Thursday March 9th, 2017

For diagnosis of the the “Water in the transformer (paper or oil)” criticality, Sea Marconi uses its own diagnostic metrics, namely:

visual signs on the transformer (and those from any internal inspection) are interpreted;

through analysis of the oil and papers (if available following internal inspection) the symptoms are identified and co-factors evaluated. One of these is undoubtedly the temperature, both sampling and ambient temperature. At the same time it is necessary to evaluate the degradation of the oil, the type of oil (between paraffins and naphthenics there are different degrees of solubility, which is even more evident for natural esters, which can reach solubilisation up to 10 times more than mineral oils ) and concentration and type of additives.

The limits recommended limits by IEC 60422 regarding oil in a just-filled transformer and before energisation are:

From Table 3 – Recommended limits for mineral insulating oils after filling in new electrical equipment prior to energization – IEC 60422 ed. 4-2013

Sempre la stessa norma a pag. 28 indica i limiti raccomandati di acqua in olio per un trasformatore in esercizio:

Starting from the analytical results of the “water in oil” test, there are several methods (with different levels of precision) for evaluating the state of cellulose hydration, that is, the water absorbed by the papers. First of all, it is necessary to distinguish between direct methods, which involve the analysis of paper samples taken from internal inspection, and indirect methods, through mathematical formulas and diagrams.

Technical standard IEC 60422, for example, enables (see table below) estimation of the moisture conditions of the solid insulator (papers). To obtain this estimate, it is necessary to calculate (see formulas below) the saturation of the water in oil, that is, the maximum concentration of oil that can solubilise in oil at a certain temperature.

After calculating the water saturation in oil in percentage terms, using the table below (Table A. 1 – IEC 60422 ed. 4-2013) it is possible to obtain an estimate of the moisture of the papers: dry, slightly moist, moist, extremely moist.

Table A.1 – Guidelines for interpreting data expressed in per cent saturation – IEC 60422 ed. 4-2013
T. V. Oommen, “Moisture Equilibrium Charts for Transformer Insulation Drying Practice”, IEEE Transaction on Power Apparatus and Systems, Vol, PAS-103, No. 10, October 1984

For integration, different equilibrium curves can be used (Fabre-Pichon curves, 1960, Oommen curves, 1983, Griffin curves, 1988, Koch curves, 2005). These are diagrams elaborated on the basis of different application case histories that permit scientific prediction of the average amount of water absorbed into the papers in relation to the average operating temperature. Subsequently, knowing the total weight of paper and oil, it is possible to make a mass balance of the total water present in the transformer, both in paper and in oil.

As an alternative to the indirect determination described above, there is also the possibility of employing a direct methodology.
In the presence of an oil-impregnated paper sample, it is possible to determine the amount of water in paper by means of a specific protocol for extracting water through a carrier gas (for example, hot nitrogen). The water extracted is determined using the Karl Fisher quantitative method (method IEC 60814). This concentration (mg/kg) makes it possible to calculate the percentage of water relative to the mass of the paper sample being analysed.

 

Sea Marconi has recently used this method for an important constructor of international transformers. Paper moisture measurement has made it possible to better understand the process of drying the transformer and optimise the entire production process

the database is used to study family or subjective case histories (in the search, for example, for failures in twin machines);

 the factors of uncertainty, speed and evolution over time (trends) of symptomatic indicators are taken into consideration and monitored during the life cycle phases; for this purpose, it is important to disengage from the sampling temperature (which changes in relation to the operating profile), renormalising the analytical result of the water in oil at a temperature of 20 °C (seeIEC 60814]. This calculation is performed using the formula [put formula] pag. 13 (IEC 60422) with correction factor (see Fig. 1, page 41, IEC 60422).

 on the basis of assessment of these key factors, the specific criticality is classified according to type and priority, and type and priority of corrective actions are identified at the same time.

Temperature is a key factor in the degradation processes of paper and oil insulating materials. Some temperature sensors (internal optical fibres or external sensors located at representative points) can monitor the transformer during its life cycle. These data, modelled using a specific “operating profile algorithm”, make it possible to diagnose the current situation more effectively and predict future evolution (prognosis) in order to prevent and mitigate specific criticalities.

Real example

TCat A transformer (see Table 2 IEC 60422), GSU elevator type generation (breathing with conservator and silica gel)
Voltage: 400 kV, Power: 250 MVA
Oil mass: 50,000 kg of non-inhibited paraffin-based mineral oil
Paper mass: 2,500 kg
Paper type: non-TU kraft paper
Cooling: ONAF
Environmental severity: normal, temperate climate height (asl) 250 m, ground level
total acidity of 0. 25 mg KOH/g (“poor” value compared with Table 5, IEC 60422),
colour = 6 dark (“poor” value compared with Table 5, IEC 60422)

Water in paper

Water in paper of new era transformer < 0. 5%. The amount of water in oil for a new transformer must be < 10 mg/kg at 40 °C.

Renormalised at 20 °C there are 4. 5 mg/kg of water in the oil, thus 0. 225 kg of water in the oil of the transformer in the example (4. 5 mg/kg x 50,000 kg = 225,000 mg = 0. 225 kg).

12. 5 kg of water in the initial paper

This gives a ratio of 55. 5, which means that for every kg of water in oil there are 55 kg of water in paper

Water in oil

Water in oil = 40 mg/kg sampled at 40 °C. That value corrected at 20 °C becomes 18 mg/kg, which means that on an oil mass of 50,000 kg, there are 0. 9 kg of dissolved water in the oil (18 mg/kg x 50,000 kg = 900,000 mg = 0. 9 kg)

Thus, moving from 0. 225 kg to 0. 9 kg, the value of water in oil is increased by 400%!

The insulating paper certainly has a degradation process under way: initially it had a mass of 2,500 kg and fell to 1,875 kg.

Using T. V. Oommen equilibrium curves, a relative oil saturation rate of about 5% is estimated, defined by the IEC as “extremely wet“.

These last two data allow us to obtain the water content in the papers: 93 kg (1. 875 kg x 5/100)

Summing up, there would be 0. 9 kg of water in the oil and 93 kg of water in the paper

It is thus shown that, in the real example, the ratio of water in oil/water in papers for a new transformer is 1 to 55; it is now almost doubled, for every kg of water in oil there are about 100 kg of water in papers

 The DP of this transformer has decreased in 35 years from 1000 to 200, understood as a mean value, which conventionally corresponds to the end of thermal life. At the same time, there has been an estimated 25% loss in paper mass; in fact, its weight has dropped from the initial 2,500 kg to 1,875 kg.

You did not find what you were looking for?

Contact us to learn more about your case





    I read the privacy information and authorize the use of my personal data for the purposes specified therein.
    Insert this code: captcha