By Dovydas Kersys · Chapter 35 of 40 in Hydraulic Troubleshooting.
Before you begin: Hydraulic equipment can contain lethal stored energy. Only trained and authorized personnel should perform hydraulic work. Isolate energy, support loads, verify the pressure state, use pressure-rated instruments and follow site-specific lockout/tagout procedures. The diagrams are simplified training illustrations; use the actual machine schematic and manufacturer instructions.
Intermittent faults are difficult because the machine may be healthy when maintenance arrives. The solution is to capture operating conditions and signals during the event rather than repeatedly inspecting a stopped machine.
In practice, the fastest diagnosis comes from deciding what the component is supposed to control and then measuring the variable on both sides of that component. Pressure tells you about load and restriction; flow tells you about speed; temperature tells you where power is being lost; leakage tells you whether clearances or seals are allowing oil to bypass the intended path.
Figure 35. Simplified diagnostic view for 35. Intermittent Hydraulic Faults.
What the technician may observe
- fault appears only after hours of running
- motion fails once every few cycles
- pressure briefly collapses
- machine resets after power cycle
- problem follows temperature or vibration
Diagnostic sequence
- Define a trigger condition that identifies the fault. Compare the reading with a known-good baseline or the machine specification rather than relying on a generic value.
- Log pressure, command signal, temperature and cycle state around the event. Record the result before changing the next variable so the test actually narrows the fault.
- Inspect connectors, pilot lines and suction joints for vibration sensitivity. Compare the reading with a known-good baseline or the machine specification rather than relying on a generic value.
- Trend filter differential pressure and fluid temperature. Record the result before changing the next variable so the test actually narrows the fault.
- Compare failed cycle data with a good cycle. Compare the reading with a known-good baseline or the machine specification rather than relying on a generic value.
- Use nonintrusive temporary sensors where permanent test points are unavailable. Record the result before changing the next variable so the test actually narrows the fault.
How to interpret the measurements
Begin by asking whether the symptom affects the whole machine or only this function. If other functions sharing the same supply remain normal, a common pump or reservoir fault becomes less likely and the local valve, actuator, plumbing and mechanical load deserve more attention.
Next, compare pressure upstream and downstream while oil is actually flowing. A large pressure difference across a component means that component or passage is consuming hydraulic energy. If flow is low everywhere, investigate pump delivery, inlet conditions, speed and major internal leakage before blaming a local valve.
Finally, repeat the test at the condition that creates the complaint: hot oil, full load, a particular direction, low speed or rapid cycling. Hydraulic clearances, viscosity, pilot pressures and valve forces change with operating condition, so a static test may look normal while the production fault remains.
Corrective actions after the cause is proven
- Fix the variable that changes at the fault event.
- Avoid replacing multiple parts between tests.
- Keep logged traces with the maintenance record.
- Verify the fix over enough cycles to cover the original failure pattern.
Common mistakes
- Adjusting the relief or compensator before recording the original setting and proving that the setting is wrong.
- Replacing a component because it is hot, even though the heat may be arriving from an upstream restriction or continuous bypass flow.
- Calling a problem “low pressure” when the actual complaint is low flow or excessive mechanical load.
- Testing only with cold oil when the production fault appears after the system reaches operating temperature.
Field rule: For 35. intermittent hydraulic faults, do not replace a component until one measurement or isolation test shows how it fails to perform its intended function.
Symptom-to-next-test table
| Observed condition | First useful test |
|---|---|
| Fault appears only after hours of running | Measure the controlled variable before and after the suspected component. |
| Motion fails once every few cycles | Measure the controlled variable before and after the suspected component. |
| Pressure briefly collapses | Measure pressure at pump and actuator under the actual load. |
| Machine resets after power cycle | Measure the controlled variable before and after the suspected component. |
Field practice
Before leaving the machine, write down three numbers that would let another technician judge whether the repair truly restored normal operation. Good choices are usually operating temperature, a pressure at a defined machine state, and either flow or cycle time. This creates the baseline that makes the next fault faster to diagnose.
PART VIII – REPAIR, COMMISSIONING AND RELIABILITY
Practical diagnosis from symptom to measurement to root cause.
Continue the series
- Previous chapter: Solenoids, PLC Outputs and Electro-Hydraulic Faults
- Next chapter: Hydraulic Commissioning After Pump, Valve or Cylinder Replacement
- Browse all Hydraulic Troubleshooting chapters
Adapted from Hydraulic Troubleshooting for Maintenance Technicians, by Dovydas Kersys. © 2026 Dovydas Kersys. Featured cover image is an AI-generated editorial illustration.