Power Transformer

Diagnostics and Condition monitoring of Power Transformer

ISOSCELES SALES & SERVICE PVT LTD

Abstract:

Various diagnostic techniques are used by utilities to carryout transformer diagnostic with a view to determine earlier failure probability. The present paper describes the result of on-sight measurement of various parameters for transformers from 11 kV to 220 kV. The measurement includes several tests like tan δ, capacitance, insulation and winding resistance, magnetizing voltage and SRFA .Significance of the tests in light with the experimental results have been discussed in the paper.

Key words:

Diagnostic, SFRA, Power transf., on -sight test

I. Introduction

Power transformers have been in service for many years under different conditions. Transformer outages have a considerable economic impact on the operation of an electrical network. Therefore it is desirable to ensure an accurate assessment of the transformer condition. Techniques that allow diagnosing the integrity through non-intrusive tests can be used to optimize the maintenance effort and to ensure maximum availability and reliability. With the increasing average age of the transformer population there is an increasing need to know the internal conditions

Experimental investigations have shown that failures [1,2] of various components of transformers occur over a period of time due to several reasons, either internal or external to the system. This led to the growing concern amongst utilities on the cost of assets and its replacements, particularly those of power transformers. In this context, monitoring and onsite diagnostics are seen as a possible way of optimizing existing assets. The main considerations are to reduce maintenance cost, to prevent forced outages with related consequential cost and to enable existing equipment work longer. Since the cost of power transformer is very high in power transmission and distribution system, it is important to develop equipment, which could aid in maintenance of power transformers. In other words, it is essential to determine the likely failures which could occur including the causes of failure. It is important to understand the differences between monitoring and diagnostics. Monitoring uses sensors to provide raw data and warning signals from the equipment under study. Diagnostics are the more important part of the equipment as it analyses the data and provides strategic information and management technique. The expectation of increased efficiency and performance in utilities today requires that less time and expense is used in evaluating the status of individual transformers and the transformer population in a more efficient manner without compromising their long-term life expectancy. Reports available from CIGRE working group highlights the type of failures and the components involved in failure. It is shown that failure of windings and terminals contribute to the highest percentage of failure.

Bushings are certainly not the most expensive piece of equipment in a substation, so the financial loss of a bushing failure is not the driving force. However, the damage that a failed bushing can inflict on its affiliated apparatus could indeed be catastrophic. Specifically, bushing failures leading to a damaged power transformer have been well documented. Most bushing failures can be attributed to internal deterioration or contamination and being able to detect these irregularities is essential in maintaining a stable system .The industry has accepted the conventional off-line power factor/capacitance test as the most reliable tool for identifying problem in bushings and current transformers. It is the success of the power factor/ capacitance measurement in bushing diagnostics and the awareness of the catastrophic results of bushing failure that have lead industry experts recommend more frequent testing of bushings.

The present paper deals with the result of several measurements carried out on 220/110 kV, 220/66 kV and 66/11 kV transformers and 100 kV CT of various utilities. The measurements include insulation resistance, tanδ and capacitance of bushings, winding resistance, turns ratio and percentage impedance. The measurement also includes various functional checks of all control components. The paper also describes the limitation of measurement in transformers where bushing tap does not exist. The comparison of tan δ and capacitance for meaningful conclusion has been discussed. Measurement using SFRA technique has been discussed with reference to movement of winding

II. Measurement and Significance

Insulation resistance measurement is known to yield very useful information regarding the status of paper oil insulation system. If there is high moisture content in the total insulation system (cellulose + oil) then it is stated that water migrates between the paper & the oil during the heating and cooling cycles of the transformer. As the transformer heats up water migrates to the oil & conversely when it cools the water goes back to the cellulose insulation. Thus moisture ppm measured from a given oil sample is a function of temperature at which sample has been taken. If one takes an oil sample & gets 30 ppm it will depend on what temperature one took the sample at. The dielectric strength can still be 50KV with that level of moisture and hence there is there is no problem in withstand characteristic. If the IR is low one needs to assess at what temperature it was taken. So if one takes the oil sample at high temperatures then carries out IR at a cool temperature then the moisture will be high in both as one is measuring some of the moisture twice because it has been migrating.

Thus it is important to do both tests at similar temperatures .That way it will see the true level of moisture. If the same readings are obtained then one would consider looking at drying out the transformer to remove as much moisture as possible.

III. Measurements and results

It is observed from the Table I that the capacitance values for all phases are almost equal and is much larger than bushing capacitance of a 220kV rating. These values are expected since the measured capacitance shows the capacitance between winding and ground rather than bushing capacitance alone. Similarly the value of Tanδ reflects the quality of oil paper insulation between the winding including bushing rather than bushing alone. It may, however be inferred that any major changes in tanδ reflects the complete failure of bushing. The statement comes from the fact that reduction in quality of oil as result of increase in moisture, sludge formation or ageing of paper and oil leads to the increase in limited value, where as failure in bushing will lead to unbalance in measurement. It is to be noted that any increase in capacitance and tan δ does not conclusively prove the presence of partial failure but unbalance in bridge will reflect the complete failure of the corresponding bushing.

TABLE I. Capacitance and tan ∂ measurement of Bushing

Bay Name: T/F No.2 (125MVA), 220/110 kV;

Voltage Rating: 220KV; Mode used for Testing: GST Mode;

Date of Testing: 27.01.2008; Mode: GST (HV BUSHING);

Phase

R

Y

B

2 KV

C (nF)

12.370

12.13

12.430

Tan (%)

7.390

12.36

8.280

Power Loss(mw)

1.204

2.343

1.361

5 KV

C (nF)

13.440

14.18

13.480

Tan (%)

18.940

41.80

21.510

Power Loss (w)

20.200

46.79

22.670

10 KV

C (nF)

12.880

9.15

12.890

Tan∂ (%)

56.900

90.00

56.600

Power Loss(w)

230.500

342.90

229.800

TABLE II. Capacitance and tan ∂ measurement of Bushing

Bay Name: T/F No.2 (125MVA), 220/110 kV;

Voltage Rating: 110KV; Mode used for Testing: GST Mode;

Date of Testing: 27.01.2008; Mode: GST (LV BUSHING)

Phase

R

Y

B

2 KV

C (nF)

11.370

11.440

11.360

Tan (%)

2.384

2.648

2.366

Power Loss(mw)

337.700

377.700

364.600

5 KV

C (nF)

11.510

11.680

11.490

Tan (%)

3.434

4.800

3.247

Power Loss (w)

3.170

4.427

2.977

10 KV

C (nF)

11.670

11.940

11.660

Tan∂ (%)

5.380

8.300

5.260

Power Loss(w)

19.740

31.840

19.430

TABLE III. Capacitance and tan ∂ measurement of Bushing

Bay Name: T/F No.1 (125MVA), 220/110 kV;

Voltage Rating: 220KV; Mode used for Testing: GST Mode;

Mode: GST (HV BUSHING); Date of Testing: 27.01.2008;

Phase

R

Y

B

2 KV

C (nF)

12.370

13.310

12.600

Tan (%)

6.600

12.670

8.610

Power Loss(mw)

1.076

2.346

1.406

5 KV

C (nF)

13.440

13.680

13.690

Tan (%)

16.770

55.800

26.080

Power Loss (w)

17.800

59.100

27.670

10 KV

C (nF)

12.490

7.477

11.760

Tan∂ (%)

62.100

96.500

71.000

Power Loss(w)

241.800

338.300

260.500

TABLE IV. Capacitance and tan ∂ measurement of Bushing

Bay Name: T/F No.1 (125MVA), 220/110 kV;

Voltage Rating: 110KV; Mode used for Testing: GST Mode;

Date of Testing: 27.01.2008; Mode: GST (LV BUSHING);

Phase

R

Y

B

2 KV

C (nF)

11.500

11.610

11.500

Tan (%)

2.448

2.884

2.436

Power Loss(mw)

363.400

458.500

370.300

5 KV

C (nF)

11.640

11.880

11.640

Tan (%)

3.413

5.420

3.516

Power Loss (w)

3.198

5.438

3.416

10 KV

C (nF)

11.830

12.170

11.830

Tan∂ (%)

5.590

9.770

5.880

Power Loss(w)

20.820

37.440

21.780

In this transformer there was no Bushing Taps and hence measurement was taken in GST Mode. Winding Tan delta is to be also taken so that comparison can be made.

Insulation resistance between winding is yet another important parameter suggesting the voltage withstand characteristic and leakage current between windings and winding to ground. The insulation resistance of a transformer depends largely on the temperature and cleanliness and dryness of the windings. If it falls below a particular value, presence of dirt or moisture may be indicated. A typical value of measured insulation resistance and corresponding polarization index is shown in Table V. It is seen from the table that both the insulation resistance as well as polarization index for LV to earth is lower compared to HV to earth and HV to LV and that polarization index of LV to Earth comes under questionable limit .A polarization index between 1.2 to 2.0 may be considered fair. More than 2.0 is assumed to be acceptable. Thus the polarization index of 1.18 for LV to Earth needs to be reviewed. Since the oil in transformer is communicating with all section of winding it is recommended that total oil should to be filtered and processed until improvement in oil quality is achieved.

TABLE V. Insulation resistance of a 66/11 kV Transformer

Voltage :66/11 KV Insulation Resistance Measurement: at Tap 17 13.22%

Connection

60 Sec

600 Sec

P. I.

HV-Earth @ 5KV

17.13 GΩ

35.50 GΩ

2.07

LV-Earth @ 5KV

3.42 GΩ

4.03 GΩ

1.18

HV-LV @ 5KV

26.10 GΩ

61.70 GΩ

2.36

HV bushing tap-gnd @500V

GΩ

GΩ

Winding resistance provides direct information with regard to the looseness in the contact between bushing rod and winding or the breakage in the winding conductor.

The looseness is considered onerous since, the normal power current can give rise to excessive ohmic loss due to increased contact resistance. The oil in the bushing being non- communicating to the bulk transformer oil can cause the excessive heating of the bushing rod and eventual softening of the Viton/rubber O – ring providing oil sealing thus causes leakage of moisture and eventual failure of bushing.

Several bushing failure in service has been linked to the bushing rod excessive heating caused due to looseness of connection, either of the winding or of the jumper from the main transmission line. Thus the measurement of winding resistance is one of the important diagnostic requirements. A typical measurement of winding resistance for a LV (11 kV) winding in Star and Delta mode is given in Table VI. In a typical case it was observed that the L.V. winding of a transformer gave a value of Ran, Rbn and Rcn equal to 7.2, 6.6 and 25.8 milli.Ω respectively when measured between line and neutral. The line to line measurement showed a value of Rab, Rbc, Rca as 13, 31.5, 31.8 milli. Ω respectively. From both measurements, it is observed that terminal ‘c' must have an improper connection giving rise to higher contact resistance. The contacts were cleaned and measured after reconnection. No significant change was observed. Thus it was inferred that the terminal contact resistance inside the transformer need to be checked and corrected.

In another measurement the resistance of all the three phases of high voltage winding is given in Table VII for all tap positions. It can be seen that the difference in winding resistance for the three phases are very small. The relatively small difference can be attributed to the contact resistance variation between various taps. Thus it can be inferred that small variation in winding resistance as seen in Table VII is within acceptable limit.

TABLE VI. LV winding resistance of transformer

Ran (mW)

Rbn (mW)

Rcn (mW)

Rab (mW)

Rbc (mW)

Rca (mW)

7.2

6.6

25.8

13

31.5

31.8

TABLE VII. Winding resistance of high voltage winding

Tap

RY (mW)

YB (mW)

BR (mW)

1

560.6

557.8

555.7

2

545.2

555.7

542.0

3

543.6

535.8

535.0

4

532.2

534.2

525.3

5

519.2

523.4

513.3

6

507.5

503.1

500.8

7

497.8

492.4

491.0

8

486.0

484.5

481.0

9

473.6

473.4

468.4

10

467.7

466.5

459.8

11

455.6

457.0

448.0

12

446.1

443.6

437.5

13

431.8

437.0

433.0

14

421.6

431.0

422.0

15

416.2

417.0

415.2

16

415.1

413.7

401.5

17

403.9

397.3

400.8

Of the several off-line diagnostic techniques used, the magnetic balance test is considered equally reliable in ascertaining the fault in the winding. The procedure followed is to apply phase or line voltage to one of the windings and measure across other two windings .It is necessary that sum of the voltages in the other phases should be equal to the applied voltage.

A typical result of primary as Delta and secondary as Star can be seen in Table VIII. It was observed that magnetic balance test gives voltages which are perfectly balanced. Let us consider a specific result of HV winding where a voltage of 435 V is applied in Delta connected HV winding say between RY .The sum total of voltage between YB and BR is found to be 438.4 V. Considering a small measurement error the values are same. Similarly in second case, an applied voltage of 435 Volts produces voltages in the other two phases whose sum total is about the same. The measurement and analysis as discussed above are applicable to the LV winding as seen in part B ‘ of Table VIII.

TABLE VIII. Results of magnetic balance test

HV Winding (A)

RY

YB

BR

I Mag. mA

435

351

87.4

3.25

265.5

435

170.5

2.05

118.5

317

436

2.07

LV Winding (B)

Rn

Yn

Bn

I Mag. mA

249.9

201.8

48.5

67.8

146.6

248.6

101.9

46.1

60.8

188.5

249.2

80.0

The sweep frequency response of transformer has now gained greater acceptability among utilities as a diagnostic tool for detecting winding movement (3,4) as well as well winding fault.

While winding movement is more clearly detectable using SFRA, disc or turn fault requires greater skill for identification. A few results are presented for a 100MVA 220/132/33 kV transformer where in a number of SRFA recordings were carried out between HV- N, IV-N and HV-IV for a given tap position. A comparison of SFRA for all the three phases of HV winding to ground (Fig 1) shows relatively better similarity. The small difference can be attributed to the difference in manufacturing accuracy of various phase windings and digitizing error. In addition the central phase always exhibit slightly different LC behaviour compared to other two outer phase. However, the SRFA between LV winding to ground shows a completely different pattern ( 2) between 100Hz to 10 kHz. Since the lower frequencies in the above range are linked to physical movement of windings, it may be inferred that there has been change in Y phase. The change can be linked either to the inter-winding shift or duct reduction in the Y-phase winding. However the statement needs to be verified by physically opening the transformer and inspecting such changes.

Case study for tan δ of bushing and transformer

The dissipation factor is a critical property in OIP bushing. It is dependent on the moisture in oil in paper and amount of contaminant in the insulation system. The value of Tan δ has dependency on temperature as well. A Typical behaviour [5] is shown in 3. Thus it is evident that with higher value of moisture higher value of dissipation factor is exhibited

A typical value for Tan δ for a 66 kV class bushing measured on a transformer working in service from several years is reported in Table 9. The test object is in Ungrounded Specimen Test (UST) mode.

Table IX : Tan δ and cap. 0f 66 kV bushing of 66 kV/11 kV Transformer


Capacitance

Tanδ %

UST - R Phase

223.1

pF

0.292

UST - Y Phase

218.1

pF

0.254

UST - B Phase

217.3

pF

0.259

It is observed that the Tan δ value in all bushing is in the range specified by the relevant standard (IS 2099).It confirms absence of moisture in the bushing system. Almost identical value of capacitance suggests the healthiness of bushing. The result suggests that if properly maintained a bushing can work successfully for longer years without replacement.A few of the measurements taken at site for transformer winding and CT are seen in Table X and Table XI. Values of tan δ as high as 15.95% and 2.069 % are observed in case of 100 kV transformer winding and CT respectively. The safe limit of tan δ in UST mode is 0.7% and in GST mode is 1%. Thus the values are far in excess of accepted value. Although the utility is advised to change or process the oil it is to be mentioned that transformer has yet been working satisfactorily. It is to be stated that even with higher value of Tan δ of transformer it can work successfully

2 kV 5 kV 10 kV

C

(nf)

Tan d(%)

C

(nf)

Tan d(%)

C

(nf)

Tan d(%)

4.799

2.25

4.799

2.24

4.799

2.23

2.580

15.95

2.240

6.16

2.502

4.81

3.393

14.11

3.509

5.2

3.468

5.8

2 KV 5 KV 10 KV

C (pF)

Tan∂

(%)

Phase

Tan∂

(%)

C (pF)

Tan∂

(%)

R

554.4

1.186

554.5

1.186

554.7

1.269

Y

563.3

1.193

563.1

1.172

563.1

1.233

B

502.1

1.975

502.2

1.984

502.1

2.069

Conclusion

The paper presents the results of measurements of various parameters related to transformers with a view to determine any weakness and long term performance characteristic of the transformer. From various measurements and analyses the following conclusions have emerged.

1) One of the terminals of 11 kV star connected windings is loosely joined to bushing terminal.It may have long term implication in bushing failure.

2) The polarization index of LV winding shows a value below acceptable level.This needs to be corrected by suitable filtering to improve the life.

3) In another transformer of 220/33/11 kV class there appears to be shift in the Y phase of 11 kV winding. The fact needs to be verified by opening the transformer and inspecting it for inter-winding or duct movement.

4) Value of dissipation factor observed for some of transformer winding is relatively high compared to safe limit and the utilities use with those higher values the life of transformer is expected to reduce.

References

[1] Z.T.Yao and T.K.Saha “Separation of ageing and moisture impacts on transformer insulation degradation by polarization measurements” CIGRE 15-304 Paris Session 2002.

[2] Pradeep Nirgude, A D Rajkumar, Channakessava, B.P.Singh and A.Bhoomaiah “Frequency response technique for diagnostic of power transformer” National Conference on Modern Trends in High Voltage and Power system Engineering, Jabalpur, July 7-8 2003.

[3] J. Christian,K.Feser “Procedure for detecting winding displacements in power transformers by transer function method” IEEE Transactioon Power DeliveryVol.19 , no1,January 2004

[4] Simon A. Ryder “Methods of comparing Frequency Response Analysis Measurements” Cof Record of 2002 Symposium on electrical Insulation, Boston,MA USA April7-2002

[5] library.abb.com/global/scot/.../2750%20515-142%20en%20Rev%200.pdf


Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!