The service file of a machine used to hold the repair orders and the test reports. A third kind of content is now finding its way in: the data from the run. It is not an extra burden, and in many cases it is simply the raw material behind the repair orders and the test reports. The open points are what to record, how to record it and what it can be used for, and there is no common practice on these yet. This article takes them one by one.
Which data is worth keeping
The first group is the process readings in steady state: working vacuum, pressure drop across the machine, oil temperature in and out, and oil level. The working vacuum of the ZYD-100 is -0.08 ~ -0.0999 MPa, and where the reading sits inside that range says more than whether it is inside the range at all. Recording the same machine several times turns the readings into a trend, and the trend carries information on its own.
The second group is the abnormal events: the time of the alarm, the type of protection that acted, the readings at that moment, and what was done about it. The value of an event record lies in afterwards, because an overpressure shutdown that left no reading behind can only be analysed by guesswork later on.
The third group is run time and energy: the hours the heating ran, the hours the vacuum ran, the running hours of the machine and the electricity consumed. The total power of the ZYD-100 is 72 + 5 kW, and the same batch of oil takes noticeably different amounts of energy in different seasons and at different temperature settings. Only a record shows that difference.
How to record so that it does not become a pile of paper
- Record the conditions together with the reading. Temperature, pressure, oil and setting values belong to the same line, otherwise the reading alone has nothing to be compared with.
- Keep the time basis of the measurement fixed. Take the reading in the same condition and at the same position every time, otherwise there is no trend.
- Use the same gauge. Gauges of the same type differ from one another, so a changed gauge has to be noted, and the baseline has to be set again where necessary.
- Keep the original values and not only the conclusion, because a conclusion can be recalculated while a raw value cannot be recovered.
- The record of an abnormal event has to include the action that was taken, otherwise the condition of the machine cannot be reproduced later.
What the data can be used for
One use is to follow the change in the sealing state. The rate at which the vacuum falls back after the pump is stopped is a direct reflection of the sealing condition, and running the same test at different times shows that a slow sealing ring and a weld on the vacuum pipework opening up under vibration show up on the curve before they show up anywhere else.
Another use is to estimate how much life is left in the consumables. The rate at which the pressure drop rises and the amount of oil filtered in total belong next to each other, and together they give a rough view of how much is left in the element instead of finding out that it is near its limit when the alarm sounds.
A third use is to build an energy baseline. The same batch of oil, the same throughput and different seasons or different temperature settings can be turned into a comparison curve, and whether an energy saving measure or a change of running method has paid off is then answered by the data rather than by an impression.
| Record | Reference value | What the trend shows |
|---|---|---|
| Working vacuum | -0.08 ~ -0.0999 MPa | Whether the degassing performance is falling |
| Pressure drop in and out | The historical value under the same conditions | How fast the elements load up and how much life is left |
| Rate of the vacuum fall after the pump stops | The baseline set by the first test | The direction in which the sealing state is changing |
| Heating and vacuum run hours | The accumulated running hours | When the maintenance items actually come due |
| Electricity used | The historical value for the same oil and setting | The difference caused by the running method and the season |
Keep it on the same line as the test reports
Runtime data on its own means little, because it has to be aligned with the test reports. A water content of ≤3 ppm (GB/T 7600), a gas content of ≤0.01% (GB/T 423) and a breakdown voltage of ≥65 kV (GB/T 507) placed next to the vacuum and pressure records of the same day show how the state of the process and the quality of the finished oil correspond to each other. The real value of a data file shows up in the comparison across years.
A single reading describes the state of the moment, a series of readings describes how the machine is changing.
