective as long as the re-built member is not significantly larger in size (stronger) than the original member. Why do I say that? Because by increasing the size of the member you are increasing its stiffness. Generally (although there are some exceptions) stiffer members in a structure are forced to carry more of the load. So by changing the size of the failed member, you are in effect changing the load distribution pattern in the structure. That is, most member forces in the revised structure are not the same as the member forces in the original structure. Therefore, the technique could lead to an over-designed tower.