A machine that holds a stable angle at the start of a shift and drifts a few tenths of a degree by mid-afternoon is not losing its geometry. The hydraulic oil has changed, and with it the relationship between valve command and ram position. This is the most common explanation for slow, one-directional angle drift during production.
1. The practical problem
The pattern is characteristic. The first parts of the shift are slightly off in one direction; after warm-up they come good; after two to four hours they begin to drift again and stay drifted. Nobody has touched the programme. Measuring the same part at the same setting in the morning and in the afternoon produces two different numbers, and the difference is often within the range that decides whether a part passes an angular tolerance.
2. Failure mechanism
Three oil properties move with temperature. Viscosity falls as the oil warms, so leakage past spools and seals increases and the flow produced by a given valve command rises. Bulk modulus falls, so a given volume of compression produces a smaller pressure change and the system becomes softer. Thermal expansion of oil in the tank, cylinders and pipework changes the effective stroke by a small amount. Together they shift the point where the ram stops relative to the material. In air bending, angle is very sensitive to penetration near the bottom of the stroke, so a small positional shift becomes a visible angle change. Closed-loop position control absorbs most of this; open-loop valve control does not.
3. Key parameters
The parameters that matter are oil grade and viscosity index, tank volume, cooler capacity, duty cycle, ambient temperature, and the control architecture. A larger tank and a properly sized cooler slow the rate of change but do not remove it. What removes it is position feedback that measures the ram rather than trusting the valve. The rate of change matters as much as the absolute value: a machine that stabilises within an hour is manageable with a warm-up routine, while one that keeps climbing for four hours points to insufficient cooling capacity or an oversized duty cycle.
| Condition | Hydraulic effect | Typical symptom | Response |
|---|
| Cold start, low ambient | High viscosity, sluggish valves | Late settle, small consistent offset | Warm-up cycle, first-piece trial bend |
| Warm-up phase | Viscosity falling | Angle moves in one direction | Do not adjust the programme during warm-up |
| Thermal equilibrium | Stable viscosity | Repeatable, reference values valid | Record as the job's baseline |
| Above recommended band | Low viscosity, seal wear | Hunting, position instability | Stop, check cooler and relief settings |
4. Diagnosis and calculation method
- Fit or use an existing tank thermometer and record oil temperature at the same moment as every first-off angle measurement.
- Plot angle against oil temperature for one job. A monotonic relationship confirms thermal drift; a step change at constant temperature points elsewhere.
- Compare a bend made cold with the same bend after two hours, using the same material, die and programme.
- Read Y1 and Y2 position at the bottom of the stroke rather than the commanded value, to see whether the shift is hydraulic or structural.
- Check that the pump pressure and valve gains were set with warm oil; a cold-set baseline keeps the whole shift offset.
- Verify the cooler is actually operating: inlet and outlet temperature of the heat exchanger should differ under load.
5. Prevention
Two measures pay for themselves. The first is a fixed warm-up routine before the first production part, so that the reference state is reproducible. The second is closed-loop position feedback on both rams, which converts viscosity variation into a control task instead of an angle error. Adequate tank volume, a correctly sized cooler and a defined oil change interval complete the picture. Our 160T/4000 synchro servo press brake positions both rams against linear scale feedback, and the requirement is the same on any machine in this class, including our 120 tonne Delem-controlled press brake.
6. FAQ
Is a drifting angle always an oil temperature problem?
No. Thermal drift is slow, one-directional and correlated with running time. A step change, a change after a tooling swap, or an error that reverses sign indicates something else and should be investigated separately. The useful test is always the same: hold material, die and programme fixed, and change only the temperature.
Would a higher viscosity oil reduce the drift?
It changes the operating point rather than removing the mechanism, and it affects response speed and cold-start behaviour. The correct grade is the one specified by the machine builder; the real fix is temperature stability and position feedback.
How long should warm-up take?
Long enough for the oil to reach a stable temperature at low load, which in practice means running the machine through its stroke several times before the first measurement. The machine's manual is the authority; the principle is that the first measured part should be made in the same thermal state as the reference.
¡Su mensaje debe tener entre 20 y 3.000 caracteres!