Q and A: Winding resistance testing
Winding resistance tests are used to assess the condition of the current carrying path between transformer bushing terminals. Experts recommend that this test is carried out routinely as, arguably, it can reveal more incipient failure modes than most other tests. Problems such as loose, defective or incorrect connections; open, partially-open (i.e., broken strands); short-circuited turns in windings; or high contact resistance in tap changers will result in a change in resistance, and therefore be indicated by an unexpectedly high, low (for short-circuits) or unstable resistance measurement. Here are some of the questions the Megger helpline frequently receives about this invaluable test.
Q: Some more experienced colleagues have told me that we have only recently started performing routine winding resistance tests because of the dangers involved. Can you explain these dangers and suggest how they can be mitigated?
A: A transformer winding is an inductor and, if an inductor is suddenly open-circuited while dc current is flowing through it, it will try to oppose the change in current by developing a high voltage. The voltage is given by V = IR + Ldi/dt, where I is the dc test current, R is the resistance of the winding, L is the inductance of the winding, and di/dt is the rate of change of current. If a test lead falls off or is pulled off a bushing terminal during the test, a dangerously high voltage will develop. For example, if the winding has an inductance of 100 H, the test current is 25 A and di/dt = 25/0.1 (that is, the time taken for the test lead to disconnect and the current to fall from 25 A to 0 A is 0.1 s), the voltage developed will be more than 25,000 V! In addition, when dc test current is injected into a winding for the duration of a winding resistance test, a lot of energy is stored in a transformer’s magnetic field. This energy has to be dissipated before leads can be safely disconnected, which can take several minutes for a large transformer. To safeguard personnel, the asset under test, and the test instrument, Megger winding resistance test instruments are equipped with multiple safety features.
These include:
- A safety circuit that provides an “escape path” for energy dissipation. (This uses current and potential leads to provide an alternate discharge path if leads are accidently disconnected)
- Protection/automatic discharge when input power is lost or when the test current is accidentally changed before discharging the existing test current.
- A test current indicator that remains functional even when instrument power is lost.
- An emergency off switch
Q: How can I be sure that my winding resistance test results are representative?
A: Make sure that the voltage (measuring) leads are “inside” the current (source) leads – using Kelvin clamps will remove any doubt about test connections but you’ll still need to be sure that the clamps make good contact with the bushing terminals. The transformer core must be saturated to obtain representative results; if it isn’t, errors in excess of 20% are possible. Having an understanding of which instrument is the best to do each job is helpful. Generally, the test current should be greater than 1% of the winding current rating but should not exceed 15%, as given in IEEE C57.152, to avoid heating of the winding. If the test current is too high, this may cause an unquantifiable error as winding resistance is temperature dependent. Check that your results compare well (within a couple of percent) between phases and with previous test results. Note, however, that when you make comparisons with previous results, your measured results, Rm, must be converted to temperature-compensated results, Rs, to take into account the temperature difference between the two tests. This is particularly important if the previous test results are factory measurements.
Q: My test results are not stabilizing and I’m questioning whether the transformer core is saturated. What should I do?
A: It is often thought that higher test currents speed up the saturation of large transformers but, in the majority of cases, this isn’t true. Voltage determines the saturation rate of the test, as given by: (flux) ɸ = Volts * t (seconds). Therefore, when selecting a test instrument, compliance voltages above 40V dc are preferred. Make sure however, that the test current is greater than 1% of the current rating of the winding under test. To improve saturation on transformer windings when charging time is slow, connecting the primary and secondary windings in series may help. This speeds up the test by providing more volt-turns of charge. It is important, however, to ensure that the windings are connected so that the current through them produces a unidirectional flux in the core. With this setup, the test current will flow through the high voltage winding and the low voltage winding in series, while the dc voltage drop will only be measured across the low voltage winding, for example. Since current will not flow into the X1 bushing terminal, be aware that the X1 bushing connection integrity will not be assessed during this variation of the test.