CA2054616A1 - Method of determining stability of insulating oil - Google Patents
Method of determining stability of insulating oilInfo
- Publication number
- CA2054616A1 CA2054616A1 CA 2054616 CA2054616A CA2054616A1 CA 2054616 A1 CA2054616 A1 CA 2054616A1 CA 2054616 CA2054616 CA 2054616 CA 2054616 A CA2054616 A CA 2054616A CA 2054616 A1 CA2054616 A1 CA 2054616A1
- Authority
- CA
- Canada
- Prior art keywords
- oil
- ionization
- concentration
- free radicals
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Landscapes
- Housings And Mounting Of Transformers (AREA)
Abstract
ABSTRACT OF THE DISCLOSURE
The usual method of determining whether failure of a transformer is imminent is to analyze dissolved gases in the insulating oil used in the transformer. The results of such analysis do not always reliably correlate with the actual failure rate under service conditions. A relatively simple, accurate method of determining the stability of an insulating oil and thus predicting transformer failure includes the steps of subjecting the oil to ionization, and determining the concentration of free radicals in the oil before and after ionization, the difference in concentration providing a quantitative indication of the gassing tendency or stability of the oil.
The usual method of determining whether failure of a transformer is imminent is to analyze dissolved gases in the insulating oil used in the transformer. The results of such analysis do not always reliably correlate with the actual failure rate under service conditions. A relatively simple, accurate method of determining the stability of an insulating oil and thus predicting transformer failure includes the steps of subjecting the oil to ionization, and determining the concentration of free radicals in the oil before and after ionization, the difference in concentration providing a quantitative indication of the gassing tendency or stability of the oil.
Description
2 ~ tr,~ > 3 ~
This invention relates to a method of determining the stability of insulating oil of the type commonly used in high-voltage transformers.
Dissolved gaseous decay products are produced in the insulating oils of high voltage transformers by incipient eleetrical failures of various types and intensities. Local over-stressing of the oil/paper insulation inside the windings can supply sufficient energy to break the strong aliphatic carbon-carbon and carbon-hydrogen bonds of some components of the complex blend of hydrocarbons found in the oil.
Aceordingly, the ability to understand and measure the gassing tendency of insulating oils under thermal and electrieal stress permits the prediction and prevention of transformer failure.
The build-up of gaseous decay products in insulating oils during serviee results mainly from the breakdown of weak valence bonds. It is important to understand the process by whieh this breakdown occurs and to identify the energy souree which initiates the breakdown.
The influence of paper, other solid dielectrics, hot spots and partial diseharges make the deeay proeess very complex. These factors all influence degradation of both the insulating oil and the cellulose insulation. It has been observed that the degree of polymerization of the cellulose deereases in the vieinity of hot spots. Formation of high moleeular weight produets (~-waxes) and their deposition in the paper insulation may reduce its permeability, resulting in impaired heat transfer and the creation of hot spots.
Since the hydrocarbon blend of insulating oils consists of more than 3000 components, excited molecules produced by the primary process of energy absorption can undergo a variety of secondary processes, both physical and chemical. Typical processes which affect the 'physical stability' of hydrocarbons are the dissociation of excited molecules into smaller fragments, their rearrangement via bond-breaking or their return to the ground state by releasing absorbed energy.
The aliphatic carbon-carbon bond energy of hydrocarbon oils is about 90 kcal mol~l (approximately 4eV).
Thermal excitation of molecules becomes sufficient to break these bonds at temperatures of 350 to 500C. Thermal cracking of petroleum products is carried out in this temperature range (see Pryor, W.A., Free Radicals, McGraw-Hill Inc., New York, 1966). Below 100C, the normal maximum operating temperature of high-voltage transformers, there is insufficient thermal energy to excite electrons of the carbon-carbon bonds.
The available laboratory methods for testing gas evolved from oils include the Merrell Test (ASTM designation:
D2298-81) which was discontinued in 1987, and the Modified Pirelli Method (ASTM designation: D2300-85) which applies to insulating oils saturated with hydrogen or other gases.
Analysis of dissolved gases in transformer oils (ASTM
2 ~
designation D3612-85) is used to diagnose the type and degree of incipient failure tsee Duval, M., "Dissolved Gas Analysis:
It Can Save Your Transformer", IEEE Elec. Insul., Vol.5, No.6, p.22-27, 1989). However, the results of this analysis do not always correlate reliably with the actual failure rate under service conditions (see Savio, J. Leo, "Transformer Fault Gas Analysis and Interpretation: A User's Perspective", Electrical Insulatinq Oils, STP 998, H.G. Erdman, Ed., ASTM, Philadephia, 1988, pp 83-88).
An object of the present invention is to provide a solution to the above problems in the form of a relatively simple and accurate method for determining the stability of an insulating oil.
Another object of the present invention is to provide a method for determining oil stability which measures the production of free radicals formed during high-voltage ionization.
Accordingly, the present invention relates to a method of determining the physical stability of insulating oil comprising the steps of subjecting the oil to ionization, and determining the concentration of free radicals in the oil before and after ionization, the difference in concentration providing a ~uantitative indication of the gassing tendency of the oil.
Research conducted by the inventors focused on the effects of free electrons, which present a very powerful source of energy. During transient (especially commutation) field enhancements (see Armanini, D. et al, "Aging and deterioration of HV current transformer insulation by very fast transient overvoltages", L'Energia Elettrica, No. 5, 1989), invisible surface imperfections or asperities of the copper windings may eject free electrons. With potential gradients of 0.3 to 0.4 kV/mm, these free electrons have the necessary energy to endanger the electronic structure of chemically-inert hydrocarbons. Although the duration of these transient electrical stresses is only a few microseconds, their cumulative effect cannot be neglected because commutation field enhancements occur during normal service conditions and because asperities in copper windings are unavoidable. In high-speed Schlieren studies of electrical breakdown in liquid hydrocarbons, no streamer was observed originating from within the liquid itself. A rapid rise in light emission occurred just before breakdown (see Wond, P.
and Forster, E.O., "High speed Schlieren studies of electrical breakdown in liquid hydrocarbons", Can. J. of Chemistry, V.55, p. 1890, 1977) which suggests that this process begins in the immediate vicinity of metal surfaces.
Forster's work at high stress on very pure and clean liquids (see Forster, E.O., "Progress in the Field of Electric Properties of Dielectric Liquids", IEEE Trans.Electr.
Insul., Vol.25, No. 1, pp. 45-53, 1390) suggests the need to clarify the role played by dissolved impurities in insulating ,~ ~3 ~ ~ 3~ 3 oils. Even at normal operating temperatures, less stable molecules among the dissolved impurities could accumulate the necessary energy for electronic transitions during the elastic collisions caused by thermal agitation. Such electronic transitions may result in the homolytical breakdown of weak valence bonds. Of the weakly-bonded molecules, peroxides are a frequent source of free radicals which can serve as initiators of auto-oxidation rèactions, as per the equations:
ROOH =~= RO ~ OH
R~I + RO ~ ROH + R
Free radicals can be transformed by reduciion-oxidation reactions into charge carriers, thus increasing the power factor of an insulating oil.
R + [cai~on ] ~ R + [cation]~
It would therefore be desirable to determine the concentration of the free radical precursors. However, no direct analytical method is available at present. The method proposed herein can be used to determine the presence of free radicals arising from these precursors, and to provide an indirect measure of precursor concentration and real oil purity.
The invention will be described in greater detail with reference to the accompanying drawings, wherein:
Figure 1 is a graph of gas pressure versus time for a variety of ionized oils:
Figure 2 is a graph of absorbance versus time for 2,2-diphenylpicrylhydrazyl before and after ionization;
Figure 3 is a graph of absorbance versus wavelength (absorption spectra) for an oil before and after ionization for different lengths of time; and Figure 4 is a graph of absorbance versus wavelength (absorption spectra) for a variety of oils before and after ionization.
The Merrell test (supra) was adopted in 1964 to determine the quantity of gas evolved during the ionization process when oil samples are bombarded with lOkeV electrons.
Using a discharge cell as described in the test, the gas evolved from two new and two used oil samples was measured (Figure 1). The samples include oxidized oil (curve A), new Shell Diala (trade-mark) oil (curve B), new Voltesso 35 (trade-mark) oil (curve C) and purified oxidized oil (curve D). Although an ionization time of 1000 minutes is prescribed in the Merrell test, decomposition after 120 minutes was sufficient to cause a significant pressure 2~ increase above the oil.
The high-energy discharge used for ionization produces not only gaseous decay products, which raise the pressure inside the discharge cell, and ionized molecules, which increase the power factor of the oil, but also, free radicals (see Tanaka, J., "Free Radicals in Electrical Insulation", Proceedings of the Nineteenth Symposium on 2 ~
Electrical Insulating Materials, Osaka, Japan, 1986). The presence of free radicals is experimentally confirmed using the very sensitive reaction of the stable free radical, 2,2-diphenylpicrylhydraæyl (~PPH). Solutions of DPPH, even at a concentration of 10-5 M, are intensely blue~violet coloured with maximum absorption at 532 nm. When the DPPH free electron forms a doublet with the electron of another free radical, the solution changes colour from blue to yellow, with a corresponding decrease in absorption at 532 nm. This change allows determination of the reaction kinetics (see Bednar, H., Sabau, I. and Silberg, I., "Purity of insulating oils", Energetica, V.26, Nr.9, 1978).
For the free radical determinations, a solution of 0.01% DPPH in benzene was prepared. Two ml of the DPPH
solution were diluted with three ml of benzene. A portion of the visible spectrum was recorded from 580 nm to 480 nm using a UV-VIS spectrophotometer. One ml of insulating oil was then mixed with two ml of the DPPH solution and two ml of benzene, and the spectrum was repeated at three minute intervals. A
significant decrease in the absorbance at 532 nm indicated a high rate of consumption of DPPH and thus a high concentration of free radicals in the oil and, indirectly, an unacceptable concentration of impurities in the analyzed oil.
A significant increase in pressure in the test chamber caused by gaseous decomposition during the ionization tests of several oils is shown in Figure 1. As shown in Fig. 2, a marked increase in the concentration of free radicals for the new Voltesso 35 oil was observed by measuring the accelerated consumption of DPPH after the ionization test.
In order for free radlcals to be formed in the oil during ionization, the oil must contain free radical precursors, which may be either impurities that were not completely eliminated during the refining process or contaminants accidentally introduced into the oil.
Tests on oil samples taken from in-service transformers with gassing problems showed the presence of a measured concentration of free radicals along with an unusually high power factor (14.7~) without a significant shift in the visible spectrum. The presence of free radicals is an important indicator that decomposition has occurred in the oil, as are the gassing problems and the increase in the power factor.
Free radicals, since they are not charge carriers, do not directly contribute to the value of the power factor or the interfacial tension. However, by capturing an electron, a free radical can readily become a charge carrier and thus increase the power factor of the oil and even cause a non-reversible reduction in the dielectric properties of cellulose insulation which adsorb these ionized decomposition products.
R: R' === R + R' R + e~ > R-2 ~ ~ ~ fi A pre-service test to measure the production of free radicals is useful to predict potential gassing problems in these oils.
As well as having the potential to increase the power factor, the presence of free radicals accelerates the aging process of the oil because diradical oxygen molecules from the naturally-dissolved air content (about 90% by volume) are able to initiate chain reactions (see Mousseron-Canet, M.
and Mani~ J-C., Photochimie et Reactions Moleculaires, Dunod, Paris, 1969). The properties arising from the peculiar electronic structure of oxygen justify efforts to reduce the contact of molecular oxygen with chemically-reactive free radicals by saturating degassed insulating oils with nitrogen and maintaining the oil in the transformer expansion chamber under a nitrogen blanket. Oxygen, a diradical molecule which in the ground state is a triplet, should be considered to be a contaminant in insulating oils, at least as incompatible with oil purity as water.
Referring to Fig. 3, the absorption spectra of an oils obtained on a UV-VIS spectrophotometer from 380 nm to 520 nm indicate that some components of the hydrocarbon blend decompose during the Merrell Test. In Fig. 3, curves E, F and G represent the results for Voltessa 35 oil before ionization, after 120 minutes of ionization at 10KeV and after 240 minutes of ionization at 10KeV, respectively. The fact that decomposition occurs is proven by significant shift of the absorption spectrum towards longer wavelengths after ~'J~j ~b~. ~J
ionization as well as by the appearance of gaseous products.
Spectral shift can be explained by the oxidation reaction of free radicals immediately after contact of the ionized oil sample with air. A comparable shift of the absorbance curve takes place after an oxidation test, as well as during the normal aging process of oil during service. An increase in the unsaturation level (number of double bonds) could be an additional chromophore, but in the presence of oxygen the two processes are overlapping and it is very difficult to establish which contributes more to the change in the absorption spectrum.
Observations suggest that the majority of components in the oil are very resistant to the action of high-energy electrons and only a small number of soluble impurities undergo decomposition. The impurities are those components which absorb enough energy to undergo electronic transition which can then be converted to kinetic energy of the resultant free radicals. After a sample of new Shell Diala A oil was purified, the yield of gaseous decay products during the Merrell Test declined and the shift of the absorption spectrum in the visible range was smaller. In this connection, reference is made to Fig. 4, in which curves H, I, J and K
represent the change in absorbance for new Shell Dialaoil, purified oil, new ionized oil and purified, ionized oil, respectively.
2 ~
The mechanism of decay presented here is a simplified model. In liquids, excited molecules can lose their energy by complex means such as by emitting a photon, by intersystem crossing, by photochemical reactions, by internal 5 conversion or by transfer of energy to other molecules.
Insoluble x-wax decay products may be formed by chemical reactions between free radicals formed during previous homolytic separations. The probability of such reactions is small however, since they occur only when the concentration of impurities is unacceptably high, as it is for the oxidized oil sample shown in Figure 1 (curve A).
Measurements of the chemical stability of an insulating oil are very important in assessing the quality of the oil, since resistance to oxidation ensures a long service life and good protection of cellulose insulation. However, at today's high voltages, an increase of the power factor of transformer oil sometimes occurs unexpectedly in the early stages of the aging process. This suggests that another ionization mechanism (in addition to reduction-oxidation reactions) is involved and that additional tests to measure the physical stability of oils are necessary. Since ionization processes are enhanced in the presence of free-radical intermediates, the concentration of such radicals is an important indicator of the insulating quality of an oil.
In addition to determining the gas evolved from an insulating oil, the modified Merrell test proposed herein can 2~
be used to determine the free radical concentration and absorption spectrum in the visible range of oil before and after a 120-minute ionization. Measuring free radical concentration with DPPH before and after ionization in the Merrell test discharge cell has given interesting and useful results. The sensitivity of this fast and precise analytical method allows discrimination between the real and apparent degree of purity of new, in-service or recycled oils. Because high-energy electrons in the discharge cell produce both homolytic and heterolytic breakdown of weak valence bonds, this method has been found to be an appropriate test for the physical stability of insulating oils.
This invention relates to a method of determining the stability of insulating oil of the type commonly used in high-voltage transformers.
Dissolved gaseous decay products are produced in the insulating oils of high voltage transformers by incipient eleetrical failures of various types and intensities. Local over-stressing of the oil/paper insulation inside the windings can supply sufficient energy to break the strong aliphatic carbon-carbon and carbon-hydrogen bonds of some components of the complex blend of hydrocarbons found in the oil.
Aceordingly, the ability to understand and measure the gassing tendency of insulating oils under thermal and electrieal stress permits the prediction and prevention of transformer failure.
The build-up of gaseous decay products in insulating oils during serviee results mainly from the breakdown of weak valence bonds. It is important to understand the process by whieh this breakdown occurs and to identify the energy souree which initiates the breakdown.
The influence of paper, other solid dielectrics, hot spots and partial diseharges make the deeay proeess very complex. These factors all influence degradation of both the insulating oil and the cellulose insulation. It has been observed that the degree of polymerization of the cellulose deereases in the vieinity of hot spots. Formation of high moleeular weight produets (~-waxes) and their deposition in the paper insulation may reduce its permeability, resulting in impaired heat transfer and the creation of hot spots.
Since the hydrocarbon blend of insulating oils consists of more than 3000 components, excited molecules produced by the primary process of energy absorption can undergo a variety of secondary processes, both physical and chemical. Typical processes which affect the 'physical stability' of hydrocarbons are the dissociation of excited molecules into smaller fragments, their rearrangement via bond-breaking or their return to the ground state by releasing absorbed energy.
The aliphatic carbon-carbon bond energy of hydrocarbon oils is about 90 kcal mol~l (approximately 4eV).
Thermal excitation of molecules becomes sufficient to break these bonds at temperatures of 350 to 500C. Thermal cracking of petroleum products is carried out in this temperature range (see Pryor, W.A., Free Radicals, McGraw-Hill Inc., New York, 1966). Below 100C, the normal maximum operating temperature of high-voltage transformers, there is insufficient thermal energy to excite electrons of the carbon-carbon bonds.
The available laboratory methods for testing gas evolved from oils include the Merrell Test (ASTM designation:
D2298-81) which was discontinued in 1987, and the Modified Pirelli Method (ASTM designation: D2300-85) which applies to insulating oils saturated with hydrogen or other gases.
Analysis of dissolved gases in transformer oils (ASTM
2 ~
designation D3612-85) is used to diagnose the type and degree of incipient failure tsee Duval, M., "Dissolved Gas Analysis:
It Can Save Your Transformer", IEEE Elec. Insul., Vol.5, No.6, p.22-27, 1989). However, the results of this analysis do not always correlate reliably with the actual failure rate under service conditions (see Savio, J. Leo, "Transformer Fault Gas Analysis and Interpretation: A User's Perspective", Electrical Insulatinq Oils, STP 998, H.G. Erdman, Ed., ASTM, Philadephia, 1988, pp 83-88).
An object of the present invention is to provide a solution to the above problems in the form of a relatively simple and accurate method for determining the stability of an insulating oil.
Another object of the present invention is to provide a method for determining oil stability which measures the production of free radicals formed during high-voltage ionization.
Accordingly, the present invention relates to a method of determining the physical stability of insulating oil comprising the steps of subjecting the oil to ionization, and determining the concentration of free radicals in the oil before and after ionization, the difference in concentration providing a ~uantitative indication of the gassing tendency of the oil.
Research conducted by the inventors focused on the effects of free electrons, which present a very powerful source of energy. During transient (especially commutation) field enhancements (see Armanini, D. et al, "Aging and deterioration of HV current transformer insulation by very fast transient overvoltages", L'Energia Elettrica, No. 5, 1989), invisible surface imperfections or asperities of the copper windings may eject free electrons. With potential gradients of 0.3 to 0.4 kV/mm, these free electrons have the necessary energy to endanger the electronic structure of chemically-inert hydrocarbons. Although the duration of these transient electrical stresses is only a few microseconds, their cumulative effect cannot be neglected because commutation field enhancements occur during normal service conditions and because asperities in copper windings are unavoidable. In high-speed Schlieren studies of electrical breakdown in liquid hydrocarbons, no streamer was observed originating from within the liquid itself. A rapid rise in light emission occurred just before breakdown (see Wond, P.
and Forster, E.O., "High speed Schlieren studies of electrical breakdown in liquid hydrocarbons", Can. J. of Chemistry, V.55, p. 1890, 1977) which suggests that this process begins in the immediate vicinity of metal surfaces.
Forster's work at high stress on very pure and clean liquids (see Forster, E.O., "Progress in the Field of Electric Properties of Dielectric Liquids", IEEE Trans.Electr.
Insul., Vol.25, No. 1, pp. 45-53, 1390) suggests the need to clarify the role played by dissolved impurities in insulating ,~ ~3 ~ ~ 3~ 3 oils. Even at normal operating temperatures, less stable molecules among the dissolved impurities could accumulate the necessary energy for electronic transitions during the elastic collisions caused by thermal agitation. Such electronic transitions may result in the homolytical breakdown of weak valence bonds. Of the weakly-bonded molecules, peroxides are a frequent source of free radicals which can serve as initiators of auto-oxidation rèactions, as per the equations:
ROOH =~= RO ~ OH
R~I + RO ~ ROH + R
Free radicals can be transformed by reduciion-oxidation reactions into charge carriers, thus increasing the power factor of an insulating oil.
R + [cai~on ] ~ R + [cation]~
It would therefore be desirable to determine the concentration of the free radical precursors. However, no direct analytical method is available at present. The method proposed herein can be used to determine the presence of free radicals arising from these precursors, and to provide an indirect measure of precursor concentration and real oil purity.
The invention will be described in greater detail with reference to the accompanying drawings, wherein:
Figure 1 is a graph of gas pressure versus time for a variety of ionized oils:
Figure 2 is a graph of absorbance versus time for 2,2-diphenylpicrylhydrazyl before and after ionization;
Figure 3 is a graph of absorbance versus wavelength (absorption spectra) for an oil before and after ionization for different lengths of time; and Figure 4 is a graph of absorbance versus wavelength (absorption spectra) for a variety of oils before and after ionization.
The Merrell test (supra) was adopted in 1964 to determine the quantity of gas evolved during the ionization process when oil samples are bombarded with lOkeV electrons.
Using a discharge cell as described in the test, the gas evolved from two new and two used oil samples was measured (Figure 1). The samples include oxidized oil (curve A), new Shell Diala (trade-mark) oil (curve B), new Voltesso 35 (trade-mark) oil (curve C) and purified oxidized oil (curve D). Although an ionization time of 1000 minutes is prescribed in the Merrell test, decomposition after 120 minutes was sufficient to cause a significant pressure 2~ increase above the oil.
The high-energy discharge used for ionization produces not only gaseous decay products, which raise the pressure inside the discharge cell, and ionized molecules, which increase the power factor of the oil, but also, free radicals (see Tanaka, J., "Free Radicals in Electrical Insulation", Proceedings of the Nineteenth Symposium on 2 ~
Electrical Insulating Materials, Osaka, Japan, 1986). The presence of free radicals is experimentally confirmed using the very sensitive reaction of the stable free radical, 2,2-diphenylpicrylhydraæyl (~PPH). Solutions of DPPH, even at a concentration of 10-5 M, are intensely blue~violet coloured with maximum absorption at 532 nm. When the DPPH free electron forms a doublet with the electron of another free radical, the solution changes colour from blue to yellow, with a corresponding decrease in absorption at 532 nm. This change allows determination of the reaction kinetics (see Bednar, H., Sabau, I. and Silberg, I., "Purity of insulating oils", Energetica, V.26, Nr.9, 1978).
For the free radical determinations, a solution of 0.01% DPPH in benzene was prepared. Two ml of the DPPH
solution were diluted with three ml of benzene. A portion of the visible spectrum was recorded from 580 nm to 480 nm using a UV-VIS spectrophotometer. One ml of insulating oil was then mixed with two ml of the DPPH solution and two ml of benzene, and the spectrum was repeated at three minute intervals. A
significant decrease in the absorbance at 532 nm indicated a high rate of consumption of DPPH and thus a high concentration of free radicals in the oil and, indirectly, an unacceptable concentration of impurities in the analyzed oil.
A significant increase in pressure in the test chamber caused by gaseous decomposition during the ionization tests of several oils is shown in Figure 1. As shown in Fig. 2, a marked increase in the concentration of free radicals for the new Voltesso 35 oil was observed by measuring the accelerated consumption of DPPH after the ionization test.
In order for free radlcals to be formed in the oil during ionization, the oil must contain free radical precursors, which may be either impurities that were not completely eliminated during the refining process or contaminants accidentally introduced into the oil.
Tests on oil samples taken from in-service transformers with gassing problems showed the presence of a measured concentration of free radicals along with an unusually high power factor (14.7~) without a significant shift in the visible spectrum. The presence of free radicals is an important indicator that decomposition has occurred in the oil, as are the gassing problems and the increase in the power factor.
Free radicals, since they are not charge carriers, do not directly contribute to the value of the power factor or the interfacial tension. However, by capturing an electron, a free radical can readily become a charge carrier and thus increase the power factor of the oil and even cause a non-reversible reduction in the dielectric properties of cellulose insulation which adsorb these ionized decomposition products.
R: R' === R + R' R + e~ > R-2 ~ ~ ~ fi A pre-service test to measure the production of free radicals is useful to predict potential gassing problems in these oils.
As well as having the potential to increase the power factor, the presence of free radicals accelerates the aging process of the oil because diradical oxygen molecules from the naturally-dissolved air content (about 90% by volume) are able to initiate chain reactions (see Mousseron-Canet, M.
and Mani~ J-C., Photochimie et Reactions Moleculaires, Dunod, Paris, 1969). The properties arising from the peculiar electronic structure of oxygen justify efforts to reduce the contact of molecular oxygen with chemically-reactive free radicals by saturating degassed insulating oils with nitrogen and maintaining the oil in the transformer expansion chamber under a nitrogen blanket. Oxygen, a diradical molecule which in the ground state is a triplet, should be considered to be a contaminant in insulating oils, at least as incompatible with oil purity as water.
Referring to Fig. 3, the absorption spectra of an oils obtained on a UV-VIS spectrophotometer from 380 nm to 520 nm indicate that some components of the hydrocarbon blend decompose during the Merrell Test. In Fig. 3, curves E, F and G represent the results for Voltessa 35 oil before ionization, after 120 minutes of ionization at 10KeV and after 240 minutes of ionization at 10KeV, respectively. The fact that decomposition occurs is proven by significant shift of the absorption spectrum towards longer wavelengths after ~'J~j ~b~. ~J
ionization as well as by the appearance of gaseous products.
Spectral shift can be explained by the oxidation reaction of free radicals immediately after contact of the ionized oil sample with air. A comparable shift of the absorbance curve takes place after an oxidation test, as well as during the normal aging process of oil during service. An increase in the unsaturation level (number of double bonds) could be an additional chromophore, but in the presence of oxygen the two processes are overlapping and it is very difficult to establish which contributes more to the change in the absorption spectrum.
Observations suggest that the majority of components in the oil are very resistant to the action of high-energy electrons and only a small number of soluble impurities undergo decomposition. The impurities are those components which absorb enough energy to undergo electronic transition which can then be converted to kinetic energy of the resultant free radicals. After a sample of new Shell Diala A oil was purified, the yield of gaseous decay products during the Merrell Test declined and the shift of the absorption spectrum in the visible range was smaller. In this connection, reference is made to Fig. 4, in which curves H, I, J and K
represent the change in absorbance for new Shell Dialaoil, purified oil, new ionized oil and purified, ionized oil, respectively.
2 ~
The mechanism of decay presented here is a simplified model. In liquids, excited molecules can lose their energy by complex means such as by emitting a photon, by intersystem crossing, by photochemical reactions, by internal 5 conversion or by transfer of energy to other molecules.
Insoluble x-wax decay products may be formed by chemical reactions between free radicals formed during previous homolytic separations. The probability of such reactions is small however, since they occur only when the concentration of impurities is unacceptably high, as it is for the oxidized oil sample shown in Figure 1 (curve A).
Measurements of the chemical stability of an insulating oil are very important in assessing the quality of the oil, since resistance to oxidation ensures a long service life and good protection of cellulose insulation. However, at today's high voltages, an increase of the power factor of transformer oil sometimes occurs unexpectedly in the early stages of the aging process. This suggests that another ionization mechanism (in addition to reduction-oxidation reactions) is involved and that additional tests to measure the physical stability of oils are necessary. Since ionization processes are enhanced in the presence of free-radical intermediates, the concentration of such radicals is an important indicator of the insulating quality of an oil.
In addition to determining the gas evolved from an insulating oil, the modified Merrell test proposed herein can 2~
be used to determine the free radical concentration and absorption spectrum in the visible range of oil before and after a 120-minute ionization. Measuring free radical concentration with DPPH before and after ionization in the Merrell test discharge cell has given interesting and useful results. The sensitivity of this fast and precise analytical method allows discrimination between the real and apparent degree of purity of new, in-service or recycled oils. Because high-energy electrons in the discharge cell produce both homolytic and heterolytic breakdown of weak valence bonds, this method has been found to be an appropriate test for the physical stability of insulating oils.
Claims (5)
1. A method of determining the physical stability of insulating oil comprising the steps of subjecting the oil to ionization, and determining the concentration of free radicals in the oil before and after ionization, the difference in concentration providing a quantitative indication of the gassing tendency of the oil.
2. A method according to claim 1, wherein the oil is ionized by subjecting the oil to a high voltage field of approximately 10 KeV.
3. A method according to claim 2, wherein the concentration of free radicals in the oil is determined from absorption spectra of the oil before and after ionization of the oil.
4. A method according to claim 3, wherein the oil is dissolved in a solution of 2,2-diphenylpicrylhydrazyl, and the absorption spectra of resulting solution in the visible range is measured for oil before and after ionization.
5. A method according to claim 3, wherein a first sample of oil is dissolved in 2,2-diphenylpicrylhydrazyl, the absorption spectrum of the resulting first sample solution is measured, a second sample of the oil is subjected to ionization for approximately 120 minutes, the second sample is dissolved in 2,2-diphenylpicrylhydrazyl, and the absorption spectrum of the resulting ionized second sample solution is measured, the change in spectra providing a quantitative indication of the free radical concentration and consequently the gassing tendency of the oil..
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA 2054616 CA2054616A1 (en) | 1991-10-31 | 1991-10-31 | Method of determining stability of insulating oil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA 2054616 CA2054616A1 (en) | 1991-10-31 | 1991-10-31 | Method of determining stability of insulating oil |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2054616A1 true CA2054616A1 (en) | 1993-05-01 |
Family
ID=4148677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 2054616 Abandoned CA2054616A1 (en) | 1991-10-31 | 1991-10-31 | Method of determining stability of insulating oil |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2054616A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7049922B2 (en) | 2001-12-05 | 2006-05-23 | Insoil Canada Ltd. | Method and apparatus for decreasing gassing and decay of insulating oil in transformers |
| WO2013059032A1 (en) * | 2011-10-21 | 2013-04-25 | Abb Technology Ag | Sensor structure for online monitoring of furans in power transformers |
| CN113203699A (en) * | 2020-12-10 | 2021-08-03 | 国网上海市电力公司 | Method and device for detecting antioxidant content in transformer plant insulating oil |
-
1991
- 1991-10-31 CA CA 2054616 patent/CA2054616A1/en not_active Abandoned
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7049922B2 (en) | 2001-12-05 | 2006-05-23 | Insoil Canada Ltd. | Method and apparatus for decreasing gassing and decay of insulating oil in transformers |
| US7205874B2 (en) | 2001-12-05 | 2007-04-17 | Insoil Canada Ltd. | Method and apparatus for decreasing gassing and decay of insulating oil in transformers |
| WO2013059032A1 (en) * | 2011-10-21 | 2013-04-25 | Abb Technology Ag | Sensor structure for online monitoring of furans in power transformers |
| CN103930770A (en) * | 2011-10-21 | 2014-07-16 | Abb技术有限公司 | Sensor structure for online monitoring of furans in power transformers |
| US9383315B2 (en) | 2011-10-21 | 2016-07-05 | Abb Technology Ag | Sensor structure for online monitoring of furans in power transformers |
| CN113203699A (en) * | 2020-12-10 | 2021-08-03 | 国网上海市电力公司 | Method and device for detecting antioxidant content in transformer plant insulating oil |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ferguson et al. | Suspended particles in the liquid insulation of aging power transformers | |
| Hamdi et al. | Stability of mineral oil and oil–ester mixtures under thermal ageing and electrical discharges | |
| N'cho et al. | The gassing tendency of various insulating fluids under electrical discharge | |
| Loiselle et al. | Comparative studies of the stability of various fluids under electrical discharge and thermal stresses | |
| CA1326576C (en) | Electrical tree suppression in high-voltage polymeric insulation | |
| Sabau et al. | The side effects of gassing in transmission power transformers | |
| CA2054616A1 (en) | Method of determining stability of insulating oil | |
| Wada et al. | Method to evaluate the degradation condition of transformer insulating oil-establishment of the evaluation method and application to field transformer oil | |
| Sokolov et al. | Evaluation and Identification of typical defects and failure-MODES Of 110-750 KV Bushings | |
| Buchacz et al. | Stray gassing of oil in HV transformers | |
| El-refaie et al. | Prediction of the characteristics of transformer oil under different operation conditions | |
| Oumert et al. | Comparative study of the degradation rate of new and regenerated mineral oils following electrical stress | |
| Sabau et al. | The electrochemical stability of mineral insulating oils | |
| Fofana | The gassing of insulating fluids | |
| Boudraa et al. | Study of spectral response of transformer oil under low electrical discharge and thermal stress | |
| Manea et al. | Accelerated Ageing Method of Mineral Oil Under High Electric Field and Partial Discharges | |
| N'cho et al. | Effect of electrical discharge on oil decaying process | |
| Aarathy et al. | Spectral analysis for condition assessment of Transformer oil under different aging conditions | |
| Ghalkhani et al. | Influence of aging byproducts on the gassing tendency of transformer oils | |
| Sabau¹ | The Electrochemical Stability of Mineral Insulating Oils Reference: Sabau, J., Stokhuyzen, R.," The Electrochemical Stability of Mineral Insulating Oils," Electrical Insulating Materials: International Issues, ASTM STP 1376, MM Hirschler, Ed., American Society of Testing and Materials, West Conshohocken | |
| Kim et al. | Study on decomposition gas characteristics and condition diagnosis for gas-insulated transformer by chemical analysis | |
| Akshatha et al. | Study of copper corrosion in transformers due to sulphur in oil using chemical methods | |
| Neettiyath et al. | Life expectancy estimation of thermally aged Cu contaminant-diffused oil impregnated pressboard | |
| Mehta et al. | Partial discharge study by monitoring key gases of power transformers | |
| Sabau | On line reclamation of aged transformer oils |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Dead |