ZFC's test protocol, in plain English.

What the test actually is

ZFC's rig is a real bicycle drivetrain running on a Tacx Neo trainer under a steady, repeatable load — same wattage and cadence on every test. The protocol is built for comparative benchmarking: products tested years apart can be ranked against each other within ZFC's published ±5% test variance, and Adam himself notes that close results (within ~10% deep into the test) should be treated as effectively the same. Each chain is brand new at the start. The lubricant is applied per the manufacturer's instructions — important; ZFC is testing the product as it's actually meant to be used, not stripped to some lab-pristine baseline.

The main test runs the chain through five 1,000 km blocks (so 5,000 km total): clean → dry contamination → recovery → wet contamination → recovery. Each contamination block introduces a different profile: 5 g doses of fine sandy loam in the dry block, 500 ml water spray plus 5 g sandy loam in the wet block. Re-lubrication happens at fixed intervals per the manufacturer's instructions — roughly every 333 km on clean blocks, with the cadence doubled on contamination blocks because that's how riders actually treat their chains in dust and rain.

Chains with sufficient wear allowance left at the end of Block 5 can then continue into an additional Block 6 extreme contamination block — 1 L of water and 10 g sandy loam, applied twice as often as Block 4. Block 6 is an extension, not part of the main test, and only a handful of products have ever reached the end of it. ZFC's cost-to-run modelling on the league table is built from the first five blocks, not all six.

The crucial design choice: the chain is never cleaned during the protocol.No wipe-downs, no flushes, no degreasing between blocks. Whatever abrasive paste forms from contamination plus lubricant plus chain wear stays put. That's why ZFC's test is as unforgiving as it is — the lubricant has to fight off ongoing degradation, not just a single wet day.

How wear is actually measured

A chain elongates rather than literally stretches — the steel plates stay the same length. The wear lives in the pins thinning and the inner-plate bores widening: cumulative microscopic play between every link adds up, and the chain physically lengthens. That's chain elongation, and it's the standard the cycling industry uses to decide when a chain is finished.

ZFC measures elongation right on the rig (or with the chain briefly removed for a manual reading), at seven or more points along the chain's length. Chains wear unevenly, so single-point measurement misses the real story; multi-point averaging captures it.

The tools that actually work

Most cheap drop-in chain checkers are misleading by design — they push adjacent rollers in oppositedirections, which lets the rollers themselves move on their pins and bundles roller wear in with the pin-and-bore wear you're actually trying to measure. The reading comes out artificially high. Adam Kerin has been clear about this: most checkers in your local bike shop give you a worse signal than no measurement at all.

What's actually worth using:

• What ZFC uses in the lab — KMC Digital Chain Checker, in pairs. A digital caliper purpose-built for chains. Slides into the links with hard physical index points and gives absolute elongation readings in millimetres to 0.01mm precision. ZFC uses a minimum of two per measurement, averaged across seven sections of the chain — that's the data the published wear figures are built from.
• What Adam recommends for at-home checking — Shimano TL-CN42. A 3-pin pass/fail checker that pushes adjacent rollers in the samedirection (the way a drivetrain does under load), so it isolates pin-and-bore elongation rather than bundling roller wear in. The other reason Adam specifically points to this one: it's laser-cut — not stamped from sheet steel, the way essentially every other cheap chain checker is. The precision of the cut is what lets the reading be honest.
• For SRAM flat-top chains — Park Tool CC-4.Same same-direction principle as the TL-CN42, sized for SRAM's differently-shaped rollers.

When to measure — wait for the wax to break in

A freshly hot-melt-waxed chain has solid wax filling the pin-and-bushing tolerances a chain checker is trying to read. Measuring immediately after dunking risks an artificially low reading because the wax is taking up the play the tool is sensing.

The numbers themselves are then scaled to the industry replacement threshold:

• 0.5% elongation is the cycling-industry chain-replacement standard — past that, the chain damages cassette and chainring teeth fast.
• ZFC calls 0.5% elongation "100% wear" — i.e. the chain has used up its useful life.
• So a block reading of 0.146 means 14.6% of the way to chain-replacement during that block. A reading of 1.0 means the chain is finished. Anything above 1.0 means the chain wore out faster than the protocol could finish testing it.

That framing is what lets ZFC compare lubricants on real-world economics: how many kilometres can you ride this chain before you're replacing it? Lubes that hold the chain at low percentages across all six blocks save you cassettes, chainrings, and chains.

The six rides, visualised

Each block represents a different style of ride. The wear-by-block charts you see across this site (the small six-bar mini-chart on every lubricant) are reading those six numbers, left to right. Hover any bar for the exact value; the colour walks from green (very low wear) through amber (moderate) to red (failing).

• Block 1 — Clean ride.Fresh application, no contamination yet. ZFC's framing: assess any initial penetration issues plus clean-road performance. Almost every lubricant looks fine here — Block 1 tells you the baseline penetration and clean-road wear profile, not real-world durability.
• Block 2 — Dusty ride. Dry abrasive dust kicked at the chain — the kind of thing you ride through on gravel and dry trails. Differences start showing up here.
• Block 3 — After-dust recovery. Clean conditions following the dusty block. Tests how well the lubricant holds up once the dust stops.
• Block 4 — Wet ride. Water and dust together. This is the most informative block in the whole test — wet contamination is where chain lubricants genuinely separate. If you compare two lubes and only have time to look at one number, this is it.
• Block 5 — After-wet recovery. Clean conditions following the wet block. Tells you how a lubricant copes with drying out.
• Block 6 — Worst-case torture. The most punishing conditions ZFC throws at the chain — heavy contamination, sustained. Mostly relevant if you ride genuinely punishing terrain regularly.

The four numbers that actually matter

ZFC publishes a lot of figures. For most riders, four of them carry almost all the signal:

1. Block 1 wear — initial penetration plus clean-road performance. Useful for indoor-trainer riders and time-triallists where every watt counts on a clean chain. Wear correlates loosely with friction here, but ZFC measures elongation, not watts directly — see the wear-as-friction-proxy caveat below.
2. Block 4 wear — the wet-conditions number. The biggest spread between products, and the truest test of whether a lubricant earns its place. If you ride outdoors at all, this is the one to weigh.
3. Single Application Longevity (SAL)— how far one application lasts before the chain dries out, in real-world km. Tells you how often you'll be reapplying, which feeds directly into how much hassle a given product creates.
4. Cost-to-run per 6,000 km— the drivetrain economy number. Combines lubricant cost with how much chain wear it's preventing, as drivetrain spend per 6,000 km of useful life (rendered in your selected currency, converted from ZFC's published AUD figures via ECB-sourced rates). The eye-opener: cheap lubes often work out the most expensive once you factor chain replacement.

Everything else (other blocks, recovery numbers, sub-categories) is context. If you're comparing two lubricants, look at Block 1 and 4 together with SAL and cost — that's 80% of the decision.

Caveats worth knowing

• ±5% test variance. Two runs of the same lube can land ~5% apart. Treat any small difference between products as a tie.
• The Choice Matrix "Tier 1".ZFC publishes a curated short-list separately from the raw block data — products that perform across multiple test dimensions and aren't just edge-case winners. When this site shows a "ZFC Tier 1" chip, that's ZFC's curation, not ours.
• Methodology adjustments are sometimes made.ZFC occasionally varies the protocol for a specific product — most notably extending the rewax interval for additives like Silca's Endurance Chip, which are designed to be used with longer intervals. When that happens, this site flags the product with an * Extended-interval test chip and the chart's bars render as outlines rather than filled. Don't compare their numbers head-to-head with standard-protocol products.
• Wear is a useful proxy for friction — but a deliberately blunt one. Adam himself notes the correlation isn't tight for small differences: a slightly-higher-wear lube couldbe more efficient in clean conditions thanks to better stiction or viscous-friction performance. The relationship becomes solid once products diverge by 10%+ deep into the test, when too-rapid wear of hardened steel takes too much friction to be ignored. ZFC measures elongation rather than friction directly because it's repeatable to high precision (digital multi-point averaging); friction is the strong implication when wear differences are large.