Towers B/C

[DarthBuilder]

Sorry, I'm a bad reader lol.

[Raleway]

Anyway to build a one piece jig balsa? Sorry storming you with questions.

Random

It's just one piece- get a really, really nice woodworker to do it for you or if you have a laser cutter, that could also work. You just make a filled mold (or put some mdf panels together) for what it would look like inside your tower. If you get a woodworker though, make sure that guy knows how to cut very, very, precisely and accurately and also be able to put them together really well... it would be terrible for you to realize after 10 towers that the error is the jig, not you. Best of luck!

[Balsa Man]

Balsa
In last years discussion you posted the leg buckling vs weight, on that graph you had multiplied it by 2.0 and 2.3 fe's based on fixed fixed end conditions


You then stated the all x design with no ladders seemed to have a fe of .4.


Did you mean this is the reciprocal (the effective length? I can't make it jive with the 2 and 2.3
But a .4 as effective length and a 2.5 fe would make sense.

FIRST, AN IMPORTANT CORRECTION ON EFFECTIVE LENGTH FACTOR FOR ALL Xs BRACING. My bad, I should have looked at the spreadsheets rather than going from memory- THE CORRECT, APPROXIMATE VALUE IS AROUND 0.55, NOT 0.4

I'll get something posted soon that will help unravel the mystery. Very briefly, we found, using a ladders and Xs bracing, that if you measure BS by doing 'single finger push-down' testing (which is essentially measuring under pinned-pinned end conditions), you have to multiply the SFPD (pinned-pinned) value by an 'effective length factor' of 2.3 to get to what the BS will be under fixed-fixed end conditions (essentially what you get/create when leg segments are braced with ladders and Xs). That meant, for a C tower, bracing at 1/5 intervals. When we switched for Nationals to an all Xs bracing approach, we needed/used a 1/11 bracing interval. We got to the 2.3...theoretically (digging deeply into Euler's buckling theorem), and it worked very precisely in practice- actual tower performance/in-place leg buckling strength was very close to what theory said it would be. The 2.0 you mention was from early in the process of figuring out how effective length factors work- that it was somewhere between 2.0 and 2.4). To get to the effective length factor for all Xs, we back-calculated/reverse engineered from some detailed data I was provided on a... high performing, non-circle bonus, all Xs tower (using Xs at 1/10 interval, from 1/32 sheet, cut to 1/16 width. The data I got was the (narrow) range of SFPD 36" BS measurements for the legs used. It held full load. Knowing that, it was simple to calculate what the effective length factor had to be. This reverse engineering approach worked- tower held full load at Nationals. I'm still working on understanding the theoretical basis, and will share when I figure it out.

What this all means, in terms of figuring out how strong legs need to be for a given (all Xs) bracing interval, is you measure SFPD at 36", you multiply that number by 0.55. You take that value and do inverse square calculation to see what the BS of (an all Xs) braced segment will be. Example- leg wood tests (at 36"/91.6cm) at 26.5gr. Multiplied by 0.55, that = 14.5gr. Then inverse square calc; leg length (from last year) = 61.8cm. Looking at a 1/10 bracing interval, braced segment length = 6.18cm. 6.18 (braced interval) is 0.067 of 91.6 (BS test length); 1 over 0.067 squared = 219.69 (in other words, a 6.18cm segment (of a stick with 14.5gr in-place strength) will be 219, almost 220 times as strong. 219.69 x 14.5 = 3179gr. That's not quite strong enough; with the leg angle in a circle bonus tower, each leg is seeing 3810gr loading. So, bump the bracing interval to 1/11 (=5.618 braced segment length); do the inverse square calc, and you see a 1/11 interval gives you 265.83 the strength, so the calculated in-place buckling strength goes up to 3846gr- just over what's needed. This example is based on last year, with single piece, straight legs. This year, you'll have to figure out leg strength and bracing intervals in both the upper chimney section, and the lower base section, but its the same calculational approach.

[Balsa Man]

Anyway to build a one piece jig balsa? Sorry storming you with questions.

Random

It's just one piece- get a really, really nice woodworker to do it for you or if you have a laser cutter, that could also work. You just make a filled mold (or put some mdf panels together) for what it would look like inside your tower. If you get a woodworker though, make sure that guy knows how to cut very, very, precisely and accurately and also be able to put them together really well... it would be terrible for you to realize after 10 towers that the error is the jig, not you. Best of luck!

Understand, when I said "one piece jig", what I was talking about was a single jig holding both upper/chimney section legs, and lower/base section legs, as opposed to two separate jigs, one for each section, where you build the sections separately, then mate the chimney section on to the base section.

As I noted, and Raleway reflected, there are a lot of ways this could be done. My take at this point is that the easiest way to get a really precise jig will be using four jig 'panels', as I described/used last year, put together at 90 degrees from each other (when you look down on the jig from the top). Looking at a single panel, from the side, instead of being trapezoids (last year), they'll need to be ...two trapezoids sitting on top of each other; the bottom portion wide with significant angle, the upper portion narrow with almost no angle. With a laser cutter, cutting such a....compound shape (from one piece of plastic) will be very easy. The alternative for hand cutting if you can't get access to laser cutting would seem to be to cut upper and lower pieces separately and glue them together to form a single 'compound shape.'

[cool hand luke]

Another question, when you are doing all x's how are you making the x's? I experimented with 3 ways and thought they all had major drawbacks.

1. one on the inside, the other diagonal on the outside of the tower. This stunk because it as difficult and time consuming to glue them on the inside after I pulled it off the jig.

2. both on the outside, meaning they were warped, which seemed to really reduce there strength, but that could be a bad tower that I attributed to the non planar x's

3. one on outside, one butt jointed across the span. This took a lot of precision cutting to get the angles on each side for the butt jointed piece.

[Unome]

Another question, when you are doing all x's how are you making the x's? I experimented with 3 ways and thought they all had major drawbacks.

1. one on the inside, the other diagonal on the outside of the tower. This stunk because it as difficult and time consuming to glue them on the inside after I pulled it off the jig.

2. both on the outside, meaning they were warped, which seemed to really reduce there strength, but that could be a bad tower that I attributed to the non planar x's

3. one on outside, one butt jointed across the span. This took a lot of precision cutting to get the angles on each side for the butt jointed piece.

Hm - #2 except laminating a thin piece where one of them connects, so it doesn't have to bend when it crosses over the other? Not sure how this would affect the strength of the joint, if at all.

[Balsa Man]

Another question, when you are doing all x's how are you making the x's? I experimented with 3 ways and thought they all had major drawbacks.

1. one on the inside, the other diagonal on the outside of the tower. This stunk because it as difficult and time consuming to glue them on the inside after I pulled it off the jig.

2. both on the outside, meaning they were warped, which seemed to really reduce there strength, but that could be a bad tower that I attributed to the non planar x's

3. one on outside, one butt jointed across the span. This took a lot of precision cutting to get the angles on each side for the butt jointed piece.

We used #2- both on the outside. As I noted, they were cut from 1/32" sheet, 1/16" wide, and they got a tiny drop of thin CA where they crossed. For the(1/64" x 1/16") X strips we used for ladders and Xs configuration, they were NOT glued at the crossing point, btw). Yes, that meant there was a very slight bow in all of them. At that cross section, they don't have much buckling strength, and it would seem that having a slight bow would reduce that strength further - an aspect of all Xs bracing that I don't yet understand. But, it works; the vast majority of State winners and Nationals top 15 used this approach; what can I say...

[Balsa Man]

Balsa
In last years discussion you posted the leg buckling vs weight, on that graph you had multiplied it by 2.0 and 2.3 fe's based on fixed fixed end conditions


You then stated the all x design with no ladders seemed to have a fe of .4.


Did you mean this is the reciprocal (the effective length? I can't make it jive with the 2 and 2.3
But a .4 as effective length and a 2.5 fe would make sense.

FIRST, AN IMPORTANT CORRECTION ON EFFECTIVE LENGTH FACTOR FOR ALL Xs BRACING. My bad, I should have looked at the spreadsheets rather than going from memory- THE CORRECT, APPROXIMATE VALUE IS AROUND 0.55, NOT 0.4

I'll get something posted soon that will help unravel the mystery. Very briefly, we found, using a ladders and Xs bracing, that if you measure BS by doing 'single finger push-down' testing (which is essentially measuring under pinned-pinned end conditions), you have to multiply the SFPD (pinned-pinned) value by an 'effective length factor' of 2.3 to get to what the BS will be under fixed-fixed end conditions (essentially what you get/create when leg segments are braced with ladders and Xs). That meant, for a C tower, bracing at 1/5 intervals. When we switched for Nationals to an all Xs bracing approach, we needed/used a 1/11 bracing interval. We got to the 2.3...theoretically (digging deeply into Euler's buckling theorem), and it worked very precisely in practice- actual tower performance/in-place leg buckling strength was very close to what theory said it would be. The 2.0 you mention was from early in the process of figuring out how effective length factors work- that it was somewhere between 2.0 and 2.4). To get to the effective length factor for all Xs, we back-calculated/reverse engineered from some detailed data I was provided on a... high performing, non-circle bonus, all Xs tower (using Xs at 1/10 interval, from 1/32 sheet, cut to 1/16 width. The data I got was the (narrow) range of SFPD 36" BS measurements for the legs used. It held full load. Knowing that, it was simple to calculate what the effective length factor had to be. This reverse engineering approach worked- tower held full load at Nationals. I'm still working on understanding the theoretical basis, and will share when I figure it out.

What this all means, in terms of figuring out how strong legs need to be for a given (all Xs) bracing interval, is you measure SFPD at 36", you multiply that number by 0.55. You take that value and do inverse square calculation to see what the BS of (an all Xs) braced segment will be. Example- leg wood tests (at 36"/91.6cm) at 26.5gr. Multiplied by 0.55, that = 14.5gr. Then inverse square calc; leg length (from last year) = 61.8cm. Looking at a 1/10 bracing interval, braced segment length = 6.18cm. 6.18 (braced interval) is 0.067 of 91.6 (BS test length); 1 over 0.067 squared = 219.69 (in other words, a 6.18cm segment (of a stick with 14.5gr in-place strength) will be 219, almost 220 times as strong. 219.69 x 14.5 = 3179gr. That's not quite strong enough; with the leg angle in a circle bonus tower, each leg is seeing 3810gr loading. So, bump the bracing interval to 1/11 (=5.618 braced segment length); do the inverse square calc, and you see a 1/11 interval gives you 265.83 the strength, so the calculated in-place buckling strength goes up to 3846gr- just over what's needed. This example is based on last year, with single piece, straight legs. This year, you'll have to figure out leg strength and bracing intervals in both the upper chimney section, and the lower base section, but its the same calculational approach.

This spreadsheet should help explain things. Note for others, the "fe" cool hand luke refers to is 'effective length factor. This sheet is what we used last year to come up with the bracing interval for both the B and C towers I worked with. With the "two part" shape this year's rules requires, it'll need to be revised to look at both upper/chimney section legs and lower/base section legs. But that's just a matter of copying to get an upper and lower leg sheet, and then plugging in correct leg (section) lengths, and leg section strength needed.

2016-17 all Xs inverse square table set.xlsx

https://docs.google.com/spreadsheets/d/ ... 1227084431

I've put boxes around the four sections of the sheet, and added notes - in red explaining where numbers are coming from, and how they're being used. Section 1 just sets up the leg strength needed. This is based on the angle the legs lean in from vertical. Section 2 sets up the effective length factor, which, as I noted above was back-calculated from leg stick buckling strength and bracing interval data I got from a couple of high performing all Xs towers. Section 3 just provides conversion factors for a given buckling strength measured at 36" to other lengths. Section 4 is where it all comes together to calculate, using an inverse square table, the braced leg segment buckling strength, to see what bracing interval, at what measured leg stick buckling strength you need to get a leg strength at or above the load it will see at full tower load.

I could spend a lot more words explaining in great detail explaining what's going on, and why, but I'm not going to at this point. I'm sure we'll get into Q and A and discussion as we move ahead. Take time to look it over, think about it, see/understand the formulas its structured with, and I think it will make sense, and you'll be able to understand how to use/modify as a starting point for figuring out leg strength and bracing intervals for this year's towers.

[DarthBuilder]

Hello,

BalsaMan can you accept my request to look at the spreadsheet?

Thanks.

[cool hand luke]

I just requested access too. I was halfway done with creating something similar, and would love to have a way of double checking my setup (and therefore my understanding)