What material is used in the sprocket

Is steel really more durable than aluminum?

Issue 10 February 2010

This article as a PDF

by Rainer Mai

Wolfgang Grabmer relies on the premise that chainrings made of steel are more durable than those made of aluminum. I would like to counter this with my decided yes and no.

I admit that I am not really competent in arguing because I have not yet dealt with this question scientifically and was too lazy to research the (possibly readable somewhere) findings of smarter people. I'm just chatting off the cuff here - from the sewing box of elementary materials science and, more importantly, that of my own experiences.

Yes, steel is more durable than aluminum

Both steel and aluminum come in different strength alloys. In principle, both groups of materials can be hardened, both by heat treatment and by strain hardening (forming, e.g. rolling sheet metal or drawing wire).

Well-hardened steels are much stronger than the strongest aluminum materials. With most components, this difference can be compensated for and even overcompensated by using more material. The specific weights are in a ratio of 3: 1, with the same weight, aluminum parts are often more resilient than steel parts.

But with chainring and sprocket teeth, aluminum cannot make use of this lightness advantage, because the dimensions (tooth width, shape and thus the load-bearing surface of the chain roller) are given by the chain and the available space is exhausted regardless of the material. In addition, there are poor friction properties: aluminum tends to seize (heavy material removal).

A good (hardened or even tempered) steel chainring is therefore more durable than a corresponding aluminum chainring.

No, aluminum is more durable than steel

So much for the theory. The disdainful practice looks different: The good steel chainrings do not exist. In any case, with one exception, see below, I don't know of any good, but only cheap items that are apparently punched from relatively soft mild steel sheets.

In the case of pinions (ring gear, rear) there is much stronger, sometimes even properly hardened material, e.g. B. Dura-Ace or the good old three-speed pinion. But hardening can also be exaggerated: harder and firmer also means more brittle. "Dead hardened" pinions tend to lose their teeth. In the event of high loads (e.g. due to a new chain whose pitch does not match that of a sprocket that is no longer new and must first run in), the teeth can literally be levered off: They tear at the tooth base and fail, for example Dura-Ace Uniglide, see pictures 2 and 3.

Now I don't know what Dr. Blendadent currently recommends to prevent tooth decay. But beyond advertising, the hardened material must retain a certain residual toughness (flowability). This works best with heat-treatable steels - the best-known example: chrome-molybdenum steels.

The only steel chainring I know that seems to meet these requirements to some extent - the highest possible strength with sufficient residual toughness - is the middle (32) blade of the current Shimano XT crankset FC-M770. It is a forged part with subsequently milled details (tooth shape). My scratch tests with steel and carbide scribers showed that the material is really hard - but apparently not quite as hard as the "dead hardened", carious Dura-Ace pinion. So one can hope that this relatively new XT hand will do well. The price is also bearable (e.g. € 27 at Rose).

But even this supposedly good steel blade has serious disadvantages. It is only available with 32 teeth and is only emblazoned on the MTB set, not on the XT-770 trekking crank. The other two chainrings are made of aluminum. I find this acceptable with the large blade because it is less subject to wear and tear and that is where the greatest weight can be saved. But I have little understanding for the fact that the 22 series, which is by far the most heavily loaded, is not made of hardened steel or at least offered as an option. This combination, practically only useful for fast-paced flat land use, is more of a gimmick for a large part of the target group (uh, what did "mountain bike" mean again?).

I also get this impression from the edging of the chainring with short fiber reinforced plastic, advertised as "carbon" and probably injection molding: It supports the chain guide on the climbing aids. The stiffening effect, on the other hand, should not be great - and with the approximately 1.9 mm thick, briefly anchored steel blade in the "slipstream" of the large blade, it shouldn't be necessary either. The chain guide could have been made of steel with a suitable construction (shaping of the blade) without any weight disadvantage. Otherwise, it does not seem to me to deserve an environmental award in view of the presumably not non-toxic substances produced when the plastic is burned. So far, chainrings have been absolutely or at least almost completely pure, and in my opinion that should stay that way - unless the composite construction offers a real technical advantage, which I cannot see here. The same applies to the old-fashioned eighties tuning look of the tiny 32 relief holes: Looks fast, but the weight advantage is almost zero - a shame about the effort.

So my steel market conclusion is: Only a single (middle) chainring in sight that (probably) will last a long time - and apart from that, only relatively short-lived, soft cheap rings as far as the eye can see.

In the aluminum sector, the perspective looks more friendly: There is a wide range of chainrings with medium to good durability. In my experience, the former also last longer than the (admittedly "simple") steel blades that I have tried so far.

However, there is a wide range from very soft, short-lived to fairly hard, permanent materials. The extremes are rather rare. For example, the old TA chainrings with embossed lettering, which as far as I know were no longer manufactured, were extremely soft, although they were beautifully polished to a high gloss and remained so beautiful even in winter salt (corrosion resistance is an advantage of relatively pure, soft aluminum alloys), but their teeth melted away like Butter in the sun. In contrast to the current TA sheets made of a much harder material, manufacturer's name »Zicral«, the inscription is no longer embossed (probably because it is too hard for it) and lasts accordingly longer. In the absence of suitable records, I cannot say anything specific about the difference, it feels like the new TA sheets last twice as long as the old ones.

Particularly solid blades can be found in the high-end racing and MTB component groups of various brands (but here, too, there are apparently exceptions). Some of these particularly strong aluminum sheets are also hard-anodized ("hard-coated"). This is an interesting option: the hard anodized layer is only a few tens of micrometers thin, but harder than the hardest steel and very wear-resistant. However, this treatment is only useful for very solid base materials, because soft ones under the thin anodized layer would be squeezed away under the enormous pressing force of the hard chain rollers.
A typical representative of this design is the Regensburg brand Mountain Goat, whose entire chainring range is hard anodized. At least with the small 20 "Stambecco", which is widely used as a conversion sheet among weak mountain bikers like mine, I have the impression that the anodized layer contributes to its good durability (for this size).

What is actually happening there?

So far, I haven't read anything about the nature of chainring wear (but I'm grateful for references to relevant research works). So I just advise:

First there is the influencing factor friction. The chain rollers gnaw at the contact surfaces - with more or less "friction dust" or friction paste in between, consisting of metal abrasion and the, probably more disadvantageous, because harder, (also unavoidable on asphalt) absorbed mineral dirt.

The rollers rub radially when the chain meshes (loaded upper chain strand) and, with derailleur gears, laterally, at least when changing gears. The changing chain skew ensures that the rollers of the tensioned chain tilt back and forth on the teeth. The friction surface on the back of the tooth changes with every switching process and is smaller than the theoretical bearing surface. In addition, the inner links of the chain rub against the sides of the teeth and thin them out - the greater the skew, the more intense.

Mineral dirt and rust, both harder than steel, are likely to significantly promote frictional wear because they increase friction, on the one hand between the roller and the tooth, and on the other hand on the inside of the roller (which becomes stiff as a result, i.e. it tends to slip in the tooth bed) and these hard foreign matter have an abrasive effect between the friction surfaces.

The second influencing factor is the pressure exerted by the chain rollers on the back of the tooth, which I refer to as pinching. This type of wear is particularly clearly visible on (too) soft chainrings - see pictures 5 and 6: The material is literally squeezed to the side and then rubbed off by the chain links that rub against it.

At least on soft leaves, it seems to me that the pressure contributes more to wear than the friction. It may be the other way around for very hard ones, but I'm not so sure about that. That would be what a microscopic examination ...

Conclusion of this consideration: A wear-resistant chainring must be as hard as possible. This reduces frictional wear, but above all - and this property seems more important to me - the squeezing away of the material on the back of the tooth, which I see as the main cause of the rapid death of "butter leaves" made of structural steel or too soft aluminum.


My impressions of wear and tear come from mountain biking, among other things: A "good" test field, because mineral dirt combined with a high proportion of uphill sections puts a lot of pressure on the drive components. And of course from the everyday bike, which is less useful because less dirt and, above all, a much lower proportion of uphill sections.

For these test machines I can say: aluminum blades, apart from the very bad ones (the old TA butter soft described above), always lasted longer than steel blades. I have to admit that the latter were always simple models, corresponding to Shimano FC-M3xx or below. Possibly "higher-ranking" steel blades, e. B. M4xx or M5xx, better. But I doubt that they are really good. It may well be that Wolfgang Grabmer's new steel blades (FC-M443) wear out even faster than his original aluminum blades. I am looking forward to his report in the distant (bicycle) future ...

But the most meaningful test machine is and will be my good old touring bike, which after 23 years has about 150,000 km under its belt, completed with TA Cyclotouriste ("large" half-step blades, 36 and 40 teeth) and small chain rings I made myself - see the the following photos and the FC article "Making circuits fit for the mountains". The bike sees a lot of altitude, in the middle and high mountains. The latter in the annual summer vacation, usually a good 40 kilometers in altitude, with full camping luggage. The loaded machine weighs almost exactly one hundredweight, the driver two. I crank slight inclines on the 36 mm aluminum sheet. A lot comes together, but the lion's share of the altitude difference and thus the potential wear and tear has to be handled by the mountain sheet. I only need the 40er downhill and on the flat - under light load, so it lasts forever. The horizontal distance covered of only about 4,000 km per year can be neglected as a first approximation in this wear analysis. ;O)

The wear report for the two large aluminum sheets is brief: TA old "Butter soft" (see above) unsatisfactorily large, TA new "Zicral" acceptable.

The Bergblatt is more exciting: it has to manage an estimated 40 kilometers of altitude per year (with a low mountain range), and that with a much greater chain pulling force than the large blades: With the same pedaling torque, the chain force only depends on the size of the chainring: the smaller, the bigger. But the precondition just made does not apply: The drive torques are not as small as on the middle blade, because I struggle up the small inclines of up to 20 percent - with pedaling force peaks that I never reach on the middle blade, because I then use this would have downshifted long ago.

The evolution of the mountain leaf size followed the respective level of self-awareness of the modest abilities of the bio-engine: Started with 30 teeth (completely out of the question, very quickly discarded), then, used for several years, 24 - 21 - 19 - 16 teeth. The general finding: The wear of the chainring, but also of the chain itself, increased with the reduction in a super-linear manner - more than (reciprocally) square.

My first 24er was a stroke of luck: a hard-anodized aluminum pinion for a racing sprocket, French make. In view of the silly, characteristic longitudinal groove in the tip of the tooth, which tends to create "intermediate idle movements" when the gear is inaccurate because the link plates mesh in it, I guess Maillard.

While editing, I noticed that the material was unusually hard for aluminum. Filing was only possible with a very fine cut (key files), sweat and patience after I had first broken through the hard exal layer. I didn't like the brownish anodized color of the sprocket sides, so I tried to remove it. That was my first experience with hard anodizing. After vigorously sliding around on one side with the orbital sander for a good half an hour, the transparent layer was only a little thinner. It just shimmered a little brighter and just laughed at my amateurish attempted murder. So I gave up ...

From a technical point of view, that was also smarter, as I later realized: The pinion was unexpectedly durable. It took a long time until the anodized layer was through, and then a long time again until the hard base material showed serious signs of wear. I only understood how good it really was when I replaced the finally well-worn noble part with a TA "soft butter" sheet of the same size (one of the "Cross" sizes, there were 19 to 25 teeth in 5-hole, matching the Cyclotouriste crank): It was finished in no time, it only reached about a fifth of the mileage.

The next evolutionary stage, 21 teeth, also a TA butter soft cross sheet, only lasted for a short time: As expected, it wore out even faster than the 24 tooth.

It continued with 19 teeth. It was a Sachs three-speed pinion with an elaborate stainless steel spider construction attached to the large bolt circle (6 holes) of the crank. That was doing pretty well considering its size.

But gradually I was not only getting older, but also more ambitious. The fact that I had to descend and push the last 100 meters on the Wurzenpass (a 500 m low bump with a measly 18 percent) stank me tremendously. The 16 was needed. I have not regretted the choice; I've been coming up with it for 5 years. Well nearly everything - in the French Pyrenees there are common asphalt passes with ramps up to 25 percent (which clearly speaks in favor of the ultimate 13 chainring; o).

My 16s are again three-speed sprockets. Despite the mild hardening, wear is rapid: a pinion typically lasts 2 years and wears 4 to 6 good quality chains in the process. But it only lasts that “long” if I change the chains in good time, with Rohloff caliber display 0.075, and if necessary grind away the undercuts of the “shark” teeth in between. What is not always trivial on the go ...

As an example, the previous history of suffering of my current 16s, which only lasted one summer: Completely new drive, HG 90 chain, the hardest selected from my collection of 16s (Shimano, my Fichtel & Sachs are softer) and gently soldered with low-melting 55 percent silver solder (Picture 10). With it I drove a tour of the Pyrenees, a good 40 kilometers in altitude and 2,300 km. After almost 20 kilometers of altitude, the HG-90 chain was ready.The only eight-fold replacement (the cassette is still six-fold UG, light gray spacer) that I was able to find in a larger mountain village in the following days was an HG40 - shifts well, but doesn't last that long. With a good second chain, I would have just got home.

But after another 12 kilometers (a week later) I needed a replacement. Again no choice, it became a TaYa. I took it down again after assembly, because of chainsucks (wrapped around the small chainring). The division of the TaYas is obviously much narrower than with Shimano and KMC. Grinding the chainring teeth would have been a practicable solution, but I didn't have the time and the necessary leisure for this uprising (find an open car workshop with cooperative staff, remove the crank and concentrate on working with the Flex): I wanted to go to Geneva, and on the way there were a few more indispensable, beautiful low mountain ranges ...

The next only eight-speed chain I could find, this time in the two big bike shops in a larger city, was a cheap SRAM PC-830 with uncranked plates (shifts poorly). The division of the SRAM chains is between KMC / Shimano and TaYa. Again there was chainsuck with serious stress on the small leaf, but much less pronounced than with the TaYa. So I went for retraction on the big leaves. Fortunately, this was possible because I had a day of flat land in front of me - unfortunately with a few tough bumps on which I had to struggle with the middle blade. In the evening it really went uphill - and lo and behold, the chain had finally broken in and was accepted by the three-week-old Bergblatt.

Why am I writing all this from Extremistan? Because I want to show that there are people who have much bigger problems with wear and tear than Wolfgang Grabmer. And that's why they're at least as interested as he is in chainrings that are as wear-resistant as possible. My 16-tooth three-speed pinions, only mildly hardened and probably tempered by soldering (partial softening) are just a hobbyist's solution with great potential for optimization. I'm interested in how a hard-anodized 16 mm aluminum sheet in Mountain Goat quality would perform. Or even an optimized specimen made from tempered steel ...

How most practical?

In terms of knowledge, there is actually nothing better than a practical comparison test. But that would be very difficult in this case because the chainring wear depends on many parameters: chainring size, topography, individual driving style (speed, effort, switching points), wear condition of the chain and the remaining drive components, dirt and moisture (depending on the year and season) , ...

The only halfway sensible test arrangement came to mind: Take a triple crankset, remove the large blade, and replace it with one of the same size and type of the middle one (both new, of course). In order to be able to shift this without any problems, either a (possibly modified) racing front derailleur must be used and / or a chain guard ring that prevents the chain from being thrown off the outer blade. With it you drive the usual routes, every day on the other sheet of paper. To compensate for the different chain lines, the two test sheets must be at regular intervals, e.g. B. be exchanged every beginning of the month.

This procedure can take years to come to usable wear and tear statements. To speed things up, hang a child trailer filled with well-fed, not-too-small passengers (which would at least achieve some degree of compatibility with the title topic of the current issue; o) on your everyday bike, with which you can barely manage the daily inclines.

Not very attractive. That wouldn't be my thing. So we come to ...

Laboratory test

The racing bike magazine "Tour" occasionally carries out comparative tests of racing bike groups from different manufacturers. For this purpose, among other things, the Vickers hardness of the chainrings is checked in order to be able to make statements about the wear resistance that correspond to practical experience.

A practicable procedure, which however has the small disadvantage here that only racing blades are checked on the »Tour« where, due to the conditions of use (large blades, correspondingly low chain forces, smaller overall weights, little dirt, predominantly good weather operation with mostly above-average care and maintenance) chain and chainring wear is not a world-shattering issue. Even less good racing chainrings usually last tens of thousands of kilometers. Most of the common everyday, touring and mountain biking drivers, however, have little of that because they cannot pedal such giant wheels.

So better check yourself. I am planning a series of hardness tests for publication of the results in the next issue. The chainrings available here - see table - come into question as test items. If you have other unnecessary sheets, you can send them to me at the editorial address to expand the test series. For this, old, worn chainrings from the junk bin are ideal, which you don't necessarily want to see again. ATTENTION: I will only send in test items exceptionally send back - and only if that is specific agreed has been.
Please send inquiries directly to me: [email protected]

Fichtel & Sachshardened steelPinion for hub gears
Mountain Goat StambeccoHard anodized aluminum20 teeth
Shimano 400 LX, FC-M400steel28 teeth Biopace, approx. 1991
Shimano FC-M3xxsteelM330, quality level Acera
Shimano LX FC-M582 MTBAluLX Hollowtech, 2008
Shimano Nexushardened steelPinion for hub gears
Shimano XT FC-M770 MTBHardened steel / aluminumcurrent XT
StronglightAlu44 teeth, LK 110, original packaging, without type designation
Suntour Winnerhardened steelFreewheel sprocket, fits on "Limbo Spider"
TA Cyclotouriste oldAlu, softEmbossed lettering (see article text)
TA Cyclotouriste newAluZicral, white lettering, eroded / lasered
Vuelta USAAlu34 and 46, LK 110

This collection lacks z. B. still Shimano LX steel pinions and XTR. Please do not send me unsorted noname sprockets. The types should be named as precisely as possible. With the Shimano initial assembly with the component number of the crank (to be found on the inside, mostly stamped in the crank eye). I will probably not get around to checking everything, but will limit myself to a reasonable selection.

Of course, I have a pretty clear idea of ​​who will be the test winner and which chainrings will end up in the front and in the back, but I'm not going to reveal that yet. ;O)

To the author

Rainer Mai is a bicycle expert in Frankfurt am Main, mechanical engineer, everyday and touring cyclist, co-founder and supervisor of a self-help workshop, co-initiator of the "AG Verflixtes Schutzblech".

Legal Notice

All information on this page is given to the best of our knowledge, but without guarantee. The authors and the association Bicycle future decline any liability for direct or indirect damage caused by following or not following the advice given on this page.

All photos, graphics or tables on this page come from the authors. Exceptions are marked.