A water content of ≤3 ppm (GB/T 7600) after treatment is often taken as the pass line for the acceptance of a machine. But there is a gap between "acceptable" and "suitable", particularly when the oil is going to be filled back into a transformer that has been in service for many years.
What order of magnitude is 3 ppm
ppm is a mass ratio of one part per million, so 3 ppm means that the water left in the oil is roughly three parts per million of the weight of the oil itself. This quantity is so small that it takes a dedicated Karl Fischer moisture determination to measure it reliably, and for the same reason it demands far more from the sampling and the testing stage than a coarse indicator such as "water content below 1%".
Three kinds of water, three levels of treatment difficulty
- Dissolved water: dispersed in the oil in molecular form, it is carried away relatively easily under vacuum and heating, and it is the main target of ordinary vacuum degassing.
- Emulsified water: dispersed as fine droplets, it affects the insulation most directly and is the hardest to remove; it needs heating together with vacuum and a sufficient residence time.
- Free water: already separated out and settled at the bottom of the vessel, this is a different condition, and the settling and the separation have to be dealt with first.
The differences between the three kinds of water also decide the treatment sequence and the choice of equipment. Dissolved water is dealt with mainly by vacuum degassing; emulsified water needs temperature, vacuum and a sufficient residence time acting together, and both the difficulty and the cost are higher; free water should be allowed to settle and separate first, and only then should the clear oil on top be sent for degassing. Talking about dehydration with the three kinds of water mixed together makes it easy to underestimate the actual difficulty of the treatment.
The slow erosion happens between the oil and the paper
What really needs watching is not the water in the oil itself, but the fact that it migrates into the insulation paper. Insulation paper is a porous fibrous material with a strong affinity for water, and once the water content of the paper rises, its dielectric strength and its mechanical strength fall together. Under the effect of temperature, the water in the oil migrates into the paper at an increasing rate, and at the same time takes part in the hydrolysis reaction between the oil and the paper to form acidic products. The whole process advances over years, and it normally comes with no alarm signal at all, until an overhaul finds that the paper has already become brittle.
Two common sources of distorted data
The first is the sampling stage: a container that has not been dried, residual moisture on the bottle mouth and humid air falling into the sample all make the reading too high and mask the real problem. The second is storage and transport: an unsealed drum mouth, drums stored through the rainy season and condensation on the inside wall of the container all put the moisture back in after the treatment has finished. The constant temperature range of 20–80 °C and the continuous running capability of the ZYD-100 decide that it can hold the water content steadily below the limit over a long period, instead of producing one momentary good figure from a short treatment run.
Damp insulation paper cannot be brought back by dehydrated oil. Treating the oil and drying the paper are two different things: the first is easy, the second needs vacuum drying.