Light'ning my bike

After first noticing it in the Pez coverage of Interbike 2007, I've really been attracted by the Lightning Crankset. Mass competitive with the 580 gram Zipp Vuma Quad but with conventional 5-bolt chainrings.

Chainring compatibility is a big deal: I really like the flexibility of 110 mm BCD "compact" chainrings, but Vuma only supports the 34-50 combo: a 47% jump which when combined with the 103% range of an 11-23 rear results in a relatively small big ring-little ring overlap which can require a lot of front shifts to avoid running up against the cross-chained gears. Except for extremely steep climbs, 36-50 is the better choice, with a reduced 39% jump in the front, improving shifting performance as well as reducing the cross-chaining problem.

Additionally, the Lightning has a replaceable spider. This adds flexibility: it can be converted to 130 mm or 96 mm BCD cheaply and easily. But additionally it promises future compatibility with the Quarq power meter.

A key attraction of these cranks was, despite their newness, that the core design is already proven. Tim, the real brains behind Lightning Cycle Dynamics (best known for their excellent recumbents) patented his two-piece "Hirth" crank design in 2002. The design was then adopted by Specialized for their S-Works BB30 crankset. The S-Works crankset has been criticized for its chainrings, but these were improved last year, and this year are being used by the Saxo Bank team.

There's an ongoing debate about which is the better choice for cranks: metal or carbon? The data are frankly mixed, and I don't think there's a clear answer. For example, Fairwheel Bikes did an excellent review of lightweight cranks, comparing stiffness and weight as well as qualitative factors. I analyzed some data from that test in the following plot (warning: I think there's a scale factor error on the y-axis, and my gnuplot skills are too limited to go back and fix it, but that's unimportant):

It's easy to see that the #1 factor affecting stiffness/mass of a crank is the spindle diameter. None of these tested cranks are BB30, requiring the wider diameter bottom bracket shells standardized by Cannondale and later embraced by Specialized and others. Instead all are for BB26, the prevailing standard. A key feature of the cranks with the best stiffness/mass ratio is that they conspire to fit the widest possible diameter spindle into the BB26 bottom bracket shell.

For the cranks with 30mm spindles (labeled on the x-axis as 900 mm²) the best stiffness/mass ratio is exhibited by an EE prototype which doesn't have a shred of carbon fiber in it. The EE crank, however, has an advantage of a bit more beef than the others with 30mm spindles, checking in at 669 grams versus the lighter Vuma, Storck, and THM cranks, and at some point when you push the limits of low mass you start to affect the stiffness/mass ratio as well as just stiffness. Considering other spindle diameters as well, neither metal nor carbon fiber demonstrates a clear advantage in anything except cost, where metal wins.

None of these 30mm-spindle cranks are noodles, though, and I honestly doubt that crank deflection on any of them has any significant affect on drivetrain efficiency for someone with my power and mass (this is potentially a long discussion). I admit to preferring metal on components which are the only thing separating my body from surgery. Call me silly that way. But I think that despite their failure to demonstrate a tangible advantage versus metal, carbon cranks have sufficiently proven themselves over the years. Lightning also uses a 30mm spindle, and it will be interesting to see where it ranks when Fairwheel publishes the next round of their test data, including the Lightning.

Unfortunately Tim's success in prototyping was tempered by frustration in going to manufacturing. Month after month ticked by, the crankset perpetually promised for "next month". Eventually it was time for Interbike 2008, a year later. Yet still it was unavailable. But clearly progress had been made. I hadn't given up, despite the warnings of many skeptics.

Tim finally came through, and this month, a year after I'd first hoped, I got one thanks to the help of a friend from the Low-Key Hillclimbs. The first thing which happens with any new component is it goes on the scale! The numbers here:

component	mass
drive-side assembly	215.7 g
non-drive-side assembly	147.6 g
bearings	89.2 g
SRAM 36 ring	29.1 g
SRAM 50 ring	89.5
Shimano chainring bolts	9.7 g
total	580.8 g

So there it is. Right at the claimed mass for the Vuma Quad. I could go lighter by switching to Extralite chainrings, verified at 106 grams for rings + bolts (but available only in 34-50), compared to 128.5 grams for the SRAM rings with Shimano bolts. I'd only consider switching the big ring, as I want the 36 inner.

Installation was easy and the instructions were very clear. Here's the result:

installed crank. Nice! Please ignore the hand.

So there it is. I look forward to giving them a spin when I finish gluing my new wheels. Of course, before I ride it, my reassembled bike goes on my Ultimate digital scale. I know I'll like what I see there. That's the important thing, isn't it? The actual riding of the thing seems almost irrelevant. After a protracted post-travel illness, I've got to get my energy back first, in any case. So for now equipment has to satisfy my cycling fix.