Iec 60085 Pdf Free Download

During their lifetime operation hydro generators are subjected to several types of stressors such as electrical, mechanical and thermal. Currently, machines that are required to operate in demanding cyclic load regimes are challenged with a possible additional stressor associated with the frequent cycles of start-stops. In this context, the purpose of the present work is to demonstrate how the insulation system has to be evaluated to cover such demands. Typical results covering the electrical, mechanical, thermal and thermal-mechanical performance are presented to illustrate which properties should be expected from a reliable technology.

Figures - uploaded by Camila S. Gonçalves

Author content

All figure content in this area was uploaded by Camila S. Gonçalves

Content may be subject to copyright.

Discover the world's research

• 20+ million members
• 135+ million publications
• 700k+ research projects

Join for free

1

fernando.pereira@voith.com , Tel: (11) 3944 5159 Fax: 3944 5182

Friedrich Von Voith, 825 – São Paulo – SP – Brazil

How to evaluate an insulation system designed to operate

in demanding load regimes

F. Pereira M. J. da Silva T. E. Lamas C. S. Gonçalves J. F. Cordeiro

T. K. Aoki E. J. Faria T. Hildinger

ABSTRACT

During their lifetime operation hydro generators are subjected to several types of stressors such as

electrical, mechanical and thermal. Currently, machines that are required to operate in demanding cyclic

load regimes are challenged with a possible additional stressor associated with the frequent cycles of

start-stops. In this context, the purpose of the present work is to demonstrate how the insulation system

has to be evaluated to cover such demands. Typical results covering the electrical, mechanical, thermal

and thermal-mechanical performance are presented to illustrate which properties should be expected from

a reliable technology.

KEYWORDS

Insulation system, dissipation factor, partial discharges, voltage endurance and thermal cycling

1. INTRODUCTION

A typical concern on the quality of stator bars/coils arises from the difficulties to assure, based upon

diagnostic tests, that the performance of any insulation system under evaluation is adequate to withstand

the long-term in-service stressors. To shed light on the subject and provide a reference for comparisons,

the present work reports the performance of a time-proved epoxy-mica VPI (vacuum-pressure

impregnation) system known in the industry as Micalastic® insulation system. Once such system is in use

for more than 40 years in the industry equipping hydro-generators (among other rotating machines) under

service regimes of base-load, peak-load and pumped-storage whose voltage levels range from 6kV up to

23kV, it may serve as a reliable reference to define the test results and properties in diagnostic tests that

can be expected for any system designed for the same or similar applications. In this context, the purpose

of the present work is to disclose the performance of the system in the most used diagnostic tests to serve

as a "performance-reference" for the industry.

The study was carried out using original Roebel stator bars produced over recent years to equip a variety

of hydro-generators worldwide. Depending on the nature of the test or evaluation the Roebel bar itself was

used as test-object. That is usual for electrical characterizations such as the ones according to IEEE 1443

and IEC 60270 (partial discharges), IEEE 286 (dissipation factor), IEEE 1043 and IEEE 1553 (voltage

endurance test), etc. In other instances such as those typically used for mechanical and thermal

characterization (i.e. ASTM D638:2010, IEC 60216, etc) not a complete bar, but samples or segments of

mainwall insulation extracted from real bars are most commonly used as test-objects. In either case

worldwide accepted international standards were used in the present work as guidelines for the

performance evaluation tests.

2. ELECTRICAL PERFORMANCE

There are several methods available in the industry to characterize the insulation performance of stator

bars and coils. The present work will provide such characterization using four of the most used techniques:

(i) tangent delta (or dissipation factor), (ii) partial discharges (PD), (iii) electrical lifetime (or voltage

endurance) and (iv) electrical strength (or electrical breakdown).

2

Tangent delta [1] is perhaps the most used technique to characterize the performance of groundwall

insulation of Roebel bars and multi-turn coils. Such measurement determines the dielectric losses within

the insulation such as the losses by electric conduction, dielectric absorption (or polarization) and

ionization of internal voids in the insulation (partial discharges). In fact most of the technique diagnostics

comes not out of the tangent delta value itself – which is a not-known combination of the several losses

mechanisms - but rather from the evaluation of how it varies as a function of the applied voltage. The

reason to focus on the voltage-dependence other than on the absolute value is that the dependence of the

tangent delta with the applied voltage is governed mainly by the growing number of ionization events of

internal voids into the insulation and does not vary much with the other loss mechanisms (conduction and

polarization). Therefore, the evaluation of how the tangent delta varies with applied voltage is a

measurement of how much of the dielectric loss is due to the presence of voids or delaminations within the

insulation volume. If the value of tangent delta does not change significantly with the increase of the

applied voltage, then the insulation is understood as not having significant voids or air inclusions. On the

other hand, if such variation is large, then the insulation may contain a concerning level of internal defects

(voids, delamination, etc).

Figure 1 presents a typical experimental curve showing the dependence of tangent delta (or dissipation

factor) as a function of the applied voltage where the tangent delta data is presented in percent (%) and

the applied voltage is expressed in terms of a percentage of the machine rated voltage (0.2Un, 0.4Un,

etc).

Figure 1 – Typical dissipation factor curve as a function of test voltage for a 19 kV rated machine.

In practice, as the measurement is performed in a large number of electrical components, each component

curve is evaluated based upon only two quantities: (a) the so-called initial tangent delta, which is the value

of tangent delta at a test voltage of 20% of the rated voltage (0.2Un) and (b) the tangent delta tip-up, which

is a measurement of the tangent delta variation between two voltage levels. The initial tangent delta is

typical of a system and may be different for different insulation systems, but once the system is fixed, such

quantity determines the state of groundwall cure. The tangent delta tip-up characterizes the voltage-

dependence of the tangent delta and therefore it is a more important parameter if one is interested in

evaluating the insulation quality. Figure 2 presents the experimental data of initial tangent delta (0.2Un)

plotted in Weibull scale of a complete set of production of Roebel bars for a 19kV machine. The Weibull

average (63-percentile) is about 0.6-0.7%.

3

Figure 2 – Initial tangent delta (0.2Un) plotted in Weibull distribution for a 19kV machine.

Figure 3 presents the results of tangent delta tip-ups measured in the same set of bars as those of Figure

2. The results show a variation between 0.01% to about 0.08%. In this plot the tip-up is defined as the

maximum variation in tangent delta found in any 0.2Un-wide interval of applied voltage (sometimes called

the maximum delta tangent delta).

Figure 3 –Tangent delta tip-up (0.2Un-wide interval) for a 19kV machine.

Although the measurement of tangent delta as a function of voltage is ultimately a method to characterize

the existence of voids or delaminations within the insulation by detecting their associated losses caused by

internal electrical discharges, another technique exists that directly measures the apparent flux of charge

during each event of an occurring discharge in or around the insulation volume. Such technique is known

as partial discharge (PD) analysis [2]. Figure 4 presents the results of PD amplitudes found in two sets of

bars, one of 19kV components (4a) and the other for a 6.9kV rated machine (4b). The results indicate that

the values expected will reach up to about 1000 pC and the average will occur around 400-500 pC.

4

Figure 4 – Stator bars partial discharges values plotted in Normal distribution for a 19kV (a) and 6.9kV (b)

machine. The PD equipment used was a LDIC LDS-6 model with 1500pF coupler capacitor.

Tangent delta and PD provide information on potential defects that may have influence on the

performance or lifetime of the groundwall insulation. It is hard, however, to establish a direct connection

between those quantities and the expected lifetime of the electrical component [3].

A technique known in the industry as voltage endurance test (or electrical lifetime test) is an attempt to

characterize the electrical lifetime of the component as a whole. In such test a voltage significantly higher

than the one expected to exist during operation is applied and the time to failure (by electrical breakdown)

is measured. Several test routines with specific voltage levels can be used each one corresponding to a

different life expectancy [4 and 5]. Figure 5 shows a typical result where 14 bars rated 13.8kV are allowed

to age until their electrical breakdown under a fixed electrical stress corresponding to three times its phase

voltage (41.4kV). The results indicate that the sample lifetime in this particular test condition varies from

50h to 200h.

(a)

(b)

5

Figure 5 – Stator bars voltage endurance test results plotted in Weibull distribution for a 13.8kV machine.

A similar way to evaluate the ability of the groundwall insulation to withstand voltage is performed by the

so-called electrical breakdown test [6]. In this test the electrical stress is continuously increased (by

increasing the applied voltage) until insulation electrical breakdown occurs. The output parameter of the

test is therefore the maximum voltage withstood by the component. Figure 6 shows the typical Weibull-

scale results distribution for a 13.8kV set of stator bars. Two reference lines at 5Un and 6Un are indicated

as commonly practiced limits in the industry. The experimental results show breakdown voltages within the

range of 95kV to 120kV for 13.8kV rated components.

Figure 6 – Stator bars breakdown voltage test results plotted in Weibull distribution for a 13.8kV machine.

Although the number of possible electrical tests used in the industry to characterize the quality and

performance of insulation in bars or coils may be huge, the ones presented in Figures 1 to 6 are the most

used and therefore provide a valuable source of reference to allow a comparison between the presented

system and any other system designed for similar purposes.

3. MECHANICAL PERFORMANCE

Among all the possible static quantities that can be used to characterize the mechanical behavior of

groundwall insulation of Roebel bars or multi-turn coils, the present work based its evaluation in the

flexural strength [7], where the deforming force is applied perpendicularly to the mica tape plane as

6

illustrated in Figure 7, and tensile strength [8] where the deforming forces are applied along the mica tape

plane within the insulation as illustrate in Figure 7.

Figure 7 – Flexural and Tensile test setup for samples of groundwall insulation. The arrows indicate the

direction of the applied force.

The relevant parameters for such type of mechanical characterization relies on the determination of the

maximum force supported by the insulation in terms of the flexural and traction strength, the maximum

elongation until break and the elasticity modulus (Young Modulus). Table 1 summarizes such results.

Property Unit Result

Flexural strength N/mm

2

50

(*)

Tensile strength 90 - 105

Young modulus 54300

Elongation to break % 0.21 - 0.25

(*)

The sample preparation that includes the removal of insulation from a real bar can produce measurements artifacts.

Table 1 – Mechanical properties of the groundwall insulation.

4. THERMAL PERFORMANCE

One of the most important properties of the groundwall insulation concerning its thermal performance is

the rate of material decomposition and consequent weight loss as a function of temperature. Such

characterization is standardized by IEC 60216 [9] and IEC 60085 [10] and gives rise to a reference

number commonly referred in the industry as weight loss thermal index. The thermal index is defined as

the temperature at which a certain amount of material is lost by thermal aging after 20,000 hours. In the

present work the fraction of weight loss used as reference for the estimation of the thermal index was 3%

of the original weight. Such weight loss level (3%) is more stringent than most of the criteria used in

literature [11].

Figure 8 presents the results of weight loss for the 3% criterion. The curve indicates the time required for

the mainwall insulation to lose 3% of its original weight as a function of temperature. The extrapolation for

the 20,000 hours aging [10] gives 168

o

C as the thermal index for the insulation system as indicated in the

graph of Figure 8. A second extrapolation, that goes far beyond the range of experimental data and

therefore has smaller rigor suggests that a timescale of about one million hours (more than about 100

years) would be the timescale required for a relevant weight loss of mainwall insulation under the typical

hydro generator operating temperatures. Such results suggest that in nominal operating conditions of

properly designed machines the thermal aging, which may be one of the main stressors for the groundwall

insulation, is not expected to be a limiting factor in the system maximum lifetime.

7

Figure 8 – Bar insulation weight loss for the 3% criterion to determine the Thermal index value.

5. THERMAL-MECHANICAL PERFORMANCE

An additional important property of the groundwall insulation is related to its ability to withstand successive

cycles of thermal expansion and contraction. Such property is particularly relevant for machines whose

operation involves load cycles such as those expected to occur in pumped-storage generators or some

peak-load machines. A method to simulate such demands is proposed by IEEE 1310 [12] where a

controlled current source provides the means to produce repetitive thermal cycles in a series of individual

bars or coils. Figure 9 shows the experimental set-up used in the present work.

Figure 9 – Thermal Cycling test setup according to IEEE 1310.

The cyclic thermal loads, however, are designed to simply provide a method of thermal-mechanical aging

and do not actually fail the insulation. The concept of the test is such that some complementary evaluation

must be used during or after the completion of the thermal cycles to in fact diagnose the performance.

Among all the possible post-thermal-cycling diagnostics (tangent delta tip-up, PD, etc) the present work

focused in the execution of electrical lifetime tests as a diagnostics to evaluate eventual relevant damages

to the insulation caused by the thermal-mechanical aging. In that sense, IEEE 1553 voltage endurance

test has been used to evaluate individual components after they have been subjected to previous thermal-

mechanical aging according to IEEE 1310. All results indicate that the components of the present system

can pass both 400h and 250h test schedules even after severe thermal-mechanical tests.

6. CONCLUSIONS

The presented data discloses the performance of a time-proved insulation system used to equip a variety

of generators worldwide for more than 40 years. The results illustrate the system performance in the most

used diagnostic tests currently applied by the industry. Once most of the characterization presented was

based upon the application of international well-known test standards any other insulation system can be

8

compared in terms of performance with the results here presented. The knowledge of such results may be

useful for the utility having to decide between competing insulation systems or to the researcher of a new

system, in need of a time-tested reference to which the new developing system can be compared with.

References:

[1]

IEEE Std 286-2000, IEEE Recommended Practice for Measurement of Power-Factor Tip-Up of

Rotating Machinery Stator Coil Insulation.

[2]

IEEE Std 1434-2000, Guide to Measurement of Partial Discharges in Rotating Machinery. IEC 60270 -

High-voltage test techniques – Partial discharge measurements

[3]

G. C. Stone, H. Sedding, B. Lloyd, B. Gupta, "The ability of Diagnostic Test to Estimate the Remaining

Life of Stator Insulation", IEEE Transactions on Energy Conversion, Vol. 3, Issue 4 (1988);

[4]

IEEE Std. 1553-2002, "IEEE Trial-Use Standard for Voltage-Endurance Testing of Form-Wound Coils

and Bars for Hydrogenerators"

[5]

KEMA S13/14: "KEMA specification for hydrogen, liquid and air-cooled, synchronous a.c. generators

with rated voltage 5 kV and above"

[6]

IEEE Std. 4 – 1995 "Standard Techniques for High-Voltage Testing"

[7]

DIN EN ISO 178 - 2006 – "Determination of flexural properties"

[8]

ASTM D638:2010 – "Standard Test Method for Tensile Properties of Plastics"

[9]

IEC Std. 60216 (all parts), "Electrical insulating materials – Properties of thermal endurance"

[10]

IEC Std. 60085–2007, "Electrical insulation – Thermal evaluation and designation"

[11]

R. Bruetsch, R. Schwander, F. Wolf, M. Naegelin "New materials and thechniques for the of coils and

roebel bars" 9th INSUCON International Electrical Insulation Conference, Berlin (2002)

[12]

IEEE Std. 1310-2012, "IEEE Trial-Use Recommended Practice for Thermal Cycle Testing of Form-

Wound Stator Bars and Coils for Large Generators"

ResearchGate has not been able to resolve any citations for this publication.

• G.C. Stone
• H.G. Sedding
• B.A. Lloyd
• B.K. Gupta

Stator windings from three generators and one motor were subjected to a wide variety of diagnostic tests. Diagnostic tests included insulation resistance, polarization index, capacitance, dissipation factor tip-up, partial discharge magnitude, and discharge inception voltage. The destructive tests included breaking down individual coils with either AC, DC, or impulse voltages. Since the results of the diagnostic tests did not correlate with the breakdown voltages, it is concluded that remaining life cannot be predicted on the basis of diagnostic tests alone. Equations and other relationships recently developed in Japan to predict the remaining life of the stator groundwall insulation systems could not be confirmed. Diagnostic tests are thought to be most useful for indicating the trend in insulation aging in a particular machine

Standard Techniques for High-Voltage Testing

IEEE Std. 4 – 1995 " Standard Techniques for High-Voltage Testing "

Electrical insulation -Thermal evaluation and designation

• Iec Std

IEC Std. 60085-2007, "Electrical insulation -Thermal evaluation and designation"

Determination of flexural properties

• Din En
• Iso

DIN EN ISO 178 -2006 – " Determination of flexural properties "

IEEE Trial-Use Standard for Voltage-Endurance Testing of Form-Wound Coils and Bars for Hydrogenerators

• Ieee Std

IEEE Std. 1553-2002, "IEEE Trial-Use Standard for Voltage-Endurance Testing of Form-Wound Coils and Bars for Hydrogenerators"

New materials and thechniques for the of coils and roebel bars

• R Bruetsch
• R Schwander
• F Wolf
• M Naegelin

R. Bruetsch, R. Schwander, F. Wolf, M. Naegelin " New materials and thechniques for the of coils and roebel bars " 9th INSUCON International Electrical Insulation Conference, Berlin (2002)

Guide to Measurement of Partial Discharges in Rotating Machinery. IEC 60270 -High-voltage test techniques -Partial discharge measurements

• Ieee Std

IEEE Std 1434-2000, Guide to Measurement of Partial Discharges in Rotating Machinery. IEC 60270 -High-voltage test techniques -Partial discharge measurements

4 -1995 "Standard Techniques for High-Voltage Testing

• Ieee Std

IEEE Std. 4 -1995 "Standard Techniques for High-Voltage Testing"

IEEE Trial-Use Recommended Practice for Thermal Cycle Testing of Form-Wound Stator Bars and Coils for Large Generators

• Ieee Std

IEEE Std. 1310-2012, "IEEE Trial-Use Recommended Practice for Thermal Cycle Testing of Form-Wound Stator Bars and Coils for Large Generators"

Source: https://www.researchgate.net/publication/267765119_How_to_evaluate_an_insulation_system_designed_to_operate_in_demanding_load_regimes

Posted by: jeromyadelmanssa.blogspot.com