Any device that uses power semiconductors to control AC power can be damaged by transient voltage spikes on the AC line. Caused by electrical activity elsewhere in the power system, these spikes last only nanoseconds, but their magnitude may be great enough to destroy semiconductors. Manufacturers of solid state motor starters and SCR controls therefore always include some type of protection against transient spike damage in their product designs.
SCRs and diodes are both unidirectional semiconductors: they conduct from anode to cathode and block voltage from cathode to anode. When a transient voltage spike exceeds the cathode-to-anode blocking capability, known as the peak inverse voltage (PIV) rating, the semiconductor will fail shorted, similar to a welded contact in an electromechanical contactor. Power company studies show that such voltage spikes are common occurrences, and their magnitudes are neither predictable nor limited. The long-term reliability of any solid state control is therefore directly dependent on the effectiveness of the transient voltage protection incorporated in its design.
Transient voltage protection methods can be grouped into three general categories:
- Selecting the PIV rating of the semiconductors to withstand transients without being damaged;
- Using a suppression mechanism to shunt transients around semiconductors;
- Self-protection, using an SCR characteristic to prevent destructive inverse voltages from being impressed on the semiconductors.
A. Peak inverse voltage ratings
This method of transient protection is the simplest and, at the same time, the most limited. SCRs and diodes are selected to have a certain minimum PIV rating, which is “guess-timated” to be sufficient to withstand the transient spikes to which they may be subjected. Since there is no way to predict the magnitude of the transient spikes that may be encountered, there is no sure way to select a PIV rating other than pure guesswork. Is 1000 PIV enough? 1200? 2000? Quite simply, there is no way to know.
Transient voltage spikes of 3 kV or greater are not uncommon on industrial power mains, while the PIV ratings of power semiconductors range from 600 to perhaps 2000 V. Clearly, positive transient protection requires more than just the selection of a discrete PIV rating in the power semiconductors.
B. Suppression
Suppression schemes use metal oxide varistors (MOVs) to shunt transient spikes around blocking semiconductors, limiting the inverse voltage impressed on them. MOVs are bidirectional solid state switches that block voltage until their threshold level is reached, at which time they switch to a conducting state and remain there as long as the applied voltage stays above that threshold. Connected in parallel to an inverse-parallel pair of power semiconductors, and selected for a turn-on threshold below the semiconductor PIV ratings, MOVs will limit the inverse voltage levels impressed on those semiconductors.
The limiting factor of any suppression scheme is the total energy rating of the MOVs used to shunt the transient spikes. The greater the spike magnitude, the greater the energy content, so the greater the MOV energy rating must be to handle it. Should that rating ever be exceeded, MOVs fail, sometimes literally exploding, leaving the semiconductors unprotected against the next transient event. So what energy rating is required? Since there is no way to predict spike magnitudes, the energy rating selection is pure guesswork. No matter what value is finally selected, a spike with a higher energy content is not only possible, but likely.
C. Self-protection
The greatest degree of transient voltage protection comes from a non-destructive forward turn-on characteristic of all SCRs known as voltage breakover (Vbo). When a forward-blocking SCR is subjected to a high enough forward voltage, it will “break over”, turning on without being commanded to do so by the control circuitry. Proper coordination of the Vbo and PIV ratings of an inverse-parallel pair of SCRs ensures that, in the event of a transient voltage spike above a certain level, the forward-facing device turns on non-destructively to shunt the spike around the reverse-facing one before the latter’s PIV level is reached. This method is known as Vbo clamping.
Vbo clamping ensures that the inverse voltages impressed on a semiconductor will never exceed a predetermined level, and that the semiconductors themselves carry the full energy of a shunted transient spike. Protection against system-generated transient spikes is ensured, regardless of PIV ratings.
Summary
The reliability and long-term viability of a solid state control in an industrial environment is in large measure determined by the effectiveness of the transient voltage protection incorporated in its design. The methods used are:
- a. PIV ratings
- Semiconductors are selected for their high reverse-blocking capability. The problem: how high is high? PIV ratings range from 600 to 2000 volts; regardless of the PIV selected, there is a good chance a higher transient will be encountered.
- b. MOVs
- Transients are shunted around semiconductors so destructive inverse voltages are not impressed on them. The problem: if a high-magnitude spike exceeds the MOV energy rating, the MOV will fail, eliminating all transient protection. Selection of MOV energy ratings is pure guesswork, since there is no way to predict spike magnitudes.
- c. Voltage breakover
- The forward turn-on characteristic of SCRs is coordinated with PIV ratings to cap the maximum inverse voltage impressed on semiconductors below destructive levels, at spike magnitudes greater than PIV ratings. Positive, repeatable protection against any normal system-generated spikes is provided.