THE TRUTH SHALL SET YOU FRICTION-FREE

CONSIDER THE HUMBLE LIKE CHAIN. IT'S AN OFTEN OVERLOOKED BICYCLE PART COMPARED TO FLASHIER COMPONENTS WHEELS, CRANKS, AND DERAILLEURS. BUT THE CHAIN IS THE CRITICAL LINK IN THE DRIVETRAIN BETWEEN THE RIDER'S POWER AND THE BIKE'S REAR WHEEL.

- DAN CHABANOV

It's also a massively complicated bit of tech, with hundreds of metal pieces that form the 110 or so chain links, all of which are in constant motion articulating around the drivetrain's gears. With that much metal-on-metal contact, chains can wear quickly, sometimes needing replacement several times a year.

Most bike mechanics estimate that you should replace your chain every 2,000 to 3,000 miles. But depending on your riding conditions and lubricant of choice, your replacement interval can be as short as 600 miles or more than 10,000. If you miss that critical service interval for your chain (defined as 0.5mm of chain wear), it could lead to your cassette and chainrings requiring replacement this can increase your repair bill from just the price of a chain to (potentially) quadruple that cost.

When I first started looking more deeply into chain lube, I realized that lube is one of the most problematic things to review as a tech editor. Proper and empirical testing of a single chain lube can take months, and that's if you have the equipment to do it. There is no standardized testing for lubes among manufacturers. To make matters worse, the marketing claims of some brands range from over-exaggeration to basically science fiction.

One of the first independent chain lube testing efforts was by Jason Smith of Friction Facts. Smith was doing independent testing of chain lubes before Ceramic Speed (a company that makes parts to optimize drivetrain efficiency) purchased Friction Facts in 2016. Smith's efforts focused on chain lube efficiency by measuring drivetrain losses in watts. This testing required custom-built machines that could set input power and measure output power. They needed to be extremely precise at both ends, with a reported accuracy within 0.02 of a watt.