Sweep FrequencyResponse Analysis

FRAX

Transformer Diagnostics

Diagnostics is about collecting reliable information to make the correct decision

Making the correct decisions saves money

Oil analysis

SFRA

FDS Winding Resistance

SFRA Basics

SFRA history (1)

1960: Low Voltage Impulse Method. First proposed by W. Lech & L. Tyminski in Poland for detecting transformer winding deformation.

1966: Results Published; “Detecting Transformer Winding Damage - The Low Voltage Impulse Method”, Lech & Tyminsk, The Electric Review, ERA, UK

1976: “Frequency Domain Analysis of Responses From L.V.I. Testing of Power Transformers”, A.G. Richenbacher, 43rd Doble Conference

1978: “Transformer Diagnostic Testing by Frequency Response Analysis”, E.P. Dick & C.C. Erven, Ontario Hydro, IEEE Transactions of Power Delivery.

SFRA history (2)

1978: FRA test developed at Ontario Hydro, Canada 1980’s: Further research carried out by Central Electricity

Generating Board in UK 1988 - 1990’s : Proving trials by European utilities, the

technology cascades internationally via CIGRE, EuroDoble and many other conferences and technical meetings

2004: First SFRA standard, ”Frequency Response Analysis on Winding Deformation of Power Transformers”, DL/T 911-2004, is published by The Electric Power Industry Standard of People’s Republic of China

2008: CIGRE report 342, ”Mechanical-Condition Assessment of Transformer Windings Using Frequency Response Analysis (FRA)” is published

Transformer mechanics basics

A transformer is rated to withstand certain mechanical forces.

However, these forces can easily be exceeded • during transportation • short circuits close to the transformer

Transformers mechanical strength weakens as the transformer ages

• Less capability to withstand mechanicalstress

• Greater risk for mechanical problems• Greater risk for insulation problems

Detecting Faults with SFRA

Core movements Faulty core grounds Winding deformations Winding displacements Partial winding collapse Hoop buckling Broken clamping structures Shorted turns and open windings Etc

SFRA = Fingerprinting

SFRA testing basics

Off-line test Transformer is a complex RLC filter circuit Changes in this circuit can be detected and

plotted as a response curve when test signals at different frequencies are applied over a winding

Changes can be compared over time, between test objects or within test objects

The method is unique in its ability to detect core problems, mechanical winding problems and other electrical faults in one test

Test results – always comparisons

Different problems can be seen in different parts of the curve

Software analysis makes it easy to detect deviations

Low frequencies• Core problems and

shorted/open windings

Medium frequencies• Winding deformations

High frequencies• Tap connections and other

winding connection problems

Core + windings

Winding deformations

Taps and connections

Comparative testsTransformer A

Transformer A Transformer B

Time based

Type based

Design based

Comparisons

Time Based (Tests performed on the same transformer over time)

The most reliable test Deviations between curves are easy to detect

Type Based (Tests performed on transformer of same design)

Requires knowledge about test object/versions Small deviations are not necessarily indicating a problem

Design based (Tests performed on winding legs and bushings of identical design)

Requires knowledge about test object/versions Small deviations are not necessarily indicating a problem

Measurement philosophy

New measurement = Reference measurement

Back in Service

New measurement ≠ Reference measurement

Further Diagnostics Required

Reference measurements

When transformer is new• Capture reference data at commissioning of

new transformers

When transformer is in known good condition

• Capture reference data at a scheduled routine test (no issues found)

Save for future referenceStart Your Reference Measurements ASAP!

SFRA measurements – When?

Manufacturing test Commissioning test Transport test Incident test - after incidents where you

suspect electromechanical changes• After transport• Short-circuit faults

Catastrophic events• Earth quakes• Hurricanes, tornadoes

Trigger based test – transformer alarms• Vibration• DGA• High temperature

FRA Methods

Impulse FRA vs. SweepFRA

Impulse FRA Injects a pulse signal and

measure response Convert Time Domain to

Frequency Domain using Fast Fourier Transform (FFT) algorithm

Low resolution in lower frequencies

SFRA Injects a single frequency signal Measures response at the same

frequency No conversion High resoultion at all frequencies

Impulse FRA

Comparing Impulse & SweepFRA

SFRA (Sweep frequency response analysis) provides good detail data in all frequencies

Black = Imported Impulse measurement (Time domain converted to Frequency Domain)

Red = SFRA Measurement

Deviations Low Frequency = Method Deviation High Frequency = Cable practice

Zoom View of impulse vs. SFRA

Impulse instrument sample rate limts frequency resolution to 2kHz.

SFRA Measurements

SFRA test setup

FRAX measurement circuitry

Considerations when performing SFRA Tests

or How do I maximize my investment in time and money when

performing SFRA measurements?

Test results – always comparisons

Reproducibility is of utmost importance!

Core NOT grounded

Core grounded

Example of reproducible results

105 MVA, Single phase Generator Step-up (GSU) transformer

SFRA measurements with FRAX 101 before and after a severe short-circuit in the generator

• Two different test units• Tests performed by two different persons• Test performed at different dates

Before (2007-05-23) and after fault (2007-08-29)

LV winding

HV winding

105 MVA, Single phase GSU

Measurements “before” and “after” were virtually identical

Very good correlation between reference and “after fault”

Conclusion: No indication of mechanical changes in the transformer Transformer can safely be put back in service

Potential compromising factors

Connection qualityShield grounding practice Instrument dynamic range/internal noise

floorUnderstanding core property influence in

lower frequencies in “open” - circuit SFRA measurements

Bad connection

Bad connection can affect the curve at higher frequencies

Good connection

After proper connections were made

FRAX C-Clamp

C-Clamp ensures good contact quality

Penetrates non conductive layers

Solid connection to round or flat busbars

Provides strain relief for cable Separate connector for single

or multible ground braids

Proper ground connection ensures repeatability at high frequencies

Good grounding practice;use shortest braid from cable

shield to bushing flange.

Poor grounding practice

Shield grounding influence

C. Homagk et al, ”Circuit design for reproducible on-site measurements of transfer function on large power transformers using the SFRA method”, ISH2007

FRAX cable set and grounding

Always the same ground-loop inductance on a given bushing

Instrument performance

Small transformers have higher attenuation at first resonance

Inherent instrument noise is often the main limiting source, not necessarily substation static

Test your instruments noise floor by running a sweep with “open cables” (Clamps not connected to transformer)

Internal noise level – ”Noise floor”

”Open”/noise floor measurementsRed = Other brandGreen = FRAX 101

Example of noise floor problem

H1 – H2 (open & short) measurementsBlack = Other brand

Red = FRAX 101

Influence of core

Try to minimize the effect, however, some differences are still to be expected and must be accepted.

Preferably: perform SFRA measurements prior to winding

resistance measurements (or demagnetize the core prior to SFRA measurements)

Use same measurement voltage in all SFRA measurements

Run winding resistance test after SFRA!

After demagnetization

After winding resistance test

40

Effect of applied measurement voltage

2.8 VOmicron

10 VFRAX, Doble and others

41

FRAX has adjustable output voltage!

Omicron (2.8 V)

FRAX, 2.8 V

Field Verification Unit

Field verification unit with known frequency response is recommended in CIGRE and other standards to verify instrument and cables before starting the test

ABB Transformer Diagnostics

Summary

The basis of SFRA measurements is comparison and reproducibility is of utmost importance

To ensure high repeatability the following is important• Use of a high quality, high accuracy instrument with inputs and

output impedance matched to the coaxial cables (e.g. 50 Ohm)• Use same applied voltage in all SFRA measurements• Make sure to get good connection and connect the shields of

coaxial cables to flange of bushing using shortest braid technique.• Make good documentation, e.g. make photographs of

connections.

45

FRAX The Features And Benefits

FRAX 101 – Frequency Response Analyzer

FRAX 101 – Frequency Response Analyzer

BluetoothOn FRAX101

USB PortOn all models

Power Input11-16VDC

Rugged Extruded Aluminum Case

Active Probe Connector on FRAX101

All Connectors Panel Mounted

GeneratorReferenceMeasure Connectors

Not only the smallest, but also the most feature rich and accurate SFRA

unit in the world!

SFRA test setup

Industrial grade class 1 Bluetooth (100m)

USB for redundancy

Optional Internal BatteryOver 8h effective run time

Easy to connect shortest braid cables

Search Database Feature

Data files stored in XML format

Index function stores all relevant data in a small database

Search function can list and sort files in different locations

Import formats

Less points where it takes time to test and where high frequency resolution is not needed

More points wherehigher frequency

resolution is useful

Traditional test about 2 min

vs.FRAX fast test< 40 seconds

Fast testing

Decision support

Unlimited analysis

Unlimited graph control Lots of available graphs Ability to create custom

calculation models using anymathematic formula and themeasured data from all channels

Turn on and off as needed Compare real data with

calculated model data

Mathematical modeling

FRAX 101 transport case

Cable compartment

Padded product bay

Rugged case 14kg/31lbs incl. Cables

FRAX-150

As FRAX-101 except: Internal PC/stand-alone No internal battery option No Bluetooth

FRAX-99

As FRAX 101 except: No internal battery option No Bluetooth Dynamic range > 115 dB Fixed output voltage 9 m cable set No active probes

Active Probes, extending the application

Active Impedance Probe• Measures Transfer functions between

two grounded connections• E.g. between winding and tank or

bushing flange

Active Voltage Probe• Measures objects with higher input

impedance than 50Ω• Allows for longer cables

FRAX product summary

Light weight Rugged Battery operated Wireless communication Accuracy & Dynamic Range/Noise floor Cable Practice Easy-to-use software Export & Import of Data Complies with all SFRA standards and recommend Only unit that is compatible with all other SFRA

instruments

Sweep Frequency Response Analysis

Application Examples

Time Based Comparison - Example

1-phase generator transformer, 400 kVSFRA measurements before and after

scheduled maintenanceTransformer supposed to be in good condition

and ready to be put in service…

Time Based Comparison - Example

”Obvious distorsion” as by DL/T911-2004 standard (missing core ground)

Time Based Comparison – After repair

”Normal” as by DL/T911-2004 standard (core grounding fixed)

Type Based Comparisons (twin-units)

Some parameters for identifying twin-units: Manufacturer Factory of production Original customer/technical specifications No refurbishments or repair Same year of production or +/-1 year for large units Re-order not later than 5 years after reference order Unit is part of a series order (follow-up of ID numbers) For multi-unit projects with new design: “reference” transformer should

preferably not be one of the first units produced

Type Based Comparison - Example

Two 33/11 kV, 10 MVA, manufactured 1977 Put out of service for maintenance/repair or scrapping “Identical” except for slightly different tap-settings

(could not be fixed at site due to missing tool…) SFRA testing and comparing the two transformers

came out OK indicating that there are no electromechanical problems in the transformers (identical problems highly unlikely…)

Type Based Comparison – LV windings

33 kV, 3-phase Ynyn transformer (30 years old) ”Normal” as by DL/T911-2004 standard

-15

-20

-25

-30

-35

-40

-45

-50

Ma

gn

itu

de

(d

B)

1 k 10 k 100 k 1 MFrequency (Hz)

[X1-X0 (open)] [X3-X0 (open)] [X1-X0 (open)] [X3-X0 (2)]

Type Based Comparison – IW tests

33 kV, 3-phase Ynyn transformer (30 years old) ”Normal” as by DL/T911-2004 standard

-20

-25

-30

-35

-40

-45

-50

-55

Ma

gn

itu

de

(d

B)

1 k 10 k 100 k 1 MFrequency (Hz)

[H1-X1 (IW)] [H3-X3 (IW)] [H1-X1 (IW)] [H3-X3 (IW)]

Design Based Comparisons

Power transformers are frequently designed in multi-limb assembly. This kind of design can lead to symmetric electrical circuits

Mechanical defects in transformer windings usually generate non-symmetric displacements

Comparing FRA results of separately tested limbs can be an appropriate method for mechanical condition assessment

Pending transformer type and size, the frequency range for design-based comparisons is typically limited to about 1 MHz

Design Based Comparison - Example

132 kV, 60 MVA transformer, manufactured 2006

New transformer never in serviceNo reference FRA measurements from factorySFRA testing, comparing symmetrical phases

came out OKThe results can be used as fingerprints for

future diagnostic tests

Designed Based Comparison – HV windings

132 kV, 3-phase YNd1 transformer (new) ”Normal” as by DL/T911-2004 standard H1-H0 vs H3-H0

-20

-25

-30

-35

-40

-45

-50

-55

-60

-65

Ma

gn

itu

de

(d

B)

1 k 10 k 100 k 1 MFrequency (Hz)

[H1-H0 (open)] [H3-H0 (open)]

Designed Based Comparison – LV windings

132 kV, 3-phase YNd1 transformer (new) ”Normal” as by DL/T911-2004 standard X2-X1 vs X1-X3

-10

-20

-30

-40

-50

Ma

gn

itu

de

(d

B)

1 k 10 k 100 k 1 MFrequency (Hz)

[X2-X1 (open)] [X1-X3 (new test) (open)]

Designed Based Comparison – IW test

132 kV, 3-phase YNd1 transformer (new) ”Normal” as by DL/T911-2004 standard H1-X1 vs H3-X3

-20

-25

-30

-35

-40

-45

-50

-55

-60

Ma

gn

itu

de

(d

B)

1 k 10 k 100 k 1 MFrequency (Hz)

[H1-X1 (IW)] [H3-X3 (IW)]

Design Based Comparison – After Suspected Fault

Power transformer, 25MVA, 55/23kV, manufactured 1985

By mistake, the transformer was energized with grounded low voltage side

After this the transformer was energized again resulting in tripped CB (Transformer protection worked!)

Decision was taken to do diagnostic test

Design Based Comparison– After Suspected Fault

HV-0, LV open A and C phase OK, large deviation on B-phase (shorted turn?)

-80

-70

-60

-50

-40

-30

-20

-10

0

10 100 1000 10000 100000 1000000

Frequency (Hz)

Res

po

nse

(d

Bs)

Design Based Comparison– After Suspected Fault

HV-0 (LV shorted) A and C phase OK, deviation on B-phase

-60

-50

-40

-30

-20

-10

0

10 100 1000 10000 100000 1000000

Frequency (Hz)

Res

po

nse

(d

Bs)

And how did the mid-leg look like…?

Insulation cylinder

Core limb

LV winding

Sweep Frequency Response Analysis

Standards

SFRA Standards and Recommendations

Frequency Response Analysis on Winding Deformation of Power Transformers, DL/T 911-2004, The Electric Power Industry Standard of People’s Republic of China

Mechanical-Condition Assessment of Transformer Windings Using Frequency Response Analysis (FRA), CIGRE report 342, 2008

IEEE PC57.149™/D4 Draft Trial-Use Guide for the Application and Interpretation of Frequency Response Analysis for Oil Immersed Transformers, 2007 (Draft)

Internal standards by transformer manufacturers, e.g. ABB FRA Standard v.5

SFRA Standards - Summary

Frequency range Dynamic range Accuracy Signal cable grounding

EPIS PRC DL/T 911 1 kHz - 1 MHz -100 to +20 dB ± 1 dB @ -80 dB

Wire, shortest length to transformer core

grounding not stated

Defined correlation in three frequency

bands

CIGRE WG A2.26 min 50 Hz - 2

MHz

-100 to +20 dB measurement

range ± 1 dB @ -100 dBGrounded at both ends

using shortest braidTest circuit with a

known FRA response

Not stated (DL/T 911 recognized as one

standard)

IEEE PC57.149/D4

"Calibrated to an acceptable

standard"Grounded at both ends

using shortest braid

Standard test object with a known FRA

response

Plot inspection, difference plots,

general correlation techniques

ABB FRA Technical Standard v.5 10 Hz - 2 MHz

Better than -100 to +40 dB ± 1 dB @ -100 dB

Grounded at both ends using shortest

wire/braid

"Condition control of FRA device, including

coaxial cables, is strongly

recommended"Plot inspection, difference plots.

StandardInstrumentation

"Suffi cient dynamic range, over the frequency range in order to accommodate all transformer

test objects" (-120 dB…?)

Self-test Analysis

Instrumentation

Frequency range – All major brands are OK Dynamic range

First transformer circuit resonance gives typically a -90 dB response. Smaller transformers may have a first response at -100 dB or lower

Note that CIGRE recommends measurement range down to -100 dB. This implies a “dynamic range”/noise floor at about -120 dB.

Accuracy ± 1 dB at -100 dB fulfills all standards.

All FRAX instruments fulfills all standards for dynamic range and accuracy!

Cable grounding practice

The “shortest wire/braid”-practice is now generally accepted All European equipment manufacturers have adapted to

this practice

Recommended grounding practice (CIGRE) Bad grounding practice (CIGRE)

Instrumentation verification

Verification of instrument including cables• Measurement with “open” cables (at clamp) should give a response

close to the noise floor of the instrument (at lower frequencies, pending cable length)

• Measurement with “shorted” cables (at clamp) should give close to 0 dB response (pending cable length)

• External test device with known response (FTB-101 included in FRAX standard kit)

Calibration at recommended interval• FRAX; Minimum every 3 years, calibration set and SW available

FRAX Field Verification Unit, FTB-101

FRAX - Benchmarking

Measurement voltage and internal noise

Measurement voltage and internal noise/dynamic range for common SFRA test sets

Highest dynamic range, -130 dB

Internal noise (dynamic range)

Internal noise (open) measurementsGreen – FRAX-101

Red – Other SFRA 1Blue – Other SFRA 2

Measurement range

Internal noise (open) measurementsGreen – FRAX-101

Blue – Other SFRA 1

-100 dB measurement(CIGRE standard)Black – FRAX-101

Red – Other SFRA 1

Dynamic Range – Comparison (1)

Neutral to capacitive tapRed – FRAX-101

Black – Other SFRA 1

End-to-end openGreen – FRAX-101

Blue – Other SFRA 1

Dynamic Range – Comparison (2)

H1 – H2 (open) measurementsRed – FRAX-101

Grey – Other SFRA

Dynamic Range – Measurements at first resonance

Blue – FRAXPurple – Other SFRA 3Red – Other SFRA 1

Jiri Velek, “CEPS SFRA Market Research”, October 2006

FRAX - Compatibility

93

FRAX vs Doble (1)

5 MVA, Dyn, H2-H3 measurement

Blue – DobleOrange – Frax

94

FRAX vs Doble (2)

YNd, H1-H0 measurement

Blue – DobleOrange – Frax

95

FRAX vs Tettex and Doble

H1-H0 (short) measurement

Blue – FRAXPurple – TettexRed – Doble(Doble high frequency deviation due to different grounding practice)

Jiri Velek, “CEPS SFRA Market Research”, October 2006

96

Frax-101, 2.8 vs 10 V meas voltage

2.8 V

10 V

97

Frax (2.8V) vs FRAnalyzer

Omicron (2.8 V)

PAX, 2.8 V

Summary - conclusions

SFRA is an established methodology for detecting electromechanical changes in power transformers

Collecting reference curves on all mission critical transformers is an investment!

Ensure repeatability by selecting good instruments and using standardized measurement practices

Select FRAX from Pax Diagnostics, the ultimate Frequency Response Analyzer!