To build a wheel you must measure your hub and rim to determine spoke length. The better your measurements, the more accurate the spoke length prediction. Here’s a tool to make that easier.
To build a wheel you must measure your hub and rim to determine spoke length. The better your measurements, the more accurate the spoke length prediction. Here’s a tool to make that easier.
The goals of wheel building and racing are the same—speed and efficiency. I am not a fan of most side stressing schemes for wheelbuilders because I believe they add time but not quality. The intention is noble—to make wheels more stable by relieving pent up stress in the freshly assembled structure. This is often done with the wheel supported at the rim on a table and a piston applying side load to the hub or all spokes facing the piston. This isn’t a cheap or compact tool. Granted, some of their users are very smart but… Read more →
A severe side (lateral) load on a bicycle wheel is bad news. Wheels are strong for radial (vertical) and torsional (twisting) loads but vulnerable for lateral. Fortunately, high lateral forces are uncommon with single track vehicles that lean through turns. Still, they account for most wheel failures and the actual dynamics remain poorly understood by many.
Let’s examine an extreme lateral load and discover how a wheel responds. How can a tensioned structure with balanced, triangulated bracing warp instantly into an unrideable potato chip shape?
A lateral force usually arrives from the tire-ground contact, perhaps from skidding. The load moves the tire to one side which, in turn, pushes the rim to the side. The rim resists the force with both bending and torsional stiffness. The force is felt by spokes in the immediate region.
A spoke on the the force side tends to maintain its tension. A spoke opposite the force loses tension. The rim does not budge until the opposite side spoke reaches zero tension. At zero tension (all within milliseconds) all forces still balance out, and the rim is undeformed.
Once the opposite side spoke buckles, the rim begins moving. Would be helpful if the force side spoke, tight from the beginning, could increase its tension to resist the force. However, due to the low relative strength of bicycle rims, the rim begins moving away from the load in an arc whose radius is the force side spoke. The rim section swings at the end of the force side spoke as a yo-yo swings on its string going “around the world.”
In practice, the force side spoke cannot increase tension to keep the rim stationary. The rim is too elastic and swings rather than remaining the same distance from the wheel’s theoretical center. In this example, the lateral force is extreme and our wheel now has a pronounced side bend at the site of impact. Neighboring off side spokes are unloaded as the rim moves their way. Their reduction in tension makes for a positive push that creates two bends that are reverse of the side load. Such “wings” also occur next to broken spokes.
Depending on the magnitude of lateral force, the flexibility of the rim and the amount of spoke tension; a sine wave can propagate around the wheel, driven by the reduction of spoke tension in half, or more, of the spokes. All energetic states seek a lower potential. The loss of spoke tension drives the sine wave, bending the rim into a potato chip shape. This has been triggered by the initial extreme lateral force, but the eventual collapse is an implosion of the tensioned structure.
This potato chip shape, also known as a pretzel or taco, is technically a hyperbolic paraboloid. You can trace the edges of one on the Dumbo ride at Disneyland. When Dumbo swings low, its radial arm acts like a spoke that is maintaining tension. Swinging high is the same, but like an opposite side spoke.
In wheels with extremely springy rims that do not easily take a permanent set, a potato chip wheel can sometimes be fixed by banging sideways on the ground. A reverse extreme lateral force can cause a sine wave that neutralizes the deformation. It’s a chancy strategy but if you witness one, you’ll know it was special. Steel and wood rims are better for this than aluminum or carbon.
If you travel around the rim like a bug (or Dumbo) it will be obvious the rim is deformed left and right but also twisted. It’s a torsional deformation, not simply left-right. Rims with great torsional rigidity resist taco shapes much better. Torsional rigidity increases as the 4th power of the radius of the largest inscribed circle that fits in the rim section.
The rim above (from 15yrs ago) can fit a large circle in its cross section, giving it extraordinary torsional rigidity. It is nearly impossible to twist into a taco shape. It’s not just the little dude inside making this rim strong. Many deep carbon fiber rims have high torsional rigidity. Pushed by an extreme lateral load they either bounce or break. Notice how many of today’s more enlightened rims feature large open space in their profiles? Good design!
A taco wheel often displays four lobes, two in one direction, two in the other; and one is usually larger than the others. Safe to assume the extreme side load was delivered at this spot and the other three lobes formed as a consequence. Trying to true this wheel is wasted time if the rim has taken a permanent set.
Loosening spokes will reveal how extreme the bend is. Once de-tensioned, the true rim shape appears. If you must use the rim again (think: remote location expedition) try and make it as straight as possible without spoke tension. Be very careful to not introduce kinks or dents as you lever the rim to flat. Once the rim is flat, it will support spoke tension again and have no residual tendency to return to a taco. Truing without first bending the rim makes an unbalanced structure prone to deform again.
Before we leave the world of tacos, pretzels, and potato chips, let’s notice one further way a wheel can assume a multi-lobed deformation—excess tension. A rim can only support so much total tension, in proportion to its mass and shape. We used to build a favorite rim to 100kgf per spoke, with 36 spokes. The same rim with 48 spokes could not support 100kgf on each spoke.
An over tensioned wheel can “pop” suddenly into a wavy pattern. It’s the same principal as extreme side loads. Such a deformation will have many (not just 4) lobes, each small deflections. If you lower the tension total, with luck, the rim may behave as if nothing happened. It was just trying to reach a lower energy level. In its wavy state some spokes lost tension, others maintained. The total was lower.
Whether you ever undertake such truing, it is important to understand the forces involved. A wheel can suddenly become a taco shape with an extreme side force. In slow motion, this begins at one spot and a sine wave of deformation travels around the wheel often leaving a 4-lobed taco shape.
When it comes to metal bending and tension, most everything is reversible so, even if impractical, such wheels can be repaired. With these principles in mind, try your hand at it. You may not succeed but the exercise is always informative. Experience at the fringe of possibilities is what makes developing world bike mechanics so resourceful. Hone your own skills, don’t let them have all the fun!
Trick procedures are as important to your tool box as the individual tools. Please share them whenever possible. In that spirit, check these:
Proper dial indicator use
Indicators magnify movement and provide numbers to better judge trueness. However, your standard dial probe indicator is not made for rapidly moving surfaces. Even when an indicator is fitted with a bearing, the measure surface (rim brake track or edge) should only move slowly for two reasons:
1/ The probe is not designed for sideways force. Its accuracy is wrecked if the probe bushing wears.
2/ The probe return spring is as weak as possible. Consequently the probe can jump off the test surface with a tiny bump and vibrate with a pattern of roughness. To navigate a rapidly moving surface, the probe would need a strong spring and a dampening mechanism to maintain contact.
Spinning a wheel on a truing stand is normal when a light gap is used to watch trueness. A gently spun wheel can turn at 60 rpm. With a full sized wheel, this is 4800”/min, the same as a 1/4” drill at 6,000 rpm. This is not slow speed!
So, move your wheel at any speed for visual truing but turn it slowly with dial indicators. Don’t let your measuring instrument use cause machinists to cringe!
Builders deal directly with hub bearing play. Play interferes with truing, even when it’s too little to bother riders. In some hubs, play can be adjusted to zero for the build and then returned to the recommended amount for use. In most cases, however, it must be tolerated.
How to measure play? Axle play is magnified 10X at the rim but quantifying is delicate work. Hold the rim where a dial indicator is located. Give it a slight lateral force, left then right. The wheel is easy to flex so your finger force must be extremely light to reveal bearing play only.
What is reasonable? No single answer exists for all wheels. One number many experts would approve is 0.008” (0.2mm) at the rim (TIR—total indicated runout). Such a reading at the rim can be produced by less than 0.001” (0.02mm) movement at the axle. These numbers are at the very extreme of manufacturing tolerances for consumer products. More accuracy may be needed by NASA, but not us.
Heat Guns rock
A must around any shop is your standard 1500W heat gun. Like anything with voltage and heat, special care is important—flammables must be far away and good ventilation present. Some of my favorite uses:
1/ Removing adhesive vinyl stickers from nearly anything. Vinyl stickers lift off effortlessly with the right heat. Use less until you discover the perfect amount. Be careful not to damage your surface.
2/ Heating metal so stuck screws or bearings can be removed. Heat makes metal expand, each material with its own CTE (coefficient of thermal expansion). Aluminum expands much faster than steel. But even in steel-to-steel assemblies, a larger unit will expand faster than a smaller (like a stuck screw). Frozen nipples are a good example.
3/ Drying touchup paint, adhesives, spoke prep, rinsed chains, etc.
4/ Removing old sewup glue from metal rims. Heat then scrape or wipe with steel wool.
5/ Lubing old leather saddles. Apply a preservative cream or wax, heat the area and watch the saddle inhale the lube. Be sparing, it’s easy to make a crispy old saddle way too oily.
6/ Paint removal where the substrate can take the heat. This works on metal or wood. Paint softens before burning and can be scraped or wiped off.
7/ Applying shrink wrap. Handy shrink wrap is available is many colors and diameters, used extensively in electronics. Find ways to employ it on bikes—bar tape or cable end finishing, for example.
Got some other tricks? Wheel specific? Please share!
As gravel beckons, where are we with disk road wheels?
Disk brakes have come late to road bikes and, until recently, have met a cool reception. Two questions arise:
1/ Will the feel and response match the best rim brakes?
2/ How well will road wheels support disk brake loads so fundamentally different than rim braking?
As we northern climes head into damp weather and cyclocross, issues of corrosion deserve attention. Corrosion cannot be ignored but, for cycling, most of its challenges can be addressed.
Beware viewing extreme cases of corrosion on the Web. An incident only matters in terms of frequency. Being one of 1,000 witnesses to a wild case may not have statistical validity. Shape your practice around incidents that are representative.
Corrosion ~ oxidation ~ redox ~ rust
Beware, I’ll be interchanging these terms, most unscientific! To begin, corrosion is universal. Everything oxidizes under some conditions, life depends on this chemistry. It simply must be controlled and limited to an acceptable rate so we get good use from our wheels.
In general, corrosion is not a main driver of component failure, but is always present at the scene and often a player. Where a failure has occurred, note all the detail you can but be cautious assigning relative importance to each factor. A badly rusted hub bearing may show extreme wear but as it takes time and/or negligence to blow bearings, the coincidence does not define causality.
Corrosion is related to both material and environment. Warm, humid, coastal areas with onshore breeze and mist are hugely more corrosive than elsewhere. Components designed for average conditions need more care (rinsing) in FL or HI. In those areas, brass nipples can turn black with corrosion and crumble to dust. If you respond by cleaning and lubricating you might not feel aluminum is such a bad nipple material. Its corrosion is similar to brass and, maintained, can be reliable.
Corroded hub shells are ugly. Chalky stains and pits can be impossible to remove. Anodizing provides a first line of defense. For hubs that are uncoated, a high polish is more corrosion resistant. Campagnolo’s early hubs were polished and builders cleaned and shined them before a build so subsequent maintenance was easier. Surface treatments like original WD40 go beneath water, clinging to metal surfaces. A light spray or wipe can minimize corrosion.
Today’s better spokes are made of stainless steels, corrosion resistant enough to not require protective coatings. These steels can corrode so don’t be surprised to see a light haze of rust. Wipe it off with an oiled rag. The black color applied to many stainless spokes is decorative only and not as corrosion resistant as the underlying metal. A light haze of rust on a black oxide coating is easily wiped off with an oiled rag and will not rust again until the oil is removed.
Remember that multiple stainless alloys are used for spokes, each with different metallurgy and corrosion resistance. Likewise, there is no single black or color coating. Integrity of a coated spoke is difficult to predict without access to manufacturing details.
For non-stainless spokes, galvanizing is a common corrosion prevention. It begins as a shiny, attractive layer, quickly becoming dull, industrial gray. This is normal. On show motorcycles and antique bicycles we see chrome or nickel plating. Neither has much corrosion resistance without regular wiping and application of a protective polish. I’ve seen bad rust on non stainless spokes but rarely breakage for which the rust could be blamed. Of course, it happens but is rare on in-use wheels.
Wiping spokes with a rag lightly soaked in an oil of your choice is good practice. Pressure washing removes dirt but dirt doesn’t cause corrosion. Washing removes the protective coatings all components need. We’re in a modern world of materials and coatings but superficial oils and waxes still play a key role for weather resistance.
The debate over brass vs aluminum nipples is eternal. The issue should be decided on mechanical attributes, not corrosion resistance (IMO). Both are quite similar and a well anodized aluminum nipple can equal brass with a nickel plated finish. Brass is more ductile and self lubricating. High grade (2024 and 7075) aluminum is stronger, can be brightly colored, and a bit lighter.
Check this chart for comparable corrosion. Notice columns for aluminum and brass. For exposure to sea water, aluminum is better. For distilled water, both are excellent. For vinegar, they are equivalent. For whiskey, brass is better. Two metals with different outcomes depending on environment. I’m for the whiskey.
In navigating the “controversy” remember that much unwelcome behavior by any single batch of nipples may depend on specific conditions. What, if any, lubricant is used? What process shaped the nipple (forged, cast, machined)? What plating and specification (depth, penetration) was used? Silver nipples are not necessarily anodized. Are yours? Notice also that many millions of aluminum nipples have been used on production wheels over the past 20 years; most outlasting their spokes, hubs, and rims.
Brass contains lead for formability. Recent law restricts the amount but the historic presence of lead in brass taints its otherwise classical charm. Regardless of nipple material, regular lubrication is key. If you soak nipples in heavyweight oil prior to building, the coating can last years. If your wheels see pressure washing, relubrication is required.
To lube a built wheel, touch a drop of light oil to the spoke as it enters the nipple. Watch it wick down the bore. A second drop belongs at the nipple-rim contact, applied from the same side. Spinning the wheel creates centrifugal force to drive lube into the assembly. Many of today’s chain lubes are ideal for this as they are thin for application and penetration but as their carriers evaporate the remaining formula becomes dense and long lasting.
Cracking of aluminum rims under tension at nipple holes can be accelerated by corrosion. After anodizing, rims are purged of liquids used in the electrochemical process. Inadequate flushing can leave seeds of corrosion deposited inside the rim. Anodizing can be too deep and hard, leading to surface cracking where the rim may be deforming from spoke tension.
Tire sealants can also be corrosive agents and leak into rim interiors, especially those with ammonia.
Carbon rims are neutral for corrosion but can act as anodes when in contact with metals, stimulating corrosion. It is relatively easy to isolate and protect such galvanic combinations with anodizing, washers, and lubricants. For better or worse, the vast majority of carbon wheels are built with aluminum nipples, showing the combination is not automatically bad news. Enve’s recent switch from aluminum to brass for internal nipples does not indict all use of aluminum. Many of the problems they observed are owed to ammonia from tire sealants and poor anodizing on nipples. Substantial numbers of carbon wheels show no aluminum nipple corrosion. There is often more to the story than simply nipple material.
Storage batteries in use or awaiting disposal can create electrical fields that produce very corrosive conditions. When you see an “impossible” corrosion example, wonder about the context. With growth in e-bikes, e-cars, and home energy storage, we can expect more such cases.Electrical fields…?
In following corrosion, make careful observations and notes, utilize magnification and photo images, and use components with highest material and coating properties.
This upcoming La Niña winter should be fun, not corrosive!
Nipples aim from their rim holes in order to align with spoke angle. The goal is to have no bend in the spoke as it enters the nipple because threading there is a weak spot. Spokes can fracture in thread roots after many loading cycles. Nipple and rim shape determine the amount of spoke angle that can be accommodated. The angle is a product of hub and rim size, spoke number, and crossing pattern.
Today, rims are frequently smaller (650 instead of 700, 20” for folding, cargo, and e-bikes, etc.). Hubs can be larger (generator, internal gear, e-motors, etc.). These combinations produce spoke angles that are more concerning.
A perfect spoke angle is 90deg with no bend in the spoke. Side angles generated by hub width and dishing are rarely below 80 and most nipple-rim combo’s handle them. The angles we need to address are in the rim plane and a function of rim and hub size.
Grin offers a great spoke calculator, among many valuable resources. It will determine the spoke-rim angle and, incidentally, works with paired spoking (another topic). Nipples can easily aim for entries of 80-90 degrees. Less than 75 may be accommodated. Below 70 is beyond most components and requires special attention. Solutions:
1/ Kink the spoke with a plier or wrench so it enters the rim at the nipple’s angle, a slow but effective process.
2/ Drill holes larger so nipples are freer to pivot, not an option or advisable on many rims.
3/ Reduce the lacing cross number to make an angle closer to 90. Despite sub-optimal torque geometry, many builders are doing it. As Grin says, “In spite of the popular wisdom not to use radial lacing on drive wheels, empirical experience has been that this isn’t really an issue with the large hubs in small rims.” The burgeoning e-bike scene cannot be slowed down even though appropriate rims are not available.
4/ Rims could address this issue, for example, with a bulge at each nipple. Motorcycles figured this out 100 years ago. Drill the rim to accommodate the required angle. Here is a solution with optimal spoke angle and torque transmission. Cycling will certainly figure this out soon.
Anticipate spoke angles and plan accordingly. An engineered solution to the situation requires initiative from rim makers. Let’s hope it is sooner than later so I can stop envying motorcycle wheels!
You’re building along when a nipple runs out of spoke thread. You feel an increase in friction as unthreaded spoke shaft encounters nipple threads. This can occur with a slightly too long spoke (or too small rim). What to do? Consider turning the nipple further into this stiff zone. Why?
1/ Nipples are softer metal (brass and aluminum) than spokes (steel) and their threads yield with minimal resistance. Also, spoke thread is rolled so peaks are above the spoke surface and valleys, below. When the nipple encounters the spoke shaft, only 1/2 of its thread must yield; far less interference than between two machined threads.
2/ Spoke threading, as with most thread forms, carries the entire tension load on 3 or 4 threads. The rest are available but not load bearing. Driving a nipple down a spoke a few turns past threading has inconsequential effect on load carrying capacity.
3/ Testing shows that nipples support spoke threads. When a few spoke threads are not engaged and when the spoke endures riding induced load cycles, fracture can occur in those empty threads. Full engagement increases spoke fatigue life for thread fracture. It would be an enlightened building strategy that forced all nipples to be turned further so no spoke threads were empty.
4/ A nipple driven further down a spoke develops friction that helps prevent vibration induced loosening, a good outcome for any hard-used wheel.
Beware of excess spoke protruding from the nipple inside the rim, it may interfere with the tire. Otherwise, please remember the possible benefits of tightening nipples past the spoke threads. Not all bad, in fact, there are some interesting benefits.
Now is a good time to introduce a new product, a better spoke ruler than any before. This ruler is by Pi Spokes, a very interesting project of which you will hear more in the future.
1/ At long last, a ruler marked in 0.5mm increments. These finer increments are easy to read (or ignore). Let’s face it, spokes are not all made exactly to whole millimeter lengths. Builders deserve to round up or down with complete accuracy. Spoke machines like our Morizumi can also cut precisely to 0.5mm so a finely graduated ruler is perfect.
2/ The spoke lies in a groove below the scale so the issue of visual parallax is minimized. Lengths can be more quickly established, even in poor light.
3/ One side is for J-bend elbows, the other for straight pull. The ruler is made of highest grade aluminum, scale is permanently laser etched over bright purple anodizing. Price is $20.
For current or aspiring bicycle mechanics, this is an exciting time. E-bikes, advanced materials, fun new categories, more diversity, mobile service, better tools, wow…the list keeps growing.
Check out the, just announced, technical workshops held by PBMA (our fledgling but dynamic mechanic association).
Coming later this year and next to Colorado, Portland, and Virginia. Most of the leading industry players are participating.
There is room for just 120 mechanics at each so early registration is the only way to guarantee going. There are scholarships available, so be bold and apply!
Let’s talk rotating weight. Rotating weight is a big issue for wheel builders. Why? We make choices that determine wheel weight, its total and location. Builders must understand this topic.
Rotating weight directly affects inertia, so the topic is really inertia. Inertia is the resistance of a mass to acceleration. Moment of inertia (MOI) characterizes this resistance and depends on rotating and non-rotating mass. Builders should be measuring wheel MOI.
I’ll show you how to measure moment of inertia (MOI), plus I’ll share a spreadsheet to shortcut the math supporting MOI measurement and its effect on riding. Plug in numbers and get usable wattage estimates. Read more →