One safety feature of the Circulo servo nodes, which were specially developed for cobots, is the smart break function. Another important feature is the torque-based power control. © Synapticon

The level of safety established in mechanical engineering corresponds to the basic requirement for many applications. The basic functions STO (Safe Torque Off), SBC (Safe Brake Control) and SS1 (Safe Stop 1) are essentially all "hard-wired", which results in corresponding disadvantages. For example, only safe shutdown is possible, but nothing else. In addition to the existing cabling for motor cables, encoders, non-safe fieldbuses, I/Os, etc., a further cable (STO line) is added. The same applies to each function (STO, SBC, SS1). In decentralized systems, this is an even bigger problem than in centralized systems, where only one cabling is required in the control cabinet.

Stage 2: STO (SBC, SS1) via field bus.

The circular architecture of the servo node hardware is particularly suitable for small cobots. © Synapticon

Due to the amount of cabling involved, implementing the STO function (or SBC, SS1) via fieldbus is a logical step - especially with a decentralized controller, as the STO cables run through the machine and are not just located in the control cabinet. This approach reduces complexity and costs in mechanical engineering. However, it becomes more complicated for the controller manufacturer, as this requires an additional, redundantly designed safety processor in the controller, which significantly increases the complexity of the electronics and software. In addition, certification in this context is much more complex than with "hard-wired" STO. On this point, it remains to be seen whether customers will accept the additional costs in view of the advantages they can benefit from. Conversely, the question arises: are the effort and additional costs for STO alone worth it? Or would it not make more sense to take the entire machine to the next level of safety?

Level 3: Safe Motion (SLS, SP and SLP).

Safe motion means not only being able to switch off safely, but also being able to monitor safe motion sequences and guarantee safe motion parameters - for example speed and position. The decisive factor here is that a safety PLC can use safe encoders to monitor the speed and position of axes, for example, and switch off via STO if something deviates. However, it cannot guarantee that only the specified speed will be traveled safely at all times. This requires the SLS (Safely Limited Speed) or SLP (Safely Limited Position) function in the controller to directly ensure that only certain values are set. In combination with the safe fieldbus, this is currently the benchmark for safety technology in servo controllers. The user receives additional added value for the extra cost of the controller compared to pure fieldbus STO (level 2).

Level 4: SLT with focus on cobot requirements.

From Synapticon's perspective, the future safety level includes the SLT (Safely Limited Torque) safety function with a focus on cobot requirements - i.e. based on torque sensors - in order to avoid the problem of gear certification.

SLT, i.e. safely limited torque, currently represents the high-end league in the safety discipline. Only a few, very safety-sensitive applications have this requirement - and only very few controllers can implement it. However, the demand for SLT is likely to increase as the desire for even more automation leads to more and more force-controlled applications. Where previously the sensitive actions of humans were necessary and humans were still responsible, cobots & co. are now taking over. This offers the prospect of automating these even more sensitive areas. However, this also means that a safe dosage of force must be guaranteed.

Safety through safely limited torque

Today, manufacturers base the safety qualification of their products on aspects of dimensioning danger at system level. In other words, the robot is designed to be so weak or mechanically slow that there is no danger. However, this is becoming increasingly risky and costly for system operators when it comes to qualifying their systems, as they have to have their production line approved in terms of occupational safety. This increases the demands on component suppliers to design and certify functional safety technology more easily.

This results in the strategy of robot manufacturers to add appropriate safe motion functions - especially SLT in the case of cobots. In addition to safe force dosage and safe reaction to collisions, the special feature of the interactively working cobot is that the robot not only acts, but also reacts to external guidance. This means that situations can arise in which the external system - i.e. the workbench - feeds something towards the robot and the robot has to give way safely so as not to endanger humans. This cannot be realized in the conventional way due to the gears in the axes, which initially buffer the external force against the motor and thus the safely controlled drive and thus remove it from the "perception" and safeguarding possibility of the controller. In this case, the controller would measure the torque via the motor phases. Instead, additional sensors (torque, force) are required. The safe design of this sensor technology by the sensor manufacturers and the evaluation of such sensors by the controller suppliers is therefore currently the key issue in the field of safe motion and functional safety.

The industry generally demands cheaper and less complex - i.e. "leaner" - safety solutions, for example with less cabling. New approaches are needed here in particular. Small safety modules that are integrated decentrally into the control system on each axis, but are very powerful and flexible in order to meet future requirements up to the described expansion stage 4, are very promising.

Nikolai Ensslen, CEO at Synapticon; Andrija Feher, CTO at Synapticon / ag