[Public-List] Structural strength of knees

[Anders Bro via Public-List]

I have had a lot of edukatin as I have read all the contributions about 
the knee structure/bolts/tabbing. And although I am absolutely no 
authority in these matters, there are a few points I would like to 
present for contemplation? First of all, my background is as an engineer 
(rock engineering and rock testing) and we in the rock engineering field 
are very concerned with shear strength, as it is what keeps those big 
blocks of rock in place so as to prevent land slides, dam abutment 
failures and the like... but I digress.

My first concern is that the term modulus has been brought up, and the 
various values of the materials involved (steel, epoxy, wood). The 
modulus is generally associated with the elastic properties of 
materials, and as such typically involves very small (elastic ie 
recoverable) strains. I expect that we are really dealing not with 
elastic properties but rather yield. As such, the strength of the 
materials is was governs the failure of the structure, not the 
small-strain elasticity. So it might be more appropriate to evaluate the 
relative strength of the materials and not the moduli.

Then there is one point that might be relevant - the plastic (yield 
deformations) that result in the sharing of the loads. If say the wood 
is replaced by a very strong material, then the shear strength on an 
individual bolt may be exceeded before the stress on the next bolt 
begins to build. So it might be better to have a weaker (although not 
weak) foundation rather than a super strong one?

This concept of sharing the shear load among the bolts may also be 
misplaced. In the field of rock engineering, we use rock bolts to 
stabilize large rock masses. However these bolts are never used in shear 
but in tension. The idea is that the tension applies a normal component 
of stress to the interface and thereby strengthens the shear interface 
(by means of the coefficient of friction of the interface). So in our 
case of the chain-plate interacting with the knees, if the normal force 
applied by the bolts becomes small (due to rot and compressibility of 
the underlying wood) then there is very little shear resistance that 
develops and the bolts (rather than the interface) are placed in shear - 
their weakest mode of failure. Again referring back to the rock 
engineering field, we actually do not orient the bolts normal to the 
shear plane, but rather at an angle so that if a small shear 
displacement does develop, then the bolt tensile load increases, thereby 
increasing the normal load and thus the shear strength of the interface. 
If the bolts are oriented perpendicular to the place, no normal load 
increase can develop due to a shear displacement. In light of this 
experience, one possibility would be to angle the chain-plate bolts 
downward?

One idea has been proposed to really beef up this junction. That is 
certainly one approach, but I expect that the weak link in most of our 
boats is (obviously) rot, and the weak bond between the plywood and the 
tabbing. (I had a Southern Cross 28 in which the wood was totally rotted 
underneath the tabbing, but not in the exposed portion of the bulkhead. 
I excavated the rotten plywood back to good, and then laid in numerous 
layers of fiberglass and then extended the tabbing about about 6".) I 
expect that if the bond and the plywood is in good shape, there the 
structure is probably adequate (?). I personally like the idea of going 
with the 5/16" bolts, but expect the problem is low stress on the 1/4" 
bolts that leads to the bolt shear. The problem I can see is that as the 
wood shrinks and expands, the bolt stress will become low and so one may 
need to rely on the shear strength of the steel rather than the shear 
strength of the knee/chainplate interface.

My 2 cents...

Anders


On 02/07/2016 08:14 AM, Don Campbell via Public-List wrote:
> Stephen:
>    Don't rely on your impression when it comes to engineering 
> strength. You might want to read some of the Gougeon's book on boat 
> construction for an explanation in more lay terms than engineering 
> texts, but the strength is in the truss system which relies on the 
> core being structurally sound and fully bonded to the laminates on the 
> three sides for the knees. Polyester lay-ups are not strong on their 
> own, and epoxy lay-ups are not really much better on their own.
> Don
>
> -----Original Message----- From: Stephen Gwyn via Public-List
> Sent: Sunday, February 07, 2016 12:32 AM
> To: public-list at lists.alberg30.org
> Subject: Re: [Public-List] Public-List Digest, Vol 3193, Issue 2
>
> Hi,
>
> I just pulled, cleaned and rebedded half my shroud chainplates (the three
> on the starboard side). It took about 5 hours. Everything was very 
> wet. But
> all the metal was all rust-free and the wood was unstained and quite hard
> when I poked it with a knife. The chain plate for the upper shroud was
> quite damp, but as far as I can make out, it was damp with salt water 
> from
> the last time I had the rail underwater. Lots of shiny crystals. Probably
> too salty for rot.  I used 3M4000 (polyether). Much better handling and
> much faster cure time than Boatlife polysulfide. We'll see how long it
> lasts.  I gave the port chainplates a hard look and they appear much the
> same.
>
> My impression of the fibreglass/wood knees for the lower shrouds is that
> the wood is just there as a spacer. The strength is all in the 
> fibreglass.
>
> SG
> _______________________________________________
> These businesses support your Association:
> http://www.alberg30.org/store/A30supporters.html
> Please support them.
> _______________________________________________
> Public-List mailing list
> Public-List at lists.alberg30.org
> http://lists.alberg30.org/listinfo.cgi/public-list-alberg30.org
> _______________________________________________
> These businesses support your Association:
> http://www.alberg30.org/store/A30supporters.html
> Please support them.
> _______________________________________________
> Public-List mailing list
> Public-List at lists.alberg30.org
> http://lists.alberg30.org/listinfo.cgi/public-list-alberg30.org