The ISO 4406 cleanliness code decoded digit by digit: the scale-number table (particles per millilitre at 4, 6 and 14 micron), target codes per hydraulic component from servo valves to gear pumps, the NAS 1638 relationship, how online and portable particle counters like the Parker icountLCM20 fit into a monitoring programme, bottle sampling discipline that does not contaminate the result, and what a cleanliness spec means for filter purchasing.
| Scale number | Particles/ml (more than – up to) | Scale number | Particles/ml (more than – up to) |
|---|---|---|---|
| 24 | 80,000 – 160,000 | 16 | 320 – 640 |
| 23 | 40,000 – 80,000 | 15 | 160 – 320 |
| 22 | 20,000 – 40,000 | 14 | 80 – 160 |
| 21 | 10,000 – 20,000 | 13 | 40 – 80 |
| 20 | 5,000 – 10,000 | 12 | 20 – 40 |
| 19 | 2,500 – 5,000 | 11 | 10 – 20 |
| 18 | 1,300 – 2,500 | 10 | 5 – 10 |
| 17 | 640 – 1,300 | 9 | 2.5 – 5 |
Reading 18/16/13 with the table: up to 2,500 particles/ml at ≥4 μm, up to 640 at ≥6 μm, up to 80 at ≥14 μm. The three sizes exist because they damage differently — silt-size (4–6 μm) particles wear servo spools and lap valve clearances, the ≥14 μm fraction scores pumps and jams poppets. Two codes with the same middle number can behave very differently at the ends; report and specify all three digits.
| Most sensitive component in the system | Typical target (ISO 4406) | Notes |
|---|---|---|
| Servo valves (motion control, test rigs) | 16/14/11 | Manufacturer lists govern; some demand cleaner still at high pressure |
| Proportional valves (e.g. D1VW-class directional stacks with proportional trim) | 17/15/12 | The common industrial "clean" target |
| Axial piston pumps/motors, variable displacement | 18/16/13 | The workhorse target for modern hydraulics |
| Vane pumps, fixed piston units | 19/17/14 | — |
| Gear pumps, cylinders, general industrial | 20/18/15 | Still twice as clean as typical uncontrolled systems |
| New oil as delivered (typical, unfiltered) | Often 21/19/16 class or worse | Filter on fill; verify, do not assume |
Raise the target one code (cleaner) for any of: pressure above ~250 bar, safety-critical function, high duty cycle, or expensive downtime — the halving-per-code arithmetic means each step buys a disproportionate life extension in rolling-contact and sliding parts. Published component-life models and filter makers' field data agree on the direction and roughly on the magnitude: moving two codes cleaner commonly doubles hydraulic component life in dirty baselines.
| Method | Instrument class | Strengths | Limits |
|---|---|---|---|
| Permanent online monitoring | Inline laser counters — Parker icountLCM20 class | Continuous ISO codes on live pressure lines, trends and alarms; catches ingress events and filter breakthrough as they happen | Needs correct sampling point and flow/pressure window; water and aeration read as particles if the installation ignores them |
| Portable survey counters | Service-case laser counters (icountPD class and peers) | One instrument audits a whole plant or fleet; before/after proof for flushing jobs | Operator discipline decides repeatability |
| Bottle sampling to a lab | Sample pump + certified-clean bottles | Full oil analysis alongside the count (wear metals, water, additives) | Days of latency; one dirty bottle or sloppy draw invalidates the count |
| Airborne counting (for completeness) | Fluke 985 class | Cleanrooms, HVAC commissioning, filter leak surveys | Not an oil instrument — listed here because search engines and storerooms confuse the two |
Bottle discipline, because most "failed" samples are sampling failures: draw from a live, agitated system at operating temperature, never from the tank bottom or a dead leg; flush the sampling valve and tubing generously before the bottle opens; use certified-clean bottles only; fill two-thirds and cap immediately in clean surroundings; label with running hours and point ID. One ritual, every time, or trend data is noise. For online units, mount per the maker's pressure/flow window, and give the commissioning survey a portable cross-check so the baseline is believed.
| NAS 1638 class | Rough ISO 4406 neighbour | Context |
|---|---|---|
| NAS 5 | ≈ 16/14/11 | Servo-grade cleanliness |
| NAS 6 | ≈ 17/15/12 | Proportional-grade |
| NAS 7 | ≈ 18/16/13 | Modern piston-pump target |
| NAS 8 | ≈ 19/17/14 | General industrial |
| NAS 9 | ≈ 20/18/15 | Uncritical systems |
The honest caveat every conversion table owes you: NAS classes count particles in differential size bands, ISO 4406 counts cumulatively at three thresholds — the mapping is a practised approximation, not an equation, and specifications should be written in one system rather than converted back and forth. Legacy aviation and some OEM documents still speak NAS; modern industrial practice, and every current counter including the icount family, speaks ISO 4406.
Where the code belongs in your paperwork: filter RFQs — specify the target code at the component and let the filtration (β-ratio at the sizes that matter, e.g. β5 or β7 ≥ 200 class elements) be selected to hold it, including offline/kidney-loop capacity for systems with high ingression — our MRO primer and CJC-class offline filtration lines serve exactly this; new-oil purchasing — state delivered cleanliness and require the certificate; flushing contracts — acceptance = measured code at named points, witnessed, not hours pumped; warranty files — a logged trend from an online counter is the difference between a goodwill discussion and a paid claim; and commissioning — new builds and refits sign off at the target code, because construction debris is the dirtiest oil a system will ever see. Monitoring hardware for all of this is quotable equipment: icountLCM20 online counters, portable survey units, sampling hardware and bottles, and the filtration to act on what they find.
How buying works at cballast. We quote counters, sampling equipment, filter elements and offline filtration against a stated target code, and we put the code on the quote line so acceptance is measurable. Accounts are vetted at onboarding; first order prepaid; established accounts on credit terms with repeat discounts; public bodies and utilities invoiced against purchase order from the first order. RFQs to sales@cballast.com.
Standards and sources referenced: ISO 4406 (coding), ISO 11171 (counter calibration), ISO 11500 (counting method), NAS 1638 (historic classes); component-manufacturer cleanliness requirements and filter manufacturers' β-ratio data.
Questions on a target code, a counter installation or a failing trend: write to sales@cballast.com. We answer whether or not you buy from us.