Wood

[MadCow2357]

Hi,
I am ordering balsa wood soon, but I am not sure which densities of balsa I should get. I know Balsa Man said some things about wood density, but I was a little overwhelmed at the amount of information in one of the posts (The math is too complicated for me at this point, but I will try to figure it out). I know I will be using 1/8s, 3/32, 1/16, and possibly 5/32. Any recommendations on Balsa Wood densities?

Thanks,
MadCow2357

[Balsa Man]

Hi,
I am ordering balsa wood soon, but I am not sure which densities of balsa I should get. I know Balsa Man said some things about wood density, but I was a little overwhelmed at the amount of information in one of the posts (The math is too complicated for me at this point, but I will try to figure it out). I know I will be using 1/8s, 3/32, 1/16, and possibly 5/32. Any recommendations on Balsa Wood densities?

Thanks,
MadCow2357

Hi, MadCow,
Welcome to the tower forums. The fact you’ve come here looking for info and understanding is a good thing. Ditto that you understand that wood density matters is a good thing, and a good first step.
Wish there was a simple answer to your question. Unfortunately there isn’t.
First, it depends on your goals – how “good” a tower do you want to get to? Good enough to not be in the bottom 10% at invitationals/regionals, good enough to be in the top 25%of the pack at invitationals/regionals, medaling/winning at State? Top 10 at Nationals?. How hard are you willing to study and work?
Any ‘recommendations’ on density would have to be based on what you’re trying to do- the higher your goal, the deeper you will have to get into, and come to understand that scary math. Before really getting into the math, it is important to get your head around the basic concepts, then dig deeper to understand in more detail, and understand how the math helps you figure things out.
You really shouldn’t be ordering wood at… whatever density specifications until you have at least to some extent nailed down basic design. The basic design decision is the tradeoff between stiffer/heavier legs needing less bracing (a wider bracing interval), and floppier/lighter legs with more bracing (a tighter bracing interval). It is the STRENGTH, specifically the buckling strength, of the leg wood that matters; higher density (and larger cross section) gets you more strength. Basically, you’re looking for the lightest wood that has enough strength for the tower to hold near full load. To be able to do that, you need to a) know the forces the legs will see at full tower load, b) be able to calculate the bracing interval needed (which will depend on the style/type of bracing you use) to brace a leg of a given strength into short enough intervals so that the leg has greater buckling strength than the load it has to carry.
Density/stick weight then comes into play. As I’ve said many times, there is a relation of density to buckling strength – higher density means/ will get you higher buckling strength, but there is significant variation within that ‘trend’ you can have a bunch of sticks at some given density/stick weight, and their buckling strength will vary considerably (+/- 10, even 20%). So, when you have figured out the buckling strength you need, then you know the density range within which you should be able to find sticks of sufficient strength. There is a graph posted that shows you buckling strength v stick weight (for 1/8” sticks) data. There is also an ‘inverse square’ table posted that shows you what bracing interval at what 36” stick buckling strength gets legs braced to sufficient strength to hold full load.
If you take the time to read, study, think about all the info that’s here in this forum, you will be able to figure this out. Feel free to ask questions as your understanding increases.

This said, if all this is…..just too much, if, for your base section you use “medium” density 1/8” for legs (1.5, 1.6, 1.7gr/36”), and brace it with all Xs (using medium density 1/16), at a bracing interval of ¼, and for the chimney use the same medium density braced at ¼ interval, it will probably carry full load (if you have used a reasonably decent jig to assemble it.

[dholdgreve]

Hi,
I am ordering balsa wood soon, but I am not sure which densities of balsa I should get. I know Balsa Man said some things about wood density, but I was a little overwhelmed at the amount of information in one of the posts (The math is too complicated for me at this point, but I will try to figure it out). I know I will be using 1/8s, 3/32, 1/16, and possibly 5/32. Any recommendations on Balsa Wood densities?

Thanks,
MadCow2357

Rather than specifying exact densities (and spending as much as double what random densities cost), I'd suggest using the same budget and buying double the amount of wood, then carefully sorting what you have by mass, matching the higher density columns to fewer X braces, and the lighter density columns to more X braces. The key to this competition to find the magical ratio that will provide the highest possible efficiency. This can't be done if ordering just certain densities.

[Balsa Man]

Hi,
I am ordering balsa wood soon, but I am not sure which densities of balsa I should get. I know Balsa Man said some things about wood density, but I was a little overwhelmed at the amount of information in one of the posts (The math is too complicated for me at this point, but I will try to figure it out). I know I will be using 1/8s, 3/32, 1/16, and possibly 5/32. Any recommendations on Balsa Wood densities?

Thanks,
MadCow2357

Rather than specifying exact densities (and spending as much as double what random densities cost), I'd suggest using the same budget and buying double the amount of wood, then carefully sorting what you have by mass, matching the higher density columns to fewer X braces, and the lighter density columns to more X braces. The key to this competition to find the magical ratio that will provide the highest possible efficiency. This can't be done if ordering just certain densities.

I’m very reluctant to say someone who has (successfully) been coaching the ‘balsa building events’ for many years may be wrong on something. We each have developed our understandings, and ways/techniques that ‘work.’ Dan and I are…. ‘on the same page’ on so many things. I know very well I don’t know ‘everything’, and, like all of us, can certainly be wrong at times.

The reason I believe it is important/smart to incorporate some density… range/boundaries in wood ordering is because a) there is a relationship between density (stick weight) and buckling strength, and b) it is the buckling strength that matters. Sticks of the same weight/density can and will have significant variation in buckling strength. To get the best tower performance, you want to get/find the lightest set of sticks that have your ‘design’ buckling strength.

You can easily calculate – in fact you can just look up in the inverse square table I posted a link to, what buckling strength of legs, at what bracing intervals, will get you to a braced leg buckling strength that will carry the load. You can also look up, for 1/8” sticks, what stick weight range will get you sticks with the buckling strength you need.

If you’re really pushing/fighting for those last few grams, or tenths of grams of tower weight, you can buy/order (from a place like Specialized Balsa), sticks in a fairly narrow range, like some 1.1gr/36”, some 1.2gr/36”, some 1.3gr. These numbers are just “for example” numbers. If you’re not pushing as hard to be seriously competitive, you can order in a broader range, like “light”, “medium”, “heavy”, and by weighing/testing find leg sets that will “work” at a given buckling strength and bracing interval, and use the lightest in a competition tower. The…basic engineering information I’ve posted about is real; the approach of being able to calculate strengths needed works, and the data on the weight range of sticks that will have some sticks at a given strength is real. Just my perspective…

[dholdgreve]

Again, 2 different schools of thought... nothing wrong with that.
Another area where Len and I look at differently...
When we receive a shipment of column material, the first thing we do is to cut them from 36" lengths to 12" lengths. Then we separate by weight (to the nearest tenth of a gram), then further stratify by separating the individual weight piles by bending strength. We have all noticed how, if stripping your X braces from sheets, the weight can vary wildly from one side of the sheet to the other. The same holds true in bending strength. Testing a 36" long stick for bending strength assumes that the stick has those same properties throughout its length. In reality, the stick is more like a chain, and can bow at its weakest area, or more accurately, the weakest area can precipitate a bow in the middle of the stick. Also, if you were to take a 36" stick and cut it into (3) equal 12" lengths, it is unlikely that all 3 pieces will weigh exactly the same. With different weights will likely come different bending strengths. By blocking your potential columns into individual 12" lengths, you can analyze each column independently, without the influence of adjacent pieces in a 36" stick.

The downside to this might be that most super sensitive scales that can weight to .01 grams have a limit of 500 grams. This can be an issue on some 12" sticks in the .5 gram weight and up, as the BS may exceed the limits of the scale.

[Balsa Man]

Again, 2 different schools of thought... nothing wrong with that.
Another area where Len and I look at differently...
When we receive a shipment of column material, the first thing we do is to cut them from 36" lengths to 12" lengths. Then we separate by weight (to the nearest tenth of a gram), then further stratify by separating the individual weight piles by bending strength. We have all noticed how, if stripping your X braces from sheets, the weight can vary wildly from one side of the sheet to the other. The same holds true in bending strength. Testing a 36" long stick for bending strength assumes that the stick has those same properties throughout its length. In reality, the stick is more like a chain, and can bow at its weakest area, or more accurately, the weakest area can precipitate a bow in the middle of the stick. Also, if you were to take a 36" stick and cut it into (3) equal 12" lengths, it is unlikely that all 3 pieces will weigh exactly the same. With different weights will likely come different bending strengths. By blocking your potential columns into individual 12" lengths, you can analyze each column independently, without the influence of adjacent pieces in a 36" stick.

The downside to this might be that most super sensitive scales that can weight to .01 grams have a limit of 500 grams. This can be an issue on some 12" sticks in the .5 gram weight and up, as the BS may exceed the limits of the scale.

Yeah, we look at some things a bit differently, but what we're actually doing in practice is very similar. It gets to and uses the same basic engineering principles; they are what they are.

Dan is absolutely right about the variation in both density and buckling strength/stiffness within a 36" stick. And it is, indeed, important to measure buckling strength at a length close to what leg segment length will be, and use that more exact value in calculation. Allows you to use a significantly smaller safety factor. For B towers, we're testing chimney legs at 11", and base legs at 12". I've said before you can expect +/-20%, both in buckling strength vs stick weight at 36", and in variation of density and buckling strength in shorter pieces cut from one stick. The data we're seeing in our latest wood orders is actually showing, in about 10% of sticks, variation pushing 40%. This is why it is critical to go by measured buckling strength to pick leg pieces; if you just go by density, there's a very good chance one or more sticks in a set will ...seriously under-perform.

If you're approaching tower design by figuring out the tradeoff in tower weight to get to an optimal leg strength and bracing interval- so you know the leg strength you need (and then it becomes a problem of finding the lightest sticks that will have that strength), what data on 36" stick strength vs stick weight gives you is a real/valid starting point- a fairlynarrow range. Its still a statistical game- I call it 'balsa bingo'- to get a few sets that are strong enough and really light, but by using/ordering in an appropriate density range, you can minimize the number of sticks you need to order to get a set, or a few sets of really high performance sticks, and it offsets the higher cost per stick for specified density. Its a very different situation if you're looking at buying a bunch of wood - your school has 5, 6, ...8, or 9 teams; you're not one builder looking for wood specific to one settled-on design; you're not trying to 'hit' on one bingo card, you're trying to get decent hits on many bingo cards.

[dholdgreve]

Again, FWIW, using the column bending strength as the design basis, then X bracing with the calculated number of tiers, we have produced a number of towers this year. As previously mentioned, we have used a safety cage to determine the primary point of failure. I was surprised that our initial volley of towers overwhelmingly failed at the tension band around the bottom of the towers, most of them very near the end adjacent to the columns. In our second iterations, we heavily reinforced the tension band (original design x 4!) our scores jumped up substantially, but then the X braces began to fail. We had been using 1/16 x 1/32 X braces of light to very light densities. Of the dozen or so tests that we have ran so far this season, we have not had even ONE primary column failure! We would not have known this without the use of the safety cage, and would now be trying to increase column strength instead of working on what really needs it... the X braces!

[nambui0701]

Do you have any images of the 'safety cage'? I'm trying to construct one, it'd be really helpful to have some examples.

[Balsa Man]

Do you have any images of the 'safety cage'? I'm trying to construct one, it'd be really helpful to have some examples.

No, but glad to explain the basic idea, and provide general directions. It is for stopping the load block from being able to drop if/as the tower fails under load. You want the load block to only be able to drop far enough for first failure mode to happen- somewhere between maybe 1/16” and 1/8” By the time the tower has deformed enough to be 1/8” shorter than it started, multiple things will almost certainly have failed. In a 'well constructed'/symmetrical tower may be enough. The cage needs to be plenty strong to carry 15kg load.

So you want to create a rectangular frame structure with square base.

1x2 or 2x2 (or a combination) for lumber (we’re using 2x2 legs and 1x2 bracing pieces). Everything held together with screws and glue. 4 vertical legs, cut carefully to the same length. The ends of these legs, when sitting vertical, should form a square; a 12” (inside dimensions) square provides plenty of clearance on a 29cm circle bonus tower base. Just like you do on a tower, you’ll want the legs to be joined by bracing (which can be screwed and glued on the outside leg surfaces – ‘ladders’- horizontal braces top and bottom, and a diagonal brace (between the ladders) on each of the four cage sides. If the top edges of the top ladders are mounted flush with the top of leg ends, you can use two on opposite sides to support a load-bearing bar.

That bar needs to be strong enough to carry at least a 15kg load put on the middle of it. We’re using a piece of 1 ¼” x 1/8” steel bar. You could use oak 1x2, or probably get away with pine 1x2. For a really solid setup, you can put pieces of 2x2 sitting on top of the leg ends, on two opposite sides.
The bar needs to have a hole drilled through it, in the center, that’s significantly bigger than the ¼” eye bolt that carries the load bucket. We’re using a 9/16” hole. If you use 1/8 steel bar, you can use a 3” long eye bolt; if you use (thicker) piece of wood, you may have to go to 4” eye bolt.
So, looking at the eye bolt, and going up from the eye at the bottom – ½ to 1” above the top of the eye is the bottom of your load block; about ¾” further up is top of load block; on top of it is standard ¾” diameter ¼” washer and wingnut. Allow ½” clearance above the top of the load block (as it sits on the tower) for the washer and wingnut, and that is the height you set/build the top of cage at

So, when you have the tower on the test platform, and the safety cage put down over it, and have things (the tower, the load block, and the cage) aligned so vertical centerline of tower is aligned on vertical centerline of the cage, you put your top bar on, with the eyebolt coming up through the hole in the bar, roughly centered in the hole. Then you drop a large washer with a ¼” hole over the eyebolt.(these larger washers are often called ‘fender washers’ – 1, 1 ¼, 1 ½ diameter – Home Depot and Lowe’s carry them). Then you put a second wingnut on, and run it down the eyebolt. The gap you leave between the bottom side of the eyebolt and the top side of the fender washer is (when the tower breaks), how far the load block will be able to fall before the load comes onto the top bar, which stops the load block from being able to go down any farther. You’ll have to play with things to confirm just the right gap.

When set right, the first thing that fails will put the load on the top bar, and nothing else happens. You will be able to see that first failure- where it is, what kind of failure it is, frozen in time and space. One of the neat possibilities with this tool is you can unload the tower, take it out and repair and reinforce the break (and perhaps corresponding parts), set the tower back up I the cage, and re-load it to see what breaks next – you can then do the repair/reinforce again, and so on….. big time saver to get to optimal wood selection.

[dholdgreve]

Do you have any images of the 'safety cage'? I'm trying to construct one, it'd be really helpful to have some examples.

PM me your email address and I'll send you some pics of our cage.