Application of Temperature Correction Factors for dissipation factor

Measurements for Power Transformers – A case study

R.K.Tyagi S. Victor N.S.Sodha

Chief Manager Dy. General Manager Addl. General Manager

Power Grid Corporation of India Limited

Synopsis

In POWERGRID EHV Transmission Network, 520 nos. of Power Transformers and Reactors
are in service with some of them being 22 years old. A variety of condition monitoring
techniques/ checks are implemented for testing of these Transformers. Capacitance and Tanδ
Measurements for Windings and Bushings is one such important test. This paper deals with
monitoring of Tanδ and Capacitance values and their rate of change with temperature. If
testing is carried out at winding temperatures other than standard temperature i.e. 20ºC then
correction factor is applied to convert Tanδ value to the standard temperature. Document
IEEE-C57.12.90 mentions temperature correction factors which are generally applied by
Testing Engineers in power utilities.

In POWERGRID, a R&D project was undertaken during 1998/99 to compute temperature

correction factors for Tanδ measurements. 4 nos. 400/220kV Auto Transformers of different
makes were selected for this project. On these Transformers, Winding temperature was
increased to about 90 degree centigrade by regulating the load and then Capacitance and Tan
Delta measurements were carried out at different temperatures during cooling cycle. It has
been observed that Tan delta values are affected by change in temperature and the rate of
change of Tanδ values for different makes of Transformers. It was also observed that the Tanδ
values for new and old Transformers do not follow the same pattern.

KEY WORDS

Dissipation Factor- Condition Monitoring – Ageing – Temperature Correction Factor- Residual Life
Assessment

INTRODUCTION

Paper impregnated with oil is widely used as insulation material for windings/ bushings of Power
Transformers/ Reactors. Oil-paper insulation degenerates with time, and this degeneration process
depends on thermal, electrical, chemical and other factors within the equipment. Condition of
Insulation affects the reliability and ultimate useful life of the Transformer. The ageing process of oil/
paper insulating systems is very complex. The primary objective should be to ensure that the
insulation of electrical apparatus is in healthy condition during manufacture and at the time of putting
into service (acceptance test). Thereafter, during its useful life span, periodic condition monitoring
checks/ tests are carried out to ascertain the ageing process and the residual life.

Due to higher ambient temperatures in India, typically the normal operating temperature range
for Power Transformers in India lies between 70-100°C. At this temperature, paper degrades
slowly with breakage of the long cellulosic chain. The water content in paper has a great
influence on the life and loading capacity of a transformer. In a new transformer, the water
content in paper is less than or around 0.5% and this gradually increases during service on
account of natural ageing of insulation and also moisture entry from various contributing
factors such as atmospheric air, oil leakages, site repairs etc. Tan Delta values are normally
temperature dependent and increase in temperature leads to increase in Tan delta values also.
If the Tan Delta Test is carried out at different temperatures then the values need to be
corrected to base temperature which is generally taken as 20 degrees centigrade. Document
IEEE-C57.12.90 mentions temperature correction factors which are generally applied by
Testing Engineers in power utilities. Application of temperature correction factors has always
been under deliberation world wide since long time. It is learnt that NGC, UK and Doble,
USA have carried out studies/ tests to verify authenticity/ application of the correction factors
as indicated in IEEE- C57.12.90 for insulation used in Power Transformers. It is generally
observed that these factors can not be applied to all types and makes of Transformers/
Reactors on uniform basis.

CONDITION MONITORING OF POWER TRANSFORMERS IN POWERGRID

In POWERGRID, many condition monitoring techniques have been adopted which include
Dissipation factor & Capacitance measurement, Frequency Response Analysis, Insulation Resistance
Measurement, DGA of Oil, Oil parameters, Winding Resistance Measurement, Turns Ratio Test etc.
In this paper, experience on Dissipation Factor measurement has been discussed.

(A) DISSIPATION FACTOR AND CAPACITANCE MEASUREMENT

1.0 Dissipation Factor(Tan Delta) And Capacitance Measurement

Dissipation factor (tan delta) measurement is indicative of dielectric losses in the insulation. The origin
of the dielectric losses are colloidal (glue) or polar (water or ageing products) or ionic (conductive
particles) contaminants in an insulation. Tan delta is temperature and electrical stress dependent. The
power dissipated in total insulation as a result of dielectric losses in an alternating field is expressed by
the below mentioned equation:

δ Watts
P = ω C V2 Tanδ

From above, it is very clear that dielectric losses are directly proportional to Tanδ value and hence
Tanδ value is true indicator of health of the insulation and its monitoring is helpful in condition
assessment.

2.0 δ and Capacitance Measurement

Factors affecting Tanδ

(a) Significance of Temperature

Most insulation measurements have to be interpreted based on the temperature of the specimen. The
dielectric losses of most insulation increase with temperature. In many cases, insulation have failed
due to the cumulative effect of temperature i.e. a rise in temperature causes a rise in dielectric loss
which in turn causes a further rise in temperature etc. The relationship with temperature is given
below:

δT0 e α(T-T0)
δT = Tanδ
Tanδ

Where,
α = Temperature coefficient

= (ln TanδT - ln TanδT0)/ (T - T0)

(b) Significance of Humidity

The exposed surface of the bushings under high relative humidity conditions may result in
accumulation of surface moisture, which can have a significant effect on surface losses and
consequently the Tan Delta results.

Surface leakage errors can be minimised if dissipation factor measurements are made under clear and
sunny weather conditions when the relative humidity does not exceed 80%. In general, best results are
obtained if measurements are made during late morning through mid afternoon. It is necessary to
consider possibility of moisture deposit due to rain or fog on equipment and measurements should be
avoided in such conditions.

(c) Surface Leakage

Any leakage over the insulation surface of the specimen will get added to the losses of the volume
insulation and may give false impression about the condition of specimen. Surfaces of bushings and
insulators should be clean and dry when making measurement (Exhibit-1)

On bushing provided with power factor or capacitance test tap, the effect of leakage current over the
surface of porcelain bushings may be eliminated from the measurement by testing in UST mode.

(d) Electrostatic Interference:

When tests are carried out in energised substations, readings may be influenced by electrostatic
interference currents resulting from the capacitance coupling between energised lines and the bus work
to the specimen/equipment. In Exhibit-2, jumpers are connected to Bushing which are required to be
opened while carrying out the measurements. In EHV s/s, the effect of electrostatic interference
currents can be cancelled by using the interference suppresser circuit. ‘Normal’ and ‘Reverse’ polarity
readings should be taken to cancel any residual interference currents. Trouble from magnetic fields
encountered in high voltage sub-station is very unlikely.

Exhibit-1( Bushing Surfaces should be cleaned

before measurements)

The effect of electrostatic interference can also be nullified/ minimised by using double
shielded/screened HV Cable/Leads. If any problem is encountered in the shielding arrangement of the
HV lead, measurement shall be practically impossible.
 Exhibit-2(Jumpers are to be removed before measurement)

Basic Guidelines for Measurement of Dissipation Factor and Capacitance Values:

Following guidelines are to be followed while carrying out the test in EHV substation:

1. The kit should have Interference Suppression Unit.

2. The HV lead should be Double Shielded.
3. Clean Surface of the Bushings.
4. Jumpers to be disconnected from Transformer/Reactor windings.
5. All phases of a particular winding to be shorted together and also with neutral to minimise the
effect of inductive currents during measurements.
6. Bushing Dissipation factor and capacitance values are to be measured in UST mode (Ungrounded
Specimen Test mode)
7. Measurement for windings shall be carried out as per the following combinations:

• Winding to winding in UST mode.

• Winding to Ground in GSTg mode with other winding guarded. While carrying out
measurement for HV winding LV winding should be guarded whereas for LV winding HV
winding shall be guarded.

Effect of Temperature on Dissipation Factor Values:

The dissipation factor value for new equipment at room temperature should be about 0.004 and
maximum allowed as 0.005 (IEC-44(1), IEEE/C57 and IEC 44-2). During service, the value may go
upto 0.007. Value greater than 0.007 indicates deterioration of insulation leading to high dielectric
losses and it may lead to accelerated ageing and ultimately pre-mature failure of the insulation.

In case the test is carried out at higher temperatures, the test results need to be corrected. The
temperature correction factors are not easily available which can be applied. Moreover, temperature
correction factors depend on the moisture content in the paper. For a new Transformer, moisture
content may be of the order of 0.5% whereas for transformers in the age profile of 15-20 years, the
moisture content may be as high as 1.5% to 2.0%. The correction factors applicable for new
Transformers may not be applicable for aged Transformers. In view of this, the correction factors as
given in IEEE/C57.12.90 can not be applied for all equipments. To verify the applicability of these
corrections factors for Transformers in POWERGRID, a field testing programme was undertaken in
POWERGRID which has been discussed below.
CASE STUDY

Effect of Temperature on Dissipation Factor and Capacitance Values:

A project was undertaken in POWERGRID to evaluate the effect of oil/winding temperature on

dissipation factor and capacitance values. Four nos. 400/220/33 kV Transformers of different makes
having capacity 315 MVA and also 156MVA (one no.), were selected for this purpose. The active
power was regulated such that temperature could be raised upto 90-95°C. To achieve the temperature,
fans and pumps were switched off under watchful eyes.

After achieving the temperature, transformers were taken under shutdown and all jumpers were
disconnected from all the Bushings. The measurements were carried out at different temperatures after
an interval of every 5°C fall in temperature. Test kits of different makes were used and results were
found to be comparable with all the kits.

The Tan delta and Capacitance values for Transformer Windings and Bushings for all the makes of
transformers are given in Table-1 and Table-2 respectively.

Table-1 for Transformer Windings

Sr.No. Temp Make-I Make-II Make-III Make-IV

Tanδδ Cap. Tanδδ Cap. Cap. Tanδδ Tanδδ Cap.

(%) (pF) (%) (pF) (pF) (%) (%) (pF)
1. 25 0.23 1811 - - - - - -
2 30 0.23 1810 0.22 6016 0.18 5421 - -
3 35 0.23 1809 0.24 6016 0.18 5418 - -
4 40 0.23 1808 0.25 6016 0.18 5415 0.15 2127
5 45 0.24 1807 0.27 6016 0.18 5410 0.14 2125
6 50 0.27 1806 0.31 6016 0.18 5410 0.13 2123
7 55 0.35 1805 0.35 6016 0.18 5410 0.13 2122
8 60 0.425 1803 0.22 6016 0.21 5422 0.12 2111
9 65 0.485 1802 0.22 6016 0.25 5400 0.12 2117
10 70 - - 0.22 6016 - - 0.12 2111
11 75 - - - - 0.12 2108

Table-2 for Transformer Bushings

Sr.No. Temp. Make - Make-II Make-III

I
Tanδδ Cap. Tanδδ Cap. Tanδδ Cap.
(%) (pF) (%) (pF) (%) (pF)
1. 25 0.34 382 - - - -
2 30 0.35 382 - - 0.36 587
3 35 0.36 382 0.36 479 0.35 587
4 40 0.37 382 0.36 479 0.34 588
5 45 0.40 382 0.36 480 0.32 588
6 50 0.41 382 0.36 480 0.31 588
7 55 0.42 382 0.36 482 0.30 589
8 60 0.47 382 0.36 482 0.30 589
9 65 0.49 382 0.41 483 0.27 589
10 70 - - 0.42 483 0.27 590
From above test results (Table-1 and 2), following observations are made:

1. Dissipation factor values are affected by variation of temperature. In most of cases, Tan delta
value increases with increase in temperature. But rate of change is different for different make s
of the Transformers.
2. For windings of HVDC Transformer (Make-IV) and Bushings of Make-III, Tan delta value
decreases with increase in temperature which could possibly be due to use of different type of
Transformer oil (Nepthanic based oil used whereas other Transformers use Paraffinic based oil).
3. The Temperature correction factors (for correcting tan delta values at room temperature) are
different for different makes of Transformers. Hence Temperature Correction factors as given in
IEEE/C57.12.1982 can not be applied to these Transformers.
4. The value of Dissipation factor upto 40 °C is almost constant for all makes of the Transformers
after which it increases at higher rate.
5. Capacitance values, are almost constant and do not vary with change of temperature.

Outcome of the R&D Project:

1. Tan delta values are affected by temperature variation but are almost constant up to 40°C.
Hence, this test should be carried out when ambient/oil temperature is about 20 to 40°C and no
correction factors need to be applied for such measurements.
2. Temperature coefficient is about 0.008 to 0.013 (ref. formula 2.0(a) above) and hence insulation
is healthy.
3. Temperature Correction factors for different makes/types of transformers are different and
correction factors given in IEEE/C57 may not be applicable to all types of Equipments.
4. In POWERGRID, temperature correction factors are generally not being applied (in case test
results are within limits).

CONCLUSION

Monitoring of Transformer Insulation health using measurement of Tan Delta and Capacitance has
proved very useful in POWERGRID. As per the outcome of the R&D Project undertaken in
POWERGRID, temperature correction factors are different from these given in IEEE/C57.12.90. Even
for a same make of Transformer with different type, different correction factors need to be applied.

In POWERGRID, presently no correction factors are being applied for dissipation factor wherever test
results are within limits. In case of violations in the test results are observed then decision is taken
considering all relevant facts and other confirmatory/ supplementary tests are performed.

REFERENCES:

[1] G.CSEPES, I.HAMOS, J. SCHMIDT, A. BOGNAR, “ A DC Expert System (RVM) for

checking the re-furbishment efficiency of high voltage oil paper insulating system using
polarization spectrum analysis in range of Long time Constants”, CIGRE August,94(12-206).
[2] Mark F.Lachman, Doble engineering Company, “Application of Equivalent Circuit Parameters
to offline diagnostics of Power Transformer (Preview)”.
[3] Jagdish Singh and R. K. Tyagi, “Monitoring Health of EHV Transformer/Reactors Bushings
and Windings” Conference on Testing procedures for EHV equipment at Siliguri, India,
March,99.

ACKNOWLEDGEMENT
The authors are thankful to the POWERGRID Management for permitting them to prepare this paper
for Doble, India 2006. The views expressed in this paper are of the authors only and not necessarily
that of the POWERGRID Management.