From: atm{at}misterg.fsnet.co.uk (Andy Gray)
To: atm{at}shore.net
Reply-To: atm{at}misterg.fsnet.co.uk (Andy Gray)


On Tue, 19 Aug 2003 06:28:24 -0700 (PDT),  Ross Sackett
 wrote:

>Up to loads where buckling is an issue the squashing
>or extension of truss members in tension and
>compression is controlled by the cross-sectional area
>multiplied by the modulus of elasticity.

...and length, and axial load.

(The axial load depends on truss geometry for a given external load - tight
angles bad, fat angles good.)

I seriously doubt whether the axial stiffness of the truss members is a
real concern here. The devil is in the design of the joints to ensure the
load on the truss members is purely axial, and in making the structure at
either end of the truss stiff enough. The classic "mirror box" of
an open truss dob looks like an open topped shoe box - next time you have
your hands on one try twisting opposite sides and see how stiff it is. Is
this a good foundation for your large diameter, tapered wall, pin jointed
carbon fibre truss ? Don't expect the truss to stiffen up the mirror box!
If the design is good, you can build it out of paper and straws and it will
be as stiff as hell (but not necessarily strong).

As to undersized connectors:

Their function should be to *clamp*, making their properties largely
irrelevant to the stiffness of the structure, since you're reliant on
friction generated by this clamping force (or pre-load in butt[1] joints).
This does require well designed joints, though.

[1] This would not cause laughter in the UK - don't know what US equivalent term is!

Buckling of truss members is a real concern (more from accidental damage /
ease of handling than pure performance), as is resonant vibration - hence
large diameter, thin wall tubes.

As to measuring the extension of a truss member:

The measurement of deflection of the whole assembly is a reasonable target,
and could be done with a weight, some string and a dial gauge. The
extension of each member can be calculated very accurately by analysing the
structure, and plugging in relevant material properties, but if you were to
do this, you would likely find that other factors dominate.

In short, good mechanical design is always worth much more than exotic
materials, or even craftsmanship. Neither of the latter can be exploited
without the former.

I have seen some absolutely beautifully made structures that will *never*
perform well, simply because there is no
*design* behind them. As an avowed coward, I can't face
pointing this out to the proud creators! The design principles and even the
analysis of truss structures isn't particularly difficult, but it does have
to be learnt. There are many engineering texts that cover it.

(OK, rant over...

..you can come out now...

... hellooooo.....)

Andy (actually a Mechanical Engineer!)

(Andy Gray, Dyserth, N. Wales, UK)

--- BBBS/NT v4.01 Flag-5