There is an old expression, ‘‘Oil and water do not mix.’’ Thus, oil is not usually thought of as having a great affinity for water, and in fact it doesn’t. However, the kraft paper insulation found in most power transformers has a tremendous affinity for water.
In fact, cellulose is often used as a drying agent or desiccant. If there is moisture present in the transformer, it will usually wind up in the kraft paper insulation. Moisture not only weakens the insulating properties of the kraft paper, it also accelerates the rate of aging.
Therefore, in order to prolong the life of a transformer, moisture must be monitored. Since samples of the insulation cannot be taken while the transformer is in service, water-in-oil analysis is used to monitor the moisture content of the kraft paper as a surrogate.
There is a known equilibrium between moisture concentrations in the kraft paper versus the moisture concentrations in the oil based on the temperature of the paper and oil. The equilibrium is expressed by the so-called Piper chart, shown in Figure below.
Notice that as the temperature increases, water is driven from the paper into the oil. At elevated temperatures the oil is able to dissolve more water than at lower temperatures. The relationship can be expressed by the following equation.
T = 31.52 - 26.605 Ln pct + 17.524 Ln ppm
where
T = temperature (°C)
pct = % water in paper
ppm = ppm water in oil
When doing a water-in-oil analysis, a syringe sample of oil is taken from the drain valve. Care must be exercised so that the oil is not exposed to the atmosphere. (Any exposure to the atmosphere will cause the oil to quickly reach equilibrium with the air.
Since ambient air usually contains quite a bit of moisture, this will generally immediately saturate the oil with water and produce a meaningless analysis.) The oil temperature is recorded at the time the sample is taken and the sample is then sent to a chemical laboratory to analyze the ppm water in oil.
From the ppm in the oil sample and the temperature of the oil, the Piper chart can be used to get an approximate indication of the percent moisture in the kraft paper.
Note that the temperature of the oil/paper interface has a significant effect on the equilibrium moisture concentration, but the temperatures of the oil and the paper vary depending on location. We would then expect the equilibrium moisture concentration to vary as well, which it does.
Generally, the insulation near the hottest spot will have less percent moisture than insulation exposed to cooler oil at the bottom of the transformer. An ‘‘average’’ value of the percent moisture concentration could be calculated from an ‘‘average’’ temperature; however, this may result in a misleading assessment of the transformer’s state because of the wide variation in moisture concentrations.
A conservative assessment would base the percent moisture on the oil temperature at the bottom of the tank. According to the Transformer Maintenance Institute, 2% is the absolute upper limit for acceptability for percent moisture in kraft paper.
Generally, if the percent moisture is less than 1%, the transformer is considered ‘‘dry.’’ There is also an equilibrium equation between vapor pressure of water in air (humidity) and % water in paper.
T = 40.17 + 22.285 Ln pct + 14.056 Ln vap (8.8.2)
where vap vapor pressure, mmHg.
Since the dew point of air is related to the vapor pressure, a dew-point measurement of the space inside a transformer before oil filling is a very good indication of the amount of water locked in the paper. This will determine whether oil filling should proceed or further drying is necessary.
No comments:
Post a Comment