TechArchive

Original article date: April 1999

There are approximately three billion relays produced annually world-wide – and there are about three around for every person on Earth! But which types should you be specifying? This extract comes from a free book from Matsushita

Several semiconductor devices (triacs, thyristors and power transistors) can be used in a similar way to electromagnetic relays to switch DC or AC circuits by applying low level control pulses. By combining such a semiconductor device with an appropriate control circuit, a power solid-state relay (SSR) can be developed.

But there are a number of disadvantages of semiconductor power relays.

• Only alternating voltage loads (AC) can be switched, since in the absence of a control signal, the triac switches off automatically at zero voltage transition

• Since none of these semiconductor switch elements offers galvanic separation between control and load circuits, a suitable isolation element is needed, such as an optocoupler or reed relay

• And since the isolation element can transmit a control signal but no load, the power required for switching must come from the load, thus making it very difficult to switch very low loads

• To avoid self-firing during rapid voltage rise in the load circuit, an RC element must be inserted. This can conduct alternating current. In the OFF state, there is a residual current of about 1 mA at 230V, 50 Hz.

In fact, electromagnetic relays, semiconductor relays and the various semiconductor switches complement each other and are rarely interchangeable.

Due to their contact arrangement, electromagnetic relays (EMRs) can switch several circuits, whereas a semiconductor relay or switch can switch only one. Exceptions are CMOS switches.

EMRs are most suitable for use in the following circumstances:

• With overloads where no elaborate protection measures are needed
• For extensive immunity to electrical faults
• For switching independently from the current direction (DC and AC up into the GHz range)
• With low switching losses
• When galvanic isolation between all contacts and the coil is needed

Semiconductor switches are particularly suitable when the emphasis is on:

• Switching times <0.2ms
• Freedom from bounce
• Life being independent of the number of switching operations
• Loss-free switching during zero transition of load current
• Low sensitivity to shocks and vibration.

With semiconductor switches, the volumetric resistance is defined as the relation of forward voltage drop to forward current. The insulation resistance of relays is a measure both of the galvanic isolation of input and output circuits and of the isolation of open circuits. It is several times higher than the corresponding barrier layer resistance of semiconductor switches which have no electrical separation between input and output. Such separation exists only in photoelectric relays and optocouplers.

Contact capacitance is an indicator of the suitability of a relay for use at high frequency and is usually measured in pF (pico-Farads). Distinction is made for capacitance between: open contact; contact pairs one to another, and between contact and earth. On semiconductor switches, contact capacitance corresponds to the no-load barrier capacitance, which is voltage dependent and can amount to 5 to 400pF.

The switched current range of relays depends, among other factors, on the contact force and on the type, cleanliness and geometry of the contacts. For semiconductor devices, it is mainly determined by the chip size and the thermal resistance.

The switched voltage range for relays is determined for closed contacts by the size of the contact resistance (minimum voltage) and with open contacts by the contact gap (maximum voltage). For semiconductor switches, it is limited by both the saturation voltage and by the maximum collector/emitter blocking voltage. With EMRs, the switched load range can extend over a range 10¹³. With semiconductor switches, it is determined by the safe operating area (SOA) and thus is limited to a range of 10⁵.

The pull-in power consumption for an EMR is that consumed by the coil which just results in the relay operating. In semiconductor switches, it is the product of input voltage (U_BE) and input current (I_B) which will just effect saturation. I_Bis dependent on the switching current (I_S).

The power consumption of EMRs is higher than the pull-in power by a safety factor that takes into account wear phenomena, environmental influences and manufacturing tolerances. In semiconductor switches it is equal to the pull-in power plus a safety margin plus collector dissipation. The power consumption of semiconductor devices rises much more steeply in relation to the collector current that it does with EMRs in relation to contact current.

The maximum bounce time of an EMR is the period from the first to the last closing or opening of a relay contact during the changeover into another switching position. Bouncing causes short-term contact interruptions. This does not occur in semiconductor devices.

The maximum electrical life of an EMR is the maximum permissible number of switch operations at a specified contact load under specified conditions, with an operating reliability of 95%. In semiconductor devices it is unlimited.

The permissible ambient temperature of an EMR is the difference between the temperature which results from coil and contact heating and the upper limit temperature which is determined by the type of plastic used in the S-relay. In semiconductor switches, it corresponds to the permissible barrier-layer temperature.

The disadvantages of the semiconductor relays are avoided by the photoelectric relay (PVR, photo-MOS-relay). Here, the light from an LED in the input circuit produces a voltage in a solar cell which then drives the gate of a bi-directional MOS-FET. It is thus possible to switch an AC circuit connected between two drain connections or two DC circuits connected between a drain and the source connection. The gate voltage produced by the solar cell is completely independent of the switched load voltage. The advantages of the photovoltaic relay are:

• It can switch both DC and AC loads.
• Its switching range extends from the micro-volt/nano-ampere range up to 400V at 0.5A or 60V at 4A.
• It has a stable, constant volumetric resistance throughout its entire life.
• The leakage current is less than 1uA.
• It shows no self-firing during rapid rise of the load voltage.

• Matsushita Automation Controls
• Ian Purcell
• 01908 231555

April 1999