MC286 with SRAM 1x

SRAM has been pushing their 1x groups for road use lately, and has taken it a step further with the recent release of the SRAM Apex 1x group. One important aspect of the new group for anyone trying to build a cyclocross bike on a budget was the release of a new version of the PG-1130 cassette: a 11-42 cassette that is compatible with the 10-speed freehub standard that is found on older road wheels and is still in use on MTB wheels.

At the time I was buying the frames and parts for these bikes, I noted in my post about components that my choice to buy Mavic's budget-oriented Crossride MTB wheels forced me to use only 10 of the 11 cogs in the Shimano 105 cassette because MTB wheels are still using the 10-speed freehub standard. Building a cyclocross bike with the Crossride wheels and SRAM 1x at the time would have required using an XD cassette and buying the XD driver that is available for this particular Mavic hub (which retails for $100). The ability to use a wide range cassette that is compatible with the built-in freehub on stock MTB wheels reduces the cost of a SRAM 1x build significantly.

Since Heather likes the SRAM shifters on her road bike better than Shimano, this summer I decided to convert her cyclocross bike to SRAM 1x so I could use the Shimano 105 group on another bike. For the conversion, I used SRAM Rival brake levers and right shifter, Rival crankset with 42t chainring, Apex derailleur, and the aforementioned PG-1130 cassette.

The original setup consisted of a 50-34 crankset and 11-32 cassette. The new setup results in a lowest gear (42/42 for a 1:1 gear ratio) that is slightly lower than the original 34/32. The highest gear, a 42/11, is approximately equivalent to a 50/13, meaning that the new setup replicates the original setup minus the highest gear of 50/12 (since the 11 tooth cog was never installed). Having ridden my cyclocross bike for over a year, I don't believe I've ever used the 50/12, so I don't believe this is particularly important for the type of riding we do (or most riders do) on this kind of bike.

Gearing range is clearly not a problem, but the other possible caveat to a 1x conversion is gear spacing. I did some spreadsheet analysis of the gear ratios on the original 2x10 setup compared to the 1x11 setup and found that the difference is not that large. The 12-32 cassette has a maximum jump in gear ratios of nearly 16%, between the 19 and 22 tooth cogs. The 11-42 has a maximum jump of 18%, between the 13 and 15 tooth cogs. Perhaps more importantly, that larger jump is closer to the edge of the range in less used gears, where the largest jump was previously in the middle. The difference should not be particularly noticeable to most riders.

Based on our experience so far, I think the SRAM 1x groups are an excellent choice for cyclocross and gravel bikes. It's not surprising to see so many bikes coming out with this setup from the bike bike brands.

X-Peria 5200 Tandem

It's now been over 5 years since I bought the Bike Friday Family Tandem. I have no regrets about it, but I'm ready for something new. I eventually gave up on the conversion to drop handlebars, because even with the various tweaks I made, shifting never worked quite as well as the original shifters did, so I put the original flat handlebars back earlier this year. The telescoping seatpost was important when my kids were shorter, but now the kids are tall enough to fit some full size tandem frames. I also found the Bike Friday to be fairly heavy (around 50 pounds) and I felt that the frame lacked stiffness, a feature that can be partly attributed to its geometry.

My requirements for a new tandem were:

• Full size frame that's a reasonable weight.
• $1500 or less. I don't know how long my kids are going to be interested in riding with me on the tandem, so I want to avoid putting too much money into something that might not be used much.
• Fit a tall rider (me) in front and a shorter rider in the rear.

The Cannondale Road Tandem 2 comes in over $3000, and newer used ones routinely sell on eBay for nearly 70-80% of that. One other option is the KHS Milano, which has a list price of $2000 for an aluminum road tandem with disc brakes and Shimano Tiagra. The only way to get a good road tandem for less than that is to buy used. The problem with most older used tandems (with the notable exception of Cannondales) is that most 1990s and early 2000s tandems had level top tubes that work great for two riders of similar size but would be much to tall in the back for me. Sloping top tubes that can accommodate a larger height differential have become much more common on tandems in the last 10 years, but many of those newer tandems are usually out of my price range. There is a steady stream of used tandems on eBay, but the number of older used road tandems of the appropriate size and price range turns out to be relatively small, and the number of those sellers willing to ship a tandem (a necessity when you live in west Texas) is smaller still.

After searching eBay for some time, I eventually stumbled onto a listing for the X-Peria 5200 tandem frame, which was being offered new for somewhere around $500 by the French online bike shop CR2V. The frame is offered in two sizes, the larger of which is pretty close to the correct measurements for me. At this price I decided I could build my own for something close to my target price. CR2V has a number of eBay listings for frames that they offer to ship internationally, but there are more options available directly on their web site, which is how I eventually ordered my frame. The site is completely in French, so I had to navigate it with the help of Google translate at times. It turns out the $500 price tag included the eccentric bottom bracket but did not include a fork, however they do offer a package with BB, fork and headset for about $90 more. Shipping to the USA is not available as part of the normal checkout process on their site, but after some emails back and forth with them, they offered to ship to the USA for an additional 100 EUR. Total cost was just over 700 USD including shipping. When I placed the order, they had just run out of stock, so there was some delay while a new frame was ordered from the manufacturer, which took nearly 3 weeks. After the frame was finally shipped, it was only a little more than a week until the frame arrived.

At first glance, the frame has a lot in common with a Cannondale, but a closer inspection shows that the welds don't look nearly as clean as can be expected on a Cannondale. I would compare it to the quality of frame one typically sees on lower end aluminum road bikes in a local bike shop. While X-Peria is clearly a French company, it's not clear to me whether the frames are made in France or somewhere in Asia. Regardless, as long as the frame holds up, it suits my purposes just fine. The combined weight of frame, fork, headset and eccentric bottom bracket is about 9 pounds.


The specifications on X-Peria's web site do not specify whether the fork is aluminum or steel, but I believe it is steel because the fork blades look too small to be aluminum, and the junction of the steerer tube (which is clearly aluminum) and fork crown suggests an interface between two different metals rather than a single continuous piece of metal.

The frame has post mounts for 180 mm brake discs front and rear. I used 203 mm brake discs with 180-203 Shimano adapters and Shimano BR-R317 cable-actuated brakes.

The rear dropout spacing is 135 mm, so it's compatible with 29er MTB wheels. Nearly all higher end 29ers now use through axles, so most quick-release 29er disc brake wheels are targeted at less expensive bikes, which gave me a number of good options at a reasonable price, particularly since cost and durability were higher priorities for me than weight. I ended up with these 32-spoke WTB SX19 wheels. 36 and 40 spoke wheels are common on tandems, but it's difficult to find pre-built wheels with more than 32, and I think these will be sufficient for me. The tires shown here are 37 mm Continental Contact, though their actual width appears to be a bit less than 37 mm. Both wheels fit in the frame easily. The fork could easily fit 40 mm tires or bigger, but the rear doesn't have much extra space.


For the drivetrain, I bought used Shimano Ultegra 6603 shifters (3x10 speed) and a 9-speed Shimano Deore XT rear derailleur. (Shimano 10-speed road shifters will work with 9-speed MTB derailleurs but not 10-speed MTB derailleurs). I set it up with single-side drive, which means I'm only using 2 of the front shift positions, but I have the option to use a triple in the future. The reasons for single side drive were keeping cost down (I already had one of the cranks I needed) and size options. I am using a 180 mm front crank and 165 mm rear. When the kids get a bit taller, I will probably eventually go to 170 rear and 175 front with regular tandem cranks.

The rear crank is made by Origin8. It's inexpensive and also comes in even shorter lengths, an ideal option for someone setting up a single-side drive tandem for a short stoker. I used 39 tooth chainrings for the timing chain because I already had one in that size. The inner two rings pictured here are 48 and 32 tooth rings, which have been paired with a 11-32 10 speed cassette. This gives a gearing range just slightly lower than a 50/34 compact road crankset with similar cassette. The 1:1 ratio of the low gear should be enough for moderate climbs and the high end is more than enough for how I anticipate using the bike.

The finished bike comes in at 38 pounds as pictured, a full 12 pounds lighter than the Bike Friday. I don't have an exact cost total, but it came in somewhere between $1600 and $1700. It is probably possible to build one of these in a similar way with true tandem cranks (such as Sugino XD) for under $1800.

The frame has a few quirks that I didn't expect. None of them are serious flaws, but things that might have been helpful to know about in advance:

• shift cables are routed for a top-pull front derailleur. As far as I know, this means most road derailleurs won't work.
• only 2 water bottle cage mounts.
• no guide for brake cables on the fork, as would be expected on a disc brake fork.
• There are double threaded eyelets on the rear dropouts, but no rack eyelets at the top of the seatstays. There is, however, a threaded hole underneath the seatstay bridge and chainstay bridge, which is presumably for mounting a fender. I'm working on a custom bracket for attching a rack I have that uses this hole.

The completed MT-MC286 bikes

I have finally completed assembly of both bikes, so here they are. First, the 58 cm bike that I will be riding.

I will eventually put SPD pedals on this before doing any serious riding. The flat pedals are for purposes of riding around the neighborhood while I get the shifting and brakes adjusted. This bike has the narrower than expected Continental Contact tires that I wrote about previously. They will probably be a good tire for riding on roads, but that's not the primary purpose of this bike, so I will probably eventually get something wider. For those who are interested in such things, I weighed the bike and it came in a little under 22 pounds as pictured. The handlebars were wrapped with Bike Nashbar's handlebar tape. I used two rolls of it (white and "nuclear green") to get the striped pattern, spacing them fairly close together so that I used about 2/3 of each roll. That has the effect of making the padding about 50% thicker than normal. The green color turned out to be remarkably close to the green color we used in the frames.

Here is the 48 cm bike that Heather will be riding:

The two notable differences on this bike are the fatter tires and a big set back on the seatpost and saddle. I wanted to go with carbon seatposts because they are supposed to be more comfortable (I don't have personal experience--I'm still using an aluminum seatpost on my carbon road bike), and I knew this frame was going to require significant set back because of the steep seat tube angle. The geometry chart for this frame can be seen at Miracle Bikes web site. The 58 cm frame has a 73 degree seat angle (the same as my road bike), but the 48 cm frame has a much steeper 75.5 degree angle. That has the effect of moving the seat forward about 2.5 cm. So, for the 58 cm frame I bought a zero set back seat post and for the 48 cm frame I bought a post with about 2 cm of set back. I originally got a different saddle for this bike, but I was not able to get it back far enough, so I eventually swapped it out for this one that came stock on a Giant road bike and has longer rails. I understand these kinds of compromises are necessary on shorter frames, but in this case I don't think the seat tube angle needed to be this extreme.

This bike has a pair of 700x40C WTB Nano tires. They were tried with some success by someone on the forum discussing this frame at mtbr.com, so I thought I would get a pair and try them out on one of the bikes. So far I've only bought one pair, but if we like them after using them a bit, I'll probably get a pair for the other bike. Miracle Bikes specifies the maximum tire width as 38 mm, so this is technically exceeding that by 2 mm. The fork is clearly not the limiting factor here, as it appears to have more clearance than in the rear.

In the rear, things are a bit tight behind the bottom bracket, but I think it will work.

The seat stays also offer barely enough clearance, which I find strange because there is no particular reason not to make the clearance in this part of the frame more generous.

So, I think this frame will work with a 40 mm tire in dry conditions. Mud might be another story. The frustrating thing about the clearance in the rear is that there is plenty of clearance between the seat tube and rear tire, implying that the unusually steep seat angle on the 48 cm frame was not actually necessary. I think it could have been 1/2 to 1 degree less steep without making the gap between seat tube and rear tire the limiting factor in clearance, which would offer much more flexibility in choosing saddles and seatposts.

Based on my experience assembling the bikes, here are some things that I would change about the frame design:

• Reduce the seat tube angle by 1/2 to 1 degree in the smaller frame sizes.
• Make more tire clearance between seat stays. This is easy and doesn't compromise anything else in the frame.
• Make a little more clearance between chain stays. This may not actually be possible without some very creative engineering such as that being done by Open Cycle, but I'd love to put an even fatter tire on this bike if it were possible.
• Move the exit hole in the left chain stay for the brake cable forward a bit. Right now the cable has to make a bit of an awkward bend to get to the rear disc brake (Avid BB7 on these bikes).
• Taller head tubes. Nearly everyone on the mtbr forum thread discussing this frame that has posted pictures of their completed bikes has had 3-5 cm (or more) of spacers between the top of the headset and stem. I think this is a sign that the head tubes on this frame are probably shorter than they need to be. However, this frame is designed for cyclocross, and I don't really have experience with cyclocross frames, so it's possible that this is normal geometry for cyclocross racing, and that I and many others who are buying this for dirt road riding just have higher handlebar preferences than your average cyclocross racer. 

Having said that, none of these are serious problems, and I fully anticipate that we'll have a lot of fun riding the bikes.

MC286 component considerations

The logical choice for components for the MC286 is to install a road drivetrain and a set of mountain bike wheels because 142x12 rear wheels for mountain bikes are readily available for reasonable prices. It turns out one of the great challenges of making this frame work on a reasonable budget is the incompatibility between 11-speed road cassettes and mountain bike hubs.

When 11 speed was introduced to road components a couple of years ago, it required a slightly longer freehub body. (Note that this is all just from reading rather than experience; my road bike is still at 9 speed.) Mountain bike components didn't introduce 11 speed until more recently, and so far they have stuck with the old freehub body dimensions. This means that using a standard mountain bike wheel won't work with an 11 speed road group. There are very few road-specific wheels available with a 142x12 rear axle at this time, though I'm sure that's likely to change soon enough. The 142x12 standard is relatively new to cyclocross bikes and to my knowledge is not used anywhere (yet) on regular road frames. I suspect most cyclocross bikes being sold by the major brands that include 142x12 rear axles are using proprietary wheels. Because this standard is being introduced on higher end bikes, there aren't likely to be many budget-oriented options right now.

I did a lot of research and considered several options for wheels:

• Use a mountain bike wheel with the new Shimano XTR 11 speed cassette and a road group. Aside from the high cost of XTR, this may not be workable because it's only available with a 40 tooth large cog, which likely exceeds the capacity of the road derailleurs. This could become viable in the future if Shimano were to offer more options (in price and gearing ranges).
• Use a Campagnolo 11 speed cassette. According to this article by Lennard Zinn, the Campagnolo 11 speed freehub bodies can fit on most 10 speed road hubs. However, this requires finding a hub that has both 142x12 and a Campagnolo freehub body available as options. I don't know whether this exists, and if it did, it would be very expensive and also quite restrictive in the available gearing ranges.
• Find a wheel specifically built for cyclocross and through axle wheels. Right now the only thing I can find off the shelf is the Stan's Iron Cross wheelset. The new DT Swiss 240S DB hub is also designed exactly for this purpose, but I don't know what the cost is, and haven't yet found any wheels for sale based on this hub. In both cases, however, I think the cost is higher than what I was hoping to spend.
• Use a standard mountain bike wheel and install only 10 of 11 cogs from the road cassette.

As far as I can tell, only the last two options are even likely to work, but the last option has many more price options available because there are so many wheels available for 142x12 29er mountain bikes. I ultimately chose the Mavic Crossride wheelset because it looks like a good reliable set of wheels and is half the price of the Stan's wheels.

For components, I considered Shimano 105 and SRAM Rival and eventually settled on Shimano 105 because in part because I was able to get a great price on the complete group without brakes from Starbike.com in Germany. They were one of the few retailers I could find anywhere that offered their road groups with brakes optional, and they also had a great price on the Mavic wheelset I wanted. Starbike's prices are such that I'm sure they would have been competitive even a year ago with shipping from Germany to the US, but the strong US dollar vs. the Euro right now made the price even better. For two wheelsets, two component groups and several other miscellaneous parts, shipping was only an extra 10 Euro for what turned out to be two very large boxes. The parts arrived in less than two weeks.

This is the drive side of the Mavic rear wheel. The freehub came with the sticker on the right wrapped around it, warning against using cassettes that use individual cogs as opposed to a carrier. I got a 11-32 cassette with my Shimano 105 group, which uses individual cogs for the first 8 cogs and a carrier for the last 3 cogs. The reason for this warning is the aluminum construction of the freehub body. Individual cogs will eventually dig into the splines of the aluminum feehub body, where the carrier spreads out the load and avoids this. Using the cassette that I have chosen will therefore have a risk of damage to the freehub body over time, particularly in the middle cogs, which will be subjected to more torque than the smaller cogs. For now I'm going to accept the risk and try it for a while.

This is the complete 11 speed cassette set in place on the freehub without the lock ring. The last cog spins freely in place because the splines are not tall enough to engage it. In addition, the threads do not extend far enough to properly engage the lock ring. It's possible to get it on, but it will only catch about 1/4 turn worth of the thread. So, it's clear this won't quite work.

The excessive stack height of the 11 speed cassette is made clearer by attempting to install the wheel with the whole cassette on it in the frame. The picture on the left shows all 11 cogs in place, and the picture on the right shows only 10 cogs. With 11 cogs, the lock ring rubs against the dropout so that it would not be able to spin with the cassette, and the smallest cog is so close to the dropout that even without the lock ring problem, the chain would likely rub against the frame. On the right, there is enough space for everything to work properly, just with one fewer speeds. Because the first two positions each require special cogs, doing this properly requires buying both a 12 tooth first position cog and 13 tooth second position cog, and using them to replace cogs 11, 12 and 13. However, someone wishing to not buy the cogs could theoretically just remove the 13, leaving a 11, 12, 14 ... combination, which would leave an unusually big jump in gearing between the 12 and 14. The picture above shows the 12, 13, 14 combination, which added an extra $20 or so in costs, but made it possible to keep the total cost of the bike down by quite a bit.

The last important consideration for this bike is tires. The Miracle Bikes web page for this frame specified that the frame could handle a 38 mm tire. I went ahead and bought some 37 mm tires and installed them on the rims. I looked at a lot of tires trying to find a balance between smooth center for riding on pavement and some kind of tread on the sides for riding off pavement. I decided to try the Continental Contact, which is really more of an urban/commuting tire. Here it is installed in the frame.

My first reaction is that I wanted something wider than this and there appears to be a lot of room. By my measurement the tire is closer to 32 or 33 mm. I measure about 50mm between the stays, and similar clearance at the bottom bracket and top of the fork. This made me wonder if I could have gone wider and knobbier, so I'm going to try a set of 40 mm tires instead.

Arrival of the MC286 frames from Miracle Bikes

The MC286 frames arrived earlier this week.

The total time from placing the order to arrival at my doorstep was a little over 7 weeks. Most of that time was the custom paint job. I was warned it would take a while, but it took a little longer than I was planning. After about 5 weeks, I emailed my contact at Miracle Bikes asking about the status. She sent me a photo of one of the completed frames, and a couple of days later sent me a shipping number. The shipping numbers can be tracked by the US postal service, but often there is a lag (in this case close to a week) between receiving the number and the USPS tracking system recognizing it. The time from receiving the shipping number to arrival was just over two weeks, so I imagine a buyer not wanting custom paint could receive a frame in less than three weeks.

The bikes look great, a nearly perfect rendering of the artwork I sent them, and the quality of the paint looks really good. The through axles I had added to the order were already installed in the dropouts. They seem to be specific to this frame design (which is also the consensus on this thread at mtbr.com), any prospective buyer is advised to order them with the frame. I didn't think to ask about it during the process of purchasing the frame, but headsets were included. I'm waiting for a few more parts and then will be ready to start assembling them.

MT-MC286 from Miracle Bikes

Last summer, I moved from central Illinois to west Texas. I've done some riding since then, but not as much as I'd like to because I just haven't found the riding very enjoyable here, the primary reasons being narrow roads with fast traffic and a surprising number of loose dogs. Like central Illinois, the area is crisscrossed with small farm roads, but the farm roads here are mostly unpaved. The dirt roads are readily accessible from where I live, and I've found them to be much more traffic and dog free, so I've concluded that riding the dirt roads is my future, and that I want a bike designed for such riding. That is the origin of my next project, a cyclocross bike, or rather two cyclocross bikes so Heather and I can ride together.

My first plan was to find some kind of inexpensive cyclocross bike from someone like bikesdirect.com, where aluminum frame cyclocross bikes with disc brakes start at about $900. We both like our carbon road bikes a lot, and since my last experience ordering a frame with custom paint job direct from China last year was relatively painless, we eventually decided to spend a bit more and go in the direction of carbon again.

I had three requirements for the frames: the size range needed to go as far as 48 cm and 58 cm (60 would be preferable, but I can make 58 work), disc brake compatibility, and through axle dropouts. Because I haven't owned a mountain bike made more recently than the mid 1990's, I really don't have much prior experience with disc brakes and through axles, but based on all of my research, this is the direction cyclocross bikes are going, with road bikes likely to follow soon. Some major brands, such as Giant, are already selling cyclocross bikes with through axles. My primary motivation for doing this is forward compatibility: through axles and disc brakes make it more likely that the wheels and frame will be interchangeable with future wheels and frames.

It took some time to find someone selling a frame that met all three criteria. Nearly everyone has a disc frame, and several manufacturers appear to have recently introduced through-axle compatible frames, but I could only find one that offered all of the sizes I wanted: the MT-MC286 from Miracle Bikes. The frame is also listed on Alibaba.com, which is how I contacted them. I got a reply pretty quickly, and I went back and forth for a few days with the person about what I wanted. The base price of the frame was quoted to be $480. I don't know whether this represents any kind of a discount over the price of only buying one frame. A custom two-color paint job was quoted at $70. Shipping is effectively two for the price of one because the frames will ship in a single box. (The frame I bought last year was in a box probably big enough for two.) They requested payment via PayPal, and add 4% to the cost to cover the fee that PayPal deducts. Total for the two frames came to $1300. I was told that they have a bit of a backlog of up to 20 days on their painting (apparently due in part to Chinese New Year). All of this is pretty consistent with my previous experience, so I wasn't too worried.

Before sending money, I thought I should read some more about them. I didn't know anything previously about Miracle Bikes, but after doing some research I did find at least one blog that discusses importing frames directly from Chinese manufacturers, and he lists Miracle Bikes as one of the ones he considers to be reputable. Given that and the built-in protections that come from using PayPal, I decided I was ready to give it a go.

We've decided to get matching frames, so we came up with a design, and sent the following image to Miracle Bikes:

I placed the order a little over a week ago, and I'm hoping that it will ship within about 2 weeks.

I'm in the process of finding components to go on the frames, but one persistent question in my mind has been what kind of through axle to use. It's my understanding that there are different types of through axle designs, and they need to match the dropouts of the frame, but there's very little information online about what is compatible with this frame. I eventually found this very helpful thread on the forums at mtbr.com. A number of commenters on that thread have already bought this frame from Miracle Bikes, and some of them bought through axles from Miracle Bikes with the frame. I contacted Miracle Bikes and was quoted a price of $29 for a set of axles, so I have added that to my order. Now, I just wait...

Tire clearance on the VB-R-027

I've become a believer in riding on the widest possible tire for the application. After many years riding very skinny tires, I decided to try 25 mm road tires a few years ago after fixing one too many flat tires at the side of the road. The reduction in pinch flats since the change was quite dramatic, so when I bought my new frame from Velobuild, I hoped it would allow me to go one step further to 28 mm tires, given that their model VB-R-027 was a bit more relaxed in its geometry than some of their other frames.

I can now report that 28 mm tires work fine, as long as they aren't really 28 mm. I got a new pair of Continental Grand Prix 4 Season tires this winter in the 700x28C size.

The problem is, I don't think they are actually 28 mm wide. They seem to be about the same size as the old tires I removed, a less expensive pair of tires that were marked 700x26C. A relatively rough measurement using a ruler and eyeballs seems to confirm that they are about 26 mm wide. (This also appears to be confirmed on various forums I've read).

Here is what the Continental tires look like in the fork:

There are maybe 2-3 mm to spare all the way around. Clearance is similar behind the bottom bracket:

The worst clearance on the frame is the rear brake bridge, which in my opinion should have been 1-2 mm higher.

I forgot to photograph the seat tube clearance, but it's about 2-3 mm. The 28 mm tires that are really 26 wide work just fine on this frame. A true 28 mm tire might work, but it would be very, very close. For this reason, I'm perfectly OK with the tires being slightly smaller than advertised.

I have occasionally had problems getting the rear wheel in and out of the frame. The angle of the dropout combined with the size of the nut on the very old quick release skewer I'm using forces the wheel forward enough that the tire rubs a bit on the seat tube and chainstay bridge. I compared it to another carbon frame made by Giant and discovered that its dropout is angled differently and avoids this problem. This isn't a very big deal for me, but if someone were worried about quick wheel changes in a race, it could be a problem.

VB-R-027: the completed bike

I finished assembling the bike just a couple of weeks after my last posting about the frame's arrival, but I have been very busy (including moving to a new house) and didn't get to writing about it until now.

Assembly

Assembly of the new bike was pretty easy. I had read online about headset installation and expected to need special tools to get the headset bearings installed into the fork crown. I eventually figured out that the headset included with the frame (which did not come with any instructions) just snaps into place with no tools required.

The shift and rear brake cables are all internally routed, and have liners built into the frame. The shift cables were very easy to install, while the brake cable hit a bit of a snag (literally) because the liner didn't seem to line up perfectly with the exit tube. Eventually I discovered that I could get a cable into the liner from the reverse direction. Using a reverse inserted cable to hold the liner in place while I threaded the cable through in the forward direction finally got the cable in relatively easily.

As noted in a previous post, my old frame had a 59 cm top tube, while this one appears to be only 57 cm (despite VeloBuild's geometry chart claiming 58 cm), so I replaced my 110 mm stem with a 130 mm stem, and have been able to replicate my position on the bike pretty well.

At the moment, I've got quite a stack of spacers on the steerer tube. I'm still considering going 1 cm higher with the stem, and then I may trim about 1 cm from the top of the tube to reduce the number of spacers.

The components to build the bike were taken off my 2005 Fuji Roubaix Pro, which had a 9-speed Shimano 105 group. The old bike was an aluminum frame with clamp-on front derailleur, but this frame has a bolt-on derailleur mount, so I needed a compatible derailleur. Shimano Sora is now 9 speed (and probably made with the same machinery as the old 105 parts), and the derailleur can be bought for about $25 on Amazon, so that's what I bought.

Finished Product

Here's the completed bike.

I've taken it out for a few rides now. With careful measurements, I succeeded in replicating the position I had on the old bike pretty well, so I haven't really had to adjust anything. It's a very different feeling from my previous aluminum frame--the increased stiffness and responsiveness to pedaling is immediately apparent. I don't have many points for comparison in terms of ride comfort. It seems at least as good as my old bike (aluminum frame with carbon seat stays), but I don't doubt that more expensive frames could further improve on it. It's a fun bike to ride, and I have no regrets about buying it.

Arrival of the new frame from Velobuild

My new frame arrived from China last week, about 4 weeks after placing the order. As far as I can tell, the shipping only took just over a week, and much of the remaining time was spent on the custom paint job. I suspect an order without custom paint might be quicker. After sending Chris at Velobuild a picture of the paint job I wanted, he contacted me about two weeks later to ask whether I wanted glossy or matte finish. I responded, and then heard back again about a week later with shipping confirmation and tracking information.

I had to go to the post office to pick up the box, though I think they would have delivered it to my door if anyone had been home when they came.

I got a large box that was taped shut very securely, but despite prominently displayed arrows pointing which way goes up, the side of the box was a bit dented, and featured several prominent foot prints.

Despite my occasional frustrations with the US Postal Service, I'm inclined to trust that they are more competent than that, so I'm blaming this one on China Post. The frame inside was wrapped with some foam padding, and as far as I can tell, did not suffer any damage from the mistreatment of the box.

Here's the finished frame. For the most part, the frame looks great. The colors look more or less like what I was expecting, and the paint looks like it should be durable. A comparison to the picture I sent them reveals that they made one small mistake: the blue region on the chain stays is a little too long compared to the blue sections on the other tubes, but I think it won't be too obvious when the complete bike is assembled.

The only other disappointment so far is that I don't think the frame perfectly matches the claimed geometry specifications. The specifications for the frame on Velobuild say the the top tube length on this frame is 58 cm, which is one centimeter shorter than my existing frame. After measuring the frame, I think it's closer to 57 cm. I think I can compensate for this by getting a stem that is 1 cm longer than I was planning, and still get the same position on the bike that I had before.

Aside from those two details, I don't have any other complaints about the frame or the process of buying it. I'm still awaiting a couple of parts I need to assemble the new bike. I've already got a new seatpost and stem for the new frame, as well as a new set of brake and shifter cables. I forgot until I'd already got everything else that I'll need a new front derailleur, because my old one is the clamp-on type, which is not compatible with this frame. When that arrives, assembly will begin.

My next bike: a VB-R-027 from 梅 辉辉

Over the last couple of years Heather and I have been riding together more, and we've been doing more rides with the kids on the tandems. More riding inevitably makes me start thinking about how much fun it would be on the latest new bike. Last year Heather upgraded her old aluminum bike to a new one with a carbon fiber frame, and she thinks it's a substantial improvement in ride quality over the old one. I'm not going to be able to buy my dream bike any time soon, but I've become curious enough about carbon fiber frames to look around at some of the less expensive carbon frames. People who ride and write about the super expensive bikes are all convinced the extra cost buys a better quality ride, but carbon fiber frame construction has advanced so much that I think even the inexpensive ones may now be well ahead of where any type of frame was 10-15 years ago, including my current road bike, a 2004 Fuji Roubaix Pro.

At some point in the last few months I became aware of a whole new world I didn't even know existed: cyclists who buy bike frames direct from the manufacturers in China. I first stumbled into this when looking for tandem frames, thinking that if I could buy just a frame I could upgrade one or both of our tandems at a lower cost than buying the whole bike. It turns out that hardly anyone sells tandem frames except for expensive custom builders, and a few Chinese frame shops that will sell direct to consumers (in addition to the frames that get resold under other brand names). The process of buying a custom titanium frame from China is well documented by the Spanner Bikes blog, which is a fascinating read. I'd like to do that some day, but titanium is beyond what I can spend right now, so I started looking at carbon fiber frames from China.

Since I have a birthday coming up, I've decided that this year I'm going to get a carbon frame (thanks, family!). To keep costs down I'm going to (for now) put the aging Shimano 105 group off my Fuji on it. After doing a fair bit of research I concluded a few things. First, there are a few sellers that seem reputable and are even willing to offer warranties nowadays, but the process of sending a broken frame back to China is probably slow and expensive, so I'm still accepting a bit of risk doing this. Second, I decided I don't want a frame that tries to imitate a brand name frame. There are outright counterfeits and frames that are just similar looking to famous name frames, but I want to respect the R&D those companies put into their designs and I'm not doing this to pretend it's something it's not. Third, it appears the main benefits of ordering direct from China are selection and customization. You can now get a generic carbon frame from a US distributor like Bike Nashbar for as little as $500. I decided to buy from Velobuild Mall, which offers a dozen carbon fiber road frames in the range of $320 to over $500, but shipping from China is $80, so the cost is not that different, but with more frame options and custom paint jobs.

The majority of frames offered by Velobuild only go up to 58 cm, while my existing road bike is a 61 cm frame. However, some of the frames have top tube and head tube lengths that are similar to my existing frame, so I should be able to replicate my fit with the right length seatpost. I'm not interested in aero frames because they are generally considered to have a harsher ride. Those constraints left me with two or three frames that could work and I went with the "VB-R-027" model, currently selling for $349. In addition to being less expensive than the other frame I was considering, it had slightly longer chainstays, which will hopefully give me the biggest possible tire clearance. I'm happy with the 25 mm tires I'm using now, but I'd like the option of bigger tires if they will fit. Since I'm going to the trouble, I decided I might as well get a custom paint job. I went with 3 colors for $70 (1 or 2 colors cost less). I put the frame, seatpost collar, and custom paint in my cart, and proceeded to checkout, at which point the shipping and some tax (I'm not sure which government is getting that) were added. A few seconds later I got a confirmation email from PayPal saying:

You sent a payment of $517.96 USD to 梅 辉辉.

So there you have it. I just bought a bike frame from China. The checkout process didn't provide a way to specify what I wanted in my paint job, so I sent an email to chris@velobuild.com to ask. Chris advertises his email prominently on Velobuild.com, and he responded within minutes. He asked me to just email him a picture of the paint scheme I had in mind. I had already prepared artwork for my proposed paint scheme using Inkscape, so sent it off. The drawing includes a head tube logo I designed, but apparently doing that level of detail would have added extra cost, so I told him to skip that part. The finished product should look something like this, minus the head tube logo:

New shifters and derailleur on the Bike Friday tandem

Last year I switched the Bike Friday tandem to road-style drop handlebars. The Sachs (now SRAM) Dual Drive system has a 3 speed internally geared hub with 7 speed rear derailleur. The system came with Grip Shift shifters, which work great, but are for flat handlebars only. At the time, I had read a couple of sources that I consider to be reliable claiming that any 7-speed Shimano drop bar shifter is compatible, so I found a set of 1990s vintage Shimano RSX 3x7 shifters on eBay. The triple left shifter worked great on the internally geared hub, but the 7 speed right shift lever never seemed to work quite right despite several efforts to readjust it. It seems that the cable wasn't pulling the derailleur far enough to get all 7 gears, so it would work OK for 2-3 gears but not the rest.

Initially, I assumed that one possible reason for the poor shifting was that the RSX shifter mechanism was simply wearing out due to its age. At some point recently I discovered that in the last year or two, Shimano has started selling a 3x7 drop bar shift lever set for the first time in many years. The 7 speed Tourney group appears to have been available for a few years and is found on various low-priced bikes, but only last year was a drop bar shifter made available. So, my first solution was to try swapping out the RSX shifters for a brand new set of Shimano Tourney 3x7 shifters.

These shifters sell online for about $90. Amazingly, after buying these I sold off the old RSX shifters on eBay for $65, so this swap ultimately only cost me about $25. I'm constantly amazed how much some vintage bike parts will sell for. The Tourney shifters have the thumb lever like Shimano Sora did until about a year ago. I prefer the "behind the brake" location for the upshift lever, but this works fine. These also have the "indicator" that shows you what gear you're in, another feature I associate with lower end Shimano road groups, and which I always assumed I didn't need. As it turns out, it's actually been nice to have the gear indicator on a tandem because it's not as easy as it is on a single road bike to know how many gears you have left just by looking down.

The Tourney shifters didn't solve my shifting problems. At this point, I started measuring how far the shifter pulls the cable compared to the original shifter that came with this bike, when it finally dawned on me that the rear derailleur on this bike was expecting a different cable pull (known as "actuation ratio") than the shifter was pulling. I had assumed that because the spacing of the 7 rear cogs was identical to Shimano 7 speed systems, a 7 speed Shimano shifter would work with this, without giving any thought to the rear derailleur.

The original derailleur is labeled "Sachs Centera". I can't find any documentation on this old shifting system. Sachs was bought by SRAM in 1997, so I'm assuming this bike and Dual Drive system was made around that time, because I don't think it's much older than that.

Regardless of the technical details, I was pretty sure the derailleur was the problem, but didn't want to spend a lot of money on trial and error. All I needed was a derailleur that's compatible with Shimano road shifters. For $15 on Amazon, I got this "SunRace R80", which claims to be Shimano-compatible.

This is designed for an 8 speed system, but Shimano 7, 8, and 9 speed all use the same cable actuation ratio, just with the clicks spaced closer together, so in theory this derailleur should work with any of them as long as the shifter is compatible with the spacing of the cogs. With the new derailleur installed, I can now finaly reach all 7 gears! Shifting is still a little sluggish sometimes. The derailleur hanger is slightly misaligned, so I'm going to try fixing that, but it could also be that this is just how a $15 derailleur shifts. In any case, I'm happy enough with this setup that I'll stop messing with it for now.

Custom headset caps

Ever since threadless headsets became standard on most new bikes (maybe 15 years ago), headset caps have become a popular way to customize bikes. A few years ago when I needed to replace a cap on my road bike (it broke due to overtightening), I considered some of the options from Purely Custom but eventually just got a plain one. I couldn't think of any text interesting enough to put on a cap, and none of the artwork options interested me much. Heather recently got a new road bike, and I was looking for something creative for her bike on her birthday when I discovered Kustomcaps, which will make a cap with completely custom artwork in addition to various text options. In trying to come up with designs that fit the circular area, I realized that Celtic knots are well suited to this, and Heather happens to be a fan of these type of designs, as she has several pieces of jewelry featuring them. So, I came up with a couple of designs:

These designs were created using Inkscape. The procedure for creating the weave pattern is very helpfully described in this tutorial. Eventually, I settled on the second design, which allowed a little more space to add some text. Here's the result (with a blue bolt to go with it).

I liked it so much I decided I needed one for my own bike. As it happened, we recently made a trip to Ireland, and inspired by the Celtic knot on Heather's bike and my own Irish heritage, I was on the lookout for interesting and unusual designs. I found what I was looking for in the floor tiles of St. Patrick's Cathedral in Dublin.

This was one of many very interesting patterns from that particular floor that we took pictures of. I had to stretch the proportions and add or remove parts in order to get the square pattern to fit a circle better, and I came up with two different candidate designs for a headset cap.

These designs were also created in Inkscape, in this case by importing the photo and using the "trace bitmap" feature to get an outline of the pattern, which then required considerable tweaking. I ultimately chose the second of the two designs. Here it is installed on my bike.

Now I don't think I can go back to a plain cap on my bike.