VeloClub is CyclingTips’ membership program which brings us closer to our members, and connects likeminded cycling enthusiasts.
by Dave Rome
March 28, 2018
Photography by David Rome
If you’re one that likes your bike running like a finely tuned machine, you’ll likely already know that chains often need to be replaced as often as your rear tyre. But have you ever considered what’s causing this rapid wear? Simply, it’s friction. Yep, the same friction that’s costing you those precious watts.
Friction occurs in multiple ways inside a chain, but most importantly, it’s the chain’s articulating points of the rollers, plates, and pins sliding and wearing together that causes a chain to grow in length (commonly known as chain stretch). If ignored, that longer chain rides high on the teeth of your cassette and chainrings, causing devastating wear and a big service bill.
Previous tests done by Jason Smith (formerly of Friction Facts) revealed that poor chain lube selection, or even just a dirty chain, can cost you upwards of five watts (at a constant 250W effort). Pedal harder and those lost watts go up in a near equal fashion, too. That’s arguably more than any equipment upgrade available, and it’s applicable no matter how fast you go.
Australian Adam Kerin is one such person who became obsessed with the idea of how a simple chain lube selection can not only make you faster, but can potentially greatly increase the lifespan of your drivetrain. Want your chain (and drivetrain) to go far further than you’ve ever seen before? Want some free speed, too? Read on.
When Jason Smith’s Friction Facts company was acquired by CeramicSpeed in 2016, the world lost its lone independent authority on what chain lube is actually best. But using his own methods and measurable outcomes, Kerin has since resumed such independent tests, all to reveal the best chain lubes available, and shine a light on those offering little more than marketing fluff.
Where Jason Smith funded Friction Facts by selling his findings and offering brands access to his testing and expertise, Kerin’s business model is designed to find the best-performing products to sell through his online business, Zero Friction Cycling (ZFC). Over time, he hopes his testing and published content earns him the trust of shoppers looking for speed and/or longevity upgrades.
“I feel like Adam and ZFC are like I was with Friction Facts five years ago – a lone voice looking to find honest answers to a lot of unknown questions and debunk potentially misleading existing information,” said Smith.
Kerin’s tests focus on chain wear. Abrasion is friction, and friction is lost watts.
But where Friction Facts focused on efficiency, Kerin concentrates on durability. He effectively torture-tests chain lubes, with the measurable outcome being chain wear. Through a consistent and repeatable process, Adam runs chains through a simulated real-life death march. The chain lube that results in the least chain wear, despite being drowned in grit, earns the spoils (or in Kerin’s case, becomes available for sale), while the losers are given no such love.
The goal of the testing is to simulate real-world riding as best as possible. It’s not to find the most efficient lube over an hour-long time trial, but the most efficient lube for when you want to go on an all-day ride; ride in poor conditions; or simply are lazy, and don’t want to clean and re-lube with each ride. Kerin’s tests are applicable to the everyday cyclist, be they amateur road racers, gravel riders, or mountain-bike marathoners.
In Kerin’s tests, and monitoring his own and customers’ chains, he’s found lube choice and careful maintenance can be the difference between a chain being worn-out at just 2,000km, and one lasting 15,000km before it showed similar wear.
Ultimately, Kerin’s tests are designed to highlight the products that are likely to save you money through improved longevity, but there are still similarities to the work Smith did at Friction Facts. According to Kerin, chain wear and friction are inherently linked together, and by aiming to increase chain longevity, you also get the bonus of reduced friction.
“A lubricant will not be able to be a high-efficiency lubricant in the real world if it records a poor longevity test result,” Kerin explained. “If a lubricant is eating through hardened steel at a good rate, that just flat-out takes friction.”
Smith and Kerin don’t entirely agree on this theory, although that’s perhaps due more to the different approaches each has to testing.
“If contamination was removed from the equation, say, like in a track environment, or a very clean road ride, increased friction does not always correlate to wear,” Smith said. “For example, compare a newer-technology dry lubricant, such as (CeramicSpeed) UFO Drip, Molten Speed Wax, or (CeramicSpeed) UFO Chains versus a thick sticky grease. The dry lubricants, from an efficiency standpoint, will be much faster than the greased chain. However, the greased chain might show less wear over time.
“This is because of the effects of viscous drag and stiction properties of the grease,” Smith continued. “The grease’s viscous drag and stiction will increase the friction levels of the chain, but these two properties, although creating increased friction, do not increase the wear of the chain. However, the dry surfaces might create slightly more wear than the grease. When all factors are considered, the friction created by the viscous drag and stiction of the grease substantially outweigh any friction created by wear of the dry-lubricated surfaces, if indeed there are any increases.”
However, Smith explained that things change once the chain and lubricant are subjected to contamination over a long period of time.
“The driving factor of friction and wear when a chain is subjected to contamination is based on how much contamination adheres to the lubricant, or conversely, the ability of the lubricant/coating to repel contamination. When contamination enters the picture, the wear rate and friction becomes more correlated, since the contaminants heavily drive the friction levels. Therefore, in very clean conditions, I feel friction and wear are not necessarily correlated. Sometimes it is; sometimes it isn’t. Yet as contamination levels increase, I agree with Adam that friction and wear will become more correlated.”
Nevertheless, Kerin’s tests yield very similar results to what Smith has produced, with a consistently close correlation between the best wear rates and the fastest lubes previously published by Friction Facts.
Friction Facts concentrated on accuracy with its dedicated test machines and protocols, claiming measurable differences in frictional losses as low as 0.02W. In comparison, Kerin’s modified Tacx Neo Smart trainer setup looks decidedly amateur, but the science behind it seems sound. A motor, not a human rider, is attached to the crankset spindle, and provides a constant output of 250W at 100rpm in controlled riding conditions. And since the Neo Smart is mechanically unique in that it’s truly a direct-drive design with no additional belts or pulleys, Kerin only has to factor one set of bearings into his baseline equations.
Kerin sets a simulated riding distance of 400km between lube intervals (on flat terrain simulation), with the frequency doubling during contamination blocks, such as where fine sand is added. Each test block lasts for 1,000km, with alternating blocks without added contamination to help understand how well a lube works to clear out contamination. It’s a real stress test of how well a lube can go the distance.
Each test starts with a new chain, and variances for manufacturing are controlled by accurately measuring the chain before the test begins. Once the new chain measures 0.5mm of “stretch” (averaged across seven sections of the chain) over the initial starting length, the test is stopped. This is considered “100% wear” in Kerin’s tests, as the 0.5mm figure is a commonly suggested replacement point before measurable cog wear occurs.
ZFC’s main test of lubes. Once a chain reaches 100% wear (0.5mm average measure), the test is stopped. The lube that makes it through the most blocks with the lowest accumulative is the winner. Note, lower is better.
The testing blocks (explained in ZFC’s test explanation document) also provide Kerin with a clear view of where things go wrong – or in some cases, go well – and in what conditions the lube falls apart. Some lubes may be very strong for the first 1,000km clean interval, for example, but fail miserably once grit comes into play. And if you’re riding outside, grit will certainly come into play.
When asked specifically about Kerin’s testing, Smith was supportive.
“I appreciate what ZFC is doing and feel it is a good representation of longevity/contamination testing. It’s very difficult to identically recreate real-world contamination conditions, [but] if the contaminants are similar in composition to the real world, are controlled with respect to quantity and discharge rates, and are repeatable from test to test, then it’s a true scientific test in my opinion. Much longevity and wear info can be gained based on the results of his controlled testing, even if the protocol does not perfectly mimic the infinite number of unique real-world conditions.”
Done out of Kerin’s pocket without financial return (as of yet), the tests do have obvious limitations. Perhaps the biggest one is that only a single test is performed for each lubricant, meaning the repeatability and consistency of these tests is somewhat unknown. For some, that’s understandably a deal breaker; for others, one controlled test is still better than no controlled test at all.
When asked about this, Kerin explained he did almost 9,000km of baseline testing before actually measuring chain lubricant performance, to both bed-in testing protocols and also to check the repeatability of the testing.
“Chains themselves do not wear exactly the same even across the same chain,” explained Kerin. “So the variability comes down to the chain, and not the lubricant or the test protocol as that is easy to keep control over.”
A chain is unlike anything else on a bicycle, and unlike bearings, the individual links do not constantly turn or spin; instead, they repeatedly articulate and have a small range of motion. There are also an usually high number of friction points, including the pin and plate shoulder; between the roller and plate shoulders; and between the inner and outer plates on both sides of the chain.
How many, exactly? In a white paper published in 2014, Smith calculated that there approximately 40,000 individual pieces of mechanical movement in a bicycle chain when pedaling in a 53T chainring at 95rpm. With so much metal-to-metal contact, it isn’t difficult to see how a good chain lube can save you upwards of five watts of energy. Even better, the relatively low cost of chain lubricants makes them the best value upgrade you can make to your bike by far.
But in order to avoid such friction, it’s not what a lube does or looks like on the outside of the chain that matters most, but rather how it behaves inside, where the loaded friction is highest.
I mentioned earlier that while Smith and Kerin have taken different approaches to lubricant testing, the relative order of the test results support each other’s claims. Case in point: White Lightning Epic Ride was the least-efficient lube Smith tested at 8.9W, and also yielded one of the shortest chain lifespans in Kerin’s tests, lasting less than 3,000km before hitting that critical 0.5mm mark. At the opposite end, a chain treated with Molten Speed Wax ate up just 4.2W of rider energy in Smith’s testing, while outlasting Kerin’s 6,000km-long wear testing procedure (and proving in the real world, with real riders, to go past 15,000km), something no drip lube has done as of yet.
A simplified version of Friction Facts’ drip lube comparison. Note that these numbers are using sterile chains that run extremely fast, and so users are likely to experience higher numbers. Still, the relative order stands. Note that submersion wax lubes have been removed from this chart, and when this data was first published years ago, it was Molten Speed Wax at the starting end. We’ll be releasing the full data in a future article focused on efficiency.
“I have had some lubricants record over 100% wear (0.5mm) for a single contamination block, versus other lubricants that recorded 25% wear,” said Kerin. “That is a huge difference in the amount of hardened steel that has been abraded away by the higher-wearing lubricant. So no matter if the lab gives it a good result, in the real world that level of wear is bad news for performance.”
Friction Facts’ testing lifted the lid on a number of extremely popular lubes and mechanic myths. Hearts were broken as fan favourites such as Prolink, Phil Wood Tenacious Oil, and Triflow were all shown to perform poorly in terms of efficiency. However, many people in the know were willing to overlook such findings in favor of other attributes, such as clean running, corrosion resistance, dependable squeak resistance, or even just the smell.
But when you include the estimated running costs involved with that increased wear, Kerin’s test findings are much harder to ignore.
ZFC’s estimate on what it costs to run various lubes for a projected 10,000km of riding. Individual rider results will vary, but generally speaking, lubes that decrease wear will save you money. Costs are based on wear through test blocks 1-5, including Australian retail cost of lubricant, plus replacement chains, cassettes, and chainrings as needed.
Both Kerin and Smith agree that most chain lubricants traditionally described as “dry” are the worst performers for real-world riding.
“Dry lubes are mostly a carrier, which evaporates; they don’t last,” Kerin said. “They carry about 10% actual lubrication.”
Even worse, Kerin says “dry” lubes can actually make drivetrain wear worse than not using lube at all.
“[Dry lubes] can be an accelerant to wear compared to a bare chain. This is often because the lubrication that does exist isn’t perfectly dry, and so it can attract contaminants into the chain.”
If getting it on your braking surfaces or in your lungs wasn’t bad enough, prior testing by Smith also revealed aerosol chain lubes perform even worse, as they often contain even more carrier than “dry” lubes, which helps the propellant.
“From an efficiency standpoint, I suggest riders avoid the relatively thicker oil-based lubes,” explained Smith. “These lubes can exhibit higher initial friction levels due to viscous drag and stiction. Then, as the chain is subjected to contamination, these lubricants are like grit magnets, further acting to increase friction. I’d also suggest riders avoid dry lubes which don’t contain enough lubricant to be effective.”
There are exceptions to this rule, however, especially if extremely nasty conditions lay ahead, such as a muddy mountain bike race with stream crossings and a gross amount of mud – basically conditions beyond that of Kerin’s tests. In those situations, Smith says that a stickier and more viscous formula might prevail as it will be better able to physically maintain a thick layer of lube on the chain instead of getting washed away.
Based on Kerin’s and Smith’s work, one thing is clear: a well-lubricated chain that resists or flushes out contamination is superior.
In Zero Friction Cycling’s tests – and in Friction Facts’ original tests, prior to the CeramicSpeed buyout – the overall best performer is Molten Speed Wax, a modern-day version of the paraffin-wax treatment favored by riders decades ago. The additive-enhanced wax is first melted in a crockpot, and then the carefully prepared chain is fully immersed in it and allowed to sit so as to let the wax fully penetrate the chain’s interior. A chain quick link aids in re-installing the chain and repeating the process with use. CeramicSpeed’s latest UFO wax formula was developed later, Smith claims that it is measurably faster yet.
Molten Speed Wax is certainly a fast and durable lubricant. But using it requires a sterile chain and a crockpot to cook the wax through. For many, it’s just too much work. Kerin does suggest that the extra work up front balances out later, and the time spent waxing (especially if doing multiple bikes at the same time) is easily made up by the complete lack of drivetrain cleaning and constant re-lubing.
While this wax type of product is best-in-class for both durability and friction, regardless of conditions, the application process is more involved than most people would like. Drip lubes may not perform quite as well, but their ease-of-use will still make them a better choice if you don’t have a lot of time on their hands. For those riders, Kerin recommends Silca NFS.
“Silca NFS lube is the pick of the bunch” he said. “[It’s the] longest-lasting by far of the drip lubes I’ve tested.”
Friction Facts’ testing shows the original NixFrixShun (NFS) lube to be very efficient, clocking just 5.48 watts of losses. Silca’s formula claims to be a modified version of the NFS lube, and suggests as little as 12 drops are needed to keep a chain lubed after the initial application, meaning the high price per bottle is easily negated over time.
“When using little of it, it doesn’t attract nearly as much crap,” Kerin says. “And it washes off easily, unlike wax lubes.”
Just 12 drops per chain is is all that’s needed to keep a drivetrain running smoothly, according to Silca.
Kerin suggests that by using NFS’s lube and meticulously keeping on top of your maintenance, a chain may last as long as 12,000km before hitting its wear point. It’s still not the 15,000km longevity number quoted for Molten Speed Wax, or as efficient according to friction tests, but it’s extremely impressive nonetheless.
However, it has its limitations.
“Off-road, Silca’s NFS is a wet lube and so it can pick up a lot of contamination,” tells Kerin.
For where NFS isn’t suitable, South African chain lube Smoove (an item I coincidently picked in my favorite products of 2017) is Kerin’s pick as far as drip lubes go.
“Smoove is a double-edged sword; to work it into the chain is tough. But once prepped, its resistance to riding in crap conditions is amazing. However, a wet ride will need the chain stripped and cleaned afterward.”
Paraphrasing Kerin, Smoove is quite similar to another South African wax-based lub, Squirt. The wax creates a barrier to contaminants entering the chain, although Smoove has a lower viscosity and seems to layer better. Kerin’s testing proved that it’s this progressive layering, careful initial preparation, and complete chain link penetration that are the keys to getting the most out from this lube, and for that, it’s a more involved process than many other drip lubes.
Both out of South Africa, these two lubes are extremely similar, although Kerin’s testing shows the newer Smoove has the edge for durability. Exact efficiency is unknown, as Smith hasn’t tested Smoove.
Taking a look back at previous statements from Smith reveals that Squirt (recording 4.51 watts) used to be the recommended top-up lube for CeramicSpeed’s UFO chains, prior to UFO Drip coming to market. Smith hasn’t tested Smoove, but given Kerin’s findings, there’s enough evidence to suggest it may be even more efficient than Squirt with the right preparation.
Either way, both Kerin and Smith’s testing show that a lube that doesn’t collect grit is going to be best in real-world conditions. This is what makes melt-on paraffin wax treatments so effective, as little sticks to the hard coating that results. And in fact, it was this principle that was a key design driver with CeramicSpeed’s UFO Drip lube; grit literally bounces off of it.
CeramicSpeed’s new UFO Drip lube is one of the most expensive on the market, but Smith swears it was designed to beat everything else for when efficiency matters.
While Kerin’s advice considers more affordable products, Smith’s advice is understandably quite narrow.
“Aside from the very extreme conditions, for both road and MTB, I always recommend a very fast 100% dry lube. I know this can be considered biased, but [CeramicSpeed] UFO Drip was designed to provide the fastest initial friction numbers (3.78w, or 0.73w faster than the next-fastest tested drip lube, Squirt), and with minimal increase from contamination.”
While that may be true, what’s also true is that very few 100% dry lubes exist beyond CeramicSpeed’s new high-priced offering.
“Squirt does not fit my definition of a 100% dry lube,” Smith continued. “When it dries, it is tacky. Because of the tackiness, it will pick up road grit. It’s very rare to find a lube, which is labeled dry by the manufacturer, to actually be 100% dry.”
Whatever lube you decide to use, keeping your chain as clean as possible is vital to getting respectable life and efficiency from it. The best lubes here excel for their ability to lubricate the inner surfaces of the chain, and keep contaminants from entering, but they’re not a replacement for proper maintenance.
“I always suggest cleaning and re-lubing after every ride, or at least re-lubing after every ride,” says Smith. “This comes from my efficiency background.”
If you don’t have time to clean your chain, but are worried about watts, then Smith suggests at least re-applying lube regardless.
“I’ve not seen any negative efficiency effects of at least re-lubing between rides, even if re-lubing over a dirty chain, if a user does not have time to clean their chain. I’ve taken many post-contamination dirty chains over the years, and with just a re-lube, the efficiency will always increase. It might not benefit the cleanliness, build-up, or effects on wear/longevity when a lube is applied onto a dirty chain, but from an efficiency-of-the-chain standpoint, the efficiency will increase.”
Kerin agrees, but warns that relubing over a dirty chain often brings external grit back into the links. So with some lubes, while efficiency will improve, wear rates can also increase.
“Definitely a smart thing to do is to re-lube a clean chain versus a dirty chain. Drip lubes will always be X-amount abrasive from contamination, and this liquid, with some level of abrasiveness, is acting directly on chain metal causing wear. Unless one fully flush-cleans the chain and re-lubes, the level of abrasiveness will generally continue to rise and rise. This means wear rates are not linear. Often, the wear rate in the first 1,000km will be much lower than the wear rate from 3,000 to 4,000km due to lack of full-flush cleans to reset abrasive contamination.
“On average, one can expect 3,000 to 6,000km from drip lubes versus an average of 15,000km for MSW. I have seen very well-maintained chains just hitting the replacement mark at 8,000km, and one example recently where the customer attained 12,000km to 0.8mm by degreasing prior to every re-lube.”
As the saying goes, a clean bike is a fast bike.
This article highlights just the beginnings of Kerin’s testing. So far, Kerin has revealed a number of popular chain lubes that claim the world, but at least according to his tests, fail miserably. Certainly, he’ll continue to test popular chain lubes in search of the next holy grail.
What chain lube is best for one person may be completely wrong for another depending on maintenance schedule, riding conditions, riding style, and attention to detail. However, one thing is for certain: I’m going to cook my chain in wax.
Check out Zero Friction Cycling to read the full reports on the lubes tested to date.