Hydraulics has a reputation for being difficult, and most of that reputation comes from one cause. The large majority of hydraulic failures are not mechanical failures; they are contamination failures. Dust, water, wear debris and degraded fluid account for the great majority of pump, valve and cylinder damage, and all of them are manageable with a routine that costs far less than the failures it prevents.
A hydraulic system runs on a fluid film a few micrometres thick, and a particle large enough to matter is invisible to the eye. Those particles erode pump internals, jam spool valves and score cylinder bores, and the debris they generate becomes more contamination in a self-feeding cycle. By the time a system fails visibly, the damage has usually been progressing for a long time.
That is why the maintenance effort belongs in filtration and in keeping contamination out, rather than in reacting to symptoms. A reservoir left open, a hose changed without capping, a filter bypassed because the replacement had not arrived - each is a contamination event, and each shortens the life of every component downstream.
Every hydraulic system has a cleanliness target, usually expressed as a particle-count code, and the filtration is specified to reach it. The target is set by the most sensitive component in the circuit: a servo valve tolerates far less contamination than a simple directional valve, and the filter specification follows the tightest requirement rather than the average.
Two things undermine filtration in practice. The first is a filter that is changed on a calendar rather than on condition, so it may be clogged long before its interval or replaced prematurely and opened to contamination. The second is a bypass valve that opens when the element blocks, letting unfiltered fluid through exactly when the system is dirtiest. Differential-pressure indication on the filter turns both into a measurement rather than a guess.
The only reliable way to know a fluid's condition is to sample it and analyse it. A sample taken from the right place - a live, turbulent point in the return line or from a sample valve, not from the top of the reservoir - and analysed for particle count, water content and wear metals tells you what the system is doing.
Trending those results over time is more useful than any single result. A particle count that is high but stable is a different situation from one that is rising, and wear metals that appear and increase point at a specific component. The sample programme turns the question 'is the fluid still good' into a trend that can be acted on before a failure rather than after.
A great deal of hydraulic trouble announces itself in ordinary operating readings, and a simple daily or weekly check catches it before it becomes a repair.
| Check | What to watch for | Why it matters |
|---|---|---|
| Fluid temperature | A steady rise over days or weeks | Heat degrades fluid and accelerates wear |
| System pressure | Drift from the set value under load | Points to internal leakage or a failing pump |
| Reservoir level | Slow loss with no visible leak | May be a leaking cylinder seal or a cooling loop |
| Filter differential | Approaching the change point | Turns filter changes into a planned task |
| Fluid appearance | Darkening, cloudiness, foam | Oxidation, water ingress or air entrainment |
None of these checks requires an instrument beyond what most sites already have. What they require is a baseline - what is normal for this machine - because a temperature that would be alarming on one system is routine on another.
When a hydraulic system does fail, the repair is far quicker if there is a record of what its normal readings were, when the fluid was last sampled and changed, and which components have already been replaced. A system with a maintenance history is diagnosed from evidence; a system without one is diagnosed by guesswork and by replacing parts until the symptom stops.
The record should also capture the intervention itself: what was opened, what was replaced, whether the system was flushed, and the sample result afterwards. That is what establishes whether a fault has actually been corrected or merely postponed.
On condition, not on time. Fluid changed on a calendar can be discarded while it is still good, or retained long after it has degraded. Sampling and trending the particle count, water content and wear metals tells you when a change is actually needed, and it is the cheapest part of the whole maintenance programme.
Only if the specification matches and you have confirmed compatibility. Different fluids can have different additive packages that do not mix well, and an incompatible top-up can cause the fluid to separate, foam or lose its anti-wear properties. Where the correct fluid is unavailable, the answer is to wait, not to substitute.
Contamination, followed by cavitation from a restricted inlet or air entrainment. Both are managed by keeping the fluid clean and the inlet line sound. A pump that has failed from contamination usually fails again unless the circuit is flushed, because the debris it generated is still in the system.
Hydraulic reliability is mostly a discipline of cleanliness and observation, and it rewards attention long before it punishes neglect. ZH Industrial Machinery supplies and services industrial machinery, and we are glad to review a hydraulic system's filtration, sampling and checking routine with a site before the next failure rather than after it.
The working fluid and the circuits it serves are described under hydraulic fluid and hydraulic machinery, with the removal of contamination covered by filtration. The lubrication duties that sit alongside filtration are described under lubrication, and the cleanliness levels a specification quotes come from ASTM International test methods.