12 High End Frames in the EFBe Fatigue Test
TEXT AND PHOTOS: ROBERT KÜHNEN (TOUR 10/97)
What do a blade of grass, an airplane and a bicycle have in common? The structural basics: all three things are lightweight structures, and two of them orient themselves at the common model. The "idea" of lightweight construction is the design philosophy of nature: to obtain maximum mechanical characteristics with a minimum of material and, by doing so, preserve resources. Engineers cribbed this principle of nature and turn it on a multiplicity of structures. Pleasant things such as air travel and elegant sports equipment which can be easily used are the direct result. Another result is an apparent paradox to experience: lighter can be more durable. Sound crazy? It isn't. But connecting light weight and durability requires more design effort.
LIGHT ALONE IS NOT ENOUGH
Building a frame that is merely light is no feat. But constructing a light and durable frame is more difficult. The real challenge consists of combining the lowest possible weight with sufficient rigidity and sufficient life span. Only then can one speak of successful lightweight construction. More important is to clarify what is to be understood by "sufficient". Under no circumstances should a frame fall below determined rigidity values, for example the way they are measured in TOUR tests. Otherwise the handling characteristics suffer. If one sets the rigidity of the frame in reference to the weight, then one keeps a first characteristic number for the quality of lightweight construction, what TOUR calls "STW ratio" (STW stands for "stiffness to weight"). The different materials' potentials for lightweight construction are remarkable. In the stiffness to weight ratio aluminum beats all other materials used in bicycle frame construction. However, this is not saying anything about the life span which can be expected.
So that the quality of the frames can be judged reliably, the testing method must also meet elevated claims of quality. The EFBe test (EFBe stands for Engineering For Bikes) enables this by mounting the frame realistically as well as the computer-controlled measuring and automatic control. TOUR increased the loads compared to the DIN test, which guarantees that high speed frames are confronted with forces that are relevant in sporting use. In mounting, the frame is clamped with a special measuring fork to the rear drop outs in front of a flexibly stored hinged support, which mimics the rear wheel. That way the riding conditions while standing, which represent the largest load for the frame, are simulated. Two pneumatic cylinders approach the crank at an angle of 7,5 degrees (corresponds to tilting the bike while standing). A linkage transfers the chain path to the rear drop outs. The cranks are at 45 degrees - in this position the rider applies the greatest force. First 100,000 cycles at 1.200 Newtons are applied, followed by additional 100,000 cycles at 1.300 Newtons. The measuring equipment enters the deflection of the frame under the test load and switches the test off as soon as a larger break occurs.
HOW LONG IS LONG ENOUGH?
Professional racers might already feel that a frame that lasts only one race is sufficiently long-lived, if appropriate advantages offset the short lifetime. Tour riders however require rather a reliable frame, that carries them without complaint for many years and never in the pass has in the technical point of view required both a "fixed" frame, which lasts the service life intended without damage. "Fatigue strength" is the priority request for all lightweight components. Airplanes, cars and bicycles must have safe operation. The counter design for fixed interpretation is the infinite life construction. Infinite life means after an "infinity" of load cycles no break occurs. This small difference in the definition has a large effect in the conversion. During continuous fixed constructions is fast the repeated material employment necessary. Therefore long everything does not leave itself which flies, drives and swims duration fixed to design. Infinite life of airplanes for example is not fair to their function: they would never lift off from from the ground.
There are lessons from real life about the fatigue strength of bicycle frames - particularly in the warranty departments of the manufacturers. The theoretical knowledge is however rather poor, because fatigue tests are not a standard requirement in the bicycle industry. Lately, now that new forms of frames are conceived and new materials are used, this deficit becomes more noticeable; because without machine tests of the service life no reliable predictions about the durability of new structures can be made. Not a few manufacturers press themselves around these tests as annoying obligations, and turn the Spiess simply: they are waiting, as if their products work satisfactorily in practice - but it is not to be so.
Therefore TOUR looked around for a testing method, and became interested in the specialist EFBe-Prüftechnik. Manfred Otto, the managing director of the Bochum Prüftechnik-Herstellers, is a man of the first hour, when it concerns the professional destruction of bicycle frames. The testing method for a frame durability test, Which he developed, forms one of the quintessential points of the new DIN 79100 bicycle test standard, which will enter into force at the year's end. Compared to other testing methods (see for example TOUR Fork Test, number 1/97) this classifying test is simple, but effective: the standing ride is simulated - not the only hurdle, but the highest one which must be overcome on the way to permanent durability. Practical experience supplies the best recommendation for the procedure, tested for many years: the apparent damages which are won on the test stand resemble those that occur in riding experience, according to Manfred Otto. Thus one of the most important prerequisites for a realistic testing method is met. Only one cannot forecast about possible mileage. Predictions like "this frame lasts over 20,000 kilometers" are illegal on the basis of this test. To do that a most complex fatigue strength test would be necessary.
The DIN test, for bicycles of all kinds that are used in traffic - from the shopping bike to the racing machine - calls for a minimum of 850 Newtons (scarcely 85 kilograms), at 100.000 cycles alternating on the left and right pedals. For simple bikes the new, intensified DIN test is thereby a genuine proof stone. Sporting application, however, confronts the material with quite different loads. Therefore TOUR tests the frames substantially harder: first 100,000 load cycles at 1.200 Newtons are to be completed, then the frame with this prehistory is tested in the next higher rating class at 1.300 Newtons. If it lasts through those total 200,000 load changes without breaking, the test is terminated.
Twelve frames of four materials passed through the test. From the results - without statistical certainty, however - conclusions can be drawn about the potential of the materials for building frames. Aluminum is at present the lightweight construction material with the best stiffness to weight ratio. When intelligently designed, these are also very durable frames - see Cannondale and Principia. Within the test field carbon is also outstanding - the material for the future. Titanium frames cannot be easily built to be very rigid, yet also durable. Steel is the material with the worst potential for light weight construction.
TOUR sends eight light frames in the weight class between 1.200 and 1.500 grams into competition. Products from renowned manufacturers - Trek, Klein, Cannondale, Merlin and Principia - meet frames with names less often heard, which however are just as light. An exciting initial position, because the differences in prices are enormous. The price difference reaches 5.199 Marks, from the expensive Merlin titanium frame at 1.299 Marks to the inexpensive Stevens-Aluminum frame. Steel frames are not competitive in this weight class; the 1.500 gram threshold can be undercut with the steel Nemo tube set from Columbus, but only if small tube diameters are selected (different outside diameters can be selected from the Nemo tube set). In the test size of 58 centimeters, such a frame is not rigid enough and is therefore not tested.
Nevertheless, steel must compete - as a comparison yardstick. Two lugged and two welded frames serve as references. The two lugged frames come from De Rosa and Barellia and are based on the most successful steel tube set of the last ten years: Columbus SLX. They represent the classic steel frame with standard diameter tubes. Representing modern steel frames are the very light welded Fondriest frame made with large diameter breath-thin tubes and the mid-weight, more simply held Nishiki frame which is also welded.
LIGHT AND STRONG
Pfffffft, pfff, pfffffft ..... For two long weeks the computer-controlled pneumatic cylinders in the EFBe laboratory carry out the heaviest work and trachten for the frame fauchend after the life. The best of the best take the machine two days of hard work - and still they do not give up! Three light frames make it through the torture without visible marks. After 200.000 strokes output - divided in pairs - which visibly distort even the stiffest frames, the computer switches the test off for Cannondale, Trek and Principia according to plan.
A sensational result, in view of the tender tubes and enormous test loads. Also for Manfred Otto comes the surprising result: "Those are the lightest frames that ever came across the EFBe test stand, and in addition, they are also the most durable." The Time Carbon frame only barely fails because of the hurdle, and succumbs to a break at the chain stay after 182.000 footsteps in the heart of the frame. Likewise the second lightest frame in the comparison does very well and thereby exceeds all expectations: Schmolke's frame manufactured from inexpensive Russian titanium cracks after 160.000 cycles. Klein's Quantum Race fails in fewer cycles: after 132.000 cycles the down tube breaks. Below that comes Merlin's titanium splendor, the Team Road, and - based on the price -gives the largest disappointment: the frame breaks at only 106,000 cycles. Stevens' RPR4-Modell forms the tail end of the light frames, but it is the least expensive one with distance.
Steel in crisis: De Rosa's SLX quickly breaks after 57.000 cycles and does not come half as far as Bruegelmann's Barellia frame from the same tubing. Interestingly enough both break in the same characteristic way, scarcely above the lower head tube lug - a type of break, by the way, the TOUR tester has already seen in the field. The Fondriest frame, extremely light for steel, does not make it past the first load class and breaks like the much heavier Nishiki frame after scarcely 80,000 cycles.
Numbers and facts of the tested frames
Frame Weight Size Price(1) Material/Construction Cycles(2) Break Location Frame Stiffness(3) Reference/Info Barellia SLX 2.080 58 798 steel, lugged 119.316 Head tube, lower lug 64,6/86,2 06196/750075 Cannondale 1.520 60 1.990 alu, welded 200.000 no break 91,3/92,1 0031/541589898 De Rosa SLX 1.895 57 1.650 steel, lugged 56.690 Head tube, lower lug 66,8/80,6 02871/275555 Fondriest 1.630 61 2.600 steel, welded 77.171 both chain stays 79,9/76,8 0821/27250 Klein Quantum Race 1.415 57 2.700 alu, welded 131.907 down tube, cable guide 89,0/94,5 06103/50700 Merlin Team Road 1.525 60 5.199 titanium, welded 100.595 down tube, shift boss 65,9/81,3 040/4806040 Nishiki Team 2.080 56 1.390 steel, welded 78.206 bottom bracket/seat tube/down tube 67,2/93,5 02871/275555 Principia RSL 1.460 60 1.895 alu, welded 200.000 no break 83,5/91,85 0531/2872913 Schmolke Titan 1.300 59 2.000 titanium, welded 160.356 down tube, bottle boss 64,3/70,2 06139/6735 Stevens RPR4 1.515 57 1.299 alu, welded 85.032 right chain stay 77/84,7 040/4806040 Time Helix HM 1.485 57 2.990 carbon/alu, lugged 181.966 right chain stay 66,5/86,2 07159/945930 Trek OCLV 1.200 58 2.800 carbon, lugged 200.000 no break 75,3/94,5 06103/50700
(1) Price for frame set with fork in DM. (2) tested for 100.000 cycles at 1.200 N, then increased to 1.300 N. (3) Head tube stiffness / bottom bracket stiffness.
The results are surprising and pleasing in their clarity. Pleasing, because they prove that by thoroughly thinking through the construction all desired positive characteristics are compatible together: durability, rigidity and minimum weight are not mutually exclusive properties. No question that some future purchase decisions will be shaped by this result.
Nevertheless the results require interpretation so that no misunderstandings develop. None of the tested frames is precarious or dangerous according to our estimate. Even to the "worst" frames in this test the following applies: the color of their bikes will no longer please most racers before the expected life span is reached. Strong and heavy riders, who proceed consistently with high force application, should however access the frames that can be proven more durable.
Because only one sample was tested in each case, no predictions about the quality and the scatter can be made about other frames in the series. During welding and brazing substantial deviations from the sample result are not unusual. It makes little sense to compare too closely the number of load cycles obtained. With this test it depends not on 10.000 or 20.000 cycles more or less, but on the order of magnitude: for example, the Time frame (182,000 cycles) and the Russian titanium (160.000) do battle in the same league. Beyond all imponderables of the sample investigation, however, the test altogether supports the thesis of durable lightweight construction: the results from the light frames with complex design cluster together at the top end of the results, frames of less expensive and heavier construction at the lower end.
The fact that aluminum and carbon frames in this test lasted longer than the steel frames is not in our estimate a question of the material, but the design effort. Not the material, but its skillful use gives the result. However, the manufacturers concentrate their design efforts in a logical way on frames with good potential for light weight construction - and those are made from aluminum or carbon, and only rarely (because of low rigidity) from titanium.
The results in detail: The test divides in two halves. First 100,000 cycles at a 1.200 N load are to be completed. Then the force is increased to 1.300 N and the frames are subjected to 100,000 additional cycles. As soon as the computer registers larger breaks, the test is terminated. It is important to note that the damage potential of the second level is clearly larger than that the first level. A frame that completes both levels is therefore substantially better than one which breaks after 100.000 cycles. However, altogether the racing frames show high efficiency. Compared with "simple" leisure bikes, all frames tested have a good standvermoegen. Three particularly light frames are very good: Cannondale, Principia and Trek passed through the test without breaking: the machine was stopped.
LEARNING FROM MISTAKES
The fracture patterns show that frequently small causes are responsible for the break: drillings and notches in the highly loaded down tube as well as in the chain stays should be avoided, also small welded joints for attachments are critical. Aluminum particularly is a notch-sensitive material that does not forgive such "pre-damage", but titanium also shows sensitivity to this condition in the test. The quality of the light frames is defined by the fact that none broke at a basic weld or bond line: Klein's Quantum Race broke at the derailleur cable entrance in the down tube, Schmolke's titanium frame at a bottle boss drilling, Merlin's Team Road at the base of the weld of the shifter boss. The sizing of the main tubes and their connections is thus correct.
The small weak points, which the TOUR test points out, could probably be eliminated by slight modifications. During carbon construction frequently the junction points with other materials form the weak points. Take for example the Time frame: here the solid aluminum internal lug, which holds the right chain stay, broke (very late) where it extends out from the metal bottom bracket shell. The Trek frame with its carbon lugs lets the forces - so the assumption goes - flow evenly and thereby obtains a higher life span. The breaks in the brazed and welded frames point to temperature problems when brazing, or to stress concentrations in the weld. With improved processing better results should to be obtainable. A positive page for the steel frames leaves this test unconsidered: steel is not so notch sensitive and therefore tolerates small damage better. Also the steel frames are surprisingly less critical in corrosion behavior than some aluminum frames. Steel rusts, but does so slowly. However, some aluminum alloys typically used in bicycle frames are comparatively susceptible to so-called "grain boundary corrosion", which can operate very briskly in the material and so cause a break.
The myth of indestructibility shattered: relatively early - based on the price and high expectations - a spiral break wound around the down tube on the Merlin Team Road titanium frame. Starting point: the small weld of the shifter boss.
Classic fatigue failure with Fondriest: Both chain stays of the light-weight steel frame broke in the weld relatively early. The difference in rigidity between the thin, flexible tube and the substantial bottom bracket shell is probably excessive and provides for stress risers.
Chain stays: The Stevens RPR4 frame likewise suffered a break in the right chain stay. Curving outward by the chain path, the stay failed in a place where two effects overlay: the notch effect at the weld and the depression in the tube.
Weak point drilling: Schmolke's titanium frame tore at the base of the bottle boss drilling. Stress risers at drillings are a classic break spot. However, in this case the drilling was well strengthened. Probably the weld caused the break.
Overheated? De Rosa's SLX steel frame suffered a break in the comparatively solid head tube. Among the possible causes is excessive heating of the lugs when brazing. This example shows that a lot of mass does not necessarily help much and that a processing technique adapted to the material plays a very important role.
Congratulations to the technical designers: the best four of the eight light frames are not only very light, but also extremely durable and represent lightweight construction on a high level. The rumor of early breakage of aluminum, compared to the all-time überlegenen materials steel and titanium, is disproved. Unfortunately, what lasts a long time and what breaks early can never be measured with any certainty just on the basis of what one sees. Also the TOUR testers were surprised by some results. It would be desirable that top frames get their quality certified by an independent source. It is high time for a quality seal, which certifies superior products their top performance!
back to EFBe Home Page
© 1997 Engineering for Bikes (EFBe), Bochum
Translated from the original German language version
using AltaVista Translation.
Used with permission.
Home back to Damon Rinard's Bicycle Tech Page
Send e-mail to Damon Rinard: firstname.lastname@example.org
Harris Cyclery Navigational Hub Page
Articles by Sheldon Brown and others Harris
Beginners Brakes Commuting
Frames Gears &
Tandems Touring What's