Weld seams in semi-finished products - such as tubes and strips - can hardly be avoided for economic reasons. Longitudinally welded tubes have become indispensable in many applications. Forming such tubes is a challenge because the effect of heat during welding leaves changes in the microstructure. The increase in hardness in the heat affected zone (HAZ) compared to the base material is the best known effect. Those responsible for the process are mercilessly exposed to the consequences of this inhomogeneity of the material, such as breakage, poor dimensional stability or significant deviations from the target geometry.

Risk: Fatal breakage

It is even worse if the breakage only occurs later when the component is under load and a recall is necessary. In addition to potentially immense costs, the company's reputation is also ruined. In the case of safety-relevant components, liability claims from customers usually quickly threaten the company's existence. The aforementioned risks can be drastically reduced if the weld seam is specifically placed in a bending zone that is as neutral as possible before forming. Possible negative effects are then reduced to a feasible minimum. Systems for the reliable detection of weld seams and corresponding devices for their reproducible alignment have therefore become a "must have".

Back against the wall in case of damage

However, in many cases, not all the technical measures have been taken on the machines that would be sufficient to avoid having your back against the wall in the event of damage. There is often a lack of suitable sensor technology and application know-how. During the design phase, too much attention is paid to secondary benefits that subsequently counteract long-term safe operation. These include requirements such as "non-contact measurement", "sensors as small as possible" or "as little mechanization as possible". In short: simply "do little" to solve the task.
In the end, the person responsible for the process has no choice but to assign personnel to insert the pipes in the correct position. In addition to the additional costs, a corresponding error rate has to be accepted because people tire very quickly here.

Weld seam not visible

What to do if the seam is no longer visible, as is often the case with precision tubes?
The situation in which the manufacturer now finds itself is more than regrettable, but is ultimately due to the wrong focus when designing the system. In addition to the question of which sensor technology is suitable, attention must also be paid to process capability during planning.
The choice quickly will be for a sensor that is intended for broadband applications.

This often works well in clear-cut situations, but does not have the urgently required functions on board for specific applications.Attempts are often made to solve this using a software solution, but this quickly reaches its limits due to poor signal quality.Those who then rely on AI solutions ignore the need for a good database.If this is not really well prepared, the training of the models will fail.The AI may then fail to achieve a satisfactory result.So back to physics, which quickly shows the limits here:If, for example, you think you have made a good choice with a small eddy current sensor for ferromagnetic material because your measurements on initial samples were promising, you should exercise the utmost caution.

Skin effects

Here, both the low penetration depth due to the skin effect and local differences in permeability are decisive factors that only become apparent in the large quantity of pipe samples. Then it is too late, because the inadequate system is already in place. These are just some of the reasons why a concept for process-reliable weld seam detection must put secondary advantages at the back of the list.
Proven systems are not always convenient in this respect, but they are significantly more efficient.

Good examples can be found in the product portfolio of Roland Electronic GmbH, which enable a large penetration depth even with ferromagnetic material or eliminate local permeability differences.
This pays off at the end of the process chain with high signal-to-noise ratios.For example, the NS24 sensor can be used to measure a wall thickness of up to 10 mm with magnetic leakage flux or eddy current.

Sensor technology helps

These two measuring principles are integrated in the sensor element and are analyzed in terms of signal technology in the associated SND40 evaluation unit. The GUI user interface provides the user with helpful wizards that take care of setting up the sophisticated evaluation. The SND40 also supplies the control signals for the drive unit to position the weld seam. This in the full angle range of 360°.

The operator can individually define the target angle for each component.
Its definition can then be selected both via the GUI and via the I/O or ProfiNet interface.
Process reliability is achieved via digital filters whose warning thresholds can be preselected in fine stages. The user can thus control how reliably the detection should work. This ensures process-safe forming and high quality requirements and prevents production downtime.