International Surveys on Circuit-Breaker Reliability Data for Substation and System Studies

Transcription

1 IEEE TRANSACTIONS ON POWER DELIVERY 1 International Surveys on Circuit-Breaker Reliability Data for Substation and System Studies Anton Janssen, Member, IEEE, Dirk Makareinis, and Carl-Ejnar Sölver Abstract Since the 1970s, CIGRE has conducted three worldwide surveys on high-voltage circuit-breaker (CB) reliability. The results of the last inquiry, published last year, are presented and compared with those of the former inquiries. With a focus on the CB s fundamental functions for the system, figures show the growth in reliability during the past decades. The reliability is expressed in failure per 100 CB years (CBY) or per operating cycles for the relevant failure modes. The overall major failure rate improved largely from the first (1.58 per 100 CBY) to the second (0.67 per 100 CBY) to the third enquiry (0.30 per 100 CBY). The failure rate increases with higher voltage classes; GIS CBs have been shown to be twice as reliable and live tank CBs twice as bad as the average failure rate. Although improved, the mechanical operating mechanism is still the subassembly responsible for most failures; besides, CBs applied for frequent switching purposes show a higher failure rate than average. Index Terms Failure modes, high-voltage circuit breaker (CB), major failure, reliability data, reliability definitions, system reliability. I. INTRODUCTION F AST, selective and reliable fault clearing is a prerequisite for an electric power system. One of important elements of the protection system is the circuit breaker (CB). Under normal circumstances, the CB has to carry the current and energize or de-energize sections of the high-voltage (HV) network. But at the very moment that somewhere in the network a short circuit appears, the CB is the only high-voltage apparatus to protect the network. If the short-circuit current is not cleared immediately, the backup protection systems will trip a larger part of the network, leading to an outage of more overhead lines, busbars, and substations. In such a case, the power supply for a larger area will be interrupted. Moreover, the fault must be quickly removed since otherwise dynamic stability problems in the entire power system may occur. Consequently, the requirements for CB performance and reliability are very high. These requirements are not only imposed on new CBs but also on old CBs and determine the utility s maintenance and life-management strategy. From the point of view of reliability, distinction can be made between a failure of Manuscript received March 28, 2013; revised June 27, 2013; accepted July 20, Paper no. TPWRD A. Janssen is with Liander, Arnhem 6812 AR, the Netherlands ( anton. janssen@alliander.com). D. Makareinis is with Siemens Power Transmission Division, Berlin 13629, Germany ( dirk.makareinis@siemens.com). C.-E. Sölver is a consulting engineer for STRI, Ludvika SE-77131, Sweden ( carl.solver@telia.com). Digital Object Identifier /TPWRD a CB that causes the cessation of one or more of its fundamental functions (i.e., a so-called major failure, MF or MaF) and other failures (minor failures, mf or MiF) and defects. By definition [1], a defect is an imperfection in the state of an item (or inherent weakness) which can result in one or more failures of the item itself or of another item under the specific service or environmental or maintenance conditions for a stated period of time. An MaF will result in an immediate change in the power system operating conditions (e.g., the backup protective equipment being required to remove the fault), or will result in mandatory removal from service within 30 min for nonscheduled maintenance [1]. The CB s fundamental functions (i.e., the lack of fundamental functions: MaF) are as follows: does not close or open on command; closes or opens without command; does not make or break the current; fails to carry the current; breakdown to earth or between poles; breakdown across open pole (internal or external); locked in open or closed position. Note that a difference is made between the mechanical response of the CB (the first two items and the last item), the electrical functions (making, breaking, carrying the current), and the dielectric failures (breakdowns, including breakdowns across open poles). Examples of MiF are: air/hydraulic oil leakage in the operating mechanism; small gas leakage due to corrosion or other causes; change in functional characteristics. The functional characteristics may include closing and opening times, travel characteristics, lock-out pressure levels, and automatic functions such as for pole discrepancy. Major failures are defined from the perspective of the CB fundamental functions, but from a system perspective, sometimes such a failure does not interfere with the system fundamental functions (e.g., does not open or close on an operator s command). Utilities therefore may face problems when linking their failure reporting, that is related to power interruption, directly to the information defined by IEC, which is formulated from the apparatus perspective. II. WORLDWIDE SURVEYS Conseil International des Grands Réseaux Electriques (CIGRE) is the International Council on Large Electric Systems ( that conducted three worldwide enquiries on high-voltage CB failures and defects in service. Forementioned definitions, functions, and failures have been established by IEEE

2 2 IEEE TRANSACTIONS ON POWER DELIVERY CIGRE Working Group 13.06, later Working Group A3.06, responsible for the surveys. In the meantime, similar definitions are used for other components and have been introduced in the IEC standards [1]. A. First Worldwide Inquiry The first international survey covered failures and defects of CBs with a rated voltage of 63 kv and above. During the survey, failures have been observed for four years ( ), in total, CB years, put into service after At that time, different arc extinguishing technologies have been included, but during the statistical analysis of the failures, no further distinction has been made between the technologies (air-blast, bulk oil, minimum oil, single and double pressure -gas, vacuum, etc.). A total of 102 utilities from 22 countries participated in the inquiry. By means of population cards and failure cards, the information on service experience has been collected. Apart from information about failures, also information about maintenance intervals, maintenance costs, and the number of switching operations has been collected (operating cycles). All information has been analyzed per voltage class ( kv, kv, kv, kv, 500 kv). Results from the survey and specific investigations have been published in several professional organizations. The most important publication was in Electra No. 79 (1981) [2]. The studies were the basis for new mechanical and environmental tests on CBs, such as mechanical operation tests, with an increased number of cycles, low/high temperature tests and a humidity test. B. Second Worldwide Survey The second international inquiry had the same structure, but was limited to single pressure -gas technology only. However, a distinction has been made between different technologies of the operating mechanism (hydraulic, pneumatic, spring) and its impact on the failure rates. Moreover, distinction has been made between metal-enclosed and nonmetal-enclosed CBs; equipment installed outdoors and indoors; and equipment put into service before and after January 1, The survey was conducted during the years , and covered CBs put into service after The total population consisted of CB-years from 132 utilities of 22 countries. The main objectives were to see whether the reliability, especially the mechanical reliability, of CBs has improved and whether maintenance intervals and maintenance efforts have developed in a profitable way. The final report has been published as CIGRE Technical Brochure 83 (1994) [3]. C. Third Worldwide Survey The third inquiry covered service experience in the years and CB-years from 83 utilities of 26 countries and was again limited to single pressure technology. The population of the third inquiry differs quite substantially from the former two inquiries; a fact to be considered when making comparisons. The inquiry was a part of a survey that Fig. 1. Third inquiry application in the percentage of CB population. covered the service experience with earthing switches, disconnectors, instrument transformers, and gas-insulated switchgear (GIS) [4]. The third inquiry included CBs of all ages, as a relation between MaF-rate and age was an important objective of the third survey. New was a distinction in CBs application (overhead line, cable, transformer, shunt reactor, shunt capacitor, bus coupler). More information has been asked about the enclosure, as a differentiation was made between dead-tank breakers, three-phase enclosed GIS, and single-phase enclosed GIS (see the Appendix). Also, more details of dielectric failure modes were asked, especially concerning its occurrence; that is, during closing/opening operation or during closed/open position. The outcome of the investigations on CB reliability in service is available as CIGRE Technical Brochure 510 (2012) [5]. A. Applications III. POPULATION AND FAILURE STATISTICS From the population data, it is clear that a quarter of the high-voltage CBs is applied in transformer bays. Sixty percent is applied to overhead line and cable bays (54% to line and 6% to cable bays); most cable bays are constructed in GIS technology. Ten percent of all CBs are used as bus couplers (Fig. 1). Shunt reactors and shunt capacitor banks each require only a few percent of the whole CB population, though they cover more than 20% of the MaF. Since the latter CBs are those that will be operated quite often, a comparison has been made between the MaF-rate and the average number of operating cycles per year per application; the number of cycles has been based on the information collected per MaF-failure card. An operating cycle is composed of one open and one close operation. For convenience reasons, in Fig. 2, the MaF-rate is expressed per CB years (CBY). In Fig. 2, the idea is underlined that from the point of view of reliability, a CB is mainly a mechanical device. B. Technology With the age distribution as published in the third inquiry [5], it is clear that about half of the population of -gas CBs

3 JANSSEN et al.: INTERNATIONAL SURVEYS ON CB RELIABILITY DATA 3 Fig. 2. Third inquiry MaF rate and number of cycles per application. Fig. 4. Overall MaF rate by voltage class for the 1st, 2nd, and 3rd inquiries. third inquiry (0.30 per 100 CBY). The main step forwards after the first inquiry was the improvement in arcing chamber performance so that less arcing chambers were required and, therefore, fewer parts and operating mechanisms. From the second to the third inquiry, a further rationalization of the design especially self-blast technology has led to less operating energy required per arcing chamber and, therefore, there is less mechanical stress and wear. Apart from that, modular design and production quality control are rewarded. It should also be mentioned that in the standards, more attention has been paid to the requirements for mechanical endurance, for instance, to class M2 [6] as will be discussed in Section IV. Fig. 3. MaF-rate per drive technology for failures allocated to the drive based on table [3], 2 25 [5], and 2 18, 2 110, (last two columns) [5]. covers the same period as investigated during the second inquiry. The change in products manufactured per year is as follows. Over the years, the portion of GIS CBs lays between 20 and 30% with a slow trend to more three-phase enclosed GIS applications. The remainder are live-tank and dead-tank breakers; dead-tank breakers show a strongly decreasing trend since, before 1980, its portion happened to be 40% of all CBs but it shrunk to 10% in the new millennium (see the Appendix). The technology of the operating mechanism is changing toward spring mechanisms: from a portion of 40% in the early days to 60% these days. The dominant position of hydraulic drives in the early days (50%) has over time been reduced to less than 20%; about the same portion as pneumatic drives. In Fig. 3, the MaF-rate of the hydraulic and pneumatic drives has decreased dramatically and is these days comparable with, or better than, the MaF-rate of spring-driven mechanisms (column after 1999 ). On average, all technologies show great improvements compared to the second inquiry. The MaF-rate decreased from 0.29 per 100 CBY in the former survey to 0.14 per 100 CBY in the last survey, related to MaF for which the operating mechanism is responsible. For the entire population of the third inquiry, the spring drive performs the best (Fig. 3, 2nd column 3rd inquiry ). The overall MaF rate improved largely from the first (1.58 per 100 CBY) to the second (0.67 per 100 CBY) and again to the C. Voltage Class, Enclosure, Location By voltage class, the results from the three surveys show the same tendency, as can be seen in Fig. 4. Similar to the second inquiry, in the third inquiry, the reliability of metal-enclosed CBs (dead-tank, GIS, hybrid) gives a far lower MaF rate than that of live tank CBs: per 100 CBY versus per 100 CBY [4]. In the third inquiry, the population of ME-enclosed CBs is slightly larger than the population of live-tank breakers. Almost all live- and dead-tank breakers are located outdoors as well as the largest part of the GIS CBs. D. Age In the third inquiry, failures have been collected per year of production, so that the MaF development by age can be calculated. For the population installed after 1978, the tendency shown in Fig. 5 is expected to come from both improvements in the technology of younger CBs and an aging and/or wear effect with older CBs. Aging, wear, and corrosion have been reported as the most important causes of MaF (almost 50%). Design faults, manufacturing faults, and incorrect maintenance together are mentioned as causes for 15% of the MaF. There is a slightly better performance of the CBs installed before 1979 in comparison to those installed between 1979 and This may be explained by the high probability that a major overhaul or maybe even replacement of the worst breakers has taken place, thus improving the average performance of the subpopulation, manufactured before 1979.

4 4 IEEE TRANSACTIONS ON POWER DELIVERY TABLE II PERCENTAGE OF MAF RATE AND MIF RATE PER FAILURE MODE, THIRD INQUIRY Fig. 5. Third inquiry MaF-rate per voltage class and by year of production. TABLE I PERCENTAGE OF FAILURES RELATED TO THE SUBASSEMBLY AND COMPONENT RESPONSIBLE, 2ND INQUIRY E. Components Asreportedinthefirst survey, half of the failures occurred in the components at service voltage and one-third in the operating mechanism. That changed with single pressure -gas technology. The subassemblies and components responsible for the MaF give the same distribution for the third as for the second inquiry. In Table I, details are given as reported for the second inquiry with the operating mechanism being responsible for most failures, both major and minor failures. Minor failures have not been analyzed in absolute numbers, as with the third inquiry, based on the response per country, wide underreporting of the many MiF is assumed. F. Failure Modes The failure modes or failure characteristics are quite similar for the second and the third inquiry. Since the third survey gives more details, these failure modes are listed in Table II. Most failures occurred in the operating mechanism, followed by the electrical control and auxiliary system (MaF) and high-voltage parts (MiF). The last item seems to be related to -gas tightness as the dominant MiF-mode is Small -gas leakage. Locked in open or closed position is a dominant MaF mode, that is mainly related to -gas tightness. It is also related to the operating mechanism and the electric control system. These two subassemblies led to failure modes, such as Does not open or close on command ; consuming 45% of MaF. With Locked in open or closed position, it sums up to 70% of MaF. As expected, the MaF rate with the mode Does not open or close on command is proportional to the number of operating cycles per year and with Lockedinopenorclosedposition, the relationship is more indifferent, although a light proportionality can be seen (table 2 57 [5]). Another 10% of MaF is for dielectric breakdowns and almost 10% for loss of mechanical integrity. Four percent of the MaF (6.5% in the second enquiry) resulted in an explosion or fire; mainly in relation to dielectric breakdowns and mainly with live-tank breakers. There is an intriguing distribution of the explosions among the applications of the CB (Fig. 6). CBs applied in transformer bays seem to perform well, while CBs applied to switch shunt reactors and capacitor banks show a higher explosion risk. However, it should be noted that the probability of an explosion is 0.01 per 100 CBY for the entire population. Only a small part of the major failures has the characteristic Does not break the current : 1.9% or per 100 CBY. So the mechanical performance seems to need the most attention from the point of view of reliability. However, the probability that a CB has to interrupt, for instance, a short-line fault, is orders of magnitude lower than the frequency of regular operating

5 JANSSEN et al.: INTERNATIONAL SURVEYS ON CB RELIABILITY DATA 5 Fig. 6. Third inquiry subpopulations and explosions per application in percent. Fig. 7. Third inquiry MaF-rate per failure mode including the fire/explosion rate. cycles. Still, reliable fault clearing is a very important duty of a CB, depending both on its capability to interrupt fault currents and its reliability to Open and close on command. IV. MECHANICAL ENDURANCE At the first and second international inquiry, a number of questions have been put forward about the number of operating cycles per year. The outcome showed that 90% of the CBs are asked upon to perform 80 operating cycles (one open and one close operation) per year or less (i.e., the 90% percentile of the number of operating cycles per year [2], [3]). The average number of operating cycles per year from the second inquiry was higher than that from the first inquiry: 42 versus 26.5 CO-operations per year. The reason for the difference is that in the first inquiry, the average has been determined by weighting each answer on the population cards equally and in the second inquiry, each answer has been weighted by the number of related CBs [7]. At the third international inquiry, the average number of operating cycles has not been collected, but could be estimated roughly by the information per the MaF report. The analysis reported in [5] gives an average of 69 operating cycles per year, far more than the 42 found with the second survey. However, this is the average of the answers per MaF-card and these answers have been differentiated per application, as shown in Fig. 1 (i.e., tables 2 59 and 2 60 of [5]). By using the averages per application (table 2 60 of [5]), multiplying them by the number of CB-years reported for that application (table 2 5 [5]) and dividing the outcome by the total population in CBY, the total average becomes 42 operating cycles per year; exactly the same number as with the second inquiry. Furthermore, there are no circumstances why the average number of cycles per year should have been changed, or anyways increased, from inquiry to inquiry. Based on these results for a 25-year period of service, these days seen as the interval between major overhaul, the majority of CBs has to perform 2000 operating cycles or less. Therefore, the mechanical endurance type test in the standards has been specified as 2000 operating cycles or operating sequences, as is called in the IEC standards. That is to say for CBs applied to normal service conditions, class M1. For special applications, such as shunt reactor switching, an extended mechanical endurance-type test with CO-cycles has been specified in IEC Standard [6]: class M2. The mechanical endurance-type test is relatively easy to perform and gives a very good quality check. Without energizing the high-voltage parts, the required number of operating cycles is performed under varying conditions of auxiliary voltage and operating pressure. Maintenance has to be performed to the manufacturer s instructions; that is, according to the manual. These days, maintenance during 2000 CO operations will be nil. The mechanical, dielectric, and electrical characteristics after the endurance test have to be identical to those before the test or at least within the tolerances as stated in the manual. This applies for the primary contacts as well as the auxiliary contacts. -tightness also has to be checked. For reliability assessment and to investigate the weak spots and endurance limits, manufacturers will test prototypes and subassemblies to a much larger number of cycles; up to tens of thousands. Such development tests will give proper information about maintenance intervals and actions [8]. The widely accepted hypothesis is that as long as the mechanical behavior of a CB is identical to that of the type tested sample and the -gas density is within the operating tolerances, the CB will perform as specified in the standards. Maintenance is therefore mainly focused on the mechanical characteristics. V. SUBSTATION AND SYSTEM RELIABILITY STUDIES The results of the CB reliability studies are also valuable for substation and system reliability investigations. In [7], the information of the first and second inquiry has been used to calculate per voltage class the number of MaFs per command to open/close and the number of other MaFs per year. For substation and system reliability studies, such a distinction is necessary, as reliability engineers want to know the probability of a failure per command (especially for protection functions) and per time period. The same approach will be followed to calculate the MaF rates with the information of the third worldwide inquiry. But since no information about failure modes per

6 6 IEEE TRANSACTIONS ON POWER DELIVERY TABLE III CB RELIABILITY DATA FOR SYSTEM STUDIES, ALL VOLTAGES 63 kv Fig. 8. Third inquiry comparison of the reported and calculated MaF rate per application. voltage class has been collected, no differentiation per voltage class will be made. The average number of operating cycles per year is crucial for the calculations. The failure figures will be calculated by using both 26.5 from the 1st and 42 from the 2nd/3rd inquiry, since there are arguments for either weighting method. Yet, the information for the number of cycles per year is collected by population cards (26.5) and not per CB (42). A comparison of the number of MaFs per operating cycle and per year, for all voltages, between the first, second, and third inquiry can be learned from Table III. The last but one column gives the overall MaF rate per year. All of the other columns to the left are related to MaF with a command to close or to open, resulting into a total MaF per command in the third column from the right. The last column gives the MaF per year for failures without a command to close or open (i.e., the second from right column minus the third from right column). In order to calculate the MaF rate per operating cycle (command), the failure mode Locked in open or closed position is divided among the failures without command (50%) and Does not open on command (13%), Does not close on command (37%). This procedure is the same as applied to the second inquiry [7]. The failure modes Unknown, Other, and Loss of mechanical integrity are proportionally divided over the other modes. When comparing the failure modes, much better performance can be noticed from the first to second and from the second to third survey. Only the number of failures Does not open on command has not improved that much from the second to third inquiry. When the MaF rates with and without a command for the third inquiry are plotted in Fig. 2, it can be noticed that the failure Fig. 9. Third inquiry MaF-rate per operating cycles for different applications and the total population (All). rates calculated with 26.5 operating cycles per year (the bold line in Table III) seem to fit better than those calculated with 42 operating cycles, that is to say, for the applications with a large number of operating cycles per year as shown in Fig. 8. Fig. 2 could also be used to plot per application the MaF rate per operating cycles. Fig. 9 shows these MaF rates, together with rates for the total population (All), using the average number of cycles per year of 69, 42, and 26.5, as discussed in this and the former section. The MaF rate All has to be close to those of the first 7 (or 5) columns. Thus, the average number of operating cycles will be 30 to 40 per year. VI. CONCLUDING REMARKS The reliability of a high-voltage CB is expressed with respect to the cessation of one of its fundamental functions (i.e., an MaF) and other failures (i.e., MiF). The definitions, as included these days in the standards, have been given. The results of three worldwide surveys have been compared, and a steady improvement of the CB s reliability can be noticed from 1.58 to 0.67 to 0.30 MaF per 100 CBY. For the third inquiry, GIS and ME CBs show 50% better performance (0.14 MaF per 100 CBY) and live-tank CBs have 50% worse performance (0.48 MaF per 100 CBY) as shown in [5, Table 2 27]. Despite great improvements, most failures are still related to the operating mechanism. For modern CBs, the reliability can be expressed in rough numbers as follows. Once per commands to open, a CB will fail to open. Once per commands to open, it will not break the current.

7 JANSSEN et al.: INTERNATIONAL SURVEYS ON CB RELIABILITY DATA 7 Fig. 10. Information from the third inquiry on the percentage of enclosure type per manufacturing period [5]: Figure Once per commands to close, it will not close or make the current. Once per 500 years, a CB will show another MaF. Once per 3000 years, a CB will show a dielectric failure (different from Does not break the current ). Once per years, a CB may show a fire or explosion. With respect to definitions: MaFs are defined from the perspective of the CB fundamental functions, but from a system perspective, sometimes such a failure is not interfering with the system fundamental functions. Utilities therefore may face problems when linking their failure reporting, that is related to system MaFs, directly to the information asked by CIGRE. As aging, wear, and corrosion have been reported as the most important cause of MaF (almost 50%), CIGRE has established two new working groups to study the effects of aging and the effect of overstressing old equipment in greater depth. APPENDIX Fig. 10 gives the enclosure technologies of the total population of CBs involved in the third inquiry per production period and as a percentage of the CBs manufactured in that period. Live- and dead-tank breakers are defined as a CB with the interrupters in a tank insulated from earth or in an earthed metal tank, respectively. A GIS breaker is part of a bay which is completely characterized by insulation and by an earthed metal enclosure (including hybrid switchgear). Single or three phase GIS is related to the CB compartment. ACKNOWLEDGMENT This paper is a summary of the work that during the last decades has been performed by three CIGRE working groups: the first and second WG and WG A3-06. REFERENCES [1] IEC High-Voltage Switchgear and Controlgear Part 1: Common Specifications, IEC Standard , 2011, Ed [2] G. Mazza and R. Michaca, The first international enquiry on circuitbreaker failures and defects in service, Electra, no. 79, pp , Dec [3] CIGRE Working Group 13.06, Final report of the second international enquiry on high voltage circuit-breaker failures and defects in service, CIGRE Tech. Brochure 83, Jun [4] M. Runde, Failure frequencies for high-voltage circuit breakers, disconnectors, earthing switches, instrument transformers, and gas-insulated switchgear, IEEE Trans. Power Del., vol. 28, no. 1, pp , Jan [5] CIGRE Working Group A3.06, Final report of international enquiry on reliability of high voltage equipment, Part 2 reliability of high voltage circuit breakers, CIGRE Tech. Brochure 510, Oct [6] High-Voltage Switchgear and Controlgear Part 100: Alternating Current Circuit-Breakers, IEC Standard , IEC, 2012, Ed [7] C.R.Heising,E.Colombo,A.L.J.Janssen,J.E.Maaskola,andE. Dialynas, Final report on high-voltage circuit breaker reliability data for use in substation and system studies, presented at the CIGRE SC 13 Session 1994, [8] A. L. J. Janssen, C. R. Heising, G. Sanchis, and W. Lanz, Mechanical endurance, reliability compliance and environmental testing of high voltage circuit-breakers, presented at the CIGRE SC 13 Session 1996, Anton Janssen (M 95) was Transmission Manager for a power company in the Netherlands. He then became Manager of the KEMA High-Power Laboratory, Arnhem, the Netherlands, in 1993 and became an Asset Manager in 2002 with Liander, a distribution system operator (DSO) (electricity and gas) in the Netherlands. Mr. Janssen is a member of CIGRE SC 13/A3, and active as special reporter and was convenor of several working groups, dealing with reliability (13-06), life management, asset management, distributed generation, long-distance transmission, and short/long line faults. Presently he is Secretary of Working Group A3.22/28, dealing with the requirements for ultra-high voltage substation equipment. He was also a member of several standardization working groups, both for IEC and IEEE. Dirk Makareinis is with Siemens, Berlin, Germany, and has more than 30 years of experience in the field of high-voltage circuit breakers, especially as a manager responsible for mechanical testing. Currently, he is Manager of the Sales and Order Processing Department for high voltage circuit breakers (Europe). Mr. Makareinis was/is an active member of several working groups within CIGRE: life management of CBs, reliability of high-voltage equipment (A3-06), and management of aging high-voltage equipment and possible mitigation techniques (A3-29). Carl-Ejnar Sölver received the Ph.D. degree in electrical power engineering from Chalmers University of Technology, Gothenburg, Sweden, in 1975 He joined ASEA (later ABB), Ludvika, Sweden, becoming the Manager for high-voltage circuit-breaker development, design, testing, application, and technical marketing. He retired from ABB in 2010 and joined STRI, Ludvika, as a Consulting Engineer. In 2000, he was appointed Associate Professor at Chalmers University of Technology. Prof. Sölver was a member of CIGRE SC 13/A3 and an active member or convenor of a number of working groups: switching of small inductive currents, switching test methods, and reliability of high-voltage equipment (former convenor of Working Group A3.06). He was secretary of IEC TC17 and SC17A from 1982 to 1986, and later Chairman of the working groups on electromagnetic compatibility for high-voltage switchgear, and on electrical endurance of high-voltage circuit breakers.