On or off-line? Settling the debate on partial discharge measurements

Whether it is better to measure partial discharge (PD) while the asset under test is connected to the grid (on-line) or removed from it (off-line) depends entirely on the situation at hand. While on-line is generally used for testing critical systems where continuous monitoring and early fault detection are essential, there are times when off-line is the wiser choice, such as for periodic detailed analysis or when on-line monitoring is not feasible or cost-effective.

On-line testing enables: 

• Continuous or temporary monitoring of partial discharge activity in real-time, providing up-to-date insights into the condition of the insulation system which means…
• Early detection of insulation degradation or potential faults, preventing unexpected failures and minimising downtime.
• The continuous operation of the asset, so the system can continue working as needed.
• Integration with condition monitoring systems, allowing for more comprehensive analysis of equipment health.

However, on-line tests can be challenging, requiring special equipment and expertise, which can drive up costs. They also need to be designed to withstand the operational environment and be resistant to electromagnetic interference. Indeed, on-line measurements are highly affected by electrical noise, meaning that very sensitive measurements in the pC range are typically not achievable in an on-line setting. In such instances, it may be better to implement off-line testing.

Off-line testing enables:

• More detailed analysis of partial discharge activity, since the equipment can be thoroughly inspected and tested under controlled conditions.
• Reduced electrical noise as the asset is disconnected from the grid.
• Manipulation of the power supply as the asset isn’t dependent on the grid.
• The ability to conduct other tests during disconnection, such as insulation resistance, tan delta (power factor), and winding resistance.
• Cost savings and the portability of power supplies for on-site settings.

However, off-line measurements may not capture transient or intermittent partial discharge events that occur during normal operation. They also require downtime and may not provide real-time insights into the condition of the insulation system. Indeed, most off-line tests are performed during scheduled maintenance intervals or when equipment is taken off-line for inspection or repair, so faults may go undetected for some time between maintenance.

As you can see, neither method is outright superior and the decision on whether to test an asset on-line or off-line is completely circumstantial, with both options being beneficial in different respects. If you’d like to explore this idea in action, then read our case study below: