Power Transformers Aging Diagnostic: Methanol from the Lab to the Field

Presented By:
Mohamed Ryadi
Électricité de France, France
Marie-Claude Lessard and Jocelyn Jalbert
Hydro-Québec, Canada
TechCon 2017


Transformer fleet renewal has always been a major concern for utilities. In view of the high financial investment level to be considered, the selection and the prioritization of critical units become crucial. This ranking process requires reliable and easy to implement diagnostic tools.

Aging condition of the cellulosic insulation system is a factor in the prioritization of the financial investment. Therefore, the challenges are high regarding the availability of diagnostic tools for such development. Existing tools, such as the analysis of by-products in oil, like furanic derivatives or carbon oxide compounds, have shown their operational limits. Alcohol based chemical marker like methanol, brings complementary advantages to the improvement of cellulose insulation diagnosis. For more than a decade two large utilities, one North-American and one European, have collaborated to make a significant improvement in the development and the use of methanol as an oil-dissolved aging marker for power transformers.

This paper proposes an insight into the theory and field experience about this third generation chemical aging marker. Laboratory investigations which allowed to consolidate the relevance of methanol, will be presented. Due to the high volatility of methanol, some precautions need to be taken into account when performing laboratory aging studies. Moreover, the importance and the influence of some physicochemical parameters of methanol behaviour will be described. Finally, utility feedback concerning the use and the implementation of this chemical aging marker as a reliable diagnostic tool on real transformers in operation will be presented.


1. Introduction

Following the first publication by Hydro-Québec (HQ) in 2007 on the use of methanol for the determination of cellulose aging conditions, this molecule continues to arouse interest. Many publications in scientific journals and presentations at conferences revealed the potential of this small molecule to follow the state of the cellulose insulation. Nowadays, many laboratories around the world have adopted this routine test for their clients. Efforts have been dedicated to the interpretation of the results and the standardization of an analytical method in the ASTM D27 Committee and in the ongoing IEC standard 63025. This process of standardization is intended to make globally accessible this analytical method. In addition, two working groups of CIGRE were formed (JWG A2D1.46 and A2.45). The first working group studies the performance of the different chemical markers of the cellulose degradation in transformers and the second working group establishes a procedure for sampling the paper wrapping conductors, performed either during forensic investigations of the failed equipment or extracted from equipment removed from operation for failure risk reasons.

The transfer from laboratory experiments to the field application was possible thanks to the correlation between oil analysis and samples of papers extracted from transformers during the post mortem investigations. The use of Électricité de France (EDF) experience on those markers in the diagnosis of the nuclear GSU (Generator step-up transformers) units gave relevant results which led us to consider an adapted approach for their applicability on hydro power plant GSU units. The disparity and the particularity of transformer operation (cyclic load) of the hydro plants’ fleet raised challenges to adapt the use of these markers. HQ successfully added this marker to their health-index decision protocol for the transmission power transformers by taking into account some fundamental parameters. Contrarily to GSU, transmission transformers have variable load challenging the interpretation due to the partitioning effect between the oil and the paper.

2. Thermal aging diagnosis

One of the aging processes of transformers is the degradation of cellulose insulation. The effect of excessive moisture, acidity or thermal stress leads to the breakdown of cellulosic insulation which results in a significant drop in the mechanical strength of the paper. As long as an aged cellulosic insulation is not subject to a brutal mechanical stress (forces due to inrush current or short circuit current), the risk of damage is reduced. But such a unit will be less reliable because of its vulnerability with regards to network constraints which are not necessarily predictable. Many papers and a more recent study confirmed the correlation between the variation in mechanical properties of paper and the measurement of its degree of polymerization [1].

2.1. Direct measurement

The well-established technique to assess aging of transformer solid insulation is to perform degree of polymerization measurement by a viscosimetric technique (DPv). To do so, paper or pressboard samples need to be taken on the active part. This technique is intrusive and partially destructive since it required at least a partial dismantling of active part. This is not practically feasible for a transformer in operation but applicable on failed or scrapped transformers. For the measurement of this feature, two standards can be used, the IEC 60450 and ASTM D4243. It is recognized that paper near its end of life corresponds to a DP of about 200.

2.2. Indirect measurement

Studies in the 80s have correlated the existence of aging markers to the degradation of cellulose. Indirect paper insulation diagnosis has been developed on the basis of the use of oil soluble degradation by–product markers. In contrast to solid insulation, the mineral oil used as dielectric and cooling liquid circulating in the active part can easily be sampled from a transformer in operation. The physicochemical analyses for diagnostic purposes have motivated important activities on chemical markers to evaluate the degradation of the insulation. Several avenues have been explored. For a long time, dissolved gas analysis (DGA) has been utilized by taking into account the analysis of the loading of the transformer during its regular operation. Moreover, the evolution of carbon oxides (CO/CO2) can be used to follow the cellulose degradation [2-6]. Specific aging marker research began with the use of five furanic derivatives, specifically the 2-furfuraldehyde (2-FAL), which was the subject of many studies during the recent decades [7-8]. 2-FAL can be presented as a second generation of cellulose aging marker with an advantage over the carbon oxides approach.

Indeed, its generation is specific to the cellulose degradation as opposed to carbon oxides that can come from oil degradation as well. Since almost 10 years ago, a third generation of markers known as the alcohol based markers have been investigated and have already shown great efficiency to assess paper degradation.

3. Origin of methanol

3.1. Laboratory investigations

Since 2001, Hydro-Quebec research institute (IREQ) has worked on an important R&D project related to cellulose degradation. From this project, two alcohol based molecules, methanol (CH3OH) and ethanol (C2H5OH) raised interest and a first article on the subject was published in 2007 [9]. EDF, having the same concern about cellulose aging of their transformer fleet particularly for the ones insulated with thermally upgraded papers (TUK), started a collaborative R&D project with HQ in 2006. Significant kinetics studies of cellulose degradation and reliable results on this marker have been generated in laboratory experiments and have already been published [10-12]. These investigations allowed to confirm the link between the buildups of methanol dissolved in oil and cellulose chain scissions. Today, the methanol marker has reached a certain maturity, however, ethanol is still in development and more research is needed to understand its behavior [13].

Figure 1 shows the correlation between the concentration of methanol and the number of scissions by cellulose chain (corresponding to the DP variation). The linear relationship between the content of methanol and the value of DP has already been established in previous studies [10-12]. As it can be noted, the linearity between the concentration of methanol and the DP is well demonstrated for temperatures ranging from 60 to 120°C.

Figure 1 Example of cellulose chains scission number and methanol concentrations correlation [10]

In terms of advantages and drawbacks, one can mention the limits of the 2-FAL regarding the type of insulating paper involved. Tests in laboratories and results from the field demonstrate that no significant concentration of 2-FAL marker is generated when TUK- paper is used.

3.2. Field investigations

Results from the field (EDF and HQ) have proven that methanol can be measured in power transformers [14-17] with a significant concentration in oil. Moreover, these results, as shown in Figure 2, are generated regardless of the transformer design (shell or core), type of breathing (open breathing or nitrogen blanket), oil type (inhibited or uninhibited, naphthenic or paraffinic) and winding insulating paper type (standard Kraft or thermally upgraded).

Figure 2 a) EDF [16] and b) HQ [9] Methanol concentrations detected in a large number of transformers

4. Using methanol as an aging marker- Some concerns

4.1. Laboratory aging studies

In order to perform accurate diagnostics, utilities need to have confidence in the measured data. Indeed, the analytical method needs to be precise, accurate and sensitive for the concentration level expected. In the case of methanol, an extraction by headspace and a separation by gas chromatography are used by all the laboratories. However, for the detection, two alternatives are under evaluation by ASTM and IEC i.e. mass spectrometry (MS) and flame ionization detectors. Mass spectrometry allows better sensitivity and minimizes the risk of interference peaks. However, due to some problems with the stability of the MS signal, the authors suggest using an internal standard to compensate for signal variation [18].

Another frequent problem arising from the high volatility of methanol is the loss of methanol during aging after a certain period of time due to the use of an improper vessel. Most of the time, when aging is performed in vials sealed with a septum, leaks of the volatile by-products are observed. This is why, most of the time, sealed glass ampoules are used and highly recommended. Another important phenomenon is the partitioning of methanol between the paper and the oil. When a sample is aged and kept out of the oven, a constant cool time is required to be able to compare aging experiments together. Finally, when the paper samples reach a high level of degradation (DP~200) some authors noted a decrease in the methanol evolution [9, 19-20]. The hypothesis of this decrease was first attributed to a partition phenomenon but some possibility of an esterification of the methanol due to presence of low molecular weight acids is also under investigation [21]. However, another scientific contribution reveals that under highly stable silicone oil, this decrease seems not to be present [22].

4.2. Field interpretation

The input data for the interpretation of results of all aging markers including methanol has to take into account a number of parameters. The design of the transformers is a key factor, because the mass ratio of solid to liquid insulation depends on the power, voltage and type of operation. Therefore, the concentrations of the chemical marker should be corrected to take these parameters into account.

During on-site oil sampling, the temperature of the latter depends on the load of the transformer, the cooling stage and of the ambient temperatures. Moreover, transformer oil is also characterized by acidity and moisture which depend on the age, the type of breathing and the insulation drying quality of the transformer.

4.2.1 Impact of the design

The EDF transformer fleet is composed of two type of design, shell type and core type transformers. Shell type design uses more cellulosic insulation than core type design. In consequence, the solid to liquid ratio is higher for shell than for core designs. The higher the voltage, the more solid insulation and oil are involved. It is important to note that for different transformer design, the solid to liquid ratio varies but in relatively low proportions as shown in the graph Figure 3.

Figure 3 Example of masse ratio Cellulose and Oil

4.2.2 Impact of the temperature

Nitrogen blanketed shell type transformers (ODAF) which regularly operate at full load were investigated. It was then possible as shown in Figure 4 to measure the concentration of two chemical markers (methanol and 2-FAL) versus the temperature during a varying load over a certain period of time. These results demonstrate that it is possible and necessary to use temperature correction factors. It has already been mentioned that oil temperature variation due to load, ambient temperature and cooling condition affect the marker’s concentration in oil.

Figure 4 Variation of aging marker concentration under different transformer load condition

A recent collaborative study between EDF and HQ, on small 100 KVA transformer models using paper wrapped conductors at controlled temperature (see Fig. 5a), allows the establishment of equilibrium curves for the alcohols bases and furanic derivatives markers. The authors [23-24] were able to determine that, at equilibrium for a specific temperature, correction factors (CfM) for each aging marker (M) were [M]20°C = [M]Ts*CfM.

Figure 5 a) Methanol behavior with oil temperature variation and b) CfM Equilibrium markers curves.

By applying the same fit function used in IEC 60422 to the data points obtained in this experiment (see Fig. 5 b)), the following equation in the form CfM = ae -bTs was obtained for moisture and the different aging markers. These equation parameters are reported in Table 1. It is interesting to note that in the case of moisture, there is a good correlation with the literature data [25] with a pre-exponential factor a = 2.59 compared to 2.24 and an exponential factor b = -0.04 for both cases. This figure reveals also that the equilibrium speeds of each marker studied were H20 as the fastest and 2-FAL the slowest.

Table 1 Equation parameters for some chemical markers and moisture [25]

The above results are the first step toward an accurate and comprehensive evaluation of the cellulose health inside power transformers. Correcting any marker’s concentrations that have a partition with the cellulose insulation may allow the determination of threshold values and of more accurate concentration trends during the entire transformer life. These results demonstrate again the importance of recording the transformer oil temperature during sampling. To optimize the paper/oil equilibrium, it is recommended to sample the transformer oil under load at a stable temperature as long as possible. This will allow all aging markers to be in equilibrium between the two phases (paper/oil). This correction factor is applicable for oil temperatures over 20°C. It is important to note that the effect of other parameters such as oil acidity, type of oil and paper need to be studied since they might also influence the partition.

4.2.3 Impact of the oil quality

The oxidation condition of oil is another factor that can affect the quality of interpretation of the aging markers [26]. It is well-known in the literature [27] that alcohols can react with organic acids to form esters and water. Many studies have demonstrated the formation of acidic compounds from the degradation of mineral-insulating oil and insulating paper. According to Ingebrigtsen [28], low molecular weight acids such as formic, acetic and levulinic acids are formed from cellulose degradation, in addition to stearic and naphthenic acids from oil degradation, which have higher molecular weight. A recent paper [21] proposed that the methanol may be chemically unstable in a very acidic environment. The main concern of these authors in using methanol as a marker for paper degradation was its high volatility, apart from some possible esterification in methyl acetate. This reaction was observed in laboratory experiments in presence of high low molecular weight acids concentrations (10%) (acetic and formic) added in new insulation mineral oil containing a large amount of methanol (8%) or in a very highly acidic oil (0,54 mg KOH/g ) coming from a real transformer. It can be mentioned that these extreme oil conditions are very unusual for the vast majority of power transformers.

4.2.4 Methanol behavior after oil processing

The marker’s capability to re-equilibrate with oil after oil processing, such as degassing, filtering or reclaiming, is a factor to be taken into account in order to diagnose the aging of the cellulosic insulation system. In order to compare the behavior of methanol versus 2-FAL, investigations has been done after transformer degassing. Figure 6 illustrate the results of these investigations over different durations of oil processing.

Figure 6 Aging markers recovery after oil processing.

5. Field application

5.1. Homogeneous transformers families ranking

For the EDF fleet, the operating conditions history is well known for each unit and it is possible to rank the transformers in well-identified families [10, 17]. This family ranking takes into account: the type of design (shell or core), manufacturers, power, voltage, type of use of the transformer (evacuation, transmission, distribution, etc.), paper and oil types. This step which is a kind of ranking of the transformers into homogeneous families is the first phase of the data analysis. In the current state of knowledge of the aging markers, this first selection is the initial recommendation for the interpretation of the measured concentrations. Homogeneous families of transformers have been selected to achieve meaningful comparisons; each transformer has an operating period between 1 and 30 years.

5.2. Transformers mapping

The family of 23 nitrogen blanketed transformers using thermally upgraded paper (TUK) has been investigated in measuring the marker’s concentration in the oil. In this family, there are some GSU units cooled by a standard cooling system specified by the manufacturer and other upgraded units with the same active part design but cooled with an enhanced cooling system defined by the manufacturer. The conventional markers CO/CO2 and 2-FAL don’t reveal any significant difference between units of this family. This confirms the limits of 2-FAL regarding the TUK while the methanol marker allows differentiation of specific units regardless the type of paper in the transformer windings. As shown in Figure 7, the average concentration rate of methanol is approximately the same for the 20 to 25 years old transformers as for those having operated for 5 to 10 years. An abnormal aging of the young transformers using the enhanced cooling system was suspected. Forensic investigations on a failed transformer of this family showed localized low DP values confirming poor local cooling of the winding. The use of these markers reveals also the need to monitor closely some young units as indicated with red arrows in the following graph. After these finding, it had been decided to consolidate the diagnostic in achieving another mapping after 2007 as illustrated in Figure 8, focusing on some units of the investigated transformer family. The measured concentration of methanol has been confirmed and also two units were monitored with a specific procedure.

Figure 7 Methanol marker in a homogenous family of GSU transformers.
Figure 8 Methanol marker in a second homogenous family of GSU transformers.

5.3. Input on post-mortem investigation

The DP results coming from post-mortem investigations represent key data for the reliability of the thermal aging diagnosis of cellulosic paper used in the transformers. The approach is to correlate the most possible analyses of the marker’s content in oil and the real measure of mechanical strength of insulating papers. HQ has recently published a paper on some post-mortem investigations in correlation with methanol measurement [29]. For EDF, it is important to have a similar signature of its own fleet in order to associate it to any outcome obtained from post-mortem investigations.

Case N0.1: 30 MVA core type power plant transformer (ONAF) A post-mortem investigation has been done on a unit which was in service since 1986. This transformer is not permanently loaded, but a cyclic loading is usually applied. This unit #1 has two sister units (#2-3) still in service (Table 2). The average DP values measured on the low and high voltage winding were 200 and 300 respectively confirming the end of life of the cellulosic insulation of this unit.

Table 2 Characteristics of the 30 MVA transformers

Even if it is still not possible to establish general criteria to associate the concentration of the measured aging markers to the cellulosic condition, if unit #1 has the same design and load conditions that units #2-3 have, we could expect that the cellulose insulation of these other units is in the same condition. This hypothesis is supported by the high level of both aging markers measured in both units.

Case No.2: 360 MVA shell type (GSU) transformer (OFAF) Another example of methanol effivacy is a post mortem investigation of a 360 MVA shell type GSU transformer taken out of service and investigated in June 2008. The transformer was in service for 23 years. The unit was cooled according to a specific procedure defined by the supplier. It had been monitored and oil sampled in 2007 and 2008. In this period, the methanol marker increased from 125 ppb to 460 ppb. The post-mortem investigation reported a hot spot on the magnetic shunt protecting the T bar and poorly cooled winding (Fig. 9) and indicated a scattered value of DP (Table 4) with a lowest value around 450 units. This observed thermal fault aligns with the rapid increase in two years of methanol concentration.

Figure 9 Hot spot on on the magnetic shunt protecting the T bar and poorly cooled winding
Table 4 Measured DP of the insulating paper of the scrapped 360 MVA shell type GSU transformer.

6. Alcohol-based markers: Challenges and future work

The work perform on the methanol marker during the last decade allows to increase the knowledge of the cellulose degradation mechanism. Different issues raised by the scientific community, such as the decrease of methanol concentration in an acidic environment, need to be further investigated. An aging model based on methanol is presently under development at HQ. This model already provides threshold values for assessing the state of the paper insulation. Due to differences between the laboratory and the field, the best way to validate and fine tune an accurate aging model is by performing post-mortem investigations. Other challenges will have to be taken into account such as the design (core or shell) and the influence of different physicochemical parameters affecting partitioning. Ethanol, another alcohol based marker which already has shown potential to diagnose hotspots is still under study. Establishment of threshold criteria for ethanol would be useful for utilities.

7. Conclusion

Methanol has proven its effectiveness in open breathing and sealed equipment because the majority of this aging marker stays in the cellulose insulation. Even though some loss of methanol could be observed due to its volatility, the methanol remaining in the cellulose will attempt to reach equilibrium again (like moisture). However, this volatility aspect remains a challenge in laboratory aging cell studies and when performing analytical measurements of methanol particularly with calibration and control standards preparation. Finally, the transformer oil must be sampled with the same precautions taken when performing DGA sampling.

Methanol marker can be used as a complementary tool to the conventional markers of carbon oxides and 2-FAL. Parameters such as temperature and oil quality should be taken into account. It is important to report the measured concentrations at a reference temperature for a better understanding of the evolution of the markers and to facilitate a comparison between units when it comes to performing a mapping process on a transformer fleet. The comparison is relevant for homogenous families of transformers and also at a level of one power plant where the transformers are similar or identical. Post mortem investigations to obtain a real DP reference need to be correlated to the marker’s concentration to ensure a strong reliability of the diagnostic.


[1] O. H. Arroyo, I. Fofana, J. Jalbert, and M. Ryadi, “Relationships between methanol marker and mechanical performance of electrical insulation papers for power transformers under accelerated thermal aging,” IEEE Transactions on Dielectrics and Electrical Insulation, vol. 22, pp. 3625-3632, 2015.

[2] R. Tamura, H. Anetai, T. Ishii, and T. Kawamura, “Diagnostic of aging deterioration of insulating paper,” JIEE Proc Pub A., vol. 101, p. 30, 1981.

[3] K. Goto and t. al., “Mesure de la température des enroulements des transformateurs de puissance et diagnostic du vieillissement par détection du CO2 et CO,” presented at the CIGRE, Paris, 1990.

[4] N. Dominelli, “The analysis of furanic and non-furanic compounds as a transformer diagnostic technique, preliminary paper,” presented at the Doble Conference, 1995.

[5] H. Yoshida, Y. Ishioka, T. Suzuki, T. Yanari, and T. Teranishi, “Degradation of Insulating Materials of Transformers,” IEEE Transactions on Electrical Insulation, vol. 22, pp. 795-800, Dec 1987.

[6] K. Hisao, M. Teruo, M. Yoshihiro, N. Sadao, and H. Takashi, “Absorption of CO and CO gases and furfural in insulating oil into paper insulation in oil-immersed transformers,” presented at the IEEE Conf. Rec. Int. Symp. Elect. Insul., Pittsburgh, PA, Jun. 5–8, 1994.

[7] P. J. Burton, M. Carballeira, M. Duval, C. W. Fuller, J. Graham, A.De Pablo, J. Samat, and E. Spicar, “Application of liquid chromatographyto the analysis of electrical insulating materials,” presented at the CIGRE Conf., Paris, France, 1988, paper 15-08

[8] CIGRÉ WG. D1.01.13, “Furanic compounds for Diagnosis” Technical Brochure 494, Electra, Elt-261-06, NO. 261, pp 73-79, 2012.

[9] J. Jalbert, R. Gilbert, P. Tétreault, B. Morin and D. Lessard-Déziel, “Identification of a chemical indicator of the rupture of 1,4-β-glycosidic bonds of cellulose in an oil-impregnated insulating paper system”, Cellulose, Vol. 14, pp. 295-309, 2007.

[10] Gilbert R., Jalbert J., Tétreault P., Morin B., Denos Y., “ Kinetics of the production of chain-end groups and methanol from the depolymerisation of cellulose during the aging of paper/oil systems. Part 1: Standard wood kraft insulation”, Cellulose, vol. 16, 2009, pp. 327-338.

[11] Gilbert R., Jalbert J., Duchesne S., Tétreault P., Morin B., Denos Y., “Kinetics of the production of chain-end groups and methanol from the depolymerisation of cellulose during the aging of paper/oil systems. Part 2: Thermally-upgraded insulating papers,” Cellulose, vol. 17, 2010.

[12] J. Jalbert, E. Rodriguez-Celis, S. Duchesne, B. Morin, M. Ryadi and R. Gilbert “Kinetics of the production of chain-end groups and methanol from the depolymerization of cellulose during the aging of paper/oil systems. Part 3: Extension of the study under temperature conditions over 120oC,” Cellulose, Vol. 22, pp. 829-845, 2015.

[13] E. M. Rodriguez-Celis, S. Duchesne, J. Jalbert and M. Ryadi “Understanding ethanol versus methanol formation from insulating paper in power transformers”, Cellulose, Vol. 22, pp. 3225-3236, 2015.

[14] Y. Denos, A. Tanguy, J. Jalbert, R. Gilbert, P. Gervais and P. Guuinic, “Aging diagnosis by chemical markers influence of core-type and shell type technology”, CIGRE, Paris, France, 2010.

[15] Schaut A, Autru S, Eeckhoudt S (2011) Applicability of methanol as a new marker for paper degradation in power transformers. IEEE Trans Dielect Electr Insul 18:533–540.

[16] M. Ryadi, A. Tanguy, J. Jalbert, C. Rajotte “Alcohols based aging chemical markers for the diagnosis of transformer cellulosic insulation”, CIGRE SC A2 & D1 JOINT COLLOQUIUM 2011, paper PS2-O-4, Kyoto 2011.

[17] Y. Denos, A. Tanguy, J. Jalbert, R. Gilbert, P. Gervais and P. J. Jalbert, R. Gilbert, Y. Denos and P. Gervais, “Methanol: a novel approach to power transformer asset management”, IEEE Trans. Power Delivery, Vol.27, No. 2, pp. 514-520, 2012.

[18] J. Jalbert, S. Duchesne, E-M. Rodriguez-Celis, P. Tétreault, P. Colin, “Robust and sensitive analysis of methanol and ethanol from cellulose degradation in mineral oils”, Journal of Chromatography A, Vol. 1256, pp. 240-245, 2012

[19] S. Y. Matharage, Q. Liu and Z. D. Wang, Aging assessment of Kraft paper insulation through methanol in oil measurement, IEEE Transactions on Dielectrics and Electrical Insulation, vol. 23, No. 3 pp. 1589-1596, 2016.

[20] L. Melzer, CIGRÉ Paris, plenary session, Preferential Subject No.1, Question No.5, Paris 2016.

[21] E. Sousa et al., Evaluation of the chemical stability of methanol generated during paper degradation in power transformers, IEEE Transactions on Dielectrics and Electrical Insulation, Vol.23, No, 5, 2016.

[22] M. Ryadi, J. Jalbert, CIGRÉ Paris, plenary session, Preferential Subject No.1, Question No.5, Paris 2016.

[23] J. Jalbert, M.-C. Lessard, and M. Ryadi, “Cellulose chemical markers in transformer oil insulation Part 1: Temperature correction factors,” IEEE Transactions on Dielectrics and Electrical Insulation, vol. 20, pp. 2287-2291, 2013.

[24] Marie-Claude Lessard and Jocelyn Jalbert, “Establishing Correction Factors for the Partitioning of Furanic Derivatives between Oil and Paper Insulation”, Annual Report – Conference on Electrical Insulation and Dielectric Phenomena, CEIDP, art. no. 6748297, pp. 132-135, 2013.

[25] Mineral insulating oils in electrical equipment- supervision and maintenance guidance, IEC 60422, 2005.

[26] H. Bertrand, et al. “Advanced monitoring and modeling methods for power transformer asset management”, European Journal of Electrical Engineering Subject Area: Engineering: Electrical and Electronic Engineering, Vol. 10, Issue 5-6, pp. 591-620, 2010.

[27] T.W. G. Solomons and C. B. Fryhle, Organic Chemistry, 10ed. John Wiley: USA, 2011.

[28] S. Ingebrigtens, M. Dahlund, W.Hansen, D. Linhjell, L.E. Lundgaard: “Solubility of carboxylic acids in paper (Kraft)-oil insulation system”, IEEE Conf. Electric. Insul. Dielectric. Phenomena, (CEIP), Boulder, Colorado, USA, pp.253-257, 2004.

[29] Jalbert, J. and Lessard, M.-C., “Cellulose chemical markers relationship with insulating paper post-mortem investigations”, IEEE Transactions on Dielectrics and Electrical Insulation, 22 (6), 3550-3554, 2015.

Join our email list

We use cookies to give you the best online experience. By using this website you agree with our cookie policy.