Investigation of Wire Insulation for High Temperature

0 downloads 0 Views 1MB Size Report
systems is to find and assess new insulation materials that can provide the ..... insulation systems and processing,” in IEEE Electrical Insulation. Magazine, vol. 29, no. ... he joined Invensys Brook Crompton as Head of Research responsible for ...
Investigation of Wire Insulation for High Temperature Motor Windings *

K. Tshiloz*, A.C. Smith†, P. M. Tuohy†, T. Feehally† Allegro Microsystems Europe, Stuart House, Station Road, Musselburgh, Edinburgh, EH21 7PB, UK. † The University of Manchester, School of Electrical and Electronic Engineering, Power and Energy Divison, Manchester, UK. Email: [email protected] and [email protected] provide the thermal performance along with the required electrical breakdown withstand and mechanical properties without a large increase in the wire diameter.

Abstract - There is a strong interest in the development of high temperature electrical machines, both from academic research and industrial consumers. Conventional electrical machines cannot operate much beyond 250°C and this limits their application in certain sectors such as oil, gas and aerospace. The key technology for high temperature operation is high temperature wire insulation systems for the motor windings. This paper presents an experimental study on the temperature dependence of the wire insulation resistance of a standard Class H (180°C) enamel wire, a high temperature MAGNETEMP CA-200 wire, a photonis Glass Coated wire, a mica taped VonRoll SK650 wire, a ceramic based CERAFIL500 wire and a newly-developed S-2 Glass-Fibre insulated wire. The results show that motor operation above approx 300°C for any length of time using conventional polymer based wire is difficult. It is possible however to operate motors up to 600°C using inorganic insulation coatings. The new wire using the S-2 Glass-Fibre insulation performed very well showing good insulation resistance up to 600°C: excellent flexibility and robustness; capable of very small bend diameters; and with packing factors almost identical to conventional copper wire.

This paper presents an experimental study on temperature dependence of the wire insulation resistance (IR) [6-8] on a standard Class H (180°C) enamel wire, a high-temperature MAGNETEMP CA-200 wire, a photonis Glass Coated wire, a mica-taped VonRoll SK650 wire, a ceramic based CERAFIL500 wire and a newly-developed S-2 Glass-Fibre insulated wire. The IR allows measuring the resistance of the electrical insulation between the copper conductor and an earthed stainless steel cylinder, thus detecting the thermal deteriorations of the insulation systems. The performance of the insulation resistances between wires with different insulation materials is investigated by experimental measurements using a built test-rig based on IEEE Std 43TM2013 [9-10]. The mechanical flexibility (i.e. minimum bending radius and mechanical pressure) is examined for the considered insulation systems. An assessment of the suitability of these wire types for machine insulation systems operating in excess of ≈375°C is also investigated by comparing the thermal response of these high-temperature wires along with the analysis of the leakage currents flowing between the copper conductor and the earthed stainless steel cylinder on which they are mounted.

Key words: Ceramic, High-Temperature, Mica-Taped, Insulation Resistance, Leakage Currents, S-2 Glass-Fibre, Glass Coated, Thermal Resistance and Yarns.

2. EXPERIMENTAL SET-UP

1. INTRODUCTION

The experimental work is based on the use of a built test-rig and a controllable WHT6/30 LENTON oven with an operating maximum temperature of 600°C as shown in Figs. 1-2. The test-rig was developed to test the various wire samples. The test-rig ensures that the wire insulation is in contact with the earthed stainless steel cylinder and that it remains in tension as it expands thermally as shown in Figs. 1-2. A single wire is wrapped around a stainless steel conductor and consistent tension is applied by a weighted mass of 0.5 kg. The single wire test-rig is then placed into the temperature controlled oven and the insulation resistance is measured using a FLUKE 1507 insulation resistance tester. The insulation resistance tests were conducted at a constant direct voltage of 500 V and 1000 V [9-10]. The readings of the insulation resistance were taken after the test voltage had been applied for 1 minute according to IEEE Std 43TM-2013 [9-10]. A 0.9 mm diameter standard Class H (180°C) enamel wire shown in Fig. 3 was tested initially followed by a 0.873 mm diameter MAGNETEMP CA-200 wire [11] shown in

Electrical machines play a key role in the field of electromechanical energy conversion. A considerable proportion of installed and currently manufactured electrical machines cannot operate beyond 250°C and this limits their application in certain sectors such as oil, gas, aerospace and electric vehicles. There is a strong interest in the development of high-temperature electrical machines, both from academic research and industrial consumers; and so there is a substantial opportunity for commercial exploitation of hightemperature electrical rotating machines. The key technology for high-temperature operation is high-temperature wire insulation systems for the motor windings. The reliability of high-temperature electrical machines will depend critically on the reliability of their wire insulation systems [1-3]. Machine windings failures caused by increased operating temperature can lead to stator and rotor rewinding thus increasing costs [45]. The critical issue for high-temperature wire insulation systems is to find and assess new insulation materials that can

1

M

M

M

M

Lenton WHT 6/30

Fig.1: Insulation Resistance Test-Rig

Fig.2: A Controllable WHT6/30 LENTON Oven

Fig.3: Standard Class H (180°C)

Fig.4: MAGNETEMP CA-200

Fig.5: Photonis Glass Coated Wire

Fig.6: VonRoll SK650 Mica Taped

Fig.7: Ceramic Based CERAFIL500 Wire

Fig.8: S-2 Glass-Fibre Insulated Wire operating temperature of ≈1000°C. Finally, a newlydeveloped 0.81 mm diameter S-2 Glass-Fibre insulated wire presented in Fig. 8 was considered during this work; the S-2 Glass-Fibre ‘sock’ is woven directly onto a nickel-plated copper conductor and has a thickness of 0.075 mm with a high melting temperature of ≈1056°C. The S-2 Glass-Fibre ‘knitted sock’ is formed using S-2 Glass-Fibre yarns, which consists of numerous filaments of 0.005 mm diameter, twisted to form a long continuous length of interlocked fibres of 0.075 mm diameter offering significantly more strength than conventional Glass-Fibre [15] and good fill factor not very different from conventional copper wire. The experiment work was performed with a 25 mm diameter earthed stainless steel cylinder. A stainless steel cylinder of 25 mm diameter was selected with a minimum wire bend radius of 12.5 mm to avoid any mechanical cracking during the experiment.

Fig. 4; this wire has an insulation thickness of 0.073 mm and an operating temperature up to ≈210°C. A 0.2 mm diameter photonis Glass Coated wire (GCW) presented in Fig. 5 was also tested during the experiment; this wire has a 0.05 mm thickness glass coating chemically bonded to the wire and hermetically sealed with a high heat tolerance up to ≈650°C [12] protecting the wire from oxidation and chemical attack. A 1 mm diameter commercially available VonRoll SK650 [13] mica-taped wire presented in Fig. 6 was included in the experiment; the VonRoll SK650 has an insulation thickness of 0.1 mm and an operating temperature up to ≈500°C. The copper conductor is coated in nickel to prevent oxidation. A 0.8 mm diameter commercially available ceramic based CERAFIL500 [14] nickel-coated copper wire shown in Fig. 7 was also tested; the CERAFIL500 has an insulation thickness of 0.02 mm with a bending radius of 4.10 mm and an

2

Fig.9: IR of the Standard Class H (180°C)

Fig.10: IR of the MAGNETEMP CA-200

Fig.11: IR of the Photonis Glass Coated

Fig.12: IR of the VonRoll SK650 Mica Taped

Fig.13: IR of the Ceramic Based CERAFIL500

Fig.14: IR of the S-2 Glass-Fibre Insulated Wire H (180°C) wire as the temperature increases. It is clear that the insulation level deteriorates as the temperature increases above 200°C and then rapidly as the temperature exceeds 300°C. The trend for the high-temperature wire – MAGNETEMP CA-200 in Fig. 10 is similar but the deterioration is less dramatic as the temperature exceeds 300°C. The magnitude of the insulation resistance below 200°C were measured at greater than 11 GΩ for both the standard Class H (180°C) and the MAGNETEMP CA-200 wires. Fig. 11 shows that the insulation resistance of the photonis GCW starts to reduce steadily above 350°C. The same behaviour seen in Fig. 11 can also be seen in Fig. 12 for the VonRoll SK650; the measured insulation resistance reduces steadily above 400°C in this case. The insulation resistances below 350°C for the GCW and 400°C for the VonRoll SK650 were measured to be greater than 11 GΩ, respectively. The results show that the VonRoll SK650 offers a slightly better high-temperature operation than the photonis GCW. However, the mica/cloth wrapping on the VonRoll SK650 wire is quite fragile and is prone to failure during winding of machine coils. Although Standard Class H

Fig.15: Insulation Leakage Currents

3. EXPERIMENTAL RESULTS The thermal performance of the selected high-temperature wires was verified experimentally in section 2. A number of experiments were performed on the test-rig. Results of the measured wire insulation resistance at a test voltage of 500 V and 1000 V are presented in Figs. 9-14. Fig. 9 confirms the deterioration in the insulation levels of the conventional Class

3

(180°C) enamel wire and the MAGNETEMP CA-200 would appear to be capable of operating at higher temperatures than their ratings, it should be made clear that this would only be for a much reduced lifetime as predicted by the classical Arrhenius relationship between insulation lieftime and operating temperature [16]. The rated operating temperatures are defined for an expected lifetime of 20000 hours [16]. The photonis GCW shows a better capability for high-temperature operation and has a more robust insulation coating (0.05 mm) for coil windings. Howerver, it is not yet commercially available in normal coil wire sizes and has limit on bend diameters because of the more brittle nature of the glass insulation compared to conventional resin based coatings. The recommended bend diameter of the 0.2 mm wire sample used in these tests was 12.5 mm but this is expected to reduce in higher wire diameters. The trend for the commercial hightemperature rated ceramic based wire – CERAFIL500 in Fig. 13 shows a constant resistance up to 350°C before starting to reduce. The insulation resistance of the CERAFIL500 below 350°C were measured >550 MΩ. CERAFIL500 offers excellent electrical insulation properties, however the mechanical performance does not match that of conventional resin-based wires especially in terms of mechanical flexibility and thermal expansion. Some form of binder is therefore used to hold the ceramic compound in place as an insulator; silicon is often used as a binder because it changes to an inorganic material when heated to high-temperature [17]. Fig. 14 illustrates the insulation resistance of the newly-developed S2 Glass-Fibre wire. Fig. 14 shows that the insulation resistance level starts to deteriorate as the temperature increases above 425°C; the insulation resistances were recorded to be >550 MΩ below this temperature. This hightemperature wire has excellent durability and bendability, fatigue resistance, and a very good fill factor not very different from conventional copper wire at elevated temperatures. There were some concerns relating to moisture ingress into the woven insulation and voltage withstand but it is expected that it would be used in conjunction with a hightemperature encapsulation which would provide moisture protection. Further coil tests at normal low voltage levels have also shown no issues with voltage withstand for sinusoidal or converter supplies. The newly-developed S-2 Glass-Fibre wire shows promise for high-temperature electrical machine windings.

currents of both the standard Class H (180°C) and the MAGNETEMP CA-200 are ≈1.20 and ≈0.22 mA at 375°C, respectively so we would regard the class H wire to have failed immediately at around 360°C. The MAGNETEMP wire reaches about 390°C before failing. It can be seen from Fig. 15 that the leakage current of the photonis GCW is ≈0.00045 mA at 375°C whilst the leakage current of the VonRoll SK650 was ≈0.001 mA which is higher than the photonis GCW leakage current. The leakage current of the newly-developed S-2 Glass-Fibre weave wire was only ≈0.00092 mA at 375°C compared to the leakage current of the CERAFIL500 wire ≈0.003 mA at 375°C as illustrated in Fig. 15. Although the measurements shown in Figs. 9-15 would appear to indicate that all of the wire samples tested could operate at temperatures in excess of 300°C, this is misleading because the expected lifetimes would be greatly reduced compared to operation at their rated temperatures. The general conclusion is that conventional polyester based wire insulation would find it difficult to operate much above 300°C for any significant length of time. This is due to the modification of the molecular structure of the polymer insulation at increased temperature predicted by the classical Arrhenius relationship [16]. Operation above 300°C requires a fundamental change to an inorganic type of insulation for which the temperature effects on the insulation resistance are less well researched but would be expected to reduce less quickly compared to organic coatings. The glass-coated, mica wrap, ceramic and S-2 Glass-Fibre wire samples are all examples of inorganic coatings.

4. CONCLUSIONS One of the key technologies for high-temperature operation is high-temperature wire insulation systems for the electrical machine windings. This paper details insulation measurement on 6 different wire samples; 2 conventional polymer based wires – Standard Class H wire and MAGNETEMP CA-200 wire and 4 inorganic wire samples: a photonis Glass Coated wire, a mica-taped VonRoll SK650 wire, a ceramic based CERAFIL500 wire and a newly-developed S-2 Glass-Fibre insulated wire. All wire samples showed good insulation resistance to approx 325°C but the two standard polymer based wire samples failed using at a leakage current of 1mA between 325°C and 400°C. The four wire samples using inorganic insulation maintained a reasonable level of insulation up to 500°C with the glass-coated, mica wrap and new S-2 Glass-Fibre wire samples maintaining insulation resistance up to 600°C. It is clear that motor operation above approx 300°C for any length of time using conventional polymer based wire is difficult. It is possible however to operate motors up to 600°C using inorganic insulation coatings. The challenges with these wire types are their suitability for winding motor coils in terms of minimum bend diameter, robustness and voltage withstand. The new wire using the Glass-Fibre insulation performed very well showing good insulation resistance up to 600°C: excellent flexibility and robustness; capable of very small bend diameters; and with packing factors almost identical to conventional copper wire.

The measured leakage currents through the insulation were also determined from the test voltage and measured insulation resistance. The measurements were taken 1minute after the test voltage was applied; this follows the normal recommended practise to allow the initial currents that flow to decay leaving the insulation leakage current. The leakage currents are shown in Fig. 15 for temperatures varying from 200°C to 600°C. The question that arises of course in the high-temperature insulation measurements is what level of insulation resistance would represent an insulation failure. In these tests, an insulation leakage current of 1mA is regarded as sufficient to represent an insulation failure, corresponding to an insulation resistance of 1 MΩ for a test voltage of 1000V for example. Fig. 15 demonstrates that the leakage

4

5. REFERENCES

6. BIOGRAPHIES

[1] H. Torkaman and F. Karimi, “Influence of ambient and test conditions on insulation resistance/polarization index in HV electrical machines-a survey,” in IEEE Transactions on Dielectrics and Electrical Insulation, vol. 22, no. 1, pp. 241-250, Feb. 2015. [2] A. Petit, “Temperature effect on insulation resistance of generator stator bar affected by water ingress,” 2015 IEEE Electrical Insulation Conference (EIC), Seattle, WA, 2015, pp. 515-518. [3] K. Suzuki, H. Sako, S. Maeda and K. Mio, “Temperature dependence of insulation resistance and current components for generator stator bars,” 2016 IEEE Electrical Insulation Conference (EIC), Montreal, QC, 2016. [4] A. D. Ashkezari, H. Ma, T. K. Saha and C. Ekanayake, “Application of fuzzy support vector machine for determining the health index of the insulation system of in-service power transformers,” in IEEE Transactions on Dielectrics and Electrical Insulation, vol. 20, no. 3, pp. 965-973, June 2013. [5] G. C. Stone and G. H. Miller, “Progress in rotating-machine insulation systems and processing,” in IEEE Electrical Insulation Magazine, vol. 29, no. 4, pp. 45-51, July-August 2013. [6] T. R. Gaerke and D. C. Hernández, “Understanding Stator Insulation In-Process Testing,” in IEEE Transactions on Industry Applications, vol. 53, no. 2, pp. 1704-1708, March-April 2017. [7] L. Lamarre and E. David, “Temperature dependence of the resistance of modern epoxy mica insulation of HV rotating machines,” in IEEE Transactions on Dielectrics and Electrical Insulation, vol. 15, no. 5, pp. 1305-1312, October 2008. [8] R. Soltani, E. David and L. Lamarre, “Temperature Effect on Dielectric Characteristics of Large Rotating Machines Insulation,” Conference Record of the 2008 IEEE International Symposium on Electrical Insulation, Vancouver, BC, 2008, pp. 280-283. [9] IEEE Standards Association, “Recommended Practice for Testing Insulation Resistance of Electric Machinery,” IEEE Power and Energy Society, IEEE Std 43TM-2013. [10] G. C. Stone, “Recent important changes in IEEE motor and generator winding insulation diagnostic testing standards,” IEEE Industry Applications Society 50th Annual Petroleum and Chemical Industry Conference, 2003. Record of Conference Papers, 2003, pp. 101-110. [11] www.superioressex.com/MagnetWire/Copper_Rod.aspx [12] https://www.photonis.com/en/product/glass-coated-wire [13] http://www.vonroll.com/en/ [14] http://www.cables-cgp.com/en/produits/cerafil-cerafil500/ [15] http://www.agy.com/wp-content/uploads/2014/03/636_S2_Yarn-Aerospace.pdf [16] K. Tshiloz, A. C. Smith, A. Mohammed, S. Djurović and T. Feehally, “Real-time insulation lifetime monitoring for motor windings,” 2016 XXII International Conference on Electrical Machines (ICEM), Lausanne, 2016, pp. 2335-2340. [17] H. Mitsui, “Progress in Japan in electrical insulation at high temperatures,” Electrical Insulation Magazine, IEEE, vol. 12, no. 3, pp. 16-27, 1996.

Kavul Tshiloz was born in Lubumbashi, Democratic Republic of Congo. He received the B.Eng. (Hons) degree in Electronic and Computer Engineering from The University of Bolton, UK in 2009. Dr. Tshiloz obtained his MSc degree in Electrical Energy Conversion Systems and his PhD degree in Electrical and Electronic Engineering from The University of Manchester, UK in 2010 and 2016, respectively. Dr. Tshiloz is currently a systems engineer at Allegro Microsystems Europe. His current research interests are in the area of electric machines modelling and drive designs, real-time sensor-less speed estimation in electrical machines and design of integrated circuits motor controls for automotive applications. Alexander. C. Smith received the B. Eng. Degree and PhD degrees from Aberdeen University, Aberdeen, UK, in 1977 and 1980. He held previous academic appointments at Imperial College, and the University of Cambridge. In 1997, he joined Invensys Brook Crompton as Head of Research responsible for motor technology. Since 2000, he has been with The University of Manchester (formerly UMIST), as Professor of Electrical Machines in the School of Electrical and Electronic Engineering. He is the Director of Rolls-Royce University Technology Centre on Electrical Systems for Extreme Environments, Fellow of the Institute of Engineering and Technology (formerly IEE) and Editor-in-Chief of the IET journal Electrical Systems in Transportation. His research interests include design and modelling of motors, generators and drives. Paul M. Tuohy received the BEng. (Hons) degree in Mechatronics with Industrial Experience and the Ph.D. degree in Electrical Engineering from The University of Manchester, Manchester, U.K., in 2006 and 2011, respectfully. From 2011 he has been a postdoctoral Research Associate at the Rolls-Royce University Technology Centre (Electrical Systems for Extreme Environments) at The University of Manchester, Manchester, U.K. His research interests include the design, finite element analysis, and testing of electric machines, actuators and drives for aerospace, vehicle, marine and renewable energy applications. Dr. Tuohy was the recipient of the Siemens Medal in 2006. He was elected Whitworth Scholar in 2007 and Whitworth Senior Scholar in 2012. He is also the holder of two patents.

5