We’re one of the few bike shops in Melbourne who specialise in custom wheel building and sell far more hand built wheels than factory built wheels at the medium to high end. This post is intended as a discussion of some of the variables involved, and hopefully the huge range of possibilities will highlight why hand built wheels which are custom-built for a particular application are going to do that job better than factory-built wheels. Be warned, this post is long and is not a how-to guide. It has some technical discussion of both component choices and then of the process of building itself.
Hub choice when it comes to building a wheel will be dictated by what the bike is intended to do. There are hubs for disc brakes, internally geared hubs, cassette hubs, freewheel hubs, dynamo hubs, coaster brake hubs and many other types. Often this is the feature of the wheel that drives the wheel build so the choice will be already made. For instance, we make a lot of front wheels with dynohubs to upgrade existing bikes. Once you know what sort of hubs you want on your bike you’ll still have a number of choices. The options are too numerous to discuss here, but some of the features deserve to be mentioned.
First, you will often have a choice between hubs with sealed cartridge bearings and alternatives with cup and cone bearings. Cup and cone bearings are the older option, and are becoming less common. In some ways though they are superior. For a hub of a certain size cup and cone bearings allow more ball bearings and larger ones and the combination is much more resilient. They are easier to service and can be re-greased or replaced very cheaply. Finally, because the cones contact with the bearings at an angle they are better at dealing with side loads. The downside to cup and cone hubs is that each model requires a differently shaped cone. Because cones wear from contact with the bearings (especially if they are not well maintained with good clean grease) you will eventually wear out the cones in the hub. If you cannot get the right replacements the bearings will never run as smoothly as they ought to, and after a few years the manufacturers normally stop making replacement cones for their hubs. Cartridge bearings, on the other hand, are made in standard sizes and can simply be pressed out and replaced. Furthermore, even when they are used in a poor state they do not damage any integral part of the hub.
We’ll discuss particular hubs in later posts. Generally though, we’re big fans of Shimano and Schmidt dynohubs. You cannot beat Shimano for value-for-money in rear hubs, Phil hubs are incredibly smooth and close to unbreakable and DT, Hope and White Industries make great lightweight sealed bearing hubs.
There are a huge range of rims which suit different applications. When thinking about rims it is best to have a strong idea of the sort of tyres you want to use, and the sort of brakes you want to use. The size of the tyres you intend to use determines the width the rim has to be. Whether or not you intend to use rim brakes determines what shape the sidewalls should be. Some rims are eyeleted and some aren’t. Eyelets are normally made out of stainless steel and are little inserts around the spoke holes that distribute the stress evenly. There are single eyelets which are only attach to one wall of the rim and there are double eyelets which attach to both. All things being equal, double-eyelets are better because they distribute the stresses over more area, but things aren’t always equal. There are good rims that use double-eyelets, good rims that use single eyelets and good rims that use no eyelets. (A ‘V’ section rim distributes forces in a similar way to an eyelet.) In most cases it shouldn’t really be a factor in your decision making whether or not the rim has eyelets, so long as the rim is a good one. An exception is if you are planning to use aluminium nipples, but that will be mentioned later.
Again, later posts will discuss the particular rims. We use more Mavics than any other brand. They are unbeatable value and great quality. Open Pros are our default road rim. A719s are terrific rims for touring. The XM717 and its little brother the XM317 are both great cross-country rims. We’ve though a lot about Mavic vs Velocity and overall Mavic wins for us. The material in the sidewalls seems harder and stands up to brake wear a little better, the joins are smoother and the nipple seat diameter is more consistent. For the most part we use Velocity rims where there are gaps in the Mavic range. We use Dyads for light touring bikes, Randonneurs and road bikes with wide tyres because they are wide and light. The closest Mavic equivalent is the A719, but it is considerably heavier. We use Synergies for 650b touring wheels because there are no Mavic 650b rims. We also carry VO rims, Stans rims, DT rims and others as needed, many of which fill gaps in the Mavic range.
Spokes are not something that captures many people’s imagination. They just aren’t that exciting on their own. But in a wheel they are immensely important and the right choices when it comes to spokes can separate a good wheel from a bad one.
The first question with spokes is how many to use. This obviously has to be decided on in tandem with which rims and hubs you plan on using. Some rims and hubs are only available in certain drillings. Also, given a certain amount of load, the lighter (ergo weaker) the rim is, the more spokes you should have. So where someone could get away with 28 holes in a Deep V, they might be better off with a 36 holes in an Open Pro. Ceteris paribus, more spokes will make the wheel stronger, stiffer and more reliable. The only real downside to more spokes is the increased weight of the spokes.
That said, there’s no point going way overboard, and even the small weight savings of lower spoke counts are worth something, if they come at no real cost to the reliability of the wheel. If you’re a light rider and we’re making you a back wheel for your road bike using a reasonably strong rim, then we wouldn’t suggest using 36 spokes, because the extra weight isn’t getting you much. Furthermore, since a front wheel is not dished and doesn’t have to transmit drive forces you could get away with even fewer spokes on the front.
The next issue with spokes is what gauge to use. On any wheel that will get serious use we will recommend butted spokes of some variety. One benefit of butted spokes is obvious – reduced weight. This is not to be sneezed at, since it’s rotating weight, but it’s certainly not the only reason to use butted spokes. Even without the weight saving we’d be using them for their mechanical properties. It’s counter-intuitive that a thinner spokes could make a stronger wheel, so you sometimes hear people talking like the weight saving is the only real benefit, and you even hear some people suggest that the weight benefit comes at the expense of strength. Some people build their touring wheels and MTB wheels with straight gauge spokes thinking they’re getting a stronger wheel, but it’s a big mistake.
Even though butted spokes are thinner in the middle than plain gauge spokes they make for a stronger wheel. The main reason for this is that the two ends of the spoke are concentration points for the stresses in a wheel. The shoulder of the spoke and the end of the threads are the two places where spokes most commonly break. A thinner middle section will flex more under the impact and so transmit less of the stresses to the weak points at the ends. This flex also means that the impact can be shared with more of the adjacent spokes. Both of these points add significantly to the fatigue life of the wheel.
There are only two real downsides to double-butted spokes, they are slightly more work to build with, and for wheels where stiffness matters at all costs (track and trials wheels) that last little bit of stiffness can be bought at the expense of longevity by using plain gauge spokes. The extra difficulty working with butted spokes is because of wind-up. Tension is added to the spokes by turning the nipples. As the tension on the spoke increases so does the friction between the threads in the nipple and between the nipple and the rim. At a certain point the spoke will want to twist up rather than move further in the threads. Good lubrication and careful building can get around this though.
The spokes we use are mostly DT Swiss. The spoke we mostly commonly build with is their Competition. It’s double-butted – 2.0mm at either end and 1.8mm in the middle. It’s a good choice for the majority of wheels. One either side of this are Revolutions and Alpine IIIs. The Revolutions are also double-butted but 1.8mm at the ends and 1.5mm in the middle. This makes them very light but not as strong or stiff as other spokes. The Alpine IIIs are triple-butted with 2.3mm shoulders, 1.8mm middle sections and 2.0mm threaded sections. They are very heavy duty and great for loaded touring bikes, tandems and extreme downhill bikes. There are two other spokes that we use less often because they are more expensive, though both are great spokes – Super Comps and Aerolites. The latter are partially bladed and are very very light and very strong. The Super Comps are triple butted but thinner than the Alpine IIIs being 1.8mm at the threads, 1.7mm in the middle and 2.0mm at the heads and are great for cross country disc brake wheels (among other things).
In addition to DT, both Sapim and Wheelsmith make nice spokes. Sapim’s CX-Ray is a great lightweight spoke, though it is very pricey. We will happily use either brand where they are appropriate. DT spokes are still our default though, just because we’ve been really impressed by their quality control, and there are no significant gaps in their range. When weighing up DT vs Sapim it just comes down to the particular spoke in question and how well it fits the application.
The next choice with spokes is what lacing pattern to build the wheel with. The classic choice is 3x where each spoke crosses three other spokes on its way from the hub to the rim. In most cases this is the strongest way to build a wheel, since it puts the spokes closest to 90 degrees leaving the hub (Number of spokes and size of flange have an impact on this equation though). This is best because it efficiently transmits drive forces and braking forces between the hub to the rim. There are a number of other options for lacing pattern – 2x and radial being the next most common.
The main choice here is between brass nipples and aluminium ones. The only real selling point of the aluminium ones is that they are light. But this is a worthwhile consideration because of where the nipples are – right at the outside of the wheel. This puts them in the most significant area for rotational weight and they need to be accelerated up to about twice the speed of your bike. The weight savings from aluminum nipples aren’t huge, but in the context of variations in tyre, tube and rim weight they are quite significant.
The downsides with aluminium nipples are pretty much what you’d expect from something that saves weight – they are less robust and reliable. This is for a number of reasons. First, the nipples are weaker – the heads can pop off under tension (though this can be avoided to a great extent by choosing spokes long enough to fill the nipple), the square section used to tension them can be more easily rounded out, the threads will fail more easily than with brass nipples (though if this is happening while you’re building it’s probably a sign that you’re doing something wrong) and they are less likely to survive an impact. Secondly, aluminium is less suited to being threaded than brass is. Brass is self-lubricating whereas threads in aluminium are more inclined to gall, seize up and strip threads. Suitable lubrication can help with this, but all things being equal brass is the easier material to work with. Thirdly aluminum is less resistant to corrosion than brass and is more likely, over time, to seize in place and resist truing. Finally, aluminium nipples are hard to use in direct contact with aluminium rims, again because they can gall under the friction. Eyeleted rims are usually best when using aluminium nipples.
Negotiating the pros and cons here is case-by-case and depends on the work you want the wheel to do. If you’re going touring or doing extreme downhill riding aluminium nipples are a bad idea. If you’re building Open Pros onto White Industries hubs with Revolution spokes then you’d be wasting an opportunity to save some important grams if you used brass. In between these two cases it gets less clear. As a general rule it’s best to err on the side of caution.
The manual side of wheel building has a reputation as a black art, but although it requires skill, practice and patience it’s not really that mysterious when it comes down to it. There are many slightly different methods people employ, but they all aim at the same thing – producing a wheel that is laterally and radially true within accepted limits, dished correctly, and has spoke tension that is as even as possible. The physical world being what it is, and manufacturing tolerances being what they are, getting each of these variables absolutely perfect is not an option. But the better the quality of your parts the closer you can get. A rim that is straight and round out of the box will be straight and round when all the spokes are pulling evenly. Eventually though, there will be a degree of choice in balancing the goals in order to produce the best possible wheel.
One classic mistake when balancing the variables is to blindly chase lateral run-out and produce a wheel that is as laterally true as possible, without regard for the other variables. For one thing this wheel will not be very radially true and will have little hops in it, which the rider will be able to feel while riding. But more significantly, the spoke tension will be radically uneven and the taughter spokes will be doing all the work. This means that the wheel won’t last as long (since some spokes do most of the work, and others are so loose that the slacken in use and so fatigue due to the movement from being loaded and unloaded). It also means that even the lateral trueness of the wheel won’t last very long, since the loose spokes will loosen further and the alignment of the rim will change.
With good quality parts not much compromise will be needed, but where compromise is needed, then all factors must be balanced, and the rim must be true enough and round enough, but with some precedence given to evenness of spoke tension. This will produce a stronger wheel that holds it’s shape better for longer. The only reliable way to do this is to build the wheel using a spoke tensiometer. This is a tool that measures the deflection of the spokes under a load. This deflection can then be translated into an amount of load on the spokes. Any wheel we build will have each of its spokes checked to be sure that variation in tension is minimal. There are people who will tell you that the amount of resistance on the nipple or the pitch of a plucked spoke is enough of a guide to spokes tension. However, the plucking is far too blunt an instrument to rely on since audible differences in pitch are produced by larger changes in tension than are relevant to a well-built wheel. The resistance of the nipple to turning is a measure of too many extraneous variable to be a reliable test for spoke tension, since friction at the threads and rim both affect it and so it it very sensitive to differences in the effectiveness of the lubricant used and small variations in the evenness of all the surfaces in contact.
Another common mistake is to build the wheel with insufficient overall spoke tension. This will produce a weaker wheel, prone to breaking spokes at the shoulders, and the extra movement of the rim will make it prone to cracking at the spoke holes.
There are a few techniques that help to achieve the best possible balance of evenness of tension and trueness of wheel. First, lubrication is a key factor – reducing the friction between the nipple and the spokes and between the nipple and the rim helps a great deal. This will also help to bring the overall tension high enough. As mentioned before, building with a tensiometer makes a huge difference, letting the builder achieve a higher degree of evenness of tension and letting them know when they have enough overall tension. Once the balance of tension and trueness has been achieved the wheel is stress-relieved and re-examined. Stressing the spokes does a few things. Firstly, it relieves any wind-up in the spoke. This creates a pinging sound and the spoke looses tension, because the wind-up that is being removed is effectively shortening the spoke. Secondly, stress-relieving beds the shoulders of the spokes into the flanges of the hub a little bit. Thirdly, stress relieving cold-sets the spokes. By taking the metal in the spokes past its yield point, we change the shape of the spoke, which means that the spoke isn’t trying to return to the shape it was before it was built into a wheel. If there is too much wind-up then stress-relieving will dramatically throw out the balance of the wheel. All of the results of stress-relieving effectively lower over-all spoke tension, and they don’t necessarily do so in a totally even fashion, so another round of tension checking and adjustment is needed (as well as truing). The process is repeated until everything settles in, and further stress relieving makes no difference to the wheel. This stability means that the wheel will hold it’s shape while being used.
Next, thread-locking compound is used (weak enough that it does not impede later truing), rim tape is applied and the wheel is ready.
Drop in or call us if you want a wheel built and we can talk about the best way to balance all these variables for your bike. We’re also always happy to build wheels for interstate customers. The most recent wheel we built for an interstate customer was a Phil Wood SLR Touring hub with 36 holes, DT Competition spokes laced 3x to a Mavic 719 rim.