The breakdown voltage is the indicator that produces a dispute most easily at the acceptance. On the same batch of oil one instrument reports a pass and the other reports a fail, and both sides can produce a report. In most cases it is not that one of them measured it wrong, it is that the two sides were not working with the same set of test conditions at all.
Why the electrode size affects the reading
The breakdown voltage reads the voltage that is needed to break the oil down, and the breakdown strength depends closely on the distribution of the electric field. Whether the electrode is a sphere gap or a plate, how large the sphere diameter is and how wide the gap is all determine whether the field in the gap is uniform and whether there is a distortion at the edges. When the gap grows, the breakdown usually needs a higher voltage; when the gap is too small, there is too little oil in the gap, the effect of the trace particles and the bubbles is magnified and the reading can be decided by a single defect.
So before two sets of data are compared, it has to be confirmed that the two measurements used the same type of electrode and the same gap.
What GB/T 507 specifies
GB/T 507 is the standard test basis for the breakdown voltage, and it fixes the test conditions: the form of the electrode, the gap distance, the rate of rise of the voltage, the waveform and the frequency of the applied voltage, and the test temperature. This is exactly where the meaning of the standardisation lies: only with the conditions the same are two numbers comparable.
Instruments used for the self-testing on site often differ from the standard method in the rate of rise and in the waveform. When the rate of rise is too fast, the field has no time to establish a stable breakdown channel and the reading may come out high; when it is too slow, the bubbles and the impurities in the oil have more time to migrate into the gap and the reading may come out low. The oil temperature affects the result as well, the breakdown voltage falls as the temperature rises.
The common sources of a difference between the readings of two instruments
First, the test conditions differ, including the electrode, the gap, the rate of rise and the waveform. Second, the condition of the electrode surface: an oil film, the residual impurities and the degree of cleaning all change the result. Third, the oil sample changes during the measurement: the exposure to the air takes up moisture, and standing may allow the particles to separate out and settle, so how long the sample stands after it has been taken has a clear effect on the reading. Fourth, the oil temperature differs, and comparing the numbers between the batches without bringing them to the same temperature will not match.
Fifth, two kinds of error have to be separated: the random error and the systematic deviation. The first becomes visible through repeated measurements, the second is caused by a difference in method and can only be removed by making the conditions the same.
Can the self-test data be used for the acceptance
Self-test data can be used for the acceptance only when the test conditions are the same as the acceptance method. If the contract states only that the breakdown voltage shall be not less than 65 kV, without the method, the electrode size, the gap, the temperature and the sampling procedure, the numbers obtained are not comparable and the dispute cannot be settled.
The more reliable practice is to agree the method and all the conditions before the acceptance, to take GB/T 507 as the reference, and to bring the water content, the gas content and the cleanliness into the judgement at the same time. The breakdown voltage is a combined criterion, a single number says little about the overall condition of the oil.
| Variable | Effect on the reading | Control on site |
|---|---|---|
| Form of the electrode and the gap | Determines the field distribution and the voltage required | Fix one electrode form and one gap |
| Rate of rise of the voltage | Too fast reads high, too slow reads low | Set it according to the standard method and keep it consistent |
| Voltage waveform and frequency | Affect the breakdown process | Use the same way of powering |
| Oil temperature | The reading falls as the temperature rises | Bring it to the same temperature before comparing |
| Cleanliness of the electrode and the oil cup | An oil film and the residues change the result | Clean and inspect before every measurement |
| Sampling and standing | Moisture uptake and particle settling both change the reading | Fix the sampling and the standing time |
- State the test method in the acceptance clause and cite GB/T 507.
- Before comparing on-site self-test data with laboratory data, check first whether the four conditions are the same.
- Repeat the measurement on the same sample to see whether the deviation is random or systematic.
- Judge the breakdown voltage together with the water content, the gas content and the cleanliness, do not use it on its own.
My rule is that when two numbers do not match, compare the conditions first, then the instruments, and only at the end the oil.