Like many kids growing up in the 1980’s and ’90s, I was a big fan of public broadcasting shows like Reading Rainbow that would occasionally delve into how things were made, with cool factory tours where the camera became my eyes. LaVar Burton was the smart big brother I never had. Sites like How Stuff Works do a fantastic job of feeding similar wonderment today, and we try to do our part on the mountain bike side of stuff whenever possible. Some examples would be our recent dive into how tires grip, loads of articles on trail building, carbon rim construction, and MTB frame design, to name just a few.
The other day our Chief Editor, Jeff Barber, sent me down a research rabbit hole to find out how brands select all of the fasteners that hold our bikes and components together. Why does one seat post clamp have a 4mm head, and the other a 6mm? How much does weight factor into the decision, and how accurate are torque recommendations? We emailed several industry professionals to find out.
Fastener heads selection
Based on the responses we received, it seems that bolt-head selection is largely convenience-based. Brands want to use heads that are available on most multi-tools, with the exception of some Torx heads that are selected for their higher torque capabilities. Additionally, Torx heads require a straight line to the bolt head, where a hex can be turned with the tool at an angle, making hex the better head in tight spaces.
Cy from Cotic Cycles said that “We generally try and stick with well used 4,5,6,8 Allen keys and T25 Torx only where we need access and high torque in particular locations. There’s no heavy engineering reason about this, it’s more about what’s best for customers.”
Product engineers at Ibis Cycles echoed Cy’s sentiment. “Whenever possible we only use heads that will work with what’s on most multitools. Torx are nice for high torque applications but the only size most people have is a T25. That’s too small for most frame pivot hardware so we just use hex’s everywhere.”
Mark Kikkert at PRO dug into some of the other variables that need to be considered when selecting fasteners. “The selection of the bolt tool interface is only one part of the equation. In terms of bolt heads, there’s also the choice between shapes like round, countersunk, flat, or hexagon heads. There are many other types, but these are probably the most commonly used ones in the cycling industry. Things that play a role in the selection of the bolt shape and tool interface are: bolt strength required, bolt size, accessibility to the bolt, and product design.
“As an example: countersunk bolts sit flush with the surface, but usually have a smaller tool interface. If there’s a high torque preload required then you would need to oversize everything to get proper tool engagement. For low preload torque applications, it’s usually ok to use flat or Philips screws or bolts. Hex is the better choice for higher torque levels, and Torx is even better as the tool interface has less chance of slipping and wear-out.”
Most bolts on a bike will fit in every hole that’s the same diameter, as the pitch of their threads is identical. For example, the rotor bolts will fit in the bottle cage mounts, most brake and shifter clamps, and several of other orifices. If you lose a bottle cage bolt on the trail you can stop it from flopping by borrowing one of the twelve disc rotor bolts. The rotor should stay tight with five fasteners until you get home. Back in the legit weight-weenie days, a lot of XC racers would only use three Torx bolts to hold their disks on, saving a total of not-very-many grams. Anyway, how do brands decide what the thread pitch should be for a given component?
While nearly everyone said they use a standard thread pitch wherever possible, Kikkert did call out the filth factor of bike bolts.
“For most fasteners we use standard thread pitch. Only in special cases the pitch is changed. If it requires fine adjustment or there is limited space available, like with the dropper post collar, a finer thread might be picked. There is a downside for finer threads and that is in ‘dirty’ environments there’s more risk of the thread seizing up, so for these parts we do recommend the use of anti-seize compound. Most visible bolts on the bike components will almost always be regular thread pitch.”
Component engineers from Rotor Bike Components added the brand’s perspective on finer thread pitch, stating that it works well with shorter bolts. “If you have two screws of the same metric, the fine pitch screw has a higher tensile strength as it has more contact surface, and the fine pitch screws loosen less than the standard pitch screws. For applications where you need short screws to withstand load and vibration, the fine pitch is your friend.”
A bolt’s use frequency also affects its chosen thread pitch, as the folks from Ibis Bikes noted, “coarse threads are used if the bolt is going to be taken in and out a lot by the rider because they’re more durable.” Axles and shock mounting bolts would be two examples where coarse threads can make an oft-removed component bolt last longer.
How about fastener materials? What are they made of?
You may have heard stories of some professional racers using aluminum or even plastic bolts to drop weight where possible. How is fastener material chosen for our stock bikes? Similar to thread pitch, it seems to depend on where a fastener will be used, and what forces are acting on it. Wherever possible, most companies choose lighter hardware.
Janek Ulatowski from NS Bikes shared their philosophy. “We use aluminum bolts or axles in places where we are sure it won’t affect the safety or strength of the joint. The only reason for that is the weight reduction. Recently we’ve started using extremely durable Ti bolts which are used in critical places like the shock mount. The additional value of them is the oil-slick finish, which gives an extra “bling” to the whole bike.”
Cy mentioned a similar focus on where the fastener is and what loads will be placed on it. “Depends on how much space you have. For instance, the seat stay pivots on our Droplink bikes are 8mm stainless because they can’t be any bigger diameter within the confines of the tubing we use at that location, and they need to be hollow so an M5 screw can thread into the other end. Aluminum simply isn’t strong enough with that constraint on space and wall thickness. At the other end of that linkage, the pivot goes through the seat tube where there’s plenty of space, so a nice big 15mm aluminum hollow axle is much lighter and stiffer than something smaller in steel.”
The standardized strength measurements that sometimes appear on bolt heads are a starting place for Rotor. “There are international standards that define the quality of a screw and its resistance to both traction and shear, as well as the materials. The numbers 8.8 and 12.8 that usually appear on the head of the screws inform us about their quality and resistance. The amount of force the screw has to endure will tell you which screw to choose.”
What about the materials being fastened together? How does that change things?
Now we have an idea of how fastener materials are selected, but how do the materials being fastened together affect the selection process? For example, I recently worked on an aluminum full suspension frame, and when rejoining the front and rear triangles I managed to miss a steel washer that belongs in the lower pivot. That missing washer resulted in alloy-on-alloy rubbing that sounded like the bike was eating itself.
According to Thomas Westfeldt of Öhlins it’s all about the material that a fastener is tightened into. “Generally, limitations are set by the material the bolt is attached to. In our case, it is often aluminum and then the strength of alloy will set force limitation.”
Ruben Torenbeek from Raaw MTB said that his focus is entirely on durability. “Weight gets added quickly with small bits of hardware. With our focus on durability and functionality, we have over 700 grams of hardware in one frame.” That’s a decent portion of the frame’s total weight dedicated to fasteners and bearings. There are featherweight carbon XC frames that weigh scant more than the hardware in a Raaw frame, and thankfully Torenbeek’s bikes are designed for a very different kind of fun.
Some materials can create issues when clamped too closely together. Kikkert said that “Stainless steel and titanium bolts seldom give any issues. Alloy on alloy might be a potential risk for galling so we try to avoid that combination as much as possible.” Two pieces of aluminum rubbing together can not only cause damage but are also a common source of creaking in a frame. Steel washers and fasteners between aluminum components can take care of those issues.
Torque specs: Are they accurate? Do engineers get to break stuff to find out?
It stands to reason that product designers get to break the stuff they make occasionally. Someone has to see how much force things can take and what happens when they hit the overload mark. Nearly everyone we chatted with on the topic mentioned the inaccuracy that’s inherent in most torque wrenches. Those mis-calibrated tools are something they have to account for in product spec and design.
At Rotor, there are a few steps prior to the “let’s beak stuff” party. “Knowing the forces that the materials support and the force that is going to be applied when using the bicycle (80kg in the stem screw is not the same as 2 kg in the screw of the water bottle), we would already have the theoretical values to select the screw. Then you have to do the real tests and break many to make sure they are correct.”
For steel frames things work a little differently, though they still get to break things at Cotic. “There’s a fairly well understood torque setting calculation that I use based on a percentage nominal preload, elongation, the ‘springiness’ of the items involved in the joint, but particularly when using off the shelf steel fasteners, the setting pretty much is what it is. For the threads I design myself (like on the pivot axles, for example), I’ll calculate the theoretical torque, then we’ll break a few to make sure that the breaking torque is higher.”
The folks from PRO also calculate first and break stuff second, designing things so that the bolt breaks first whenever possible. “For sure we break things to find out the limits, but initially we design them to stay in one piece within the expected load limits. Bolts are relatively standardized in a way that certain sizes, materials, and grades also have ‘fixed’ maximum load-bearing capacity. We try to match the bolt with the intended use. A recommended torque spec is usually so that a (much) lower torque might increase the risk of slipping, whereas (much) higher torque increases the risk of part failure. For example, a handlebar limits how hard you can tighten the stem clamp down before it fails, so we pick bolts that might fail before handlebars do, to minimize the risk of damaging the most expensive parts on your bike. A bolt is easier to replace.”
Torenbeek added a little comedy to the question of how far bolts can be over-torqued. “Very far. The thing that usually fails first is the tool-interface, which requires a very red face.”
On the topic of over-torquing, Cy adds that home mechanics need to use some common sense, and understand the feel of a tight fastener that doesn’t need to be turned further. “I have a bit of a pet-hate about people slavishly sticking to torque settings when using cheap wrenches. Especially in safety-critical areas like stems. I have seen bars move in stems when someone has used an exact torque setting instead of simply ‘doing it up tight.’ We’re dealing with such small torques on M5 and M6 bolts, and cheap torque wrenches (and by cheap, I mean less than £200) are often so badly calibrated that the 5Nm on that stem bolt is more like 4Nm. Now, on bigger bolts, 1Nm difference is neither here nor there. When I worked on the railway we were bolting things to trains using 250Nm wrenches. But on a stem bolt, if that cheap, uncalibrated torque wrench is 1Nm out, it’s 20% under torqued, so the bar will move, and the rider will hurt themselves. We have had it a few times when customers insist that they’ve used a torque wrench, but it has not tightened something up. We get it back, put our calibrated gear on it, and find it’s way under. Torques are there for your safety, but use your feel and your sense a bit too. If that bolt doesn’t feel tight enough at the ‘correct torque’ on your wrench, it probably isn’t. “
What about thread-lock fluids? When do we need them?
So some bolts use a finer thread pitch to keep them from rattling loose, while others like axles are more coarse so that they can easily be removed and reinstalled more frequently. At what point does a brand decide to add thread locking fluid to a fastener to keep things in place? A lot of disc rotor bolts, for example, come pre-coated in “blue 242” and having lost a few on the trail I would certainly add some if they didn’t.
For Chris Deverson of Deviate Cycles, the answer is often. “We use a medium Loctite for most of our fasteners for peace of mind. With our sealing system and bearing layout, you should rarely need to remove them. We’ve only sold one set of bearings so far.”
Rotor deals with smaller parts than the folks at Deviate are creating, and their engineers similarly prefer the peace of mind method. “There are different types of Loctite and they can be used on almost all the screws on our bike. I recommend it on those screws that can tend to loosen such as the screws of the chainring, the saddle, or the stems.”
PRO is also on board with thread-lock across the board, though not the near-permanent variety. Kikkert said “On the bike there is the risk of bolts running loose because of vibrations. Therefore we use and recommend the use of thread locking agents on bolts that are not too permanent so you can always loosen them again if needed.”
With a little spin on the subject, Ibis too wants to stop the bolts from rattling free. Their twist on thread-lock is this: “The only time we don’t use Loctite is on a bolt that’s acting as an adjuster. Mountain bikes are subject to constant vibration and complex loading so almost anything can rattle out eventually.”
What’s the deal with Torx heads?
It seems that more high-end components are coming with Torx heads these days. Are they, and if so why?
Most of the folks we spoke with said that Torx heads are harder to reach on the bike because they require a direct line to the bolt head, where hex heads don’t. Cy said that “we try to avoid Torx, mainly because unlike Allen keys, there’s no such thing as a ball end Torx. So, access has to be straight into the head, which often is a bit tricky on a mountain bike.”
The engineers at Rotor feel that the star-shaped bolt trend will prevail. “Torx screws allow you to reduce weight and volume and your wrench is more manageable. They will end up imposing themselves on the market.”
Over at Öhlins, they like the Torx for a few reasons, but prefer to use bolt heads that riders have in their pocket or pack. “Torx can manage a high torque for small screws compared to hex and Philips. It is also slightly less prone to rounding. We strive to make our products consistent and rider-friendly by using the tools you have with you on the trail.” While a lot of multi-tools have at least a T25 Torx driver for tightening 6-bolt rotors, they don’t all include the little star bits quite yet.
In a perfect world, what would be different about fasteners?
The ol’ “in a perfect world” question should be part of every discussion with product designers and engineers. If there was no user error to worry about, cost wasn’t a factor, and companies could select fasteners to achieve a perfect strength-to-weight-ratio, what would they do differently?
For Öhlins, the perfect world has no torque specs to speak of. “In a perfect world, the rider would not need to tighten any bolts to a specific torque; Super easy interfaces that click when properly fastened would be great. One challenge when developing improvements, like our floating axle design that ensures minimal friction, is that the common way of doing things that hinders progress. For example, tightening the front axle like the systems used in most of our competitors. If you overtighten it you risk damaging the lower. If you then skip the alignment process before tightening the clamp you do not get the benefit of a smother fork.” This video shows a visual representation of the axle mounting procedure that Westfeldt is describing.
Engineers at Rotor would love to simplify things further. “In a perfect world, the bolts would be the same but people would worry about checking their tightness every few thousand kilometers and would change them once a year if the bike is heavily used.”
Over at Cotic Cycles, the shift would be toward lighter and stronger in tandem. “If cost was no object, I’d use Ti fasteners in place of steel throughout just because it would save weight, but it’s hard to make the value-for-money argument!”
Too-light and too-tight mistakes
All of this testing and torque spec can’t always hold up as we hop. We thrash mountain bikes fairly hard, and occasionally things break or wiggle loose. We made sure to ask everyone for stories about their fasteners causing issues.
The folks at Ibis Ibis shared a story of overdoing the thread-lock on some frames. “We had one issue where I built all the prototypes with one drop of Loctite per bolt and they came apart just fine. We then had a guy in production who felt he was doing a really good job by thoroughly coating both bolt and nut threads in Loctite and suddenly people couldn’t get their frames apart without damaging them. Now we’re more specific with our frame assembly SOP’s [standard operating procedures].”
Issues with tightening different components can sometimes come with shifts in the industry, where not everyone has the necessary torque info right away. Over at NS bikes, the problem was with a dropper post on some of their models. “As the products and technology develop at a fast pace there have been some issues we faced. It is really hard to avoid. One of them was when dropper posts appeared in the industry. Having in mind our bomb-proof dirt bikes we tend to over tighten the seat post clamps which could result in poor performance. Once we found out the issue we introduced the [torque] procedure. “
And lastly, yet no less intriguing, Raaw bikes has had to shift their torque spec up a bit to meet forces placed on their sweet gravity frames. Torenbeek said “We had to up the torque on the main pivot, because we underestimated the clamping it needs.”
That’s the nuts and bolts of it. We’ll see you right here on Singletracks for the next installment of Riding Rainbow!
Show your Support
Become a Singletracks Pro Supporter today and enjoy benefits like ad-free browsing.
With your support we can provide free worldwide trail information and original content created by our team of independent journalists.
Love this nerdery. Thanks!
Fasten-ating, ha! I like that companies do consider the tools we already have, though it does seem the number of shapes and sizes has increased over the years.
I’d rather have torx everywhere. Considering the cost of mtb components, I think one can also put money into a torx wrench set. Better than having to deal with stripped allen heads (well, it can also happen with torx, last time I’ve resorted my bolt extractor was for a disc brake screw).
The excuse of the poor accessibility of screws on bikes seems to me … just an excuse.
Just my 2c 🙂
Just be grateful the industry has not started using torque to yield bolts!
Those can be used once, safely however, are junk after a second use.
How can those guys not know that there are ball end torx drivers?