Transformer Diagnostic

Measurements

• Transformer Overview
• Life Span and Operating Conditions
• Cigre Guide for Transformer Maintenance
• Oil Tests – Dissolved Gas Analysis, Degree of
Polymerisation (DP) & Furan Analysis
• Overview of CPC100 and Operating Software (Toolset)
• Electrical Measurements
• Dielectric Dissipation Factor – Tan Delta
• Dielectric Dissipation Factor - Moisture Content
• Partial Discharge
• Sweep Frequency Response
• Testing Current Transformers
• Summary

1
 Electrical Diagnostic
Measurements

Reference – Cigre “Guide for Transformer Maintenance” 2

 SFRA – What is it ?

Reliable and sensitive Method to evaluate the

mechanical and electrical integrity of
transformer‘s mechanical structure.

Detects Shorted turns,core faults,winding

deformation and displacement, faulty core
grounding, faulty screen connections and
damage due to transport or heavy through
faults.

3
 What is FRA?
Effective method for
detecting and
diagnosing mechanical
and electrical
deformations in the
active part of power
transformers

Leads
Core

FRA
Clamping
structure
IFRA SFRA
Windings
 SFRA – What is it ?

Sweep Frequency Response Analysis

evaulates the mechanical and geometrical
integrity of the core and windings of a
transformer by recording the electrical
transfer functions (Gain) characteristics of the
transformer over a wide frequency range.

Any change in the internal structure of

the transformer is detected by comparing a
recorded curve to a previous record of the
same or an identical transformer.

5
Model of Transformer Winding

6
 How does it Work ?

C = ε * A/d Force
C ~ 1/d

7
 How does it Work ?

C =Cε~*
A/d1/d

8
 Effects of Geometrical Changes

Geometrial Changes inside =

Change in Complex Electrical Network =
Change in the Freq. Response 'Fingerprint'

9
 SFRA Measurement Principle
3 3

2 Transformator 2

Spannung U1/U1 in V/V

Spannung U2/U1 in V/V

1 1

0 0
0 50 100 0 50 100
-1 -1

-2 -2

-3 -3
Zeit t in µs Zeit t in µs

Erregungssignal Antwortsignal
Input signal
(variable Frequenz) 3
Output signal
(sine wave of 2
1-|TF(f1)|

variable frequency)

Spannung U/U1 in V/V

1 ^
U1
^
U2
0
0 50 100
-1 ϕ(f1)/2ππf
^
U2
-2 |TF(f1)| =
^
U1

-3
arc(TF(f1)) = ϕ(f1)
Zeit t in µs

Betragsfunktion Phasenfunktion
|TFU2/U1(f)| arc(TFU2/U1(f))
Magnitude 1.0 200.0
Phase
150.0
dB
1.000e+002 1.000e+003
0.8 1.000e+004 1.000e+005
°
f/Hz 100.0
-10
150

arc(TFU2/U1(f)) in deg
|TFU2/U1(f)| in V/V|

0.6 50.0
-20
100
0.0
-30
0.4 50
1.000e+002
-50.0
1.000e+003 1.000e+004 1.000e+005
f/Hz
-40
0 -100.0
0.2
-50
-150.0
-50
-60
0.0 -200.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 -100 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
-70 f in MHz f in MHz
-150
-80

-90
NV
NV

10
 Measurement setup Clamps

Grounding
method:
Alu braids
or cooper
wires

CMC RMC A B C

Coaxial cables Power Transformer (complex RLC network)

Output Reference Measurement

Channel Channel
50Ω
Rref=50Ω Rme=50Ω
U1 U2 Injection signal
U Reference signal
Measurement signal

FRA Instrument
 SFRA Principle
x(t ) = X sin ωt

y (t ) = Y sin(ωt + φ )
1.000e+002 1.000e+003 1.000e+004 1.000e+005

Amplitude (k) -20

f/Hz

k = 20 log10 (U 2 / U1 )
-30

-40

-50

-60

-70

-80

-90

-100

dB

Phase
150

ϕ = tan (∠U 2 / ∠U1 )

−1
100

1.000e+002 1.000e+003 1.000e+004 1.000e+005

f/Hz
°

H0 H1 H0 H2 H0 H3
Frequency response of an inductance

Amplitude [dB]
-50

L=200 mH
-100 L=2 mH
L=20 H

-150 2 4 6
10 10 10
Frequency (Hz)
Rm
TF =
Rm + Z specimen

Z specimen = jX L = jωL = j 2πfL

 Frequency response of a capacitance
0

-50

Amplitude [dB]
-100
C=1uF
-150 C=20nF
C=1pF

-200 2 4 6
10 10 10
Frequency (Hz)
Rm
TF =
Rm + Z specimen
1 1
Z specimen = − jX C = − j =−j
ωC 2πfC
Parallel resonance

5
0
-5

Amplitude [dB]
-10 C=1nF
C=10nF
-15
C=50nF
-20

-25

-30 2 4 6
10 10 10
Frequency (Hz)
Series resonance

-20

-40

Amplitude [dB]
-60

-80
C=1nF
C=10nF
-100
C=50nF
-120

-140 2 4 6
10 10 10
Frequency (Hz)
Understanding the FRA response of a
transformer

Cs1 C12 Cs2

R1 L1 L2
R2
Lm
Cg1 Rm Cg2
N1 N2

> Page 17
Classification of the FRA response in
sub-bands
Typical pattern of windings of high voltage YN windings (outer phase)
A B C D E F
0
LF1 LF2 MF HF1
-20
Magnitude (dB)

-40

HF2
-60

-80

1 2 3 4 5 6
10 10 10 10 10 10
200

100
Phase (°)

-100

-200
1 2 3 4 5 6
10 10 10 10 10 10
Frequency (Hz)
 When to perform an FRA test?
After transportation On-site
Routine test
Manufacturer
After through faults
After earthquakes
Trafo
After alarm messages and protection
relays operations
After changes in monitored variables
(gasses)
Trafo After detecting abnormalities with
Quality Assurance
traditional diagnostic tests (winding
Before and after the short- resistance, etc.)
circuit test
After major maintenance in OLTC,
Before and after the Impulse
bushings, windings.
test
Transformer modelling for analysis of
Before transportation
transients
 Types of FRA Measurements
End-to-end (open circuit)

The delta of the

tertiary shall be
closed
 Types of FRA Measurements
End-to-end (open circuit)

1.000e+002 1.000e+003 1.000e+004 1.000e+005

f/Hz

-20

-30

-40

-50

-60

-70

dB

Phase C (HV) OPEN

Phase B (HV) OPEN
Phase A (HV) OPEN
 Types of FRA Measurements
End-to-end (short-circuit)
 Types of FRA Measurements
• When is it recommended to perform the FRA short-circuit
test?

– For confirming deviations found in the end-to-end

(open circuit) test
– For confirming the diagnosis provided by the short-
circuit impedance (leakage reactace) test
– When its fingerprint is available
– If the measurement is to be compared to other ones
measured at different a different output voltage
– For failure investigation (i.e. if DGA indicates a
problem)
 Types of FRA Measurements
End-to-end (short-circuit) against end-to-end (open circuit)
1.000e+002 1.000e+003 1.000e+004 1.000e+005
f/Hz
-10

-20

-30

-40

-50

-60

-70

dB

Phase C (HV) OPEN Phase A (HV) SHORT

Phase B (HV) OPEN Phase A (HV) OPEN
Phase B (HV) SHORT Phase C (HV) SHORT
Types of FRA Measurements
Inter-winding capacitive test
 Types of FRA Measurements
Inter-winding capacitive test
1.000e+002 1.000e+003 1.000e+004 1.000e+005
f/Hz
-30

B arriers
C12 -40
G12

-50

Core

-60

-70

Spacers Oil

-80

dB

A a1 B b1 C c1
 Types of FRA Measurements

• When is it recommended to perform the FRA

capacitive-inter-winding test?

– As fingerprint to be used in the future for

investigation of radial deformations

– For confirming the diagnosis provided by other

FRA measurements
Types of FRA Measurements
Inter-winding inductive test
 Types of FRA Measurements
Inter-winding inductive test

1.000e+002 1.000e+003 1.000e+004 1.000e+005

f/Hz
-20

-25

-30

-35

-40

-45

-50

-55

dB

H1 x1 (#28) (grounded)**
H2 x2 (#29) (grounded)**
H3 x3 (#30) (grounded)* (2)*
 What can be concluded from this ?

FRA-devices need a sufficient dynamic range (at

least down to -90 dB).

Impulse measurements are not able to "see" the

lower frequency range, which is expressive for
several types of failure.

30
 Basic Functionality Requirements

Ability to compare multiple tests

Import functionality for all instrument

manufacturers test files

Increased dynamic range minimum upto -90 dB.

31
Transformer Mechanical defects/Faults

32
 When to do SFRA test ?

When build completed in factory

After installation on site

If relocated to another site

When a heavy through fault has occurred.

When a weakness of the winding mechanical

pressure system is known (Old and Suspected unit).

33
 Low- frequency Response

Low frequency variation

 Core fault

 Shorted turns fault

 Open circuit or high impedance winding fault

34
 Low- frequency Response

SFRA plot of a winding with all other winding left

open-circuited Core parameter is dominant in the
frequency region up to 1 kHz.

35
 Short-circuit SFRA plot of HV winding for
two sister units

Winding inductance and resistance is dominant component up to 2kHZ

Red- With oil, Green- Without oil

36
Open-circuit SFRA plot of HV winding for
two sister units

37
Case study 1 : Confirmation of LV winding SC fault

Transformer has tripped on Overcurrent protection

Rating:
• 132 kV/11 kV, 10 MVA, 3-phase Transformer
• Winding configuration – HV(Star) / LV(Star)
• Year of manufacturing - 1979

SFRA was conducted first time after the over

current tripping

Three phase comparison of SFRA plot was

performed

38
Case study 1 : Confirmation of LV winding SC fault

3-phase HV winding Open circuit plot- After fault

39
Case study 1 : Confirmation of LV winding SC fault

Data analysis is performed on three phase

comparison basis.

SFRA plot of HV B- phase (middle phase) is different

than the Outer phases.

In case of B-Phase winding the first resonance

around 300 Hz is absent.

40
Case study 1 : Confirmation of LV winding SC fault

3-phase LV winding open circuit plot- After fault

41
Case study 1 : Confirmation of LV winding SC fault

3-phase HV winding Short circuit plot- After fault

42
Case study 1 : Confirmation of LV winding SC fault

In low frequency the difference in dB of B-phase winding

is around 0.4 dB

43
Case study 1 : Confirmation of LV winding SC fault

SFRA is able to identify that turn to turn fault is in

LV winding.

Turn to turn fault can be identified by other method

like
 No load excitation current
 Turns ratio
 Magnetic balance test

But in which winding HV or LV winding is a question

mark.

SFRA is the test which can identify the winding and

phase both, of the shorted turn fault.

44
Case study 2 : Confirmation of mutli winding fault

Transformer was tested for SFRA after it trpped on

Differential protection.

Rating:
• 132 kV/11 kV, 16 MVA, 3-phase Transformer
• Winding configuration – HV(Star) / LV(Star)
• Year of manufacturing - 1976

SFRA was conducted second time on the

transformer after the incidental fault and tripping.
First base data of SFRA was available in the past
record.

45
Case study 2 : Confirmation of mutli winding fault

After Fault
Befor
e
Fault

Before Fault After Fault

HV winding SFRA plot

46
Case study 2 : Confirmation of mutli winding fault

HV winding SFRA data analysis concluded that

transformer has developed turn to turn fault in
H3-H0 phase.

It was confirmed in conventional ratio test also,

where the ratio of H3-H0 was different than other
two phases.

47
Case study 2 : Confirmation of mutli winding fault

Before Fault After Fault

LV Winding SFRA Plot

48
 Case study 3 : Confirmation of LV winding
Short-circuit fault

25/20 MVA,132/33 KV Transformer indicated turn to turn

fault in LV winding
5. 000e+001 1. 000e+002 5. 000e+002 1. 000e+003 5. 000e+003 1. 000e+004 5. 000e+004 1. 000e+005 5. 000e+005
f/Hz

-10

-20

-30

-40

-50

-60

-70

-80

dB

100

5. 000e+001 1. 000e+002 5. 000e+002 1. 000e+003 5. 000e+003 1. 000e+004 5. 000e+004 1. 000e+005 5. 000e+005

f/Hz

-50

-100

-150

x0 x1 x0 x2 x0 x3

49
 Mid-frequency response

1 kHz to 1 MHz represent the main winding

structure of transformer.

Overall axial and Radial deformation of the HV, LV

winding due to high short circuit system fault is
detected well.

50
Mid-frequency response
(Tap 17)

51
Mid-frequency response
(Tap 1)

52
 Mid-frequency response
Capacitance and Power factor test data

53
Mid-frequency response

54
 High-frequency response

In high frequencies region between 1MHz to 2 MHz,

transformer winding behaves as a purely capacitive
element.

55
 High-frequency response

Auto-transformer, 400/220kV, Very low IR between

core and tank after movement of transformer from
factory to site.

Comparison of factory and site SFRA plot indicates

significant shift was found in frequency range of
1 MHz to 2 MHz for all 3-phase common and series
winding.

56
High-frequency response

57
 FRA Measurements
Step 1. Create a new transformer
 FRA Measurements
Step 2. Enter nameplate data
 FRA Measurements
Step 3. Create a new test
 FRA Measurements

End to-end-open circuit + End to end short-circuit +

IEEE Templates capacitive inter-winding + inductive inter-winding

OMICRON End-to-end open circuit + End to end short-

Extended circuit + capacitive inter-winding

End-to-end open circuit + capacitive inter-

OMICRON
winding
OMICRON
(FRAnalyzer 1.1) End-to-end open circuit
 FRA Measurements
User-defined templates
 FRA Measurements
Step 4. Check the settings and modify them, if necessary
 FRA Measurements

if

BW

if: intermediate
Sampling points
frequency
BW: bandwidth
 FRA Measurements

> 16 March, 2012 This is the sweep profile

defined as default.
Provides a very good
signal to noise ratio and
short measurement times

OMICRON Sweep profile: OMICRON 2.1

 -115 dB
-90 dB
-85 dB

-140 dB
-105 dB
-100 dB
-75 dB
-55 dB
-65 dB
-45 dB
-25 dB

-35 dB
-30 dB
20 Hz
RBW=30 HZ
30 Hz
RBW=30 RBW=10 Hz
RBW=30 HZ 45 Hz
HZ RBW=3 Hz
RBW=10 Hz
65 Hz
100 Hz

RBW=30 HZ

3 kHz

1 Hz 3 Hz 10 Hz 30 HZ 100 HZ 300 HZ

3 Hz 10 Hz 30 HZ 100 HZ 300 HZ
FRA Measurements

2 MHz
 FRA Measurements

20 dB

20 dB
-35 dB
-45 dB
0 dB

Attenuation Type Value New Value

< -45 dB Attenuator 20 dB 0 dB
> -35 dB Attenuator 0 dB 20 dB
 FRA Measurements

50 100 200 500 1000 50 100 200 500 1000

f/Hz f/Hz
-65 -65

-70 -70

-75 -75

-80 -80

-85 -85

-90 -90

-95 -95

dB dB

OMICRON 2.0 (adapt automatically) OMICRON 2.0 (adapt automatically)

OMICRON (FRAnalyzer1.1) OMICRON (FRAnalyzer 2.1)
 FRA Measurements
Rules for a transformer specific frequency range for extra large, large
and medium sized power transformers and distribution transformers

residual flux FRA evaluation range set-up, noise, accuracy extra large PT
> 500 MVA; > 400 kV

large PT
> 200 MVA; > 200 kV

medium PT
> 50 MVA; > 100 kV

large DT
> 5 MVA; > 20 kV

DT
< 5 MVA; < 20 kV

100 1k 10 k 100 k 1M 10 M 100 M

frequency f [Hz]
 FRA Measurements
Step 6. Make connections of the measurement setup

Case A: Single- terminal cabling Case B: Multi- terminal cabling

The grounding wires shall be as short as possible and shall not be

wrapped (DL/T 911, 2004)
 FRA Measurements
Step 6. Make connections of the measurement setup

Wrong: grounded braids are missing

 FRA Measurements
Step 6. Make connections of the measurement setup

Make sure that the grunding braids do not have contact to the
terminals

1.000e+002 1.000e+003 1.000e+004 1.000e+005

f/Hz

This is wrong. If
-10

-20

this happens, -30

you will get -40

useless results -50

-60

like this: -70

-80

-90

dB

E2E_A Ph
 FRA Measurements
Step 6. Make connections of the measurement setup

Remove the pollution layer in the

connection points of the
grounding clamps (use the file)
 FRA Measurements
Step 7. Connect the FRAnalyzer to the laptop (USB)

Do not forget to check the state of

the battery
 FRA Measurements
Step 8. Document the conditions of the test

Tap position (OLTC and DETC)

Change-over selector position (if applicable)
Oil temperature
Addtional conditions (core grounding)

Add
comments
 FRA Measurements
Step 8. Document the conditions of the test

The OLTC shall be at the tap position that

provides the highest number or turns

If there are not time constrictions, it is

recommended to perform the
measurements at the max. min. and middle
tap positions for making easier the
localisation of deformations
 FRA Measurements
Step 9. Start the measurements
 FRA Measurements
Step 10. Use signal processing tools for removing noise if necessary

30 50 70 100 200 300 500

-10 f/Hz

-20

-30

-40

-50

-60

-70

dB

°
30 50 70 100 200 300 500
f/Hz

-50
-100
-150

N W Before noise removal

N W After noise removal
 FRA Measurements
Step 11. Assess results

Assessment
Cursors table

1.000e+002 1.000e+003 1.000e+004 1.000e+005

f/Hz
-10

-20

-30

-40

-50

-60

-70

dB

°
f/Hz
1.000e+002 1.000e+003 1.000e+004 1.000e+005

-50
-100
-150

H0 H1 (6) H0 H1 (9)
 FRA Measurements
Step 12. Generate a report
Frequency Response Demonstration

16 March 2012