(The Nuances of) Measuring Buckling Strength with SFPD

[Orion66]

I've recently switched over to measuring buckling strength instead of wood density to select tower wood. As most people here probably do, I use a single-finger push down (SFPD) to measure buckling strength.

However, I find it difficult to judge when a column of wood is buckling and when it is not. Do I count buckling as even the slightest deflection in the column? Or do I count buckling as total curvature of the wood and inability to transmit further force to the scale? On a short piece of wood (~22cm), the difference between the scale reading in these two cases can be upwards of 50g, far too imprecise to use with the inverse square tables or any other calculations.

Also, if I were to switch out the scale for a force plate (that logs force as a function of time) and gradually increase my pushing force, would I see a noticeable drop or plateau in the force-time function once the wood begins to buckle? If so, I think a force plate might be a more precise and quantitative alternative to measure buckling strength, rather than guessing on a scale.

Thanks!

[Balsa Man]

I've recently switched over to measuring buckling strength instead of wood density to select tower wood. As most people here probably do, I use a single-finger push down (SFPD) to measure buckling strength.

However, I find it difficult to judge when a column of wood is buckling and when it is not. Do I count buckling as even the slightest deflection in the column? Or do I count buckling as total curvature of the wood and inability to transmit further force to the scale? On a short piece of wood (~22cm), the difference between the scale reading in these two cases can be upwards of 50g, far too imprecise to use with the inverse square tables or any other calculations.

Also, if I were to switch out the scale for a force plate (that logs force as a function of time) and gradually increase my pushing force, would I see a noticeable drop or plateau in the force-time function once the wood begins to buckle? If so, I think a force plate might be a more precise and quantitative alternative to measure buckling strength, rather than guessing on a scale.

Thanks!

A good question. Yes, if you have access to a force plate, you'll be able to get better data. "Better" as in more consistent. It is important using this technique (with a scale) to develop a consistent technique. It is easier to 'get a read' on long (like 36") sticks. But because of variability within a stick, it is important to get a read on pieces that are closer to installed length. For base leg segments in a C tower, it can be pretty tricky; easy to break in testing.

This technique, whether with force plate or scale, is approximate. Hence the need for some level of safety factor. But, this approximation will get you to leg... specifications that will 'work' much better, much more closely than..... anything else you can readily measure.

With a perfectly straight stick, aligned vertically, you'll get pretty 'clean' results. If there's curvature, it will buckle in the way its curved.... preferentially. If you bump it/push the bow in the opposite direction, you will get a higher reading. With it held straight on the jig, and braced to pretty short lengths, 'working' BS will be between the 'preferential' and 'induced' measurements.

If you read up on Euler buckling, under idealized conditions- which are the basis for primary equation - (and which aren't really there in a piece of wood), a loaded column will stay straight as force increases, and at some force, it will buckle- the outward deflection will happen. If that force is just maintained, buckling will (quickly) progress to failure. Doing SFPD testing, you'll see that the force seen on the scale varies very little once buckling is seen; from the force at which it just 'pops out' into a curve, to the point it is bowed to the limits of its elasticity- it stays essentially constant. So, how much of a bow to use in testing is a subjective and physical technique thing. 36" sticks, out 2-3"- you should get a pretty steady read on the scale. Shorter, looking for that distinct 'pop out' point- on a 12" segment, maybe a half inch of bowing. Again, consistency of technique is what gets you good useable numbers

[dholdgreve]

In addition what Len said, it is super critical that when testing, you hold the piece as perfectly vertical (perpendicular to the scale) as you possible can. Just like in politics, leaning the column candidate left or right can cause a substantial reduction in bending strength!

Also as Len mentioned, it is critical that each test is performed similarly... If you are using your index finger to apply the force, do not stabilize the stick with your thumb and middle finger. This will create a much stronger false reading, especially on shorter sticks, like 12" long.

[Balsa Man]

In addition what Len said, it is super critical that when testing, you hold the piece as perfectly vertical (perpendicular to the scale) as you possible can. Just like in politics, leaning the column candidate left or right can cause a substantial reduction in bending strength!

Also as Len mentioned, it is critical that each test is performed similarly... If you are using your index finger to apply the force, do not stabilize the stick with your thumb and middle finger. This will create a much stronger false reading, especially on shorter sticks, like 12" long.

Right-on, Dan.

A couple of other little tips
- when testing shorter (cut down from 36") sticks - be careful to make the cuts nice and square (like the ends of the full length sticks). If the end on the stick (particularly the end sitting on the scale) is cut at an angle, so the end has a point, you'll get a weaker than actual reading
-flip the stick end for end; get a second reading. If there's significant difference, go with the lower reading
-track/inventory your sticks by percent of 'design' BS (at the length tested) and percent of design weight; makes seeing the tradeoffs in the selection of a set for a build easier. Like, hmmm, for only 0.2gr more leg weight, minimum safety factor goes from 8% to 15%....

[M017]

For buckling strength, couldn't you just pin it horizontally on a piece of foam and put a 10g weight in the middle and measure the bowing?

[scioly2345]

I've recently switched over to measuring buckling strength instead of wood density to select tower wood. As most people here probably do, I use a single-finger push down (SFPD) to measure buckling strength.

However, I find it difficult to judge when a column of wood is buckling and when it is not. Do I count buckling as even the slightest deflection in the column? Or do I count buckling as total curvature of the wood and inability to transmit further force to the scale? On a short piece of wood (~22cm), the difference between the scale reading in these two cases can be upwards of 50g, far too imprecise to use with the inverse square tables or any other calculations.

Also, if I were to switch out the scale for a force plate (that logs force as a function of time) and gradually increase my pushing force, would I see a noticeable drop or plateau in the force-time function once the wood begins to buckle? If so, I think a force plate might be a more precise and quantitative alternative to measure buckling strength, rather than guessing on a scale.

Thanks!

I have an extreme extreme helpful tip!!!! So basically, you can do what is called a "hanging test", which is where you take 5-10 sticks of 36 inch balsa, and keep 3-6 inches (just eye it out) on a table and let the rest hang off. Now make sure none of the sticks are hitting each other, and let the rest of the wood droop down. Whichever wood holds itself up the best, is the strongest. What i usually do is i eliminate the wood that is dropping closest to the floor, pick the strongest wood, then weigh the strong wood on a .01 g scale and use it accordingly.

[M017]

https://docs.google.com/spreadsheets/d/ ... sp=sharing

Here's a spreadsheet that I made this week that has all of the values for a columns strength based on BS and bracing interval length

[M017]

I did forget to factor in loading bloc assembly and bucket, but nobody should be getting within 500g unless they're crazy, or have some other reason

[MadCow2357]

Do you weaken the wood after doing SFPD Tests?

[cheese]

Weakening the wood shouldn't happen if you are doing it properly. If you don't push down super hard, and just go slowly until the scale isn't going up anymore, then you stop. If you hear cracking, you've gone too far and have probably damaged it.