Re: Testing, was [CR]No more DeRosas : they all broke ?!?!

(Example: Events:Cirque du Cyclisme:2007)

Date: Sun, 07 Nov 2004 09:44:40 -0800
From: "Steve Maas" <stevem@nonlintec.com>
To: classicrendezvous@bikelist.org
Subject: Re: Testing, was [CR]No more DeRosas : they all broke ?!?!
References: <418E153A.6080108@erols.com>
In-Reply-To: <418E153A.6080108@erols.com>
cc: HM & SS Sachs <sachs@erols.com>

I know, I know, I'm a mean SOB when it comes to evaluating technical work. I frequently review technical papers for publication, and recommend rejection of probably 80% of them. Still, when I get on an airplane that takes off from Logan Airport in 50 MPH winds, as I did on Friday night, I'm thankful for the mean SOBs among us and hope fervently for their continued influence.

That said, let me be as charitable as I can to the folks in Germany who did this work.

I think that they have started, but not completed, their work. I'd like to see them conclude something like "From these results, we can estimate the mean lifetime of the frame under X level of stress to be N cycles" where X is a much lower level of stress. I don't know if that's possible from what they did; I assume not, since they didn't make such a conclusion. I can't see how one could do so from testing at only one level.

Possibly one could use a single-level test like this to compare several frames that were very similar, and of the same size. Say, a bunch of 58 cm steel frames. Then the qualitative results might be useful; for example, if many failed the same way, you could conclude that there is a particular weakness. But I see no way (unless someone who understands the underlying technology enlightens me) to use this to compare very different frames.

I have no problem intrinsically with the use of high levels of overstress or the fact that only one mode of use was examined, as long as the researchers recognize this (which they did) and don't claim that the results are valid for all types of stress. The problem is that I can't see how to use the specific results to make a more general conclusion.

Oh, by the way, I think that Dale is probably not the best person to receive all those weak, dangerous DeRosa frames, but I'll be glad to dispose of them for all of you.

Steve Maas (bummed by the rain in) Long Beach, California

HM & SS Sachs wrote:
> Gentlepeople:
> I'm a bit conconcerned about our instinct to jump on and denigrate the
> test results (and methods) shown at
> http://www.damonrinard.com/EFBe/frame_fatigue_test.htm.
>
> First, it is devilishly hard to build a test protocol that mimics "real
> life." The guys at Engineering for Bicycles have at least tried. Maybe
> we don't know enough about how the head tube was anchored, but why
> should the results otherwise surprise.
>
> Second, I found the test approach rather interesting, in assuming that
> climbing out of the saddle is the extreme load, and angling the force to
> simulate that. The second test I might have wanted in an ideal world
> would have had an impulse applied near-vertically to the rear drops, to
> simulate hitting something ugly.
>
> Third, Wilson and Papadopoulos (Bicycling Science) have given the
> properties of materials some thought and expressed things pretty well.
> If I understand them correctly, there is an evolving and influential
> school of thought that extrapolating from megacycles at low strain is
> less relevant for failure prediction than testing with extreme
> loads/strains. This is exactly what these folks have done.
>
> Fourth, I didn't find much to be surprised about. Yes, it probably
> missed some failure modes at the back end (see above), but overall it
> seemed to fit what we all think we know about notches, welding, holes in
> cylinders, and the likely effects of joining very heavy stuff to very
> light stuff.
>
> Finally, and I don't want to put words into others' mouths, I don't see
> this as affecting our love for lugged steel frames. We all know that it
> is possible to consistently manufacture lighter frames that are
> relatively reliable from other materials. But we also know that for some
> inexplicable reason we like lugged steel, and ride it confidently,
> knowing that our bikes are incredibly unlikely to fail in service --
> even though we've all seen lugged steel frames that have failed.
>
> So, I'm thrilled that someone has found time and funding to do this
> work, and to expose the results to public scrutiny. Can you at least
> clap with one hand? :-)
>
> thanks
>
> harvey sachs
> mcLean VA
>
> +++++++++++++++++ Earlier posts +++++++++++++++++++++
>
>>
> I would point out that this test assumes all the loads are applied as if
> climbing out of the saddle. And then you need to assume the test actually
> simulates those loads. I suppose if you want a bike optimized for riding
> out
> of the saddle all the time- this is your test.
>
> As for me I'm not much concerned about head tube failures in steel frames.
>
> Joe Bender-Zanoni
> Great Notch, NJ
> ----- Original Message ----- From: <OROBOYZ@aol.com>
> To: <gillies@cs.ubc.ca>; <classicrendezvous@bikelist.org>
> Sent: Saturday, November 06, 2004 4:03 PM
> Subject: Re: [CR]No more DeRosas : they all broke ?!?!
>
>
>
>>> In a message dated 11/6/2004 2:41:45 PM Eastern Standard Time,
>>> gillies@cs.ubc.ca writes:
>>>
>>> << http://www.damonrinard.com/EFBe/frame_fatigue_test.htm
>>>
>>> Quite possibly - after all, the DeRosa in this fatigue-limit test
>>> performed the worst out of 20 bikes !! And will we be stuck with TREK
>>> OCLV frames for a long, long, long time because apparently they are
>>> balanced in a way that allows them to survive many fatigue cycles ?? >>
>>>
>>> Talking about most of these bikes is off topic but this test is at least
>>> questionable and at worst bologna.
>>> It is ironic that in real world use, Trek OCLV, Cannondale and Principia
>>
>>
>>
> are
>
>>> KNOWN to break! Of all the makes listed, these guys have an acknowledged
>>> history!
>>> And then the DeRosa and other lugged bike broke in the exact same
>>> place...
>>
>>
>>
> A
>
>>> place I have never seen a bike break in over 30 years in the biz! That
>>
>>
>>
> really
>
>>> seems strange. I see that they don't show how they held the front end of
>>
>>
>>
> the
>
>>> frames in their stress testing machine. Hmmmm.
>>> I urge anyone concerned about their DeRosas to abandon use
>>> immediately and
>>> send them to me.
>>>
>>> Dale Brown
>>> Greensboro, NC
>>
>>
> _______________________________________________


>

> .