The same oil sample sent to two laboratories, and the figures on the two reports do not match. This is a common situation in oil testing. In most cases neither report is wrong; the two tests simply differ in method, in conditions or in the basis on which the readings are interpreted. The value of oil testing does not lie in obtaining a piece of paper, but in whether the figures can confirm one another, so reconciling them needs a fixed procedure.
First: align the method codes
For the same indicator there is often more than one way of performing the test. Water content is measured to GB/T 7600, gas content to GB/T 423, breakdown voltage to GB/T 507, and cleanliness is classified to NAS 1638. For one and the same water content indicator, the readings from the on-site rapid method, the Karl Fischer method and the microbial method can land on different levels. Which method the report states decides whether the two figures can be compared at all.
Recording the method code is more useful than recording the figure itself. The figure changes with the condition of the oil and with the timing of the sampling, while the method code is what tells you whether two figures describe the same thing. When the report arrives, look at the method column first rather than at the result column.
Second: align the sampling conditions
The condition of the oil sample is fixed at the moment it is taken out of the machine. A sample taken online and one taken after the oil has stood, a sample from the middle of the flow and one from the sediment at the bottom of the tank, can differ completely in gas content and water content. Whether the sampling point is downstream of the filter, whether the machine was running or stopped at the time of sampling, whether the sample bottle was sealed, and whether it was shaken in transit: all of these are written into the final reading.
- Fix the position of the sampling point and take both batches at the same point.
- Fix the state of the machine at sampling, whether the sample is taken running or stopped, and write it into the record.
- Standardise the sealing of the sample bottles and the way they are transported, so that no bubbles and no moisture change on the way.
- When the same batch of oil has to be tested several times, retain a duplicate sample before sending it for testing.
Third: align the basis of judgement
Before putting two figures side by side, first confirm that they refer to the same thing. Water content is expressed either in ppm or in volume percentage, and cleanliness has two classification systems, NAS 1638 and ISO 4406. There is no one-to-one conversion between the two systems, and a figure converted by force cannot be used as a criterion.
| Indicator | Common basis | What has to be aligned when judging |
|---|---|---|
| Water content | GB/T 7600 | The method code and the unit: check whether it is a mass ratio or a volume ratio |
| Gas content | GB/T 423 | The timing of the sampling and the state of the machine |
| Breakdown voltage | GB/T 507 | The electrode size and the oil temperature |
| Cleanliness | NAS 1638 | The classification system; it is not converted directly to ISO 4406 |
Take the processing capability of the ZYD-100 as an example. The oil it produces corresponds to a set of indicators: water content 3 ppm, cleanliness NAS 1638 class 5, gas content 0.01% and breakdown voltage 65 kV. These figures are the reference for acceptance, but on the condition that the test method and the sampling conditions are the same on the equipment side; otherwise, when they do not match, it is impossible to tell whether the problem lies in the machine or in the test.
Make the reconciliation a fixed action
When two reports do not match, the most effective action is not to argue but to lay the conditions out and compare them: method code, sampling point, sampling time, sample storage and transport, and the judgement threshold, five items aligned one by one. In most cases where the figures do not match, the cause is already visible in the third item, and the time that is left can be used to judge the condition of the oil itself.
Do not rush to suspect the equipment when the figures do not match. Align the sampling and the method conditions, and most of the disagreement shows itself.
