The Importance of Through-Fault Monitoring in Power Transformers

Presented By:
Drew Welton
intellirent, division of ElectroRent Corp.
TechCon 2024


One of the more common causes of mechanical failure in power transformers comes from exposure to through faults. Most of today’s transformer protective relay systems provide comprehensive monitoring of these potentially damaging events but are rarely activated. In this paper we will define the characteristics of a through-fault, investigate various proper settings for a through-fault monitor, and present an actual case study of a severely damaged transformer exposed to this condition. We will also see examples of detecting mechanical failure in the transformer as it relates to a through-fault condition.

I. What is a Through-Fault?

Power system faults external to the transformer zone can cause high levels of current flowing through the transformer. Through-fault currents create forces within the transformer that can eventually weaken the winding integrity.

Let’s assume that in a substation a transformer is connected to an EHV transmission line. The transformer has its own protection in the form of the % differential (87) scheme of protection. This responds to any internal fault in the transformer zone.

The EHV transmission line has its own protection in the form of, say, distance protection. When a fault occurs in the EHV transmission line the fault current flows through the transformer to the fault in the line.

The transformer protection is not expected to respond as it is not in its protective zone. This fault is known as a through-fault for the transformer. To understand through-fault condition let’s consider differential protection in the case of transformers. Figure 1 below illustrates the protective zone of the differential (87) relay.

Figure 1 -Protective zone of a transformer differential relay diagram

The transformer differential protection will only sense the faults occurring inside the range of CT’s. Through fault conditions rely on a distance, line differential, or overcurrent protection schemes. Power transformers throughout the power system experience different levels of through-fault current in terms of magnitude, duration, and frequency. The recording capability of some transformer protection relays allow for monitoring and recording of the through-fault current.

Figure 2 – Example of through-faults diagram

II. Effects of Through-Faults on Power Transformers

Transformer failures can be classified into three basic categories, insulation, electrical, and mechanical. The potential damage from through faults can result in mechanical failure.

The amount of energy flowing through the transformer during a through fault places stress on the core and coil assembly. It may not always cause an immediate failure, but the frequency of these events can have an aggregated effect on the transformer’s mechanical structure. This can result in a decrease of the transformer’s fault withstand capability and an increase in the insulation aging rate. In other words, the more these faults occur, the less likely the transformer can handle them. In Figure 3 we see some damaged windings due to through-fault conditions.

Figure 3- Damaged windings due to through-fault conditions image

Design considerations also play a part in the amount of potential damage resulting from a through-fault, as the path through which flux flows makes the zero-sequence impedance of the shell type typically higher than that presented by a core form. Through-faults are potentially more damaging to core form rather than shell form due to multiple current channels.

Figure 4-Core form vs. shell form diagram

III. Through-Fault Monitors

Through-fault monitoring, whether part of the protective relay system or other transformer monitoring systems record and store information relative to the magnitude, duration, and cumulative number of the through-faults to which the transformer was exposed. These recorded values can assist the substation crew in prioritizing transformer maintenance and testing, and, over time, provide additional information in determining problems within the transformer.

The through-fault is measure using an inverse time over-current relay, 51TF, measuring the current flowing through the transformer in magnitude and duration.

The through-fault monitor records the cumulative I2T value for each phase and compares it against a threshold to provide an alarm, as shown in Figure 5.

Figure 5- Block diagram of transformer monitoring function

ANSI/IEEE standards provide operating limits for power transformers. Initially, these operating limits only considered the thermal effects of transformer overload. Later, the capability limit was changed to include the mechanical effect of higher fault currents through the transformer. Power transformer through-faults produce physical forces that cause insulation compression, insulation wear, and friction-induced displacement in the winding. These effects are cumulative and should be considered over the life of the transformer.

Settings for the through-fault monitor depend on category 1 -4 of the transformer, and also the design as previously mentioned. Transformer OEMs should be able to provide the mechanical limits associated with the transformer.

A good reference is IEEE Std C57.109-2018-IEEE Guide for Liquid-Immersed Transformers Through-Fault-Current Duration. A characteristic curve is designed for each transformer class and sets fourth recommendations believed essential for the application of overcurrent protective devices applied to limit the exposure time of transformers to short-circuit currents. Figure 6 illustrates the through-fault capability limit curve for category IV transformers. The curve represents both the frequent and the infrequent fault occurrences. For category III and IV transformers, the mechanical duty limit curve starts at 50% of the short circuit current.

When the transformer monitor characteristic curve time is reached, the through-fault counter will be increased by 1. The through-fault counter indicates the number of events that have lasted long enough to exceed the transformer withstand capability and, thus, can help predict the remaining life of the transformer.

Figure 6-Through-fault capability limit curve for category IV transformer graph

IV. Transformer Diagnostic Testing for Mechanical Damage

Because the power-protection system operates effectively, there is not a lot of consideration given to the weakening of the transformer’s clamping system or core. It is hard to quantify the level of through faults a specific transformer can sustain and remain fit for service because there are so many variables to be considered when performing the assessment. Tests for mechanical failure include routine dissolved gas analysis (DGA), leakage reactance, and sweep frequency response analysis (SFRA). In the event the through-fault monitor has recorded a cumulative amount of through-faults since the last scheduled maintenance outage, the DGA should be closely reviewed for any abnormalities. Since mechanical damage may not be indicated with a DGA, a leakage reactance test should be performed in addition to the power factor and excitation tests.

The field Leakage Reactance test, otherwise known as short circuit impedance, is an AC (60Hz) test which is performed to detect mechanical winding movement and/or deformation within a power transformer. There are two methods for performing Leakage Reactance tests, as follows:

  1. Three-Phase (3-Phase) Equivalent Test
  2. Per-Phase Test

The Leakage Reactance measurement directly corresponds to the leakage flux. Leakage flux is flux that does not link all the turns of the winding. It is normal that some of the flux escapes. This leakage flux also helps create impedance that is used to limit short circuit current. Leakage flux creates reactive magnetic energy that behaves like an inductor in series in the primary and secondary circuits. This impedance can be easily measured, analyzed, and trended. A normal 3-phase equivalent test should not vary from the name plate impedance by +/- 3%. The per-phase test should also be within 3% of each phase.

In the event that the transformer has no source if through-fault monitoring, then a leakage reactance test should always be incorporated into routine maintenance testing.

Should the measured impedance in either of these two tests exceed the acceptable limits, a sweep frequency response analysis (SFRA) test should follow.

SFRA is a method to evaluate the mechanical integrity of core, windings, and clamping structures within power transformers by measuring their electrical transfer functions over a wide frequency range. Common mechanical deformations which SRFA can determine include core movements, faulty core grounds, winding deformation and displacements and winding collapse, hoop buckling, shorted turns and open windings.

Since mechanical failures can also result from transportation, seismic activity, or faulty generator synchronizing, SFRA should also be considered under any of these circumstances.

Two IEEE guides relevant to these tests can be found in IEEE C57.149-2012, “IEEE Guide for the Application and Interpretation of Frequency Response Analysis for Oil-Immersed Transformers” and IEEE C57.152-2013, “IEEE Guide for Diagnostic Field Testing of Fluid-Filled Power Transformers, Regulators, and Reactors”.

V. Case Study using SFRA and Leakage Reactance Following an Elevated DGA

The protective relay recorded simultaneous through-faults with an elevated magnitude beyond the mechanical threshold, the differential relay suddenly took the transformer off-line. The transformer in this case study is rated at 50 MVA 90.2kV/34.5kV, and upon immediate investigation, oil appeared to be leaking from the LV DETC. The standard test protocol was applied, which included a DGA, power factor, SFRA and Leakage Reactance. The DGA analysis showed accelerated levels of acetylene, indicating serious overheating.

Figure 7-DGA results from faulted transformer data
Figure 7-DGA results from faulted transformer

The power factor tests would not run on the low side at 10 kV, so the test voltage was lowered to 7 kV. A power factor of almost 50% was obtained for the CL insulation, as shown in Figure 8, which is clearly unacceptable.

Figure 8-DGA from faulted transformer table
Figure 8-DGA from faulted transformer

The leakage reactance results indicated an unexpected high impedance on phase B, which is typical of an open circuit. The DC winding resistance test also confirmed an open circuit. However, the other phases, A and C, produced and expected 40 mΩ.

Figure 9-Leakage reactance test results from faulted transformer table

Notice in the test results, the 3-phase equivalent test results were within the 3% range of the nameplate. However, phase B of the per-phase test was far out of tolerance.

The SFRA results indicate a high impedance fault on phase B of the low voltage winding. Both the LV open circuit and HVV-LV short test provide evidence. The trace analysis clearly supports this finding.

Figure 10-SFRA results from failed transformer graph

After disassembly, it was determined that a significant series of through faults caused extensive damage to the windings of this transformer, as seen in Figure 11.

Figure 11- Faulted winding caused by through-fault image


[1] Sample Reference Section: IEEE C57.149-2012, ” IEEE Guide for the Application and Interpretation of Frequency Response Analysis for Oil-Immersed Transformers”.

[2] Sample Reference Section: IEEE C57.152-2013, “IEEE Guide for Diagnostic Field Testing of Fluid-Filled Power Transformers, Regulators, and Reactors”.

[3] Dynamic Ratings: Webpage on Through-Fault Monitoring

Join our email list

We use cookies to give you the best online experience. By using this website you agree with our cookie policy.