Aspects of the Practical Application of Sweep Frequency Response Analysis (SFRA) on Power Transformers (Part 2)



Finally a summary of guidelines derived from a large number of successful measurements is given to put the reader in a position to achieve a good degree of repeatability, too. This is of need because all types of assessments explained in chapter 3 are based on the ability to exactly reproduce the measurement results under same conditions. Without this it is often very critical or even impossible to distinguish between measurement mistakes and real damage inside the tested transformer.

  • All connections from the transformer except the tank ground and auxiliary connections shall be removed.
  • The contacts of the bushings shall be cleaned and the connection clamps have to be tightened firmly to assure reliable electrical
  • Three shielded high frequency cables (usually coaxial cables) of exactly same length should be used.
  • It must be assured that the ground extensions of the measuring cable shields do not have electrical contact with the terminal
  • Ground extension of the test leads must be of low inductance (broad braids with large surface, made of a large number of small wires to reduce the skin effect at higher frequencies).
  • Ground extension to the base of the bushing (reference potential is the transformer tank) shall be as short as possible and with the smallest achievable
  • It is very important to ensure reliable contact between ground extension and tank. A lot of measurement mistakes are related to this
  • Detailed information about the test set-up should be stored together with the test data. This will help to reproduce the measurements for future tests. Detailed photographs are

Every test result should immediately be checked for plausibility and compared to expectations and available references. A very noisy curve is almost an indication for poor grounding. Figure 9 gives two typical examples for this phenomenon. It is important to recognize measurement mistakes on-site and to repeat the questionable test after finishing the required corrections.

hình 9


The evaluation of transformers transportation is an increasingly popular application of the SFRA method. This is reasonable due to the ability of SFRA to provide in-depth information about the core, the windings and the clamping structures, as well with one set of tests. All these parts are susceptible to transportation damage.

As for all other applications of SFRA, performing the test under the same conditions is important to get reliable results. Therefore the initial test shall be performed on the transformer in its transport configuration. Usually the transformer will be equipped with bushing cover plates or transport bushings, which is strongly recommendable to facilitate testing, and without oil (depending on size and restrictions). Thus it is obvious that normal baseline data from factory or on-site fingerprint tests can not be used for this purpose because the results will differ from each other. On the other hand it must be noted that transportation test results usually can not be used as baseline data for future tests in operational condition. The first transportation evaluating test shall be performed as end-to-end open circuit measurement with all other terminals floating. Short circuit measurements are not able to display the core area. The test shall be performed using logarithmic measurement point setting since linear scaling may not sufficiently cover the lowest frequency region related to the magnetic core which is especially vulnerable for transport damage.

After the initial test before the start of the transportation, rechecking tests can be performed at any time during transit. It is important to note that the SFRA test should be the last electrical test prior to transportation and the first test after arrival. Other tests in between, especially DC tests (e.g. winding resistance test) may change the core magnetization status and hinder a reliable evaluation of the core integrity. The status of core remanence should be noted in the test documentation. The same is valid for tap changer position and the oil level or even its absence. If the test has been performed shortly after draining the oil this fact should be noted too, because of the effects of residual oil within the insulation. A re-test on-site without oil may lead to inconclusive results since the residual oil may be dropped out during the transport which may lead to changes in the capacitance and therefore slightly shifted SFRA curves.

In addition it is important that the transportation configuration of the transformer is well documented and available to other testing personnel who have to perform the repeating measurements. It might be necessary to identify and record more than one transportation configuration since the transformer may undergo several legs of transport like truck, ship, railroad, crane off-loading etc. Any of these transportation legs can exert undue physical shock to the transformer. Testing before and after transport legs that have different legal custodies is prudent. After the receipt of the transformer at its final destination one last test in transportation configuration should be performed to evaluate the transportation. If this test confirms healthy status another SFRA test with the transformer assembled and oil-filled as required for insulation resistance testing should be performed to be used as baseline or fingerprint data for future testing. In all cases it is recommended to photograph the connections between FRA device and the bushings.

In the following example SFRA tests has been performed on a batch of transformers after their transportation from Europe to Africa. Impact recorder readings were suspicious. Additionally bumping marks were found. Figure 10 shows two example pictures of a suspect unit. No SFRA tests had been performed prior shipment but fortunately type-based and phase comparisons were possible.

hình 10

The measurement on the left of figure 11 displays the core region of the suspect transformer in the frequency range 100 Hz – 50 kHz. Contrasting on the right a sister unit with the core in healthy condition is shown in the same frequency range. The remanence information has been available from the manufacturer. The cores were demagnetized and therefore all equal-type transformers should show the same behavior in the core region below 5 kHz. Obviously the transformer on the right has a perfect match of the outer phases (red and blue). The deviation of the middle phase (green) is as expected. The curves on the left are different – the outer phases have a poor congruity. The most unlikely shape shows the curve of phase 3 in blue.

hình 11

Based on the result above and the impact recorder readings it has been decided to open the transformer for visual inspection. The findings confirmed the assessment of the SFRA results. The limb 3 was displaced to the right. The according clamping rod was found bent and the upper wooden clamping frame was partly ruptured. The clamping pressure was significantly reduced on the affected phase. Finally the transformer has been shipped back to the manufacturer for repair.

hình 12


The SFRA is a powerful method for the detection and diagnosis of defects in the active part of power transformers. It can deliver valuable information about the mechanical as well as the electrical condition of core, windings, internal connections and contacts. No other single test method for the condition assessment of power transformers can deliver such a diversity of information. Therefore the SFRA is an increasingly popular test. The value of fingerprint data is more and more recognized by users all over the world. Comparing the time and the frequency domain FRA test methods is seems to be obvious that SFRA, measuring directly in frequency domain, prevailed. Reproducibility is the key for a successful application of SFRA. Therefore highest accurateness is essential when establishing the connections. Due to the work of Cigré WG A2/26 consensus about the effective and reliable application of this method has been achieved. The authors wish to thank all members of this WG for their work.


  1.  H. Bartley, “Analysis of transformer failures”, 36. Annual Conference of engineering insurers, Stockholm, 2003
  2. The Electric Power Industry Standard of People’s Republic of China, “Frequency Response Analysis on Winding Deformation of Power Transformers”, China, 2005
  3. Cigré WG A2/26, “Mechanical condition assessment of transformer windings using Frequency Response Analysis (FRA)”, Brochure 342, Paris, 2008
  4. Cigré WG A2/26, “Mechanical condition assessment of transformer windings using Frequency Response Analysis (FRA)”, Electra N°228, Paris, 2006
This entry was posted in . Bookmark the permalink.
024 6683 0230

Send us
an Email