Transient voltage spikes can harm solid-state starters.

Nov 1, 1996 12:00 PM

Why do these devices, a though generally reliable, sometimes fail unexpectedly?

Since their introduction in the late 1950s, solid-state contactors and starters have taken over many functions that traditional electromechanical devices (EMDs) either couldn't handle or couldn't perform as well. These electronic devices are now used in many reduced-voltage, reversing, and high-cycling applications, where high reliability and long-term economy are required.

Unlike their electromechanical cousins, whose normal life span is approximately 10 million mechanical operations and 1 million electrical operations, solid-state devices have been tested to as many as 200 million operations.

Beware of transient voltage spikes

Theoretically, under ideal conditions, solid-state devices should be good for billions of operations. The key words here are "ideal conditions." Although generally very reliable, solid-state devices can be the victims of heat and load faults. Perhaps the most common threat is from transient voltage spikes, as shown in Fig. 1, that many times reach 10,000V or more.

Solid-state devices have a silicon-controlled rectifier (SCR), which is the electronic equivalent to the contacts in an EMD. But unlike the contacts in EMDs that are not affected by transient voltage spikes, solid-state devices cannot take this punishment. In essence, any solid-state device is only as good as its ability to handle transient voltage spikes.

Use voltage breakover clamping to prevent damage

The best way to overcome this problem is by a technique called [V.sub.bo] (voltage break-over) clamping. Voltage spikes can be generated from the line and the load side, either one having the capacity to destroy the integrity of the solid-state device. [V.sub.bo] clamping is a very effective method of dealing with these spikes by passing them harmlessly through the unit. Fig. 2 represents one leg of the power circuit of a 3-phase solid-state device equipped with [V.sub.bo] clamping and illustrates this technique.

SCRs have many of the same properties as a diode. One of the properties is that they are designed to pass voltage in only one direction. By definition, they have a peak inverse voltage (PIV) rating, which represents the maximum voltage in the non-conducting direction that they can resist before they are destroyed.

What protects the SCR in Fig. 2 from load-side voltage spikes is the opposing diode. Since it continuously conducts in the opposite direction, it allows load-side spikes to pass through to the mains before destructive PIV levels can build up against the SCR.

For line-side spikes, however, it's the diode that's in danger of being destroyed and must be protected. The [V.sub.bo] clamping circuit is designed so the voltage breakover of the SCR (the threshold voltage at which the device will begin to conduct without a gate signal) is lower than the PIV rating of the diode. This allows line-side spikes to pass through the SCR and into the load, thus protecting the diode. The SCR and diode are integral parts of the [V.sub.bo] clamping technique. One serves to protect the other.

There are other less effective ways of protecting the SCR from transient voltage spikes. One is to shunt the spikes into metal oxide varistors (MOVs) designed into the circuit. However, there is a possibility that once the upper energy rating of the MOV is surpassed, it can fail catastrophically and be potentially dangerous.