Nobuo and Yutaka Kondo with me as I take delivery of a set of Gokiso wide carbon clinchers with wine red hubs! |
The “makers” are many and diverse. Of course, there are globally known giants such as Toyota and Panasonic. But for each global giant, there are thousands of smaller firms. Indeed, many of the most successful Japanese firms today are not global giants, but smaller companies that have no significant brand or consumer presence, primarily “B-to-B” businesses with relentless focus on a core competency. They make components that go into other company’s products: the Nidec motors that go into almost every hard disk drive (HDD) in every PC or server in the world; the various parts of electronics or autos that--even today as Korea, Taiwan and China move up the food chain and U.S. manufacturing experiences a revival--cannot be made anywhere except Japan, or cannot be made anywhere as well as in Japan. And then there are hundreds or thousands of subcontractors who work with each of the global giants.
Nagoya-based Kondo Machinery Corporation is one such a “maker”.
Kondo Machinery Corporation's production facility |
The office, just across a busy street from the factory |
A skilled machinist opens a high tech cutting tool to reveal a shaft -- core component of a bicycle hub. |
HDD motor (left) and central spinning part. Kondo Machinery makes the machine tool that makes these parts. |
The three pieces on the top in this photo combine to make the piece on the bottom, which spins, really fast and smooth. |
So basically Kondo Machinery’s focus is on the most demanding, high speed large and tiny metal spinning part applications in the world. Everything must be "manufactured as designed". Only the tightest tolerances. (0.1 micron; 0.5 micron). “High precision” does not quite describe it. In fact, “ultra high precision” would be a better description.
Nobuo Kondo, President, and younger brother Yutaka Kondo, Senior Managing Director in charge of product development, are Japanese "otaku" engineers – technical geeks who are fanatical about the problems they take on, who like to experiment and who are, admittedly, not the most commercially minded. They want everything to be the best, and want to find solutions to problems others may not have even realized exist.
So what would happen if a company like Kondo Machinery designed and built road bike hubs and wheels?
For a cyclist whose company happens to make ultra high-precision spinning metal parts, the challenge and opportunity is obvious and apparent, even if your company has no other "B-to-C" businesses or brand presence. Is it possible to make better hubs, to make the best, smoothest rolling, fastest bicycle hubs in the world? And once you have done the work to produce these hubs, you have the testing equipment setup and have analyzed the data, what do you do when you confirm that a hub is only as good as the wheel it is a part of? Of course, you set out to see if you can make the best wheels as well.
Wheels in the production facility |
It is a challenge to get the color "just right" and consistent among different hubs. Gokiso standard hubs are now offered in black and wine red ... not these greenish grey colors. |
And the Gokiso approach – an ultra precision, aerospace quality component – was the opposite of what I had just been taught in the frame-building class I took at UBI in Portland, February 2012. One instructor there (a successful, prize-winning welder and framebuilder) would tell us not to worry too much about precision. “After all, we’re not sending these things to the moon!,” he would say. No need for aerospace precision with a bicycle frame. Use your “builder’s eye” to get things “just about right”.
And if you mess up a bit – say you use the 135mm MTB blank instead of the 130mm road one to hold your rear triangle in place as you weld the stays to the seat tube and to each other in your jig, you can “cold work” the metal later to adjust the spacing.
“Cold work” is just a polite word for application of brute force. It sounds better to the customer buying your frame if you talk about “cold worked metal" than if you say you needed to bend the damn thing with all your strength, to take a mallet to it even, until you got it right! A 5 millimeter error in the spacing of the rear stays. That would be 5000 microns. And maybe it can be reduced to 500 or 750 microns (0.5 or 0.75mm) by application of brute force. But there is a long, long way to go from 500 microns to get to 0.1 micron tolerance!
And then, there was the strong yen. In 2011~2013, just about any bicycle product made in Japan was priced out of range, even for those of us living here. This was especially so for a high-end product like Gokiso hubs. Gokiso displayed their wares at NAHBS in 2013, and the reaction, from what I can read online, was a mix of envy at the product and true puzzlement at the high cost, so much more than other hubs. Indeed, the exchange rate in 2012 was 78 yen to US$1, and the standard Gokiso hubs front/rear set at that point carried a list price of US$3800. Ouch.
At the time, I was working to develop a business, investing in projects that might or might not pay off, and was definitely putting off any luxury purchases. As time went on, I forgot about Gokiso.
One-of-a-kind set of Gokiso "climber" hub shells in Q36.5 green -- intended for David Marx/RGT |
The titanium shell of a "super climber" Gokiso hub. Each one is hand polished by Mr. Kondo! |
Then in October 2014, I happened to turn on the NHK BS-1 television channel on a weekend evening. NHK, the national broadcaster, had a one-hour special on the most famous of Japanese cycling “hill climb” amateur competitions, the annual event at Norikura and Makoto Morimoto, a previous 4-time champion.
You can watch the (Japanese language) documentary online here. I highly recommend it. (Japanese language only, but you can just watch the hill climbing sequences and a bit of the Gokiso team back in the factory to get the picture.)
Norikura is a spectacular climb on a road bike, cresting at 2700 meters elevation, essentially as high as iconic European climbs such as the Passo della Stelvio in the South Tyrol or the Col du Galibier in the French Alps.
Would Morimoto-san be able to regain his prior shape and compete effectively at Norikura? And could he do it riding the heavier Gokiso wheels? Gokiso hubs are designed not for lightness, but to be the smoothest rolling hubs imaginable. As Nobuo Kondo explains during the program, they have done exhaustive tests, and believe that their hubs roll smoother than others, and by absorbing shock in the hub shell without it affecting the shaft/axle, achieve a better result for the rider than that of a lighter hub, even when climbing.
Likewise, the Gokiso team exhaustively tested different types of rims, spokes and tires, eventually selecting carbon clincher rims of various rim heights and widths, with a lot of material in the area under the brake tracks. Contrary to what I had thought, they found that wide carbon clinchers have much lower rolling resistance than tubulars, likely because of the tire shape formed by the clincher rim. The Gokiso rims are heavier than lightweight climbing rims or standard aluminum clinchers, but they are ultra precise, “built as designed” for Gokiso. They roll beautifully and hold the line incredibly through turns. Again, Kondo-san says the test data supports this approach over one that emphasizes light weight.
Kondo Machinery has tested Gokiso wheels at 300 kph. No, I do not expect to ever ride more than one-quarter that fast, even on a straight alpine descent! But if the wheel does not develop stresses, imbalances or start shuddering at 200 or 300 kph, then it will not do so for me at 75 kph on my fastest descent. If the hub is tested for 100,000 kms, then it should work well for most riders pretty much their entire lifetimes.
Wheel testing setup at the Kondo Machinery factory. |
Here you can find a Japanese language (Cyclesports Magazine) online interview with Morimoto featuring the equipment that he used. Note that he used a Gokiso bottom bracket as well as Gokiso wheels -- another spinning part with ball bearings in it!
Instead of $3800 just for a set of hubs as back in early 2013, the list price for a pair of carbon clincher wheels with 38mm high, wide (23mm) carbon clincher rims, CX-Ray black spokes and current generation “standard” Gokiso hubs, is 386,000 yen, or around US$3200.
-----------------------
I found myself in Nagoya in early December to pick a set of Gokiso wheels. Thanks to the introduction of David Marx of RGT Enterprises, who is helping secure some foreign distribution arrangements for Gokiso, I got to spend a good part of the day at the company, getting official maintenance training, a factory tour, and a chance to meet much of the Kondo Machinery Corporation family, even a lunch with the Kondo brothers, Tsuji-san (Factory Manager), Morimoto-san and one of their specially certified machinists.
I get training on hub assembly and maintenance from the expert, Ms. A. The maintenance is actually quite easy to do after a short course. |
Re-greasing the rear hub's freewheel assembly. |
The Gokiso truing stand. Simplicity and precision. No dishing tool required. Just flip the wheel. |
Delicious lunch near the plant. No plastic food here -- instead a rusted metal sculpture featuring the meal's main ingredient. |
The hubs use spherical washers at each end next to the quick release, to eliminate any mis-formation that might otherwise occur when the quick release and fork dropout presses against the hub from each side.
And each hub uses double sets of P5 ceramic bearings on each side. Of course, the best bearings are a key to why these hubs are so smooth.
Gokiso’s own videos can explain better than I.
Hub video Part 1 (English):
Hub video Part
2 (English):
Rear hub
structure video (English):
http://youtu.be/zpSlzZROGWI
Testing video -- 100,000 kms at 100 kph; 300 kph test -- other parts of bicycle broke at 210 kph, so the test is rerun after other components are reinforced:
(Japanese speakers -- You can find similar videos with Japanese narration by searching at Youtube.)
Testing video -- 100,000 kms at 100 kph; 300 kph test -- other parts of bicycle broke at 210 kph, so the test is rerun after other components are reinforced:
Front and rear hub internals, and the shell 0.5mm off of the internal components |
Front internals |
Rear drive side/cassette area internals |
Another rear hub internal -- the original model |
Another front hub internal -- the original model |
They make a set of track hubs as well! |
Riding the Wheels -- First 800 km Impressions
I left Nagoya a very happy
camper. The next morning, I took an early winter cycling trip to the Miura Peninsula, revelling in the ease with which I could accelerate and hold a decent speed on the fast stretches.
In total, I have ridden around 800 kms on my Gokiso wheels over the past month. Five rides of 100 kms or longer: Miura first ride, Winter Solstice ride -- Kobu Tunnel, SFC "commute" during the Festive 500, Hakone to Tokyo via Ashigara/Yabitsu, and the Nishi Tokyo Brevet.
The only time I have NOT used them during this period is for commuting/in town, and one long ride where I needed a front dynamo hub/light.
How do I like the wheels? They are great.
I am noticeably faster when I ride them, and I seem to work less hard. They roll and roll and roll, and they are stable and strong. After 800 kms, they are as true as the day I got them. And if anything they roll smoother.
As regular readers will know, I have set lots of Strava "Personal Records" on these rides, even in mid-winter and even when I have not worked any harder than usual.
Of course, they are carbon clinchers, so I am using the brake pads Gokiso recommends (and supplies). The braking surfaces are perfect. I had some squealing when braking on one ride after the brake pads wore a bit, but that I fixed the issue by “shoeing in” the pads again so the front edge is closer to the rims than the rear. Issue resolved.
On Saturday, January 10, I rode my first Brevet of 2015, a Nishi Tokyo 200 km ride in a big loop from Machida around the Miura Peninsula. I did not work particularly hard, except the first 20 kms where I wanted to work to stay warm and get through the long lines of riders waiting at traffic lights; I even stopped and took the time for a sit-down seafood lunch at Misaki Guchi--on the the southern tip of Miura; .... and yet I was 2nd to finish out of around 70 riders. That never happens to me. But it did this weekend, on the Gokiso wheels.
After 800 kms, the rims are still perfectly straight, perfectly balanced. And they roll much smoother than any hubs I have ever used.
Here is the result of my amateur spin test on the front hub/wheel. The wheel rotates nearly 10 minutes, so feel free to "fast forward" through the video.
Here is another amateur spin-test I found online, this one a comparison of a Gokiso wheel with an FFWD 60mm carbon rimmed wheel with DT Swiss 240S hubs. This video is less than 2 minutes - at which point the competitor wheel has already stopped spinning, whereas the Gokiso has barely slowed at all from its initial speed.
Here are some other English language online references to Gokiso.
In total, I have ridden around 800 kms on my Gokiso wheels over the past month. Five rides of 100 kms or longer: Miura first ride, Winter Solstice ride -- Kobu Tunnel, SFC "commute" during the Festive 500, Hakone to Tokyo via Ashigara/Yabitsu, and the Nishi Tokyo Brevet.
The only time I have NOT used them during this period is for commuting/in town, and one long ride where I needed a front dynamo hub/light.
How do I like the wheels?
I am noticeably faster when I ride them, and I seem to work less hard. They roll and roll and roll, and they are stable and strong. After 800 kms, they are as true as the day I got them. And if anything they roll smoother.
As regular readers will know, I have set lots of Strava "Personal Records" on these rides, even in mid-winter and even when I have not worked any harder than usual.
Of course, they are carbon clinchers, so I am using the brake pads Gokiso recommends (and supplies). The braking surfaces are perfect. I had some squealing when braking on one ride after the brake pads wore a bit, but that I fixed the issue by “shoeing in” the pads again so the front edge is closer to the rims than the rear. Issue resolved.
On Saturday, January 10, I rode my first Brevet of 2015, a Nishi Tokyo 200 km ride in a big loop from Machida around the Miura Peninsula. I did not work particularly hard, except the first 20 kms where I wanted to work to stay warm and get through the long lines of riders waiting at traffic lights; I even stopped and took the time for a sit-down seafood lunch at Misaki Guchi--on the the southern tip of Miura; .... and yet I was 2nd to finish out of around 70 riders. That never happens to me. But it did this weekend, on the Gokiso wheels.
My rear and front hubs (and my foot) |
Front and Rear wheels |
Small decal |
Large decal |
Here is the result of my amateur spin test on the front hub/wheel. The wheel rotates nearly 10 minutes, so feel free to "fast forward" through the video.
Here is another amateur spin-test I found online, this one a comparison of a Gokiso wheel with an FFWD 60mm carbon rimmed wheel with DT Swiss 240S hubs. This video is less than 2 minutes - at which point the competitor wheel has already stopped spinning, whereas the Gokiso has barely slowed at all from its initial speed.
Here are some other English language online references to Gokiso.
1. Cycling Dirt – Gokiso at 2013 NAHBS
2. 2011 gizmag
3. Bike Rumor -- 2013 NAHBS
4. Cycling Tips -- 2014 Feb review (Australia)
http://cyclingtips.com.au/2014/02/gokiso-climber-hubs-review/
5. Tokyo Cycling Club - Pro Cycling Mechanic - Gokiso / GS Astuto review
5. Tokyo Cycling Club - Pro Cycling Mechanic - Gokiso / GS Astuto review
--------------------------------------
I will have more to say about these wheels as I use them over the coming weeks, months ... and years!
I will have more to say about these wheels as I use them over the coming weeks, months ... and years!
Fascinating. I must admit that I did not watch the full 10 minute spin test though.
ReplyDeleteDo you think they could apply the same approach to the bottom bracket? I am sure there must be room for improvement in design and efficiency there too.
Hi Graham:
ReplyDeleteI have not seen the bottom bracket and do not know if it is going to be launched commercially.
Certainly companies in the U.S. such as Chris King have done (1) hubs, (2) bottom brackets, and (3) headsets, not necessarily in that order, and there is significant commonality among hubs and BBs (parts that spin and so require bearings, and that bear heavy loads).
Best, David
Hello,
ReplyDeleteWatts. How many watts do these hubs preserve for a given rider's effort versus more conventional designs?
Or are these hubs an exercise in overengineering?
Hi Unfiltered Dregs:
ReplyDeleteYES, OF COURSE these hubs are over-engineered. This is like most things in Japan, but ESPECIALLY products that are designed by "otaku" engineers who focus on ultra high precision!
As noted, they are tested at 200kph, 300 kph ... for an application that will never need something that goes more than 75kph (for me at least ... a bit faster for a racer on a descent).
Kind of like driving a high-end German sports car around town and mostly within the speed limit (though, as I point out to my wife from time to time when I see someone driving a Porsche, Maserati or similar in Tokyo, the sum total of all the cycling gear I have purchased over the past 15+ years still has not aggregated to the cost of a single mid-level sports car).
Not a mass market product.
Of course, there are still car lovers who seem to derive great pleasure from the sports car, even sitting in traffic.
... and I think there are plenty of cyclists who will enjoy riding these hubs and experience them as a thing of beauty, who want the "best" and smoothest hubs -- that spin for nearly 10 minutes that absorb road shock, and that are designed to last pretty much a lifetime (with a 10 year warranty).
As for me, I have damaged enough cycling gear (see the "things we broke" page on this blog) so that I look at most gear as disposable. Rims, spokes, frames, brifters, etc. But I now re-use the "over engineered" components when I build up my bikes, and the more I can do so the better. See Robert Penn, "It's All About the Bike".
How many watts do they save? I have no idea, but it cannot be very many, since even my dynamo hub does not consume more than a few watts, and any "normal" road hub will be a fraction of that. That said, the combination of (1) Gokiso hub, (2) wide carbon clincher rim, and (3) low rolling resistance tire (e.g. Conti Supersonic), does add up to significant saving in watts, and for someone doing a TT or triathlon bike leg, the difference will show up in the recorded time.
Thanks Dave...
ReplyDeleteI can certainly appreciate the elegance of fine engineering.
Considering the cost it would be great to see some real world performance numbers, as well as how the engineering translates in terms of other performance factors.
That certainly is something all the great sports car companies provide.
I'd be especially interested in what the shock absorbing factors contribute while cornering at high speeds.