Make the most of HV insulation testing

Insulation testing at 5 or 10kV is increasingly seen as an important aid for investigating the condition of medium-voltage electrical plant such as large motors. Many studies have shown that testing at these voltages produces more dependable data than testing at voltages up to 1kV, but to maximise the benefits of HV insulation testing, some essential factors must be considered, particularly in relation to test set performance.

Accurate readings

The test set must, for example, produce accurate readings at high insulation resistances. To see why, consider a piece of equipment that has in the past shown an insulation resistance of 5GΩ but, when retested, shows just 2GΩ. Probably that’s still high enough for the equipment to "pass” the test, but the change may be an early indication of a developing fault. If the tester had shown all readings above, say, 1 GΩ as "infinity”, then there would have been no indication of anything being amiss.

When measuring high values of resistance, surface leakage must be considered. Without compensation for leakage currents, even a small amount of surface contamination can produce misleading results. For this reason, good HV insulation test sets have a guard terminal. This is normally connected to a bare wire wrapped around the surface of the item under test, allowing it to intercept surface leakage currents and shunt them away from the measuring circuit.

Guard terminal performance varies greatly between different manufacturers’ instruments. Some may be incapable of providing sufficient current to supply the surface leakage when the guard terminal is used, making the test results virtually worthless. Many manufacturers have also found it difficult to produce instruments that offer good guard terminal performance while meeting the requirements of the CAT IV 600 V safety rating.

Insulation testing is often carried out in locations where high levels of electrical noise are present. This can lead to inaccurate results unless the test set has high noise immunity. The best manufacturers quote a straightforward noise immunity value in milliamps of noise current: for example, Megger’s latest 5 and 10kV insulation testers for commercial and industrial applications have a noise immunity of 3mA.

A further key consideration is productivity – poorly designed test sets are slow and frustrating to use. An important factor is the current capability of the test set. An instrument with a low output current may be cheap, but it will take an interminably long time to charge a capacitive test object. A good test set should deliver an output current in the region of 3mA.

Productivity boost

Preconfigured test modes boost productivity, and the best entry-level instruments support insulation resistance (IR), timed insulation resistance (IR(t)), dielectric absorption ratio (DAR) and polarisation index (PI) testing. High-end instruments should add dielectric discharge (DD), step voltage (SV) and ramp diagnostic testing, plus internal memory for storing test results.

Finally, power supply options can have a big bearing on productivity. Internal rechargeable batteries may be convenient, but what happens when urgent testing is needed and the batteries are flat? The latest instruments address this problem by allowing mains power to be used irrespective of the state of the batteries, and by using advanced battery technology that allows the batteries to retain most of their charge for months or even years.

Chosen with care, a high-voltage insulation test set will give years of dependable and convenient service and, if it used to carry out tests regularly as part of a predictive maintenance programme, it will quickly pay for itself by reducing downtime and damage to equipment. Before making a buying decision, however, examine instrument data sheets carefully, using the guidance provided here, and invest only in the best!