The machine can be fitted with an online moisture meter. Once it is installed, the panel carries a moisture reading at all times, and it is natural on site to take that reading directly as the criterion for the treatment result. The two are by no means the same: the online reading is a process value, GB/T 7600 is the measured value given by a standard test method, and the judgement on water content ≤3 ppm (GB/T 7600) has to be supported by the latter.
The first difference between the two is in the sampling. The probe of the online meter is mounted in the pipework and measures the oil in motion, so what it reads is the state at a given sampling point at a given moment. The standard method is to sample on site, seal the sample and send it for testing; from the moment the sample is taken until the measurement is finished, the container, the sealing method and the storage time all play their part, and the laboratory gives the measured value under the conditions specified in GB/T 7600. The sampling step alone can change the result: if the sample is not sealed properly or has been standing for a long time, the reading will deviate.
They serve different purposes
The strength of the online meter is continuity. It does not only give a single figure, it also gives a trend: a slow rise in the moisture reading usually means that the efficiency of the condensation water separation stage has fallen; a sudden jump in the reading usually means that there are air bubbles near the sampling point or a change in the flow condition. Seeing the trend gives much more lead time than waiting for the laboratory result to come back before acting.
There are two conditions of use that are easily overlooked. The probe has to be in contact with oil in motion; if it is mounted in a dead leg or at a point with a very low flow velocity, the reading will clearly deviate from the true value. The probe and the sampling line need regular cleaning and verification, otherwise after the drift sets in the reading is still displayed but no longer points to the real water content.
| Point of comparison | Online moisture meter | GB/T 7600 laboratory testing |
|---|---|---|
| Object measured | Oil continuously flowing in the pipework | Oil sample taken and sent for testing |
| Sampling method | The probe measures in situ, with no manual sampling | Sampled on site, sealed and sent for testing |
| Form of the output | Continuous reading and trend | Single measured value |
| Main use | Process monitoring, early detection of deviations | Basis for acceptance judgements and for disputes |
| Relation to ≤3 ppm | Cannot substitute directly for the judgement | Basis of the judgement |
Why the online reading and the laboratory result do not agree
The most common reason for two methods to give different figures is that the state at the sampling point is not the same. The online probe measures oil in motion, while the laboratory measures an oil sample that has stood still; during that standing time dissolved gas escapes and the water redistributes on the wall of the container, so the state of the sample when it reaches the laboratory is no longer the state it was in at the moment of sampling. The second reason is the timing of the measurement: the online reading records one particular moment, and a reading taken while the treatment is still running is in any case not the same thing as the measured value after shutdown and standing.
It is therefore not surprising whether the online reading is lower or higher than the laboratory value; what matters is to fix the difference. With the same oil and the same treatment conditions, if the difference between the two stays within a stable range, the online instrument can be used for process monitoring with confidence. If the difference fluctuates, then the flow condition at the sampling point or the instrument itself has a problem, and the installation and the verification should be checked first.
How to use them on site without contradicting each other
- Acceptance testing still follows the sampling and laboratory testing of GB/T 7600. An online reading below 3 ppm cannot be written down as a pass, because it is not a measured value from the standard method.
- Establish a comparison baseline. For the same machine and the same oil, record the online reading and the laboratory result as pairs; after running for a period, the position of the deviation between the online reading and the standard measured value on that machine becomes visible.
- Write the online reading into the operating record. Trend data shows whether the treatment process is stable, and says more about the state of the machine than a single pass does.
- Clean and verify the probe and the sampling point on a fixed cycle. Both a drifting reading and a blocked sampling line make the online data meaningless, and this kind of failure is not visible.
The online instrument answers the need to find problems in time; the standard method answers the need to make the judgement. Each has its own use, and neither can substitute for the other.
