Everyone in the solid state control industry acknowledges that short circuits will destroy unprotected power semiconductors. There is almost no
agreement, however, on what constitutes protection: different manufacturers employ different methods with very different degreesof effectiveness.
The result is confusion among users of this technology as to
a) whether short circuit protection is actually possible, and
b) the relative effectiveness of the various means of protection.
The following discussion is intended to clarify these issues.....................
When a short occurs in an electrical circuit, extremely high currents flow. Because these currents will be much greater than circuit design parameters,
they can damage circuit components, and pose fire and other safety hazards. Protection mechanisms such as fuses and circuit breakers are designed
to be the weak link: they break the circuit to minimize or prevent such damage. While simple in concept, the science of short circuit protection is a
complicated one, requiring careful coordination of component withstand ratings and the interrupt capabilities of the chosen protective device(s) with
the available fault current.
When all the components in a circuit are electromechanical devices, published technical data can be used to select equipment that will provide the required
degree of protection. Introduce a device using solid state power components (SCRs, diodes, or TRIACS) into the circuit, and short circuit protection
becomes more difficult. In general, for a protective device to prevent damage to a circuit containing these components in a fault situation, it must
a) break the circuit completely in 2 milliseconds or less, and
b) limit the peak and total let-through currents to levels that the semiconductors can tolerate.
However, published technical data on the various protective devices do not provide accurate information on their performance within that
brief time period. Therefore, short circuit protection for power semiconductors cannot be selected from specification information alone.This fact is not
widely recognized even in the solid state control industry, leading to confusion among users of these products over the issue of short circuit protection.
Some manufacturers leave short circuit protection for their products to the user; others use a bewildering variety of fuses, circuit breakers, and electronic
trip circuits, which may or may not actually prevent damage should a short occur. This lack of consensus, a condition existing for years, has led to the
general impression among users that reliable, repeatable short circuit protection is not possible for solid state controls. Users, therefore, often live with a
much higher incidence of short circuit damage to their solid state equipment than necessary simply because they do not know that there is an alternative.
The key to selecting solid state controls that can survive short circuit events lies in knowing the effectiveness (or lack thereof) of the various protection
schemes. The types of short-circuit protection used with SCR controls can be grouped into five general categories:
Conventional protective devices
"Semiconductor", or "I2t" fuses
"2 millisecond" fuses
The control manufacturer makes the user responsible for short circuit protection. Because published technical data on the various protective
mechanisms is incomplete for the brief time intervals which must be considered with solid state controls, it is impossible for the user
to make an effective choice. Should a short circuit occur, semiconductor failure is almost a certainty.
2) Conventional Protective Devices:
These include most fuses and circuit breakers. Most fuses are simply too slow, and allow too high a let-through current, to have any chance
of preventing damage. Because they are mechanical devices with moving parts, circuit breakers also break the circuit too slowly. With either
of these choices, a short circuit almost always means semiconductor failure.
3) "Semiconductor" or "I2t" Fuses:
These fuses are marketed by fuse manufacturers specifically for use with semiconductors. Claims to the contrary notwithstanding, however,
most of these fuses WILL NOT provide reliable short circuit protection for semiconductors. Some allow too much let-through current; others
do not clear fast enough. Whatever the reason, the fact remains that even fuses that are sold for semiconductor protection do not actually
provide it. Remember: published data cannot be used to select fuses that will protect semiconductors.
4) Electronic protection:
Circuitry in the solid state control monitors load currents, and reacts to fast-rising levels (characteristic of a short circuit) by preventing
further semiconductor turn-on, in effect, breaking the circuitto prevent short circuit damage. Even this method, however, does not
interrupt fault currents fast enough: not because it can't react quickly enough (it can), but because it cannot shut off a conducting semiconductor.
Solid state a.c controls are line commutated: the semiconductors only turn off when the voltage waveform goes through zero. Once triggered
on by the control electronics for a given half-cycle of the a.c. waveform, a semiconductor will conduct that half-cycle of voltage and current
EVEN IF THE ELECTRONICS ARE DISABLED. Since a half-cycle of 60 Hz power is 8.333 milliseconds long, while the nominal clearing time
for effective short circuit protection is 2 milliseconds, electronic circuits cannot and do not shut down the control before damage occurs.
5) "2 Millisecond" Fuses:
These fuses are "semiconductor" fuses that have been proven to clear within the nominal 2 millisecond time frame, and limit the
peak currents and total energy let-through sufficiently to protect a semiconductor from short circuit damage. These fuses can only be
selected by empirical testing using specialized recording equipment, not from any published technical specifications. They provide the
only reliable, repeatable, effective protection for power semiconductors against short circuit damage. The clearing characteristics of these
fuses are so special that only fuses from certain manufacturers will work. There are no valid cross-references from one manufacturer
to another. Solid state controls CAN BE PROTECTED against short circuit damage, but not every control manufacturer does so. Because of the
variety of protective devices employed by control manufacturers, extreme care must be taken to specify and select controls that incorporate
those protective measures that have been proven to work. In summary:
-Avoid controls that do not include fuses.
-Avoid controls that utilize "electronic fusing" or "electronic trip" circuits.
-Avoid controls that feature "semiconductor" or "I2t" fuses.
-Consider ONLY those controls that offer "2 millisecond fuses", and which manufacturers can support their short circuit protection claims with certified test results.
From Payne Engineering