1. Frequency response analysis (FRA) is a diagnostic tool used to check for changes in the internal geometry of transformer windings and cores. 2. FRA involves applying a low voltage sinusoidal input to the transformer and measuring the electrical response to detect displacements, deformations, or other mechanical issues. 3. Many factors can influence FRA results including tap position, insulation medium, temperature, magnetization, measurement setup and direction, and physical changes to windings or components over time. Proper interpretation requires comparison to baseline measurements.

1. FREQUENCY RESPONSE
ANALYSIS (FRA)
By Raveendra Nath Macherla, Senior Electrical Engineer
(Some Images Courtesy: Megger)

2. Standard
•IEC 60076-18

3. What is FRA test?
FRA is a diagnostic tool to
check for changes in internal
geometry of the active part of
transformer for displacements
or deformations.

4. When to
perform
frequency
response?
On all new transformers for fingerprinting purposes (FAT and
Pre-commissioning)
As part of routine electrical tests
After relocation
To assess damages following a through fault or long duration
short circuit
To assess damage following a tap-changer fault (or) After
repair to winding / tap changer
To assess damage during transportation,
To assess damage following a seismic event

5. Some of the causes affect frequency response
of Transformer
Winding deformations and displacements
Shorted turns and open windings
Loose or broken clamping structures
Core displacements
Faulty core grounds
Hoop Buckling (winding compressive failure)
Partial Winding collapse

6. Why FRA shall be done immediately
after Transformer arrival
(transportation)?
• Frequency response analysis is commonly used a tool to the detection
and evaluation of damage to a transformer during transportation. The
method can provide information about the mechanical condition of the
core, the windings, and the clamping structures with one set of
measurements. All these parts are susceptible to transportation damage.
Core to frame and tank insulation resistance should also be checked for
any transportation damage detection.

7. Does electrical tests are not sufficient? Why
FRA required?
Traditional electrical tests such as winding capacitance, excitation current and
leakage reactance measurements have proven to be not particularly sensitive
to detect winding movement.
 Winding capacitance measurement can detect winding movement only if
reference data is available or if measurements can be made on each phase.
 The excitation current can detect turn to turn failure as a result of winding
movement. However, if a turn-to-turn failure is absent, winding movement can
remain undetected.
 Leakage reactance , the discrepancies from the nameplate value of 3% can be
a reason for concern. This makes accurate assessments of the mechanical
integrity of the transformer very difficult.

8. PRINCIPLE OF FRA
Principle of FRA is the measurement of steady electrical response against a
steady sinusoidal input given to the test object.

9. How to perform FRA?
To make a frequency response measurement, a low voltage
is applied to one terminal of the test object with respect to
the tank. The voltage measured at this input terminal is
used as reference signal and a second voltage signal (the
response signal) is measured at a second terminal with
reference to the tank. The frequency response is the scalar
ratio between the response signal (Vout) and the reference
voltage (Vin) (presented in dB) as a function of the
frequency. The phase of the frequency response is the
phase difference between Vin and Vout (presented in
degrees).

10. Condition of Test object
• The test object shall be fully assembled as for service complete with all bushings (but coolers and related auxiliaries do not need to be
assembled.) and transformers shall be filled with liquid or gas of the same type as that which is to be used in service.
• The transformer under test must be completely disconnected from the network
• Neutral must be removed to ensure that the transformer under test is floating
• If internal CTs secondary terminals shall be shorted and earthed, if not connected to any load.
• The core and frame to tank connections shall be complete and the tank shall be connected to earth.
• If stabilizing winding (additional delta winding), where two terminals brought out shall be shorted but earth shall be disconnected.
• If DC testing was performed, the core must be demagnetized before any SFRA measurements
• Measurement must be made at the highest tap position ( full winding)
• It is not recommended to perform SFRA while the test object temperature is changing rapidly for example immediately following oil
treatment.

11. Checks prior to
measurement
The continuity of the test leads shall be verified.
If specified, zero-check measurement shall be
carried out as an additional measurement.
On completion of the standard measurements,
the first measurement shall be repeated and
recorded for repeatability check.
Instrument performance check (either by
calibration or by measuring the response of a
known test object)

12. Data to be recorded for each measurement
• Transformer serial number, Make, Rating, Tag No.
• Date (DD-MM-YYYY) and Time
• Test Instrument Make, Model, serial No.
• Peak voltage used for the measurement
• Reference terminal, Response terminal
• Terminals connected together
• Terminals earthed
• OLTC Tap position
• Previous OLTC Tap position, the tap position from which the tap-changer was moved to reach the tap position used during the measurement.
• Insulation medium filled or not
• Each data file shall be named as “Transformer tag no_reference terminal_response terminal_tap position_date_time”
• Any additional data shall be filled as per requirement

13. Measurement configuration
• The standard measurements shall be end-to-end measurements of each
phase of each winding, with the phases and windings separated as far as
possible and with all other terminals left floating.
• Measurement must be made at the highest tap position ( full winding)
• For two-winding three-phase transformers, the standard measurements
shall be one measurement of each phase of each winding, a total of 6
measurements for a transformer without tap and 9 for a transformer with
OLTC.

14. Measurement configuration
Star Connection winding Delta Connected Winding

15. Recommended measuring frequency range
• The lowest frequency measurement shall be at or below 20Hz.
• The minimum highest frequency measurement for test objects
- with highest voltage > 72.5 kV shall be 1 MHz.
- with highest voltage ≤ 72.5 kV shall be 2 MHz.
It is recommended that a highest measurement frequency of at
least 2 MHz is used for compatibility and simplicity for all test
objects.

16. Frequency
response
region -
dominated by:
Lower frequency region (up to 2 kHz) – Core
Middle frequency region (in between 2 kHz and 20
kHz) – Interactions between the windings
Higher frequency region (in between 20 kHz and 1
MHz) – individual winding structure, internal
connections
Highest frequencies – connection leads

17. Influence
regions

18. What to identify while comparison of
frequency responses?
• Changes in the overall shape of the frequency response;
• Changes in the number of resonances (maxima) and anti-resonances
(minima);
• Shifts in the position of the resonant frequencies.

19. Typical Frequency Response for HV Star winding
• For three phase three limb core-form
transformers, the middle phase would
have a single anti-resonance in this
frequency region due to the
symmetrical magnetic reluctance paths
seen by the middle phase of the
through the other phases. The outer
phases generally have two anti-
resonances since they experience two
different magnetic reluctance paths
one through the nearest (middle)
phase and one through the furthest
phase (the other outer phase).

20. Typical Frequency response for HV winding-
Delta (D)

21. Typical SFRA graph for HV winding- Zigzag (Z)

22. HV and LV winding frequency response
characteristics
• Typically , the response of the HV winding of large power transformers with a
high winding series capacitance shows a generic rising amplitude trend with few
resonances and anti-resonances.
• On the other hand, the LV winding with low series capacitance generally shows
flat amplitude trend and superimposed by a series of anti-resonances and
resonances.
• AT the highest frequencies of above 1 MHz (>72.5kV) or above 2 Mhz (≤ 72.5 kV),
the response is less repeatable and is influenced by the measurement set-up,
especially by the earthing connections, which effectively relies on the length of
the bushing.

23. Influence of Stabilizing winding
• If the stabilizing delta winding
connection is made outside the
tank and earthed, for better
phase comparison the earth
connection should be removed
leaving the delta connection
intact.

24. Influence of delta stabilizing winding

25. Influence of Tap position on Frequency
response
• When comparing the two
extreme tap positions to each
other, the curves follow each
other very closely in the lower
frequency ranges and then
upon reaching 40 kHz, they
diverge similarly to the
comparison of the nominal to
the extreme raised tap
position, though not as
significantly.

26. Influence of measurement direction
• It is important to record in
the measurement, in which
terminal the signal is being
injected into and from which
terminal the response
measurement is being made.

27. Influence of tap reversing switch on
frequency response
• When testing a transformer with an OLTC that
has a reversing switch, even if two measurements
are carried out at the same nominal tap position,
the results may vary slightly based on how the
tap position is approached. This happens on
designs where the reversing switch contacts are
positioned differently when the tap changer is
lowered to the nominal position compared with
when it is raised to the nominal position.
• Attached graph, there is no deviation between
the curves, most probably because the
transformer has a reactive type tap
changer. Nevertheless, care should still be taken
during SFRA testing always to approach the
nominal tap position from the R1 tap, and this
should be noted on the report to ensure
consistent testing in the future.

28. Influence of Insulation medium

29. Influence of DC injection frequency response
• DC injection test (Switching
impulse tests and winding
resistance tests) can leave
residual magnetism in the core
and can cause discrepancies
between frequency response
measurements especially in
the core influence region in
the low frequency range.

30. Influence of bushings on frequency response
• It is possible that different
bushings are used during FAT
compared to those fitted at site. This
may cause difference in the higher
frequencies.
• Larger discrepancies may be
expected in the higher frequencies if
the transformer is directly connected
to the SF6 insulated busbars and the
measurement is made by connecting
to the disconnected earth
connection of an earth switch.

31. Influence of shorting lead length in short-
circuit measurement response
The shorting connections used for short-circuit
measurements should be point-to-point – that is, as
short as possible.
The length of the shorting leads can influence
measurements in the high frequency range (1 MHz)

32. Influence of test voltage for frequency response
• changing the test voltage can
affect the sweep in the low
frequency region because of
the core influence. At higher
frequencies where the winding
dominates, the sweep
becomes independent of the
applied test voltage.

33. Influence of
removing core
ground

34. Influence of temperature on frequency response
• Temperature affects the frequency
response when the temperature
variation is larger than 50°C. The
Temperature differences cause changes
in winding resistance and hence the
amplitude of the frequency response.
Changes in fluid density and dielectric
constant with temperature together
with possible physical expansion can
also cause minor but consistent shifts
in resonant frequencies across the
frequency range.

35. Influence of grounding braid length on frequency response

36. Influence of
cable shield
grounding
on frequency
response

37. Influence of test instrument grounding on
frequency response
• Even though the grounding of
the test instrument does not
affect the measurements, it
must nevertheless be always
grounded to ensure safe
operation.

38. Some factors that influence Frequency responses:
1. SFRA measurement is sensitive to DC magnetization of the core and hence it has to be carried out before
application of any DC voltage. If any DC measurement was carried out before SFRA measurement, it should be
properly demagnetized.
2. SFRA plot may distort from the actual one in case of a loose connection. (may lead to variation in amplitude)
3. Proper grounding of the test instrument is essential, it is required that the bushing flange be solidly grounded
to the transformer tank.
4. Tap positions may have changed; there may be design difference between ‘sister units’.
5. Different measurement set-up and practice
6. Different internal lead lengths
7. Different winding inter-connections
8. Residual magnetization of the core also influences the frequency response in this region.

39. Is it safe to perform FRA on a transformer
without oil?
Yes.
FRA is safe to perform on a suitably prepared transformer without oil as
the test is low voltage one.
The effect of removing the oil is to reduce the capacitance of the
transformer, by changing a key dielectric from oil to air. This alters the
position of the resonances and their amplitude.

40. Conclusion
FRA is a powerful tool for use in analyzing transformer health and mechanical integrity.
FRA simulation does not provide “exact” matches with measured results; instead, it is
a tool to aid the understanding and interpretation of the measured results.
In order to interpret a measured frequency response, a comparison is made between
the measured response and a previous baseline measurement. If the baseline
measurement is unavailable, comparison can be made with the response measured
on a twin transformer.
“Don’t jump to conclusions”.
Where there are differences between phases or between successive measurements on the
same unit, the first question that must be asked is “ what are the possible causes of
the difference?”.