Thomson Elite Seatpost Failure

Michael Press wrote:
In article
,
Andre Jute wrote:

On Apr 28, 1:23�am, Jay Beattie wrote:
On Apr 27, 5:05�pm, Michael Press wrote:

[...]

There is no protection from the weakness brought on by anodizing
an Al structural member. The anodized layer is brittle, extremely
brittle. The Al metal is flexible. When the Al flexes, the anodized
layer _breaks_ and cracks. The cracks make a stress riser that
multiplies the tearing apart force at the valley bottom of the
crack. Thus an anodized Al load bearing member is weaker than
the same member that is not anodized. That a seat post can be made
with enough excess Al to compensate for an anodized layer and still
sell is unlikely.
Is this true with all anodizing or just hard anodizing (color
anodizing versus hard anodizing)? �And to follow up on what Chalo was
saying, is polished aluminum (such as the American Classic post)
treated in any away apart from being polished? -- Jay Beattie.
Well, if treatment like anodized ali is all that bad -- and I'd like
to see global failure statistics -- and lightness is essential at any
price, there is an alternative: steel machined in ridges or waves
along the length, inside as well as outside if it can be done, which
will demonstrate the stiffness and strength of the pre-machined
component but only a part of its mass.

Al airframe parts are never anodized. There is some evidence.


not true. you need a better source of info.

On Apr 17, 8:15*pm, Chalo wrote:
wrote:

Chalo wrote:

IK wrote:

Chalo wrote:

It's worth noting that for a
cyclical load to pose a fatigue issue, it has to reach a large
fraction of the material's yield stress somewhere in the part.

This is not true for aluminum and its commonly-used alloys, of which the
Thomson seatposts are made.

If you extend the number of cycles out into the billions, you are
correct. *For ordinary human and bicycle lifetimes, you still have to
come up with a large percentage of the material's yield stress.

I think that these graphs of stress & cycle-to-failure for aluminuym
illustrate your point:

*http://books.google.com/books?id=Gt-...frontcover#PPA...
*or
*http://tinyurl.com/d6qskg

They show that reducing the stress per cycle increases the cycles-to-
failure by orders of magnitude for the aluminum samples.

For example, Fig. 1100.HT03 shows stress dropping from ~35 ksi to ~15
ksi, roughly a 50% decrease in stress, while cycles-to-failure
increases from ~10^3 to ~10^7, or 1,000 to 1,000,000 cycles, ten-
thousand times as long to failure.

That's an increase from 1000 cycles to 10,000,000 cycles, like the
rest of your text would suggest.

These numbers indicate the number of cycles that reach up to that
level of stress. *Since there is a wide distribution of force
amplitude on bicycles, most cyclic forces are much lower than their
peak values, and therefore contribute exponentially less to metal
fatigue.

Even a seatpost that has a fatigue life of only a thousand cycles at
peak load can have a functionally infinite service life. *If each of
those peak cycles represents a painful and anomalous impact, then that
one thousand cycles could represent more than a riding lifetime.

Chalo

if my rear gets even a 1000 hits while seating on the saddle that
causes the seatpost to count as 1 hit to peak load then the saddle
maybe embedded in my "you know what"