Thermographic Inspection of Electrical Equipment

Theoretical Foundations, Measurement Technology Limits, and Practical Advice

In addition to our previous articles related to thermography, with this article we aim to provide further practical advice for professionals carrying out measurements of this kind, so that they can determine the real temperatures of electrical equipment as efficiently as possible and with minimal measurement errors.

Considering that undersized or damaged conductors, poor connections (due to their increased transient resistance), and in most cases electrically faulty devices heat up to higher temperatures than usual (beyond the permissible limits), the necessity of maintenance can be assessed mainly through the detection of hot spots. However, since heating typically occurs only during operation (under voltage), a measurement method that allows for contactless temperature sensing is advantageous. Basically, two categories of devices are available for this purpose: infrared thermometers (often incorrectly referred to as "point thermometers" or even more erroneously as "laser thermometers"), or thermal cameras (or thermographic cameras, infrared cameras) suitable for graphical representation of surface temperature distribution.

Note regarding infrared thermometers: A common misconception is that they measure through the laser point, or - also incorrectly - precisely at the small point where the laser pointer is visible. This point only marks the center of the measuring surface, or in the case of 2-point or circular laser devices, the edge of the surface. It has nothing to do with temperature measurement. Temperature measurement is carried out by detecting the thermal radiation emitted from the surface using a sensitive detector, followed by converting the radiation intensity into temperature values based on the object's emissivity. For the detection of thermal radiation, neither light nor laser beams are required. The main advantage of both devices is that measurements can be safely performed from a distance - even on equipment operating at several kV - without affecting the operation of the equipment being examined. However, it is crucial to know and take into account during measurements evaluation that both types of measuring devices determine the surface temperature of the object based on the detection of infrared radiation (thermal radiation), which has very serious physical (theoretical) limitations. The practical implications are as follows: Outdoor measurements (e.g., transformers and overhead lines) can only be carried out at night (during periods without sunlight, preferably under a cloudy sky) or during the day in the presence of very thick, completely closed - but precipitation-free - cloud cover to eliminate heat reflections and interfering background radiation. (Based on experience, we prefer conducting measurements at night.) Especially with new equipment (as highly reflective, finely machined metallic surfaces have very low emissivity factors), thermal radiation reflection occurs. To eliminate such measurement errors, it is advisable to perform measurements from multiple positions. Pay attention to whether the noticeable temperature phenomenon (e.g., hot spot) "moves" with changes in the measurement position or not. If its location changes in our thermal image, it is a reflection of some point-like heat source behind us; if it does not "move," then it is indeed a high-temperature point. In the case of higher-power/voltage distributors, flat copper or aluminum busbars are common, with surfaces almost polished to a high gloss. It is easy for the temperature reading to always show 35°C exactly on the current-carrying busbar in front of which we are conducting the measurements. To avoid this, it is recommended to perform measurements not perpendicular to the surface of the busbar, but at an angle of about 75-80°. This trick eliminates the reflection of our own body temperature on the busbars. If it is impossible to carry out evaluable measurements due to reflections (e.g., in the case of newly installed, polished metal surfaces), applying partial or complete coating to the surfaces to be measured can help, typically using matte paint or thin insulating tape with known emissivity properties. Of course, all these preparatory activities can only be performed in a de-energized state. The enclosed parts of switchgear and distribution assemblies cannot be measured directly. Contactless temperature measurements cannot be conducted through plexiglass or other plastic windows/doors/covers. These coverings should be removed (in a de-energized state!) before measurements, if possible.

Figure: Incorrect Temperature Display due to Reflection on Busbars and Polished Surfaces [source: PIM]

Figure: Incorrect Temperature Display on Outdoor Insulators due to Sunlight Reflection [source: PIM]

Examples of outdoor measurements:

Figure: Transformer Inspection [source: InfraTec]

Figure: Substation Examination [source: InfraTec]

Figure: Inspection of Overhead Line Faults [source: InfraTec]

Also, during the implementation of practical measurements, it should be taken into account that both infrared thermometers and thermal cameras are optical instruments capable of detecting radiation from only a part of the measured object surface with a specific geometric resolution. In practice, this means that when measuring small objects or at large distances (e.g., on power lines), the geometric resolution provided by the non-contact temperature measuring device used (whether a thermal camera or an infrared thermometer) must be considered. For too small objects or measurements taken from too far away, averaging the background and object temperatures (assuming a higher object temperature) leads to erroneous display of lower object temperatures. The greater the difference between the object and background temperatures, the greater the measurement error!

In the case of infrared thermometers, the characteristic optics with values of 20:1, 30:1, or 50:1, or 60:1 determine how suitable they are for measuring the temperature of electrical equipment and what is the smallest measurable object or the largest measurement distance. With an infrared thermometer having optics of 20:1, the average temperature of a 1 m diameter circular surface is displayed from a distance of 20 m. At a distance of 1 m, this surface is approximately 50 mm in diameter. (In practice, an infrared thermometer with such optics is hardly suitable for measuring the temperature of electrical equipment because - due to safety regulations - measurements must be taken from greater distances for high-voltage equipment, and for low- and medium-voltage equipment, the object sizes are smaller than 50 mm. Therefore, it is advisable to use infrared thermometers with optics of 50:1 or 60:1 for condition monitoring of electrical equipment. With a 60:1 instrument, the temperature of objects as small as 16 mm can still be detected from a distance of 1 m, assuming that the measuring surface (marked with at least 2 laser points in better instruments) does not protrude from the surface of the object being measured (cable, connector, contact). When using thermal cameras, the rule to follow is that at least three elemental pixels fall on each measurement surface to correctly evaluate the measured temperature data. If this rule is not followed, it is highly likely that only pixels showing the average temperature of the object surface and background will appear in our thermal image. To determine the smallest object that can be measured with a given thermal camera and the maximum allowable measurement distance (these two values are naturally interdependent), the technical data of the thermal camera include the IFOV parameter (instantaneous field of view, smallest elemental viewing angle). This parameter indicates the viewing angle that is mapped with a single sensor (pixel). For example, a value of 1.5 mrad indicates that each measurement point assigned to a pixel - projected onto the object - has a diameter of 1.5 mm at a distance of 1 m, and at a distance of 2 m, the projected surface has a diameter of 3 mm, and so on. (This should be imagined as the beam of a flashlight, which embraces an increasingly larger circular surface depending on the distance.) If necessary, the geometric resolution capability must be adjusted to the size/distance of the object by using appropriate optics (special telephoto lens).

Figure: Field of view and geometric resolution of the thermal camera [source: Infratec]

Figure: Effect of geometric resolution on measurement results [source: PIM Ltd.]

The thermal image series below vividly demonstrates the importance of ensuring that the object being measured is at least three times larger than the unique measuring surface projected at a given distance. Deviating from this - for example, if someone were to incorrectly take an "overview" thermal image of a large switchboard or distribution cabinet - the unique measuring spots may contain not only the surface of the object but also its background. Since averaging takes place within the measuring spot, the measurement result may be higher or lower than the actual temperature of the object due to the influence of the background temperature.

Figure: Effect of geometric resolution. Radiative image taken with a 1.5 mrad value, on the left from a distance of 2 m (maximum value 261 °C), in the middle from a distance of 1 m (maximum value 320 °C), on the right from 0.2 m (maximum value 415°C) [source: PIM Ltd.]

Based on the above, it is already evident that the amount of radiation detected by the thermal camera sensor depends on optical relationships. Therefore, correct focusing deserves special attention, as failure to do so - contrary to common belief - leads not only to blurry thermal images but also to serious measurement errors.

Optical focusing works the same way as we are used to in photography: the task of the collector or focus lens inside the camera is to project the incoming rays onto the sensor surface (in traditional photography, onto the film). In the case of poor focus, the rays are collected in front of or behind the sensor plane. In such cases, the image becomes blurry. However, in the case of a thermal image, the problem is more significant: because only a part of the actual radiation falls on the sensor, the rest is projected around it. This leads to the measured temperature always being lower than the actual temperature at a high-temperature point (so-called hot-spot). The worse the focus setting, the more it deviates from the correct value.

Figure: Good focus (left side), as well as poor focus (right side) [source: PIM Ltd.]

The image on the right shows that in case of poor focus, only a part of the radiation incoming to the sensor surface (the rest of the radiation hits its surroundings). Therefore, in case of poor focus adjustment, the thermal camera always shows lower maximum values and higher minimum temperatures than the actual surface temperature of the object. This error can reach up to 20-30%.

Left image: image of a soldering iron taken with very poor focus (maximum value: 280°C) Middle image: image of a soldering iron taken with slightly better focus (maximum value: 338°C) Right image: image of a soldering iron taken with perfect focus (maximum value: 428°C)

Figure: Effect of focus [source: PIM Ltd.]

Rahne Eric (PIM Ltd.) pim-ltd.com, thermalcamera.com

Contact Us

The content of this publication is protected by copyright. Any (partial) use, electronic or printed re-publication is only permitted with the indication of the source and the author's name, and with the author's prior written permission. Violation of copyright (Copyright) will have legal consequences.