Re: Is my frame about to collapse?

"Simon Brooke" <simon@xxxxxxxxxxxxxx> wrote:

James Thomson ('yosnappyj@xxxxxxxxxxx') wrote:

if you estimate typical stresses to be some fraction of peak,

I don't. I assume that the loading is a spectrum, with rare, short
periods of very high loading (sprinting, honking) and much longer
periods of low loading (descending hills, cycling on the flat). So I
guess for a typical road bike you might have a profile which looks
like

More than 50% of the time below 50 watts

[snip]

Less than 40% of the time between 50 and 200 watts

[snip]

Less than 9% of the time between 200 watts and 400 watts

[snip]

Less than 1% of the time above 400 watts

[snip]

So when you say you don't, in fact you do. You assume that 90% of riding
time is spent producing less than 50% of peak power, though of course power
and stress aren't equivalent.

Obviously there's also the issue of the weight of the rider on
the bike as the bike encounters variations in the road surface,
but I'm assuming that this is (except for mountain bike riders
landing jumps) a relatively small issue compared to the
reciprocating forces on the bottom bracket.

That's a debatable assumption, given the very high frequency of some
road-surface induced loads, but let's not get into that now.

Let me repeat the part of my question - the most important part - that you
snipped:

if you know that a number of the frames are made of a material
that exhibits a fatigue threshold at 30-50% of yield, what would
you expect to be the effect on the ranked results of applying *only*
exceptional stresses?

Of course I'm referring to the EFBe test.

But a frame which was designed to handle 200 watts indefinitely
would be no good if it couldn't handle 400 watts occasionally,

[snip]

Naturally. So the important question is the relative frequency of those
loads (you estimate 50:1), and in fact what proportion of the applied load
spectrum falls below the fatigue threshold (if any) of any frame in the
test. That's something that the EFBe test, with its monochromatic load
spectrum entirely fails to address, and that's what makes it a Mickey Mouse
test.

James Thomson


.

Relevant Pages

  • Re: Track bike.
    ... >could add lights et al. ... Steel frame CBR600 race bike. ... Bulletproof engine and loads of spares ...
    (uk.rec.motorcycles)
  • Re: What if I cant get my bike on the train
    ... the bike (most bikes not being equipped with a handy clip). ... There are loads of places on a bike where you can wedge something like ...
    (uk.railway)
  • Re: MP criticises ID card opposition
    ... canoeing/microlighting/abeiling `holiday' in Iceland. ... some of the scariest rivers on the planet - Icelandic rivers are quite ... Whereas an old Brit bike will give you trouble almost ...
    (uk.people.support.depression)
  • Re: What a mess, eh?
    ... There's a switching valve in the case of the engine you describe - I ... | throttle and then opened the decompressor. ... It's a rare bike without something like that, ...
    (uk.people.support.depression)
  • Re: MP criticises ID card opposition
    ... turn somewhere around Chester and no-one noticed until the Watford ... My first bike ... I just used L-plates on a car licence, ...
    (uk.people.support.depression)