Jill Duplessis - Global technical marketing manager
Testing transformers at non-traditional frequencies – that is frequencies other than DC, 50 Hz or 60 Hz – can provide invaluable diagnostic information. This has led to the development of a whole range of novel test techniques. Unfortunately but unsurprisingly, many of these tests have the word “frequency” in their title, and this has frequently led to confusion, sometimes leading to missed diagnostic opportunities.
Let’s dispel that confusion by taking a look at some of the most popular of these tests, many of which have, in fact, evolved from long-established tests simply by repeating the traditional measurements as the frequency of the test source is changed.
The most notable example is the power factor or tan delta test. Traditionally, this has been performed at line frequency (50 Hz or 60 Hz) only. It is a useful tool for assessing the health of (most of) the insulation systems of a transformer, but it has several shortcomings. Many of these can be eliminated by repeating the test over a range of frequencies, as we will discover shortly when we discuss FDS, DFR and narrowband DFR testing.
A less well-known example is a short-circuit impedance measurement. This is typically used to look for winding deformation by evaluating the so-called leakage reactance. When this measurement is repeated at several frequencies, typically between 1 Hz and 500 Hz, with the focus on the resistive component of the results, unique diagnostic information becomes available. This is the FRSL technique it will be further discussed shortly.
A common mistake is to confuse dielectric frequency response (DFR) and frequency response of stray losses (FRSL) tests with sweep frequency response analysis (SFRA) tests. This is because SFRA testing is much better known than the two other techniques and, as a result, many people upon hearing the word “frequency” assume that the reference must be to SFRA testing. Unfortunately, the tests are most definitely not the same and, while SFRA provides a lot of diagnostic information, it is not a substitute for the information provided by the other frequency-based tests.
So, next time your hear frequency mentioned in the name of a test, be wary of jumping to conclusions! Remember that there are many diagnostic tests based on frequency, and each has unique diagnostic value. For transformers, the following tests are the most common.
SFRA – Sweep Frequency Response Analysis: Below is the output from Megger's FRAX instrument
This test is used to evaluate the mechanical integrity of the core, windings and clamping structures within power transformers. A small voltage signal is injected into one end of a winding and measured at the other end of the winding so that the transformer’s electrical transfer function can be determined. This test is repeated over a frequency range from 20 Hz to 2 MHz. When the test results are compared with a reference “signature”, a wide range of fault types can be detected, including core movements, faulty core grounds, winding deformations, winding displacements, partial winding collapse, hoop buckling, broken clamping structures, shorted turns and open windings.
DFR – Dielectric Frequency Response, also known as FDS – Frequency Domain Spectroscopy.
Dielectric response measurement, together with insulation modelling, is a preferred method for measuring moisture content of the cellulose insulation in power transformers. The results are normally presented as a capacitance and/or dissipation factor/power factor versus frequency curve. A typical measurement range for transformers is 1 mHz to 1 kHz).
A DFR measurement combined with modelling the response using the X-Y model is capable of evaluating the condition of the insulation system. Using DFR for determining the level of moisture is based on the comparison of the transformer’s measured dielectric response to a modelled dielectric response. An analysing algorithm rearranges the modeled response and produces a new curve that reflects the measured transformer. The results are presented as moisture content and oil conductivity for the transformer. This test can be completed in as little time as 22 minutes using an IDAX test set.
Narrow Band DFR, also known as Variable Frequency Power Factor:
This is a DFR measurement – a series of power factor/tan delta measurements, each made using a voltage source at a different frequency – but over a narrower band of frequencies, from 1 Hz to 500 Hz. The analysis does not rely on modelling capabilities and it does not provide the estimated moisture content of the cellulose insulation.
Rather, it is a much shorter DFR measurement (approximately two minutes per test) that provides earlier indication of problems than a traditional power factor/tan delta test. It also verifies that seemingly good power factor/tan delta values actually are good, and reveals when seemingly good values are not good, as well as allowing a transformer’s unique individual temperature correction (ITC) to be determined. An example of an instrument to make these measurements is a DELTA4000.
FRSL – Frequency Response of Stray Losses: This test stands alone in its ability to detect strandto-
strand short circuits in a conductor bundle. It is also sensitive to problems that have resulted in increased losses of structural components such as the transformer tank, clamping structure and tie plates. These are problems that may cause overheating of gases in the oil. It is worth noting that Megger’s SFRA test set – the FRAX – provides FRSL results concurrent with an SFRA short-circuit test measurement.
Frequency-based tests on transformers are some of the most potent – and most convenient – diagnostic tools currently available. To get the best from them however, it’s important to understand the unique benefits that each of the available tests can provide. Hopefully this short article has provided some useful guidance and also helped to dispel the confusion that often surrounds tests of this type.