Partial Discharge is a breakdown of a small area of insulation, which is subject to a high voltage. The breakdown does not span the entire distance between the two insulated electrodes; hence, it is referred to as ‘Partial’. The magnitude of such discharges is usually small; however they do cause progressive deterioration of insulation that may lead to eventual failure. The material in a well-designed insulator, without any defects, will have a nearuniform electric field, without any significant changes in the electric field gradient. A poor design, or a defect in the material will create a high level of electrical stress over the area. Partial Discharge will occur at any point where electric field strength exceeds the breakdown strength of that portion of insulation material.
PD activity may occur in an insulating material for many reasons. In solid insulation, a defect could result in a gas-filled void. In liquid insulation, this could be from bubbles in the insulation. In air insulation, this can be seen as corona. Non-intrusive partial discharge detection provides a means for identifying these potential sources of insulation failure that result not only in loss of supply to customers, but can also endanger staff.
A partial discharge emits energy in the following ways:-
There are a number of causes of Partial Discharge including voids in solid insulation often caused during manufacture, mechanical breakdowns caused during installation, particle contamination and design defects. Alternatively if you have aging assets a combination of all of these factors may be the cause of the Partial Discharge.
Internal discharge activity occurs within a void in insulation. It will generally reach higher amplitudes than surface discharge activity but will have low numbers of pulses per cycle; typically in the region of 1 – 6 pulses per cycle. Relatively strong Transient Earth Voltage (TEV) signals are produced from internal voids together with ultrasonic signals. However any ultrasonic signals will be occurring internally to the insulation and so will not be externally detectable.
Surface discharge over time will result in surface tracking. Surface discharges occur when there is a degradation of the surface of the insulator. This degradation can occur as a result of surface contamination.
When moisture combines with surface contamination, a small amount of current flowing over the surface can result. The flow of current will cause the surface to heat up, which will result in an area drying and ceasing to be conducting. A small gap may be created as a result of this heating process, which can result in a large electrical field across the newly created gap. If the electrical field exceeds the breakdown voltage, then a partial discharge activity will occur. If the insulator contains a carbon based compound, then the heating can result in carbonization, which will form a permanent carbon track on the surface.
Over time, these carbon tracks can spread and form a tree like shape. Due to a tree shaped track having a large number of discharge points, there will be a high number of potential discharge sites. This results in a high discharge rate, often in the range of 8 – 30 pulses per cycle but can be greater. The magnitude tends to be smaller than a void discharge, and the magnitude may not increase as the tracking increases towards the point of failure. There will be high levels of ultrasonic activity together with lower levels of electromagnetic activity emitted from surface discharges.
This refers to the situation where metalwork is not tied to a specific voltage potential, and is insulated from any reference voltage. It is possible for the voltage of the metalwork to drift higher or lower as voltages are capacitive coupled from the conductor. Once again, the insulating material will have to withstand the electric field strength that results from the voltage difference between the conductor and metalwork. If the electric field strength exceeds the breakdown voltage of a defect in the insulation then a partial discharge will occur.
Metalwork can float as a result of poor earth connections or poor contacts being made. Floating metalwork can produce TEV signals that are high in magnitude and typically at a rate of 30-50 pulses per cycle. For contact noise, activity can be even higher, with a much greater amplitude TEV signal produced. Airborne ultrasonic noise may also be detected.
One thing is certain, once Partial Discharge starts the damage will only increase, it never decreases.
The most practical techniques for non-intrusive testing are based on the detection of the radio frequency part of the electromagnetic spectrum and ultrasonic emissions.
Acoustic emissions from partial discharge activity are normally at a frequency too high for the human ear, i.e. ultrasonic. As the partial discharge progresses towards failure the frequency sometimes decreases into the audible range.
The most sensitive method of detection is using an airborne ultrasonic microphone. This method is very successful at detecting partial discharge activity provided there is an air path between the source and the microphone.
When partial discharge activity occurs within high voltage switchgear it generates Transient Earth Voltages (TEV) and electromagnetic waves in the radio frequency range.
The TEV signals are high frequency and therefore travel in the skin of the switchgear and can only escape from the inside of the switchgear through openings in the metal casing. These openings may be air gaps around covers, gasket materials or other insulating components.
In addition when the electromagnetic wave propagates from the partial discharge site it also impinges on the metal casing of the switchgear producing a Transient Earth Voltage (TEV).
The partial discharge activity may be detected non-intrusively by placing a probe on the outside of the switchgear whilst the switchgear is in service.
The Transient Earth Voltages caused by PD Activity were first discovered at EA Technology in the 1970s by Dr John Reeves. He established that the TEV signals have a direct relationship with PD Activity and good indication of the condition of switchgear insulation.
Partial Discharges activity provides clear evidence that an asset us deteriorating in a way that is likely to lead to failure.
If Partial Discharge is left unattended it will lead to flashover and failure.
As an asset manager the ultimate goal is to decrease outages caused by Partial Discharge, ensure staff safety and be fully aware of asset condition to enable you to make the most cost effective asset management decisions.
The location of a Partial Discharge site can be identified quickly and accurately by using an instrument which employs two coupled probes, 100mm or more apart, to measure TEV signals. The instrument calculates the time of flight variations between discharge pulses as they reach the two probes and in turn indicates the source of the discharge in relation to the probes. Using this method, the source of PD activity can be identified to an accuracy of 0.33ns (nanosecond), equivalent to 10 centimetres.
Partial Discharge activity can be quantified and recorded in several ways, depending on the sensor technologies employed:
Instruments are available to convert Partial Discharge ultrasonic emissions into audible signals, so an engineer can hear changes in Partial Discharge activity. They can also be displayed and recorded as numerical decibel values.
TRANSIENT EARTH VOLTAGE (TEV)
TEV signals are measured, which can be displayed and recorded on the appropriate instrument.
ULTRA HIGH FREQUENCY (UHF)
UHF emissions are also measured which can be displayed and recorded on the appropriate instrument.
Continuous monitoring of PD activity uses the same sensor technologies as Partial Discharge measurement, but is employed to record:-
Transportable Partial Discharge monitoring systems are typically used to record and analyse changes in PD activity over a set period. They include multiple sensors which can be sited on a variety of assets and are networked to a central server unit, with data storage facilities and remote access capabilities so that data can be retrieved for analysis online. Monitoring of this type is often used to provide a detailed picture of the health of assets which have a history of problems and/or where ongoing reliability is a critical issue.
Illustrative system layout of an UltraTEV Monitor™ System