Solid State Relays (SSR) What is a solid-state or semiconductor relay? • SSR is a semiconductor device that can be used in place of a mechanical relay to

switch electricity to a load in many applications. • SSRs are purely electronic, normally composed of a low current control side

(equivalent to the coil on an electromechanical relay) and a high-current load side (equivalent to the contact on a conventional relay).

• SSRs are typically feature electrical isolation to several thousand volts between the

control and load sides. • SSR contains one or more LEDs in the input (drive) section. The SSR provides optical

coupling to a phototransistor or photodiode array, which in turn connects to driver circuitry that provides an interface to the switching device or devices at the output. The switching device is typically a MOSFET or TRIAC.

• Several SSRs that incorporate TRIACs provide a built-in series-RC snubber network to

protect the TRIAC against line voltage surges. A snubber protects the TRIAC against small to moderate surges.

• Semiconductor relays are quite common component nowadays but they are still quite

expensive compared to their complexity.

Figure: Typical block diagrams of triac and FET based SSR circuits

Benefits of semiconductor relays

• No mechanical moving parts • No arching in contacts • No contact materials which will wear out in frequent use • No inductors on control side • No contact bounce (spring back) • No acoustical noise • No arching or sparking • No electromagnetic interference (EMI) from contact commutation • High switching speed • High reliability • Long operating life • Resistant to shock and vibration • Wide input voltage range possible • Possible to always turn on and off only at zero phase • High input-output isolation

Bad sides of semiconductor relays

• Output gets damaged quite easily by overvoltages • Typical failure mode is output short-circuit • Output has minimum voltage and current in order to work • Output has some leakage current on off-state • More expensive than normal relays • Low to moderate volumetric efficiency • Restricted to single pole, normally open (NO) configurations • On-resistance much larger than normal relays (means more waster power and

voltage, heat sink often required at high current models) • Large output capacitance (typically 1 pF for normal relay, more than 20 pF for

SSR) • Heats up noticeably when large current passes through it • More sensitive to voltage transients • Most types work only on AC current (there are also special DC SSR available) • There is some leakage current even when relay is off • If the triac inside semiconductor relay is not driven properly or faulty, it can act

like a rectifier and result in a pulsating DC to the load

Electromagnetic relays usually an edge over SSRs in applications requiring extremely high voltages and currents. SSRs can't match the current-carrying capacity and low on-resistance of the biggest electromechanical relays.

Applying SSRs to circuits

Solid State Relays (SSRs) cannot always be applied in exactly the same way as Electromechanical (EMRs) and when such is the case, caution should be taken.

• Highly inductive loads such as transformers and chokes are likely to have any significant influence on SSR operation. These loads can create large current surges and the SSR should be derated accordingly.

• Extremely high current surges are commonly associated with transformers, especially those which can saturate. The zero voltage turn on feature of standard SSRs can increase this possibility and might require that special precautions be taken.

• Dynamic loads such as motors and solenoids, etc., can create special problems for SSRs. High initial surge current is drawn because their stationary impedance is usually very low and back EMF can also add to the applied line voltage and create 'overvoltage' conditions during turn off.

• Incandescent (tungsten filament) lamps have a high inrush current, but somewhat similar to the surge characteristic of the thyristors used in AC SSR outputs, making them a good match. The typical ten times steady state ratings which apply to both from a cold start allow many SSRs to switch lamps with current ratings close to their own steady state ratings.

• Using SSRs for driving mercury, fluorescent, or HID lamps should be avoided. If they must be used, the SSR must be severely derated and thoroughly tested in the specific application.

Typical technical specifications of semiconductor relays 1. Small current DC semiconductor relays

SSRs offer several desirable attributes for telecommunication applications. Small DC semiconductor relays are typically used in new modems to replace bulky line relays. Those small SSRs are typically housed in small 6-pi DIP package. They have typical LED input like every other optoisolator and the output is typically optocoupled MOS-FET section. Typical optocoupled FET driving circuit works so that the LED shines to many series connected photocells which make the control voltage to FET gate to make it conductive. The relay can be made to pass current to both directions using pair of MOSFETs in the relay output.

This kind of MOSFET pair used in small telecom relays can provide on-resistance in order of 10 ohms. Typically this kind of relays have 200-300V output ratings, can handle 100-200 mA of DC current and have isolation of 450 megaohms at 2 kV or 4 kV. One disadvantage of SSRs in this kind of telecommunication application is that they are more prone to be damaged because of overvoltage or overcurrent than their electromechanical relays. If you are using SSRs in telecommunication systems, you must usually provide extra protection against spikes and inductive kicks on the lines.

2. AC semiconductor relays for main voltage control

AC semiconductor relays typically are constructed using TRIAC output stage and optocoupled triac driver. The triac output stages usually work at voltage range of 24-250V and can typically handled 1-4 A in small relays. There are larger relays available for higher currents (those typically need external heatsink). The TRIAC output stage in semiconductor relays typically has about 1-1.5V voltage drop when it conducts and this causes some heat generated inside the semiconductor relay (typically around 1.2W/A). Because the TRIAC output stage and some filtering components semiconductor relays have some leakage current when they are off (can be up to few mA).

Semiconductor relays typically work at 3-30V control voltage range and the input current is typically around 8-16 mA (the current does not typically change much over that voltage range).

The isolation voltage in mains controlling semiconductor relays are typically 2.5 kV or 4.5 kV. If there is bare metal for fitting heatsink that is typically isolated from the relay electronics (check the datasheets to be sure about your relay type).

What does a zero-crossing turn-on circuit refer to?

Zero-crossing turn-on and turn-off refer to the point on the AC wave form when the voltage is zero. It is at this point that an AC SSR will turn on or off. When the AC circuit voltage is at zero, no current is flowing. This makes it much easier and safer for the semiconductor device in the relay to be turned on or off. It also generates much less electrical EMI/RFI noise.

Can I use an AC SSR to switch a DC load?

No. Because of the zero crossing circuit described above, the relay will most likely never turn on, and even if it is on, it will likely not be able to turned off, as DC voltage typically never drops to zero.

Can I use a DC SSR to switch a AC load?

No. If the DC semiconductor relay is polarized, it may break down and conduct for the portion of the waveform that is reversed in polarity. There are available also non-polarized semiconductor relays which can be used on DC and AC but those are more expensive.

Can a DC SSR be used to switch an analog signal?

This is not recommended at all, for several reasons. First, the voltage drop across the relay will cause signal loss. Second, the conduction characteristics of the SSR are very non-linear at low operating voltages and currents. Use a mechanical relay; it will work much better.

Can I hook up SSRs in parallel to achieve a higher current rating?

No. There is no way to guarantee that two or more relays will turn on simultaneously when operated in parallel. Each relay requires a minimum voltage across the output terminals to function; because of the optical isolation feature, the contact part of the SSR is actually powered by the line it switches. One relay turning on before the other will cause the second relay to lose its turn-on voltage, and it won't ever turn on, or at least not until the first relay fails from carrying too much current.

How does a TRIAC-based semiconductor relay work?

TRIAC is a semiconductor relay (SCR) which can operate in both current directions: from anode to cathode and cathode to anode. An SCR is a four layer diode. It has three terminals, the anode, cathode and gate. It is non-conducting from anode to cathode until a pulse of about 5-50mA is applied to the gate (referenced to the cathode). When

the gate is pulsed, it turns on and conducts from anode to cathode until the current flowing through it drops below a certain level (usually about 20mA). Once turned on by a gate pulse, it cannot be turned off until current stops flowing through it.

Small ones come in TO-92 packages and can handle 500mA-1.5A. Most SCRs come in TO-220 packages and handle 4-8A, but larger ones in other packages are not uncommon. Very large SCRs can handle hundreds or even thousands of amps. Almost all SCRs and TRIACs will work at 100V and most are either 400V, 600V, 1000V or 1200-1500V.

How does a DC semiconductor relay work?

DC semiconductor relays typically have a FET output stage which needs a control voltage to conduct. That control voltage is usually generated using the following method: the LED in control circuit shines to series of semiconductors which generate voltage from light (like solar cells). This generated voltage is sufficient to turn on the output FET. The picture below shows a circuit diagram of basic DC semiconductor relay:

The relay will pass DC current to both directions, because one FET is conductive in one direction when relay is turned on and the protection diode inside other FET will always pass the current through. Because this SSR passes DC to both directions, it can be also used for controlling AC. The commercially available products are more complicated than this, because they have typically a constant current circuitry in the LED input section and some type of current limiting on the output section.

Solid State Relays vs. Electromechanical Relays

In a relay's most basic function, the switching of a load circuit is controlled by a low power, electrically isolated input signal. In the past, Electromechanical Relays (EMRs) have been the component of choice, largely due to price, function, and availability. Now, however, the emergence of semiconductor technology has provided the means to manufacture solid state relays (SSRs) which in many applications outperform their predecessors. Solid state relays provide the advantages of almost infinite switching lives, bounce-free operation, immunity to EMI, higher operating speeds, low level control signals, small package size, and multi-function integration. These advantages can save the design engineer board space, component count, time and money while improving product life, performance, and reliability. Solid State Optronics, Inc. has been a leader in the design, development, and production of low cost, high performance SSRs over the past 15 years. SSO offers a wide range of MOSFET & SCR based relays, complemented by a growing selection of multifunction telecommunication components. By reducing cost and package size while increasing function and performance, solid state relays from SSO now serve as a viable and superior option to electromechanical relays.

This application note will compare the operation of a typical EMR to that of a solid state relay, and examine advantages of each in different types of applications. The Electromechanical Relay

The Figure shown below illustrates the topology of a typical electro-mechanical relay. An input voltage is applied to the coil mechanism. The input voltage magnetizes the core which pulls the arm towards it. This action causes the output contacts to touch, closing the load circuit. When the input voltage is removed, the spring lever will push the contacts away from each other, breaking the load circuit connection.

Inherent in its design, the EMR must make mechanical contacts in order to switch a load. At the point of these contacts, oxidation breakdown occurs over extended life cycling (typically 106 operations), and the relay will need to be replaced. When an EMR is activated, bounce occurs at the contact site. Bounce creates a window of time where the load circuit is flickering between open and closed, a condition which may need to be considered in load design. Because there are internal mechanical components with physical dimension restraints, the package size of an EMR can limit the size of a PCB design. Isolation voltage is another area where EMRs are limited. Most EMRs are typically rated for minimum input to output isolation voltages of 1500 to 2000 VAC.

The Solid State Relay

The Figure shown below illustrates the topology of a typical 1 Form A, MOSFET based SSR. An input current is applied to the LED, which in most cases is a Gallium Arsenide (GaAs) infrared LED. The emitted light is reflected within an optical dome, generally constructed of a gel-like lensing material, onto a series of photo diodes. The photodiodes generate a resulting voltage which, through driver circuitry, is used to control the gates of two MOSFETs.

All of the components are fabricated out of semiconductor material and as a result, the solid state relay combines many operational characteristics not found in other types of devices. Because there are no moving parts, solid state relays have established switching lives of more than 1010 cycles, and exhibit bounce-free operation. The input LEDs require low signal levels (<5mA) to guarantee operation, making SSRs ideal for TTL and CMOS controlled circuits or products where low power consumption is a necessity. The input to output isolation of solid state relays is determined by material properties of the molding compound and flensing material. These properties allow for minimum isolation voltages of 2500 VAC and up to 5000 VAC in some cases. As semiconductor technology has developed smaller and smaller components, the overall package size of solid state relays has shrunk, allowing the designer to conserve PCB space, and makes them valuable in PCMCIA applications.

SSR Output Types

The most common SSR output type is the low threshold MOSFET. Low threshold devices are more easily controlled by the driver circuitry, and allow for fast turn-on times (<500mS). Design of the driver circuitry also permits fast MOSFET gate discharge, translating into quick turn-off times (<100mS). Two MOSFETs inversely connected in series allows for bi-directional control of DC and AC signals with frequencies into the RF range. Typical blocking voltages range from 250Vpp to 400Vpp, with continuous loads of up to 300mA.

A second type of SSR output is the silicon controlled rectifier (SCR). This type of output is designed for AC loads only, and exhibits tight, zero-volt switching. High dV/dt characteristics allow this type of device to control highly inductive loads (PF > 0.3), and driver circuit design prevents false triggering. Typical blocking voltages range from 400Vpp to 700Vpp, with continuous loads of up to 1.2Amps.

SSRs vs. EMRs

By the nature of design, one can see the differences between an electromechanical relay and a solid state relay. In an effort to demonstrate inherent advantages of each type of relay, the following characteristics should be examined: Service Life, Reliability, Isolation Voltage, On Resistance, Capacitance and Package Dimensions.

• Service Life: Because of solid state technology, the SSR definitely exhibits a longer operational life. Since there are no moving parts to jam, degrade or warp, the life span is virtually infinite.

• Reliability: During initial operation, both types of relay will exhibit similar levels of reliability. Over time, however, the solid state relay will gain the edge for the same reasons it has a longer service life, there are no moving parts. Also, bounce free operation increases reliability and ensures consistent load control.

• Isolation Voltage: Again, by the characteristics of construction, the solid state relay will almost always exhibit higher input to output isolation voltages than an electromechanical relay. For many telecommunication applications, a minimum of 3750VAC is desired, clearly making the SSR the optimal choice in telecom design.

• On Resistance: Electromechanical relays have an On Resistance in the range of 100 milliohms, whereas SSRs have an On Resistance in the range of 10 Ohms. The higher On Resistance of SSRs is due to the nature of the MOSFET. The low On Resistance of the EMR allows for greater load current capability and less signal attenuation.

• Output Capacitance: Electromechanical relays typically have an output capacitance of less than 1 picoFarad, whereas SSRs typically have a capacitance of greater than 20 picoFarads. Capacitance becomes an issue in high frequency signals, and EMRs are a better option for HF applications.

• Package Dimensions: The internal components of the relays control the overall package dimensions. Because there are mechanical parts (coil, core, arm, contact lever arms, spring mechanism) within the EMR, the package size is limited to the physical dimensions of functional internal components. The SSR on the other hand, is limited to only the size of the semiconductor components, and is clearly capable of being manufactured in a much smaller package.

Why Solid State Relays?

Although there are advantages to using both types of relays, solid state relays are fast becoming the better choice in many applications, especially throughout the telecommunication and microprocessor control industries. The high reliability and long life mean less field failures and better product performance. Low input signal levels are ideal for TTL or CMOS applications, and less power consumption translates to longer batter life in portable devices.

The overwhelming advantages of solid state relays lie in the isolation voltage, package dimensions and multifunction capabilities. These advantages are increasingly becoming apparent in the telecom industry.

For most tele-communication applications, especially those in Europe, high input to output isolation voltage is required. Typical standards require a minimum input to output isolation voltage of 2500VAC. Not only do solid state relays meet these requirements, most far surpass them.

The Figure presented below shows a comparison between the package volumes and footprints of an EMR and an SSR.

From the figure, it is evident just how much board space can be saved by using a solid state relay. The lower height lets the solid state relay easily fit into PCMCIA applications, making it ideal for laptop and palmtop modems. The smaller footprint means less board space, allowing more real estate for other components, and creating fewer design restraints. Finally, multifunction capabilities place SSRs in a class by themselves. Semiconductor technology has allowed the fabrication of small, multi-purpose telecommunication relays where one device can handle both hook switch and loop current or ring detect functions. Even more complex is a device which combines a 1 Form A relay, Optocoupler, Darlington Transistor, and Bridge Rectifier all within a small, 16 pin SOIC package. These multifunction relays give the design engineer unparalleled flexibility in developing new and innovative fax/modem products.

Cost Issues

In the past, there has been a rather large gap between the price of an electromechanical relay and the price of a solid state relay. For a basic 1 Form A SSR, the price was as high as several dollars more than an EMR. With continual advancement in manufacturing technology, this gap has been reduced dramatically making the advantages of solid state technology accessible to a growing number of design engineers.

Conclusion The future of solid state relays only looks bright. With further advancement in semiconductor fabrication and manufacturing technology, increased performance and functionality will emerge. Already, MOSFETs are being fabricated with On Resistance values of less than one(1) Ohm. As these devices become more readily available, low On Resistance will no longer be a deciding factor in choosing an EMR over an SSR. As semiconductor components become smaller, package dimensions will also decrease. These same advancements will mean that the price gap between SSRs and EMRs is also going to decrease.

By investing in semiconductor fabrication and manufacturing technology, Solid State Optronics, Inc. is positioned to continue leading the industry with innovative, low cost, high performance solid state relays for a growing number of applications.