A machine that lifts slowly, stalls in the dig, or holds pressure only at idle is the most common hydraulic complaint a repair shop receives. The instinct is to condemn the pump. In practice, roughly half of the pumps that arrive at a rebuild bench with a "no pressure" tag are mechanically serviceable, and the fault sits in a control, a relief valve, a suction line or a signal circuit.
This guide sets out the sequence that separates a control fault from a wear fault, using measurements rather than opinions, and identifies the parts that normally need replacement when the pump really is worn.
"It has no power" is not a measurement. Before anything is removed from the machine, capture four values at operating temperature:
| Measurement | Where to take it | What it tells you |
|---|---|---|
| Standby pressure | Pump outlet, all functions neutral | Whether the compensator or relief is set correctly |
| Relief / cut-off pressure | Pump outlet, function stalled against a load | Whether the system can reach its rated ceiling |
| Flow at pressure | Inline flow meter downstream of pump | Volumetric efficiency under real load |
| Case drain flow | Case drain line, warm oil, at rated pressure | Internal leakage through the rotating group |
These four numbers take under an hour and eliminate most guesswork. A pump that reaches full relief pressure with low flow is worn. A pump that never reaches relief pressure at all is usually being bypassed, mis-signalled or wrongly controlled — not worn out.
Each mode below produces the same operator complaint but a different measurement signature. Work down the list before opening the pump.
| Failure mode | Distinguishing measurement | Typical evidence |
|---|---|---|
| Rotating group wear (piston, bore, valve plate) | Case drain flow high and rising with pressure | Low flow at full pressure, warm case, metal in oil |
| Pressure compensator setting drift | Cut-off occurs early, at a consistent lower value | Pressure stable but capped, no metal, no heat |
| Relief valve passing | Relief opens below spec; pump flow normal when dead-headed elsewhere | Hot relief body, chatter, pressure creep |
| Load-sense signal loss | Pump stays near standby; flow only at low pressure | Blocked orifice, leaking signal line, stuck spool |
| Suction restriction or air ingress | Pressure fluctuates, noise, foamy oil | Collapsed hose, clogged strainer, low reservoir level |
| Drive coupling slip or stripped spline | Pump speed below commanded; flow collapses under load | No pressure rise, audible slip, worn spline |
Notice that only the first row requires pump disassembly. The remaining five are machine-level faults, and they are the ones most often missed when a pump is changed "to be safe" and the problem returns within a week.
Most excavator, crane and mixer pumps are variable displacement units with a pressure compensator, a load-sensing control or a torque limiter. Each deliberately reduces displacement as pressure rises, so a control that is set slightly low, or a control spool that is sticking, produces exactly the symptom an operator calls "the pump is worn out".
On Rexroth A4VSO and A10VSO units, the compensator and the load-sense cartridge are accessible without removing the pump; on Kawasaki K3V and K5V units the regulator sits on the rear cover and is equally serviceable in frame. Checking settings and fault codes before disassembly is therefore not a shortcut, it is the correct order of work.
A proper load test measures pressure and flow together at more than one working point. Three points are enough to make a decision.
| Step | Condition | Record | Interpretation |
|---|---|---|---|
| 1 | Oil at 50–60 °C, engine at rated speed, neutral | Standby pressure, case drain flow | Baseline for leakage and control setting |
| 2 | Function stalled against relief at rated pressure | Peak pressure, flow at that pressure | Volumetric efficiency under load |
| 3 | Same function at half pressure, metered load | Flow, case drain flow, oil temperature rise | Leakage trend — wear shows as flow falling faster than pressure |
Compare the flow at step 3 with the pump's rated displacement multiplied by shaft speed. A drop of more than about 10–12 percent from nominal, with case drain flow climbing as pressure rises, indicates genuine internal leakage. If flow is close to nominal but the pressure ceiling is low, go back to the control, relief and signal circuits.
Rebuilding a piston pump is not an all-or-nothing decision. Wear usually concentrates in the rotating group, and the economical repair replaces only the surfaces that carry the load:
For units built around the Rexroth A4VSO range, parts are listed for the A4VSO40 through A4VSO750 frame sizes, and for A10VSO pumps the A10VSO180 rotating group parts cover the most common large-frame repair. Linde machines running the HPV range are served by parts for HPV55 to HPV280 pumps.
Because so much of the diagnosis depends on the exact frame size and control variant, the fastest route to a correct quotation is a single message containing four items:
With those four items our engineers confirm the correct variant, state the material and hardness specification of the wear surfaces, and quote against a delivery date — rather than asking you to accept an "equivalent" part on trust.
Yes, at low flow. A badly worn pump can often dead-head to near rated pressure with almost no output, which is why pressure alone is a poor test and flow at pressure is the reliable one.
Manufacturers publish a limit per frame size, commonly around 2–5 percent of pump output when warm. Measure at operating temperature and rated pressure; a cold reading is meaningless.
Usually yes, provided the cylinder block is replaced at the same time. Replacing one face of a bedded pair typically produces a repeat failure within months.
Because the original fault was never the pump. Relief passing, suction restriction and contaminated oil all destroy a new pump in the same way as the old one, and the second failure usually arrives faster than the first.
A machine that lifts slowly, stalls in the dig, or holds pressure only at idle is the most common hydraulic complaint a repair shop receives. The instinct is to condemn the pump. In practice, roughly half of the pumps that arrive at a rebuild bench with a "no pressure" tag are mechanically serviceable, and the fault sits in a control, a relief valve, a suction line or a signal circuit.
This guide sets out the sequence that separates a control fault from a wear fault, using measurements rather than opinions, and identifies the parts that normally need replacement when the pump really is worn.
"It has no power" is not a measurement. Before anything is removed from the machine, capture four values at operating temperature:
| Measurement | Where to take it | What it tells you |
|---|---|---|
| Standby pressure | Pump outlet, all functions neutral | Whether the compensator or relief is set correctly |
| Relief / cut-off pressure | Pump outlet, function stalled against a load | Whether the system can reach its rated ceiling |
| Flow at pressure | Inline flow meter downstream of pump | Volumetric efficiency under real load |
| Case drain flow | Case drain line, warm oil, at rated pressure | Internal leakage through the rotating group |
These four numbers take under an hour and eliminate most guesswork. A pump that reaches full relief pressure with low flow is worn. A pump that never reaches relief pressure at all is usually being bypassed, mis-signalled or wrongly controlled — not worn out.
Each mode below produces the same operator complaint but a different measurement signature. Work down the list before opening the pump.
| Failure mode | Distinguishing measurement | Typical evidence |
|---|---|---|
| Rotating group wear (piston, bore, valve plate) | Case drain flow high and rising with pressure | Low flow at full pressure, warm case, metal in oil |
| Pressure compensator setting drift | Cut-off occurs early, at a consistent lower value | Pressure stable but capped, no metal, no heat |
| Relief valve passing | Relief opens below spec; pump flow normal when dead-headed elsewhere | Hot relief body, chatter, pressure creep |
| Load-sense signal loss | Pump stays near standby; flow only at low pressure | Blocked orifice, leaking signal line, stuck spool |
| Suction restriction or air ingress | Pressure fluctuates, noise, foamy oil | Collapsed hose, clogged strainer, low reservoir level |
| Drive coupling slip or stripped spline | Pump speed below commanded; flow collapses under load | No pressure rise, audible slip, worn spline |
Notice that only the first row requires pump disassembly. The remaining five are machine-level faults, and they are the ones most often missed when a pump is changed "to be safe" and the problem returns within a week.
Most excavator, crane and mixer pumps are variable displacement units with a pressure compensator, a load-sensing control or a torque limiter. Each deliberately reduces displacement as pressure rises, so a control that is set slightly low, or a control spool that is sticking, produces exactly the symptom an operator calls "the pump is worn out".
On Rexroth A4VSO and A10VSO units, the compensator and the load-sense cartridge are accessible without removing the pump; on Kawasaki K3V and K5V units the regulator sits on the rear cover and is equally serviceable in frame. Checking settings and fault codes before disassembly is therefore not a shortcut, it is the correct order of work.
A proper load test measures pressure and flow together at more than one working point. Three points are enough to make a decision.
| Step | Condition | Record | Interpretation |
|---|---|---|---|
| 1 | Oil at 50–60 °C, engine at rated speed, neutral | Standby pressure, case drain flow | Baseline for leakage and control setting |
| 2 | Function stalled against relief at rated pressure | Peak pressure, flow at that pressure | Volumetric efficiency under load |
| 3 | Same function at half pressure, metered load | Flow, case drain flow, oil temperature rise | Leakage trend — wear shows as flow falling faster than pressure |
Compare the flow at step 3 with the pump's rated displacement multiplied by shaft speed. A drop of more than about 10–12 percent from nominal, with case drain flow climbing as pressure rises, indicates genuine internal leakage. If flow is close to nominal but the pressure ceiling is low, go back to the control, relief and signal circuits.
Rebuilding a piston pump is not an all-or-nothing decision. Wear usually concentrates in the rotating group, and the economical repair replaces only the surfaces that carry the load:
For units built around the Rexroth A4VSO range, parts are listed for the A4VSO40 through A4VSO750 frame sizes, and for A10VSO pumps the A10VSO180 rotating group parts cover the most common large-frame repair. Linde machines running the HPV range are served by parts for HPV55 to HPV280 pumps.
Because so much of the diagnosis depends on the exact frame size and control variant, the fastest route to a correct quotation is a single message containing four items:
With those four items our engineers confirm the correct variant, state the material and hardness specification of the wear surfaces, and quote against a delivery date — rather than asking you to accept an "equivalent" part on trust.
Yes, at low flow. A badly worn pump can often dead-head to near rated pressure with almost no output, which is why pressure alone is a poor test and flow at pressure is the reliable one.
Manufacturers publish a limit per frame size, commonly around 2–5 percent of pump output when warm. Measure at operating temperature and rated pressure; a cold reading is meaningless.
Usually yes, provided the cylinder block is replaced at the same time. Replacing one face of a bedded pair typically produces a repeat failure within months.
Because the original fault was never the pump. Relief passing, suction restriction and contaminated oil all destroy a new pump in the same way as the old one, and the second failure usually arrives faster than the first.