Thread: The Search for the Perfect Lug Rig Boom for A goat Island Skiff

Originally Posted by Boatmik

Very roughly the modulus is proportional to the density of the timber or material (exceptions being things like carbon fibre and kevlar). It doesn't depend on compression or tension. It's the same for both.

Ok, that's the important bit for the current bunfight. If that's the case, there's no point having odd sections of mixed species. Wont help stiffness/weight.*

Knew the other stuff (but worth noting for people who didn't know).

You seem very interested in materials and structures! Can I recommend one of my very favourite books?

The New Science of Strong Materials - or why you don't fall through the floor. By JE Gordon. It's in paperback. I'm definitely not fobbing you off with this suggestion but pointing you at a book that is technically very nice, but at the same time is full of real world examples - mostly to do with boats and aircraft, but also diverging into why the british long bow was never successful south of France or why the Greeks had to prop up their chariots overnight. Or why a cook was essential to the analysis of why the WW2 Liberty ships broke up in the beginning or how the doors wouldn't shut on glassfibre train carriages. It really humanises the engineering .. showing the problems that real people had with materials. And that makes it easy to read.

Sounds good. I'll see if I can find a copy.


*ETA: Which makes me wonder why the old racing oars used to do it. Might just have been because they never did a real analysis and just thought it was a cool idea.

Originally Posted by Sumbloak

ETA: Oh and being a civil engineeer and all, can you verify what Ian Howick said about elastic modulus being the same in tension and compression, even for timber?

I'll have to do some more research on the properties of timber before I comment on that. But most materials are either stronger in compression or tension and my understanding is that bending moments (which cause deflection), in a beam, can generally be consider as part of the beam being in tension and the other part being in compression. So changing the shape of a beam or the properties of the material in parts of a beam can be used to improve the stiffness (see my example below).

Think of a steel reinforced concrete beam - concrete is much stronger in compression than in tension and it is used in a beam primarily for it's compressive strength. Steel which is much stronger in tension than concrete is used in the side / face of the concrete beam under tension to carry the tensile loads. I think this might have been the direction you where heading (perhaps indirectly - as this can effect the location of the neutral axis) in some of your earlier posts - but I don't know enough about the difference between the tensile and compressive strength of timber (or if if it is significantly different), to know how much of an effect it would have.

I'll do some googling...

And an ancillary point it that ... usually structures (even a teaspoon is a structure ... though plastic ones are even more so!) are expected to retain their shape. So design is to stay well within their elastic limits.

Your car and car springs are expected to be the same shape tomorrow!

So usually the stiffness requirement is more important than the strength for normal use.

Gordon's book has a picture of a passenger jet with the wings stressed to the elastic limit of the the material. It is truly impressive. And it makes me realise as I fly through turbulence between the Philippines and Taiwan that the wingtips are only deflecting a total of a metre or so relative to the fuselage.

To reach the elastic limit of the material the ends of the wingtips have to be about vertical.

Or another example ... it is possible to make boat decks much lighter and still have them easily strong enough ... but people don't like them to bend under their weight.

Best wishes (and thanks for prompting a great discussion)

MIK

Originally Posted by surlyone

I'll have to do some more research on the properties of timber before I comment on that. But most materials are either stronger in compression or tension and my understanding is that bending moments (which cause deflection), in a beam, can generally be consider as part of the beam being in tension and the other part being in compression. So changing the shape of a beam or the properties of the material in parts of a beam can be used to improve the stiffness (see my example below).

Think of a steel reinforced concrete beam - concrete is much stronger in compression than in tension and it is used in a beam primarily for it's compressive strength. Steel which is much stronger in tension than concrete is used in the side / face of the concrete beam under tension to carry the tensile loads. I think this might have been the direction you where heading (perhaps indirectly - as this can effect the location of the neutral axis) in some of your earlier posts - but I don't know enough about the difference between the tensile and compressive strength of timber (or if if it is significantly different), to know how much of an effect it would have.

I'll do some googling...

Engineers are interested in rather gross differences. I don't mean that unkindly (I am half an engineer). But within the average properties of a material you will find that the elastic modulus will be the same.

Because the elastic component of strength is load on the atomic level. The bonds between atoms are elastic and respond to compression and tension in the same way within the minor changes we are talking about here. You see lots of diagrams that show eetsy little springs between atoms.

Goes to show that engineers can be cutesy wootsie!

There is some component of strength that relates to the type of bond. Van der Waals, ionic, covalent. But below certain critical stresses each one is elastic - because of the small displacements and fundamental electrostatic forces that each is based on. But above certain stresses they will start to displace or dislocate and each material with have different mechanisms for deforming.

However, except in straight tension, these force are usually never realised because there will be some instability in the gross structure. Like the wood driniking straws analogy making the behaviour (and inelastic permanent deformation) different in different parts of the structure or material.

So with wood ... it is not just a material but a structure as well.

Liked Dave's comments of "working with the wood sprit" very much. My inner engineer bristles .. but at the same time a lot of the elegance of classic engineered structures is because of this working with the spirit of wood, or iron, or carbon fibre.

I'd say it is one of the great enjoyments of engineering.

MIK

Originally Posted by Sumbloak

*ETA: Which makes me wonder why the old racing oars used to do it. Might just have been because they never did a real analysis and just thought it was a cool idea.

Wood Engineering started to become scientific around the era of aircraft ... because there is such a pressure on reducing weight compared to any other type of engineering ... so your point about allowing for the different tensile and compression strengths is completely relevant for other structures where light weight is essential.

My experience of oars for racing is that the final wooden generation were as you say .. hollow with a sort of triangular apex shape.

Or maybe it was to provide a flat back to lean up against the straight pivot of an offset oarlock? And then they found that they could build them lighter with few failures that way.

Engineering informs practice and practice informs engineering.

MIK

Originally Posted by surlyone

OK - I have a Civil Engineering Degree but have never practiced in this field, so I feel like I have enough knowledge to be dangerous here (especially as I haven't been on a Goat or yet made either a boom or a yard).

My understanding is that the goal is to increase stiffness in the vertical plane of the boom/yard. So my thought is to borrow a design element from the inwales which could be used to increase stiffness in this plane (possibly a lot?) for a weight increase of 15-25%. Below is a side profile of a the boom/yard design I have been thinking about.

The 40mm by 40mm section is replaced by two 40mm x 20mm sections. In between these two sections spaces would be inserted with the thickest spaces placed where the bending moment is greatest (the image has this at the center but I think the plans indicate that this is 1/3 along the length of the boom/yard). In this example I have the "biggest gap" set to 30mm which would increase the distance from the top and bottom surfaces from 40mm to 70mm (20mm + 30mm + 20mm) in the centre. The ends effectively have an unchanged cross-section but the bending moments should be much smaller in these locations.

I don't think the stiffness in the horizontal plane should suffer (depending on how the sail is attached). However it is possible that the surface in compression (from vertical loading) could twist/deflect horizontally due to excessive loading (which can be a problem with I beams) - but I suspect the section would have to be significantly deeper for this to occur.


https://picasaweb.google.com/1153149...04258527052754

It could look neat or ridiculous - certainly different. It would take more work to make, paint and will weight more (dependent on spacer size, spacing and timber used). I don't know if there would be any problems with the rigging (either attaching or interfering with the rigging) - I have no experience in this area.

Howdy ... from spar building practice .. the tendency has been to go for tubes ... whether they are round or square or octagonal. Mostly because they turn the "hard to calculate" twisting/torsion moments into factors of no significance. Torsion is a significant problem. As well side forces are significant too.

Sometimes every few years you see a hot boat with a truss as a boom. But in the end it is much too much work when a simple tube does the same job ... well ... more simply.

Also the smaller diameter truss tubing is vulnerable when it comes up hard against a stay during an intense gybe. (hmmm ... I am feeling my bs detector going off as I write that sentence! Possibly true because you can transfer the load into the wide top or bottom face if a rectangular hollow section (as one example).

MIK

Originally Posted by boatmik

this relates to the modulus of the material. Very roughly the modulus is proportional to the density of the timber or material (exceptions being things like carbon fibre and kevlar). It doesn't depend on compression or tension. It's the same for both.

Originally Posted by boatmik

for strength there are different mechanisms on tension and compression faces.

Sumbloak - I would agree with these statements that MIK has made - my quick internet refresher coarse indicates that the modulus of elasticity is generally considered (assumed) the same in both compression and tension for most (a lot) of materials. I think there are exceptions but I couldn't find any evidence that timber fell into this category.

Table 2 of this document http://www.fpl.fs.fed.us/documnts/pdf2001/green01d.pdf lists the tensile and compressive strength of some types of timber and the tensile strength is normally much greater than the compressive strength. I think this may have been what you were originally referring to. But as MIK pointed out these refer to the stresses required for the timber to fail (or to start plastically deforming past ie past the point where it will return to it's original shape) not to the stiffness before permanent/plastic deformation occurs.

Under normal operating conditions the boom should be flexing in the elastic range so compressive and tensile strengths should be roughly equal on each face of the beam and the neutral axis should be close to the center of the of the section. Under these conditions it shouldn't matter (in regards to stiffness) if an additional timber section was added to the the top or the bottom face. If you were looking at increasing the failure strength of the section then you would add the stronger material to the compressive face.

Got to love engineering and science! I think you should ignore my steel reinforced concrete example in my earlier post - it isn't very relevant here and I think it confuses and mixes the issues of elastic and plastic deformation.

Originally Posted by Boatmik

Howdy ... from spar building practice .. the tendency has been to go for tubes ... whether they are round or square or octagonal. Mostly because they turn the "hard to calculate" twisting/torsion moments into factors of no significance. Torsion is a significant problem. As well side forces are significant too.

Sometimes every few years you see a hot boat with a truss as a boom. But in the end it is much too much work when a simple tube does the same job ... well ... more simply.

Also the smaller diameter truss tubing is vulnerable when it comes up hard against a stay during an intense gybe. (hmmm ... I am feeling my bs detector going off as I write that sentence! Possibly true because you can transfer the load into the wide top or bottom face if a rectangular hollow section (as one example).

MIK

Spoil sport! I guess I shouldn't be to surprised - there probably aren't to many radical new designs left for common parts like these.

Ok, so if the "problem" with the existing booms is too much vertical deflection, the obvious thing to do is to just make them deeper. If using a box-section oregon boom as a benchmark, it should be feasible to replace the side panels with paulownia at twice the height. Density will be about half. Presumably (ballpark geusstimate) other properties will be about half too. Net result should be a boom with the same weight as the benchmark, with the same sideways stiffness, but with around four times as much vertical stiffness. The paulownia should still be able to handle the torsion and shear easily enough.

Spot on Sumbloak.

Increase section without increasing weight ... or better ... reducing it.

Complicating factor is that wood doesn't have a mechanism for
1/ preventing splits in thin walls. The standard guideline for timber walled spars is that timber thickness is to be 20% of cross section (safe) or a 15% (a little risky).
2/ That split prevention also relates to allowable load on the glue joints ... having enough glue surface area. That can be a problem as the wall gets thin.

My own instinct is that plywood gets around both of these problems. However it has half the tensile strength which might be too great a sacrifice.

So I'm imagining a top and bottom flange of some nice looking fir or pine (or maybe paulownia is strong enough) Maybe 12mm thick like the mast walls with ply sides. With some organisation it might be possible to get the ply out of the sixth sheet used in the standard build.

Though I do think it might be fun to make something with 4mm ply sides and then try to work out how to prevent it from breaking!

That approach worked for the Tornado Catamaran - which was delicate when it came out but gradually they worked out how to build it (in timber originally) to be a reliable structure. It is still among the very fastest catamarans despite being almost 50 years old.

But 4mm means buying extra stuff.

Another approach would be to glass a full timber boom ... but I'm really against all that extra messing around. And it doubles the analysis problems.

By the way ... do you know we have been collecting data?
WIKI for setting up and tuning Lug and Sprit Rigs | Michael Storer Wooden Boat Plans

MIK

Mik, I'll bring at least two and possibly three booms to SailOK. Ply/timber, lap-joined timber and a third one (different assembly method) I'll whip up early next week.

I'm not completely sold on the ply/timber one (6mm ply) but that may result from poor craftsmanship when I built it. I don't have enough 6mm on hand right now to more carefully make a new one. I am picking up more ply early next week and I may be able to craft a new ply/timber boom before leaving for Oklahoma.

Is there any reason, other than aesthetics, not to make a taller/deeper yard in order to improve stiffness?

See you in Oklahoma!

I still have some 6mm off cuts from the GIS that are 8' long. I think I have enough to butt-join two sides of a ~12' boom. I also have a decent remnant of a 8' DF/Oregon board that could scarf to length for top and bottom caps. I'll wait to hear the SailOK feedback but I might be interested in putting together a ply-sided boom box. MIK, what if the caps were channeled out as c-sections? would the wall thickness still need to be the minimum (12mm for example) or could the cap start at 12mm with an interior lightening groove of 4-6mm? Will the remaining "flanges" retain the stiffness?

Originally Posted by Boatmik

Spot on Sumbloak.

Increase section without increasing weight ... or better ... reducing it.

Complicating factor is that wood doesn't have a mechanism for
1/ preventing splits in thin walls. The standard guideline for timber walled spars is that timber thickness is to be 20% of cross section (safe) or a 15% (a little risky).
2/ That split prevention also relates to allowable load on the glue joints ... having enough glue surface area. That can be a problem as the wall gets thin.

1/ Yup, but the wall thickness would still be the same as the benchmark spar so it should be ok. The only change is increasing the height of the side walls and decreasing the density. As long as it isn't taken to ridiculous levels (like 300 tall and 12 thick or something) it should work. Might need some extra internal blocking here and there for added support, but that's not a big deal.

2/ In my idea, the actual gluing area is the same for both spars. Stress on the glue joints between the sides and top/bottom will be markedly reduced because of the deeper section, so the less dense timber can probably handle the shear loads ok. Easy enough to check with a quick calculation.

My own instinct is that plywood gets around both of these problems. However it has half the tensile strength which might be too great a sacrifice.

Yup. You'd lose out on sideways stiffness if using plywood. Also, you'd have to rebate it into the top and bottom walls, since ply couldn't be butt joined to those (unlike solid timber). Not sure you'd gain anything.

Though I do think it might be fun to make something with 4mm ply sides and then try to work out how to prevent it from breaking!

Lotsa carbon.

By the way ... do you know we have been collecting data?
WIKI for setting up and tuning Lug and Sprit Rigs | Michael Storer Wooden Boat Plans

MIK

Saw that the other day. Good stuff.

I prefer Oxygen and Hydrogen in my carbon.



MIK