The argument against top-tier components

Shimano’s Dura-Ace 7400-series “25th Anniversary” groupset, on display at Shimano Cycling World in Singapore. Note the headset and seatpost at the far right, as well as the Seiko wristwatch.

Shimano Dura-Ace. SRAM Red. Campagnolo Super Record. All of these three groupsets are the pinnacle of what each of the three big bicycle component manufacturers offer, and can confidently be considered the state of the art.

And yet, today I’m going to tell you a few reasons why you may not want them on your bike.

“THAT COSTS HOW MUCH?!” – HIGH ACQUISITION COST

As with many things, becoming a technological showcase is going to cost you a pretty big coin. As simple as a bicycle may look, a top-tier groupset mainly exists to present to the world just how much it can improve in terms of performance, many of them frequently being refinements or marginal gains over the previous generation, as was the case with Dura-Ace 7900 over 7800.

This is why I believe top-tier components are really meant for sponsored professional athletes, who will appreciate these marginal gains more – not amateur enthusiasts like many of you reading this decidedly amateur publication.

I feel this is most onerous with Campagnolo Super Record. Campagnolo already isn’t cheap to begin with, at any level; for some people, the exorbitant purchase price of Super Record over even Shimano Dura-Ace or SRAM Red just adds insult to injury.

“THAT COSTS HOW MUCH?!!” – HIGH REPLACEMENT COSTS

Much like used supercars and luxury sedans may appear like bargains up front, the high costs of a top-tier groupset do not stop with the initial purchase. You still have the maintenance aspect to look at. As the parts wear away with time and usage, when the time comes that a chain, cassette, or chainring requires replacement, the sticker price may be a source of unwelcome financial “aftershocks.”

Why is this the case? Like it or not, part of the technological advancement brought on by a top-tier groupset is down to the use of unconventional materials, such as carbon fiber or titanium. One example is the titanium cogs on a typical Dura-Ace cassette. These materials are chosen mainly to save weight, longevity be damned. When hub manufacturer Chris King moves away from titanium as a material for its R45 rear hubs’ ratchet mechanisms, one should already suspect that something is up with the material. Titanium is great for making bicycle frames, but plays a distant second or third fiddle to good ol’ steel for wear items.

Exploded view of a SRAM Red 10-speed PowerDome cassette. Photo credit: Glory Cycles/SRAM

Apart from unconventional materials, sometimes it is the manufacturing methods themselves that are unconventional and lead to the greater expense. SRAM is the poster child for this, its trademark PowerDome cassettes CNC-machined from a single, solid steel billet. This feat of engineering is expensive to replicate, and so replacing a worn PowerDome cassette like-for-like is also going to mean at least a US$279 hit to your wallet.

QUESTIONABLE CONSTRUCTION METHODS…?

Somewhere along the way of searching for marginal gains and shaving every last gram possible from a component, bad news starts to happen. Worryingly, this isn’t restricted to just top-tier stuff – sometimes even second-tier components have problems, too.

A cut-away Ultegra 6800 crank arm, showing off the construction that makes it hollow. Photo credit: Shimano Cycling World/Hands On Bike.

A few years ago, Shimano started to hype the lightweight construction of its Dura-Ace 9000 and Ultegra 6800 crank arms. These are made hollow by forging them in two pieces, which are then bonded together. This is in contrast to the one-piece solid forging technology used on Shimano’s cheaper or lower-tier cranksets.

For comparison, the non-series Shimano FC-R565 crankset uses heavier solid-forged arms, hollowed out from behind for some weight savings.

While this newfangled construction should be all well and good from a stiffness standpoint, numerous reports of Ultegra 6800 crank arm failure have also come in. And they’re not pretty.

Photo credit: Carlin the Cyclist.

As you can see above, this particular Ultegra FC-6800 crankset cracked just above the pedal thread area, then totally came apart precisely where the two crank arm halves were bonded together. John Carlin’s experience isn’t the only one, either. This has been a worryingly common issue with Ultegra 6800 cranks – one that I hope Shimano have addressed with its R8000 successor. While the problem isn’t as widespread with the contemporary Dura-Ace FC-9000 cranksets, it has also happened to a few riders using these top-tier cranks.

A little worrying for me is that this “Hollowtech Crank Arm” technology has trickled its way down to third-tier Shimano 105 level. That said, trickle-down technology usually means it’s been proven at higher tiers, and that same technology is made more reliable and more inexpensively in order to reach a larger audience, so we’ll just have to bet that the manufacturers have indeed done their homework. At the very least, I haven’t yet heard of any 105 FC-5800 hollow crank arms separating from each other…

CONCLUSION

At the end of the day, people are going to buy what they want. I have no illusions of making people change their minds, nor was my intention to scare people off. However, my only request is to know what you’re getting into, and not be immediately dazzled by the new tech on offer. Compare and contrast, and study the pros and cons. When it’s your money on the line, you might as well make it so that you maximize what you get.

Advertisements

An angry swarm of bees vs. the sound of Scylence

Here in the Philippines, many cyclists will put a bike on a stand, shift to a hard gear, and hand-pedal the cranks to a high cadence, then suddenly let go…all to listen to, and savor, the sound of a freewheeling freehub. The louder the freehub, the better.

They even have a name for it: “tunog-mayaman,” roughly translating to “it sounds [like a] rich [person].”

The first time I heard this expression, I let out a mild chuckle. Because I’m a curious son of a gun, though, I did a bit of research as to why people say so, and why they prize the sound of a loud freehub.

For the uninitiated, a freehub is two things. Its exterior allows the mounting of cassette sprockets (usually via splines and a lock ring), while its interior is essentially a one-way ratchet mechanism.

A freehub body (left) removed from its hub shell, which contains the ratchet gear teeth (right). Note the pawls on the freehub body. Photo credit: MBR.co.uk.

While there are other variants, the most common freehub internal design sits on the rear axle on bearings, and has a number of spring-loaded fingers, called “pawls,” that will catch on ratchet gear teeth on the inside of the hub shell. When they catch and engage, your pedaling motion at the cranks gets transferred from the chain and sprockets into the hub and rear wheel. When you stop pedaling, however, the pawls will slide and skip over the ratchet gear teeth, allowing you to coast.

This high-frequency skipping of the pawls over the ratchet gear teeth is what we hear as the sound of the freehub. Conventional wisdom states that more expensive freehubs use a number of tricks — either using stronger strings on the pawls, installing more pawls in the freehub, or something else entirely — all in order to improve engagement and drive of the mechanism. The sound is really just a byproduct.

Indeed, many premium hubs do have a noise to them, and this noise is a quality in and of itself. Chris King’s R45 rear hubs, for example, are prized for the “angry swarm of bees” sound they make, but as the video above suggests, they are handily beaten by other competitors.

That hasn’t deterred many locals from asking bike groups or forums about how to make their freehubs sound louder, i.e. how to make them tunog-mayaman. Most of these modifications involve replacing the pawl springs for something with higher spring rate, although this is a questionable practice at best.

Leave it to Shimano to throw a wrench in the works, though…

Photo credit: BikeRumor.

At the end of May 2018, Shimano announced the M9100 generation of XTR, its top-tier mountain bike groupset. While I didn’t really care about the divisive new Micro Spline freehub body superseding the tried-and-true Hyperglide design that has served riders for 30 years, I was very curious about what lay inside the polarizing freehub. Instead of the classic pawl and ratchet system, Shimano used a new drive mechanism they call “Scylence.”

Exploded view of the Scylence hub drive mechanism. Photo credit: Shimano/BikeRumor.

Ridiculous spelling aside, it’s obvious what they were going for with these hubs: silent running. Instead of pawls catching on a ratchet, Scylence transfers drive via angled slots on two sides of the hub that mesh, which pull a pair of ratchet rings (yellow and green above) together. This is somewhat similar to the helical-cut gears meshing in a car’s gearbox, plus a clutch disc mating with an engine flywheel, but implemented in another way. When coasting, springs pull back on the ratchet rings and disengage the angled slots, resulting in zero friction and noise.

As XTR is Shimano’s flagship MTB groupset, the notion that these very expensive hubs are designed to be silent is hilarious in its irony. It flies in the face of the whole “tunog-mayaman” philosophy – and the one major thing that I wish ended up included into future versions of Shimano’s road bike groupsets.

LitePro’s hubs make a moderately noisy freewheeling sound.

Needless to say, I don’t really buy into the whole tunog-mayaman phenomenon. For me, it doesn’t really promise or mean anything other than the fact that your freehubs are loud. Proof? My folding bike Bino, which has gearing that tops out at a 37 km/h cruise in top gear, has a louder freehub than my fast bike Hyro

Much ado about 12-speed

A long time ago, I wrote about my dad’s Peugeot P8 road bike, manufactured circa 1981. While that was a “ten-speed” bike, it was down to having a five-cog freewheel at the back and two chainrings up front: a so-called 2×5 (two-by-five) drivetrain.

My father’s Peugeot P8 road bike, circa 1981.

Almost forty years later, we say a 10-speed drivetrain refers to ten cogs at the back…and that’s not even the state of the bicycle drivetrain art. We’re now at 12-speed drivetrains, starting on mountain bikes with SRAM’s XX1 and X01 Eagle in March 2016, and Campagnolo on the road bike side with its 2018 Record and Super Record groupsets. While retaining 11-speed on the road, Shimano has followed suit with XTR M9100 being offered with either 11 or 12 cogs on the back wheel.

SRAM’s XX1 Eagle 1×12-speed drivetrain for mountain bikes, revolving around a 10-50T cassette the size of a dinner plate. Photo credit: BikeRumor.

Why this march toward ever more cogs on the cassette though? I take a look at the arguments for and against this proliferation.

ROAD BIKES: NOT JUST FOR RACING

With road cycling and racing, the historical focus was primarily on keeping a tight spread of gears on the cassette. This is to help keep a rider’s pedaling cadence as constant as possible, with the smallest possible jumps between the next easier and harder cogs to minimize “shift shock.” Not long ago, it wasn’t unusual to have 11-23T and 11-25T cassettes, which have a very narrow gearing range (the difference in size or number of teeth between easiest and hardest cogs), but are ideal for criterium races in the flats. Heaven help you if you had to climb a long steep hill though…while you have a second smaller chainring to help, if that gearing isn’t low enough, it’s either you grit your teeth and grind your legs pedaling away, or you get off and walk.

Shotgun in San Mateo, Rizal province. Not the kind of terrain you want to climb if your bike’s cassette tops out at a 25T cog.

Lately, however, people have woken up to the realization that not all road cyclists are into racing or grinding away uphill at a low cadence. Some of them just want to be able to keep riding their bikes, staying in the saddle, and spinning their way up…slowly but steadily. I would say it’s at the 10-speed level that gearing becomes an acceptable compromise between wide range and cadence control.

Both my bikes have Shimano Tiagra 12-30T 10-speed cassettes.

Looking at Shimano’s Tiagra 4600 and 4700 road bike groupsets is a good way of seeing how things have changed. Both are 10-speed, but Tiagra 4600 topped out with a 12-30T cassette option. Tiagra 4700, by contrast, offers 11-32T and 11-34T cassettes. Just five years ago, a cassette with a 34T cog was mountain bike material.

With its professional-grade Dura-Ace R9100 11-speed groupset, launched in 2017, Shimano even offers an 11-30T cassette option too. Sacrilege?

RELIEF FROM MENTAL OVERLOAD?

At the launch of Cannondale’s revamped 2018 F-Si, its premier cross-country hardtail mountain bike. Photo credit: Cannondale/BikeRumor.

Mountain bikers aren’t as concerned with cadence control as road cyclists are. While smoothness is still key, it seems applied in a different way compared to the souplesse roadies swear by – more to maximize grip at the tire/ground interface, and less for smooth pedaling. They’re more concerned about outright gear range, instead…and the simple reality is not all combinations of chainring and cog are distinct, nor usable (due to cross-chaining). On a 2×11 groupset, for example, one usually has only 14 or so distinct gears – the rest are duplicates.

The cockpit of a Rocky Mountain Instinct 999 MSL from 2014, featuring extra switches for the Rock Shox Reverb dropper post and the remote CTD adjustment for the rear suspension. Photo credit: Singletracks.com.

The way technology has exploded on mountain bikes in the past six years, there are so many controls and levers festooned on handlebars to think about: shifters, brakes, dropper posts, and remote suspension lockouts. Bombing down a trail at high speed, it can be mental overload to simultaneously read the trail’s features, plan your approach accordingly, shave off speed, work your shifters…you get what I mean.

SRAM’s effort to kill off front shifting, with its “1x” philosophy and the release of the original XX1 groupset in 2012, was a bid to simplify things, I think. You end up performing all gear changes with just one lever. It also freed mountain bike frame designers by reducing the variables contributing to “chain growth,” where compression of the rear suspension changes the distance between bottom bracket and rear axle. I don’t know all the details, but it seems to have an effect on the handling of the bike.

A look at the alternating narrow-wide tooth profiles of an XX1 chainring. Photo credit: BikeRumor.

To ensure that such 1x drivetrains did not lose the gearing range of an equivalent 2x system, SRAM made its cassettes with huge range – 10-42T on the original 1×11 stuff, then growing to 10-50T on 1×12 Eagle. Having 11 or 12 cogs with no front shifting comes pretty close to the 14 or so distinct cogs of a typical 2×10 or 2×11 drivetrain.

To combat chain drop, it adopted chainrings with tall teeth and narrow-wide tooth profiling, as well as a one-way friction clutch mechanism on the rear derailleur, which itself is restricted to horizontal operation only. These same tricks to improve chain retention make 1x drivetrains ideal for the short but frantic action of cyclocross racing, where the longest a race lasts is an hour.

While we’re on the subject of off-road things, the growing adventure/gravel bike category is sort of a middle ground of the demands of road bikes and mountain bikes. A single-chainring drivetrain is simpler and guards against chain drop, but many gravel riders insist on a double crankset for the increased gear range and options, especially for much longer events.

COMPLEXITY AND INCOMPATIBILITY

Shimano’s Hyperglide freehub body. In its original form, it doesn’t really support more than 10 cogs.

The jump toward more cogs introduces its own problems and growing pains. Honestly, this is a bit of a mixed bag and a minefield all on its own.

First things first: Campagnolo’s 12-speed cassettes will work on wheels and freehub bodies splined for Campagnolo 11-speed. End of story there.

Shimano’s classic Hyperglide freehub body, with its 13 splines, has been around for 30 years. Part of its universal appeal is that it is an “open” design: most hub makers worth their salt use it. Heck, even their competitor SRAM does. Unfortunately, it’s got limitations.

A DT Swiss wheel with an 11-speed road freehub. It is 1.85 mm longer than the original Shimano Hyperglide design, but is otherwise identical. Photo credit: road.cc.

As drivetrains have gained more and more cogs in their cassettes, the original design no longer accepts more than 10 cogs, at least on road bikes. For 11-speed road bike gruppos, the freehub body had to be extended by 1.85 mm to accommodate the 11th cog; these hubs can still be used with 8/9/10-speed cassettes with an appropriate 1.85 mm spacer.

On the mountain bike side, Shimano did something clever: it dished the largest cog on its 11-speed cassettes so that it could still fit on a 10-speed freehub body and sit on top of the spokes. It also offers a pair of 11-34T 11-speed cassettes, designated “CS-HG700-11” and “CS-HG800-11”, which are meant to work with road bikes with 10-speed freehub bodies.

A 3D model of a rear hub with SRAM’s XD Driver freehub body fitted. The hybrid design has short splines, but also has screw threads, like rear hubs of old. Photo credit: BikeRumor.

The second limitation of the Hyperglide freehub body is it will not accept a cog smaller than 11T. When SRAM came out with XX1 and its 10-42T cassette, it introduced its XD Driver, a different freehub body with a slimmer end, which will fit even a 9T cog. The XD Driver and its XD-R road bike counterpart are also open standards, so many companies offer an XD/XD-R replacement freehub unit for their rear hubs.

To fit a 10T cog, Shimano had to come up with its new Micro Spline freehub body. Photo credit: BikeRumor.

With XTR M9100, Shimano faced the same dilemma, and thus did away with the Hyperglide freehub. They replaced it with the Micro Spline design, which has 23 smaller splines and is actually shorter.

There are two unfortunate consequences with Micro Spline: there is no backwards compatibility with cassettes made for Hyperglide freehubs; and Micro Spline is, as of this writing, a “closed” design. Shimano has chosen to withhold use of the design, leaving only itself and development partner DT Swiss as beneficiaries. This means nobody else can use it, although if you have DT Swiss hubs, you could upgrade them to Micro Spline if you want.

HOW STRONG IS YOUR CHAIN?

Another issue with ever increasing cog counts is that the chain has had to get narrower and narrower to match. This started with 9-speed, where manufacturers simply crammed one more cog in the space that used to be allotted for eight, and has continued to this day.

Another issue specific to 1x drivetrains is the chain line. We’re avoiding cross-chaining in the traditional sense of using big chainring and biggest cog or small chainring and smallest cog, but even with a single chainring up front, the sheer number of cogs out back means the chain can assume some pretty extreme angles as you shift and pedal.

While Shimano, SRAM, Campagnolo, and third-party chain makers like KMC and Taya have stepped up their game in quality, some people still think that making chains so narrow should have some disadvantage in terms of longevity. All I can say here is: your mileage may vary.

SO…WHAT DO I THINK?

Campagnolo’s 2018 12-speed Super Record cassette, rear derailleur, disc brake, and chain. Photo credit: BikeRumor.

It baffles me that Campagnolo ignored the no-compromises approach that 12-speed brings. With that many cogs, a rider should no longer have to choose between close gear spacing and wide range. However, their largest cassette option is 11-32T. Not making a wider range cassette is a huge missed opportunity, I think – one that would have endeared them to gravel cyclists. Maybe when the tech trickles down to Chorus or Potenza level…

What does XTR M9100 mean for Shimano’s road bike transmissions? Right now, I wouldn’t be too worried. While the design has other benefits, the whole point of Micro Spline is fitment of a 10T cog, which I doubt anybody would use except for die-hard 1x fans. With Dura-Ace R9100, Ultegra R8000, and 105 R7000 still relatively fresh, I feel we may not need Micro Spline on the road just yet.

For my riding, I’m not concerned about having the most cogs. There is merit to having more, but more important to me is the gearing range spread across those cogs. While I haven’t had to use my 30T largest cog very often, I’m pretty sure my heavy carcass would welcome any larger cog available if I had to climb stuff like Shotgun or Pico de Loro.