About 20 years, a German magazine published fatigue test results on a dozen frames. Four were steel, 4 were aluminum, 2 were titanium and 2 were carbon fibre. The first three frames to fail were steel and the poorest result was recorded by a DeRosa SLX at ~57,000 stress cycles. The next two were a steel Fondriest and Nishiki at ~77,000 & 78,000 stress cycles respectively. The first of the two carbon fibre frame to fail was a Time at ~182,000 stress cycles, just over 3x the score of the DeRosa. The other carbon fibre frame, a Trek OCLV, attained the test limit of 200,000 stress cycles without failing. Two other frames reached the 200,000 stress cycle limit without failing, an aluminum Cannondale and an aluminum Principia.

Now, does this mean that we should all put our steel frames, particularly our DeRosa , for sale on Craigslist, Kijiji and Ebay before they crack under the strain? Of course not. The sample sizes were far too small to derive any statistically significant data regarding the superiority of one manufacturer or material over another. However, it does illustrate that aluminum and carbon fibre can be used to produce a light and durable frame, provided they are not abused.

It also illustrates that the most important factor in fatigue resistance is not the material itself. This is borne out by the result of the 4th steel frame, a Barellia, made from the same Columbus SLX tubing as the DeRosa. It did not fail until ~119,000 stress cycles, over twice as long as the deRosa. This corroborates that there is much more at play than the properties of the material itself. The control of the manufacturing processes used during frame assembly are of critical importance in fatigue resistance.