need rudder shafts

[Stumble]

.............................Young's modulus .........mass
G5 Titanium................15.84 x 10^6 psi .... 4.50g/cc
304 stainless...............27.55 x 10^6 psi .... 7.98g/cc
316 stainless...............27.55 x 10^6 psi .... 7.87g/cc
6064 T-6 Alumi............10 x 10^6 psi ......... 2.70g/cc
RM Carbon composite...19.5 x10^6 psi ....... 1.60g/cc

Titanium has a lot going for it, but it isn't particularly stiff, as compared to its very high strength. Note that carbon is actually only about 2/3 as stiff as 316 stainless, but at about 1/4 the weight.

So sure, G5 Ti has a higher YM than aluminium, but it's about half that of stainless. This is the same reason that aluminium masts are always going to be preferred over titanium ones. You could build one sure, but it would be heavy for its stiffness.

[delmarrey]

Not going to argue the point but there is a reason why titainium is use for wing struts, landing gear & high stress components of aircraft.

[Hunky]

I asked Kurt about sizing a smaller solid shaft compared to the specified tube and he gave me this:

E x I = E x I
and
SM x Fb = SM x Fb  for each example.

Been awhile since I studied structures in college but a quick google shows me that "E" is Young's modulus and "I" is the area moment of inertia.

"SM" must be Section Modulus? and "Fb" bending force? It'll take some more googling to figure out how to apply this. I did find this (a pdf): http://www.engr.sjsu.edu/bjfurman/co..._Stiffness.pdf

But that notwithstanding, I may just go with 1.5" solid shaft in 316 or better if I find it well priced comparatively.

[jimbunyard]

Modes of failure also has a dog in this hunt.

Glass, FRP, whatever it's constituents, is prone to sudden, catastrophic failure.

Know next to nothing about titanium, having fooled with it on only a couple of occasions, but in those occasions it seemed significantly 'stiffer' or 'not ductile' than comparable thickness steel. It also seems I remember reading that titanium is 'brittle', but that seems a rather nebulous term.

Aluminum is defeated (in my book) by it's relatively low strength and corrosion tendencies.

Whilst bending is not a desirable quality, it is better to have a bent rudder than no rudder (to a point anyway), and rudders can be designed with that in mind.

'Toughness' might be an appropriate word here (though harder to quantify). I've read that at some point the 'stiffness/weight' ratio is crossed and a solid shaft is actually less stiff than a tube, but can confirm neither the math nor the accuracy nor the mechanics (physics?) involved in this claim.

The potential for corrosion-related failures rules out tubular metal construction elements in a rudder shaft for anything but a racing boat for me (and even then sailing ability would seem to trump the 'advantage' gained by shedding a few pounds in one's rudder shaft), but I'm notoriously old school and conservative when it comes to building or designing anything...

[Stumble]

Originally Posted by jimbunyard

Modes of failure also has a dog in this hunt.

Glass, FRP, whatever it's constituents, is prone to sudden, catastrophic failure.Sure it can happen, but you are talking about the build material for the hull too, no one worries much about their boatjust catastrophically failing. Glass just doesn't have the stiffness that is really required for a rudder shaft. You could build one, but it would need to be very big diameter.

Know next to nothing about titanium, having fooled with it on only a couple of occasions, but in those occasions it seemed significantly 'stiffer' or 'not ductile' than comparable thickness steel. it isn't. Those memory shape glasses you can bend in knots are CP2 (commercially pure grade 2) titanium. It's actually a pretty flexable metal. It also seems I remember reading that titanium is 'brittle', but that seems a rather nebulous term.

In engineering brittle refers to the delta between the yield and ultimate tensile strength by percentage. The yield of G5 titanium is 118ksi, the uts is 128ksi. So the delta is less that 10%. Aluminium with a yield of 18ksi and a UTS of 28ksi is pretty mailable, but note that in absolute terms they are both 10ksi. One of the difficulties in working with titanium, say bending Ti pipe I said that because it is so flexible it's hard to know when you have actually pushed it past its yield point. It isn't uncommon for new Ti fabricators to bend a piece of pipe and have it spring back to its original shape, because it was nevertheless stressed enough. Just like those memory shape glasses.

Aluminum is defeated (in my book) by it's relatively low strength and corrosion tendencies.agree

Whilst bending is not a desirable quality, it is better to have a bent rudder than no rudder (to a point anyway), and rudders can be designed with that in mind.

'Toughness' might be an appropriate word here (though harder to quantify). I've read that at some point the 'stiffness/weight' ratio is crossed and a solid shaft is actually less stiff than a tube, but can confirm neither the math nor the accuracy nor the mechanics (physics?) involved in this claim. a solid shaft is stiffer by OD than a pipe, but is much less stiff by mass. Solid rods are a very inefficient use of material, but if weight isn't important then it doesn't matter much. If weight is important, and for a rudder it should be, then increasing the OD even a tiny bit is far more effective than moving to a solid pipe. The numbers get complicated, but basically a pipe gets stiffer as an exponent of OD increase, and as a negative exponent of ID (gets stiffer slower for each additional wall thickness)

The potential for corrosion-related failures rules out tubular metal construction elements in a rudder shaft for anything but a racing boat for me (and even then sailing ability would seem to trump the 'advantage' gained by shedding a few pounds in one's rudder shaft), but I'm notoriously old school and conservative when it comes to building or designing anything...

The actual stiffness of a rod or tube is given by the formulas
I(rod) = Pi x D^4/64
I(tube) = Pi x (Do^4-Di^4)/64
Where Do is the Outside diameter
Di is the INside diameter

Flexural Stiffness = EI/L
Where E is the Young's Modulus
L is length

So even a minimal increase in Do increases the stiffness by the fourth power. But increasing the Id while helpful doesn't help nearly as much.


All this is a long winded way to say that for a given amount of material making the OD larger increases stiffness much faster than adding material to the ID. In practice you make the tube as large OD as will fit inside the rudder, then only add as much wall thickness as you need to safely contain the loads.

[delmarrey]

From a Machinist view there are 3 main properties to metal. Here's a link to a full chart that will give a good comparison.

Modulus of Elasticity or Young's Modulus - and Tensile Modulus for common Materials

[Scot McPherson]

When I needed a new rudder shaft for my seapsrite 23, I just ordered a piece of 1 1/2 inch schedule 40 #316 SS from the closest plumbing supplier I could find. I could only buy a 12 ft piece from them, but it was the cheapest option by far compared to other choices.

[Hunky]

A guy who built a 36' cat recently used schedule 80 pipe for his shafts. He said so far no problems. For now, though, I'm searching for solid round bar. I checked Online Metals and ground shipping was $250 to Anchorage so looking a bit closer to home - west coast, west Canada.

[Hunky]

So I have found .375 wall 316 tube for around $243 each plus shipping (which will be at least that amount) despite being told by one vendor that they don't make it. That is at Pennstainless (that has the tube).

I've also found round bar ranging in price from around $280 for both 6' sticks to over $680, not counting shipping.

At this point I'm thinking of going with bar anyway, even though the tube price is not out of reach by any stretch. I do like the suggestion that bar may be bent some and still get you home. Plus I'm traveling soon and will be able to pick up the bar and then pay $75 in excess baggage fees to get it home - as well as get a keyway slot cut in it at the time.

One quote I haven't dealt with yet asked me about straightness tolerance. Googling that I see references ranging from nearly a 1/4" over 5' (seems too excessive) down to about .1mm over 5'. And another resource saying 1/8" over 5'. I suppose 1/8" wouldn't be too bad? Better if less.