Sam wrote:

The Natural Philosopher wrote in message ...


A structuralk engineer is simly a DIYer who has access to more data than
the average, and has professional insurance to back it up.


No, I always get a contractor to do the work that I design, never DIY.


You missed teh irony. I was a D-I-Y electronics tinkerer with a degree
when I joined the Marconi company and found myself instantly a
'professional' designer of e.g. guided missiles.


When *I* said 'look at the way a dam is built', I did that from a sound
grounding in basic structural engineering. Water is teh worst of all
becuse it has no friction - its all weight and its all fluid. If a wall
will hold back water it will hold back soil.


OK, 100% of the vertical pressure of water act horizontally and for
soil only a fraction (about 30-40%). But waterlogged soil is heavier
(approx 2x) than water & there is still usually some water pressure
behind the wall. Engineers will always look at the case of the drains
being blocked, but with a reduced Factor of Safety. Also, soil can
have a surcharge and it can slope up behind the wall - this produces
worse pressures than flat as it is effectively extra surcharge.
Friction at the back of the wall has only a very minor effect on the
lateral load.


No argument with that.

The inverse parabolic arch with buttresses is a good design.


Only if you have somewhere to take the thrusts.

Agreed. You need to use buttresses and maybe anchors at teh end, ort cut
deeper in to leave some soil being at teh'wings'.



So is filling it with steel. Blockwork can take massive compressive
forces, but its lousy in tension.


Steel is good if placed correctly. Blockwork is not as good as you
think. You can put that much steel in a masonry wall as the blockwork
will crush long before the steel yields.

Agreed. But its still masively better n compression than tension sans
teh steel.


The art of the design process is to make sure it never does.


Unless it is reinforced with steel otherwise the reinforcement is
useless....


The other technique is to use massive weight in the wall itself and
broaden its bsae so that the total vetor sum of the forces on any block
never falls outside the wall. Thats how they built the mediaeval
cathedrals and fortresses.


Not cheap and usually means a larger overdig to get the thing in.....


*shrug* The mediaveal guys didn't have the steel. Or the concrete. The
arrival of wrought iron - able to take tensile stresses -
revolutioneised structural design in the 18th and 19th centuries.

Nevertheless the romans built massive aqueducts without it.



A 45 dgree trianguar corss sectuoj wall will never break, but it may
slide - thats fixed by keying it in underground.


But this wont stop a shear failure in the horizontal mortar planes.

Not if the wight of material is sufficient. Or of course use steel. Or
uise interloking blocks not laid as continuous courses.



You can protype a wall like this by using a few wood blocks and some
sand behind it and then hosing it. This will imediately illutstrate all
the failure modes and where the stresses come.


No, it will illustrate the first failure mode to happen, which, unless
the model is accurate, may not be representative.

Then tyou rebuild it to get to teh second failure mode.

No-one really wants to know the second failiure mode anyway if they are
buried under the first one.