How can relays keep your electronic systems safe?

As the world becomes increasingly mediated by electrical machines, reliable switching becomes more important. Think of manufacturing processes that used to rely on mechanical power and physical safety features and which now use programmable logic controllers and electric motors. If you can’t turn them off when you need to, they’re not safe. Think of the traction motors in electric transportation: it’s really important to be able to turn them on and off reliably. And think of medical equipment: a correctly administered shock from a defibrillator, enabled by reliable switching, can save lives.

This is one of the reasons why the market for electrical relays, switching signals with powers ranging from milliwatts to kilowatts, is growing. Although power semiconductors are increasingly capable and popular, for safety-conscious applications an electrically controlled mechanical switch (a relay) that can create an airgap between two conducting contacts is the favoured choice. If that relay can also be designed to indicate when it has malfunctioned, then so much the better for all our safety.

To refresh your memory, relays work by having a coil of wire that is wrapped around an armature and a yoke. When the coil is energized, a magnetic field is formed that creates a magnetic flux through the yoke and armature. This flux exerts a force on an ‘anchor’ piece that moves the armature and closes (or opens) a contact.

As relays switch and their contacts are just parting or coming together, it’s possible for very large electrical fields to develop because of the very short distances across which the signal potential is being applied. This, in turn, can create a plasma arc that erodes the contacts. At small loads, spark-based erosion is much less likely to cause operational errors than issues such as contact oxidation or dirt. At higher loads, though, it’s a different story.

Powering purely resistive AC loads, such as electrical heating coils, through relays can lead to continuous contact erosion. Careful designers will take this behaviour into account and regard a relay’s contact material as a consumable resource that is a factor in defining its service life. At the end of this life, the contact will be unable to close, as shown in Figure 1. The worst-case behaviour of such relays is usually a safe failure because the contact erosion makes it impossible for a circuit or device to be switched on.

Figure 1: These contacts are at the end of their life because a large area of the contact has been eroded

The situation is different with capacitive or inductive loads. When such loads are energised, the switching relay often experiences high inrush currents and/or voltage peaks. When such loads are switched off, they are then likely to generate powerful arcs that can damage the relay’s contacts and lead to them becoming entangled or even welded together. Figure 2 (below, right) shows the resulting typical ‘mountain and valley’ shape of the two contacts.

Figure 2: Contacts eroded by switching reactive loads can form valleys (at left) and peaks (at right)

This is dangerous because if the contacts become welded together, the connected load will not be switched off and so can become a safety issue.

One way to address this is to design the relay so that it can report when its contacts become welded together to the control circuitry that is switching the relay on and off. The reporting signal can also be used to ensure the circuit fails safely, for example by using it to switch off another relay in series with the first. In a nice irony, the signal that makes electrical systems safer is generated by mechanical means using a technique called ‘forcibly guided contacts’. The approach is heavily regulated by standards, including the IEC 61810-1 standard for electromechanical elementary relays, and the IEC 61810-3 for relays with (mechanically) forcibly guided contacts.
 

How forcibly guided contacts operate

Forcibly guided contacts work by yoking two or more sets of contacts together mechanically, so that when the signal contacts move, so do another set of contacts whose status (open or closed) can be read by control circuitry.

A relay with forcibly guided contacts must have at least one normally-open (NO) contact and one normally-closed (NC) contact. These must be connected so that both sets of contacts cannot be closed at once.

When the relay is off, the NO contact is open and the NC contact is closed. If a NO contact is closed and then gets welded in that closed position, the relay’s NC contact must have been designed so that they cannot close, and they retain a contact gap of at least 0.5mm.

When the relay is switched on, the NO contact is closed and the NC contact is open.

If the NC contacts become welded closed, the NO contact must not close and must retain a contact opening of at least 0.5mm.

In the event of a fault, such as a broken contact spring, a short circuit must not occur that would disable the forced operation.

These core specifications of IEC 61810-3 apply over the entire service life of the relay, which must be at least 10 million switching cycles.

Relays with a single NO or NC contact are available, but types with four, six or even eight contacts are more common and come in a wide variety of contact configurations.

There are two types of forced operation.

In type A relays, all their contacts are mechanically interconnected and have positive guidance. For example, if a six-pole relay has four NO contacts and two NC contacts and one of the NO contacts gets welded closed, both of the NC contacts must remain open when the relay drops out. The status of the other NO contacts will be indeterminate, because two or more of them could also have become welded shut by the event that caused the original failure.

In Type B relays, things are a little more complicated because the relay uses both mechanically connected and unconnected contacts. In the case of the Panasonic SF4D relay (Figure 5), for example, the eight contacts are connected to one another in pairs, so a NO contact is always positively driven with an opener. In addition, the internal contacts show type A behaviour in the event of errors. By making the right interconnections, it is possible to detect which path through the relay has the faulty contact.

Figures 3 to 5 show the implementation of mechanical positive guidance in type A and type B relays.
 

Figure 3: In the Type A 6-pole SFY relay, all contacts are positively guided to one another. Arrows indicate the actuator that makes the mechanical connection

Figure 4: The 6-pole SFN relay has a partially forced operation, enabled by the two-part actuator indicated by the left-hand arrow

Figure 5: The 8-pole SF4D relay also has partially forced operation, using four white actuators to mechanically link pairs of contacts

Putting forcibly contacted relays to work

The small mechanical difference between a standard relay and one with forcibly guided contacts provides a basis for better electrical safety, without actually implementing it. Safety improvements come from integrating the working contact and its associated signalling contact in the feedback loop of the system’s control circuitry, so that the host system can recognise and then respond to the danger. Relays with forcibly guided contacts form an excellent basis for this approach, and this is explicitly mentioned in the machine directive EN ISO 13849-1. 

Find out more about Panasonic Electric Works' range of safety relays, or if you would like to discuss your relay requirements, get in touch with our technical specialists in your local language.