In the field of magnetic crack detection, we are also witnessing the emergence of new innovative technologies that digital radiography have also revolutionised. Innovation can provide a significant competitive advantage to companies that can make good use of the opportunities offered by science and fine-tune their testing processes accordingly, even for the simplest process.
Magnetic crack detection, as a non-destructive material testing method, is excellent for surface and near-surface crack detection, and can detect deviations with much greater precision and confidence than ultrasonic or X-ray material testing.
The aim of this article is to take this excellent method for detecting surface defects and show you a little beneath the surface. The efficiency of a process can only be increased if we understand its principle of operation in depth, and the main trends and tendencies in technological innovation: this is customizability, automation and the artificial intelligence.
The principle of magnetic crack detection
A magnetic crack detection as its name implies, is based on the magnetisability of materials and can therefore only be used for non-destructive testing of ferromagnetic materials. Its main advantage is that it can be used for the detection of surface and near-surface material continuity defects, similar to penetration or liquid penetrant testing. The nature of the defect, its geometry and the use of the material should be considered when deciding which method to use.
Four factors are needed to carry out the test:
- on the part to be tested (with magnetisable material composition)
- magnetic material applied to the component (magnetic powder suspension - dry or liquid)
- magnetic field
- test personnel or automation that coordinates the whole test process (preparation, testing, evaluation, post-cleaning, magnetisation)
The magnetic field can be detected by applying a fine, dry or petroleum-insoluble iron powder to the component under test: the condensed iron powder indicates the exact location and shape of the defect - the lack of continuity causes the magnetic lines of force to scatter.
There are two main types: the black magnetic powder suspension can be detected by the naked eye, and the fluorescent magnetic powder can be detected by UV radiation.
Magnetic field excitation
The magnetic field required for the test can be generated by magnetic crack detectors in two modes:
- Ferroscopic or pole magnetisation method: in this case, the poles of the machine operate as DC electromagnets and are recommended for defects perpendicular to the longitudinal direction of the component.
- Ferroflux or current magnetisation process: in this case, the magnetic field is not generated by the machine, but by the component itself by passing a strong alternating current through it. It is suitable for detecting longitudinal faults.
Advantages and disadvantages of the procedure
One of the biggest Advantages, is an excellent method for detecting surface and near-surface defects and is very sensitive: it can detect defects as small as 0.001 mm wide. It provides relatively fast, immediate results, can be applied through thin coatings (non-magnetic coatings up to 0.075mm do not affect the effectiveness of the test) and allows the inspection of high and low temperature components. It is also much easier to perform compared to radiographic and ultrasonic material testing.
The magnetic crack test downside, as already mentioned, is only applicable to ferromagnetic materials, so for the detection of near-surface defects, if the test piece material is not magnetisable, the liquid penetrant method is the best solution. A further disadvantage is that the test piece requires pre-cleaning, the magnetic suspension can only be applied to a clean surface (free of grease, oil and dirt) and the test piece requires demagnetisation after the test has been carried out.
NDT 4.0: a step up in the process
Non-destructive material testing processes, including magnetic crack detection, have been transformed by the rapid, exponential advances in technology. The achievements of the fourth industrial revolution, Industry 4.0, are trickling through and increasingly redefining the future of the profession: NDT 4.0 (non-destructive testing). non-destructive material testing a new chapter has opened.
Industrial computer process support, the use of industrial robots and even artificial intelligence to support evaluation are shaping the process of magnetic crack detection, so the work of materials inspectors is closely intertwined, even if indirectly, with IT, robotics and mechatronics specialists. Interdisciplinary collaboration is the future of the profession.
Real progress can only be achieved if the human factor is taken into account in all efficiency improvement measures, a skilled workforce is the most important condition for increasing efficiency, and even in the case of self-service systems, people are still in control in the background.
Tools to increase efficiency
The key to increasing efficiency is to approach material testing tasks with a systematic mindset. Innovation can only deliver real business benefits if test factors are considered as a coherent whole, rather than in isolation. In this way, target devices and equipment optimised for the task/process can be developed for the inspection of specific components or for the inspection of large quantities of mass-produced products, with which inspection efficiency and sensitivity can be significantly increased.
Customizability and automation
The above-mentioned elements of magnetic crack detection (test piece, magnetic crack detector, magnetic suspension, test personnel) can all be integrated into a system optimised for the specific inspection task, with automation features that make the inspection much more efficient.
The programming of robotic arms allows the test pieces to go smoothly through the process: using immersion baths to apply the magnetic suspension faster and more efficiently, cameras and artificial intelligence to improve the evaluation efficiency, and post-cleaning and magnetisation to be more easily integrated into the process with specific solutions and automation that are built to suit the test task, taking into account every detail.
New trends: hysteresis and exploiting the physical principle of MMM
Magnetic hysteresis and metal magnetic memory (MMM) are two new trends in technical diagnostics that deserve to be mentioned in the context of magnetic materials testing, especially as such techniques open up new horizons in terms of efficiency.
The state-of-the-art method and tool for testing mass-produced components is the electromagnetic classifier. It is based on the property of magnetisable materials that magnetic hysteresis curve varies according to the composition of the material, its heat treatment and its mechanical processing. For the test, a flawless reference part is required, the hysteresis curve of which is compared with the hysteresis curve of the other parts.
Magnetic memory of metals (MMM)
The distribution of the magnetic field of the test piece is recorded and analysed by the MMM method. The test can be used to detect stress collection and defect locations, inhomogeneity of welds and materials. It is currently one of the most effective methods for testing the service life of equipment and requires no surface preparation like conventional magnetic crack detection. Here to read more about this procedure.
The main focus of our paper is on traditional magnetic crack detection, but we thought it important to mention new trends that are likely to receive more attention in the future, because two essential tools to increase efficiency are: the development of customisable, task-optimised automation and systems, and the continuous monitoring of the cutting edge of technology.
Is the future brighter with machine vision?
Artificial intelligence is one of the undisputed pioneers at the cutting edge, but it is important to stress that even the most advanced technology is not yet able to replace humans, and we have yet to create systems with the complexity of our brains.
However, while in theory a materials examiner is efficient at about 80% (20 times out of 100 samples wrong), in machine vision this ratio is 95-99%. In addition, it can detect defects much faster than a specialist, but it is important to note that in some cases it can find defects in pieces that a materials inspector with sufficient experience and a more complex (programmed for life and not a fixed task) view would have assessed as correct. However, this is a "necessary evil" if we consider that the faulty pieces are filtered out with excellent efficiency by artificial intelligence.
There are several variables that affect the effectiveness of machine vision in the process of magnetic crack detection, but the most important to highlight are two factors that indirectly affect several circumstances:
- choice of marking material: here it is important to ensure adequate contrast to make the images easier for machine vision to recognise and evaluate, and this is also important for naked eye examination (fluorescent dye and ultraviolet light or coloured substrate is recommended in most cases)
- type and sensitivity of the test camera, and shutter speed depending on whether the test pieces are static or in motion.
Artificial intelligence is likely to become an indispensable component of non-destructive material testing in the future, and is already one of the most widely used tools to increase efficiency.
New perspectives in magnetic crack detection
Customisable solutions, automation and artificial intelligence, including machine vision, have undoubtedly opened up new horizons for non-destructive material testing, including magnetic crack detection. However, humans are still in the background.
Super-sensor cameras may be able to compete with the eye of the materials inspector, and artificial intelligence algorithms are a huge step forward in evaluation, but human judgement is still an indispensable element of non-destructive materials testing processes. This is why it is so important to focus on training non-destructive material testers to keep up with technological innovations - to balance the relationship between man and machine.
Even if we cannot compete with the precision of machine vision, it is still only humans who can see the whole process, innovative opportunities and new perspectives - professionals are the systematic principle among the tools for efficiency.