After an oil test report is issued, two laboratories giving two different figures is not unusual. Looking back afterwards, most of the time the testing itself was not at fault; the sample was not kept under control between the two ends. From sampling to the report, the sample passes through several hands, and each step can change its condition.
Numbering: three things a sample has to state
The first is the origin: which machine, which model, which run. The second is the time: the date and the hour of sampling, and which round of this batch of oil it was taken from. The third is the sampling point: the top of the tank, the middle of the tank or the oil outlet. All three are written on one label, and the label goes on the sample bottle.
Here is why all three are needed. The machine shows which process steps this batch of oil has been through, the time shows how long it has been in the machine, and the sampling point shows which part of it the sample represents. Leave out any one of them and every later discussion of whether the result is good turns into an argument in which each side talks past the other.
Handing over: the container, the seal and the transport
The container is chosen per test item, because the water content, the gas content and the breakdown voltage do not ask the same of it, and a measurement taken in the wrong container gives a different number. The seal has to hold through the journey without leaking and without letting air through. In transit, keep the sample out of direct sunlight and out of a hot environment, and do not leave it beside a heater in the vehicle.
At the handover, pass the three items of information above together with the sample to the next person and have the receipt signed. That step looks trivial, but it is the only record that draws the line of responsibility: if the sample was exchanged, left standing too long or kept at the wrong temperature, there is something to point at.
- Choose the container per test item and write the label with a pen that does not dissolve in oil, so that the handwriting is not wiped away by a solvent.
- After sealing, turn the bottle upside down and look for a leak; if it leaks, take the sample again on the spot rather than sending it to the laboratory and having it sent back.
- On the handover note, state the number of samples, the tags, the time of handover and the signatures of both sides, one line per bottle.

| Step | What has to be confirmed | What happens if it goes wrong |
|---|---|---|
| Sampling | The sampling point, whether the fluid is stable, whether the container is suitable | The result becomes less representative |
| Numbering | Origin, time and sampling point all stated | The result cannot be tied to a specific condition |
| Handover | Receipt signed by both sides, with the time and the state recorded | No traceability when there is a dispute |
| Transport | Avoiding heat and heavy shaking | Dissolved gas escapes, the container leaks |
| Retention | One copy each side, under the same conditions | The two results cannot corroborate each other on a retest |
Retention: one copy or two
The practice is simple and necessary: split the sample into two on the spot after it has been taken, send one for testing and keep one under the same conditions. The equipment site and the testing party each keep one copy, under identical conditions: the same container, the same seal, the same surroundings.
A retained sample is of no use on an ordinary day, and it is decisive on the days that matter: two results that do not agree and need a retest, a report whose validity has run out and needs confirming, an oil problem that needs tracing. The cost of keeping a sample on site is small, one bottle of space and the moment it takes to split it.
Three indicators, three different demands on the sample
The indicators of the ZYD series are water content ≤3 ppm (GB/T 7600), gas content ≤0.01% (GB/T 423), breakdown voltage ≥65 kV (GB/T 507) and cleanliness NAS 1638 class 5. These four are not obtained from one bottle tested once. After sampling, the condition of the sample changes with the time it is exposed, and the different items differ in how sensitive they are to that change.
The order form sent to the laboratory therefore has to state which items are to be tested, because one bottle of sample usually covers one item or two. Which items to test, which samples to take and how to number them: these three are settled before the sampling, not afterwards.
A dispute over test results ends up on the sample. Numbering, handover and retention are what make two reports agree.
