Hello,
In IEEE1584 the expression trip time and opening time are used. Trip is as I understand a combination of desired time delay and the slowness in the relay it self. The opening time is the time the circuit breaker use for opening it's main contacts(which is varies a lot, some breakers only use 22ms other breakers 80ms). And then there is arcing time, which also is variable sometimes 18 ms, for some only 10-15ms.

My questions is should the opening time include the opening time + arcing time= Total interruption time?

Super happy for any response!

1584 uses "arcing time" in the equations for incident energy, so this must be the time a worker is assumed to be exposed to the arc. It is either the total interruption time or an upper limit such as 2 seconds.

In breaker terminology, "arcing time" is the time from initial contact parting to current interruption, and is included in the breaker's interrupting time. See Figure 2 of C37.010.

There is some room for terminology standardization here.

We had a few inconsistencies in the 2002 IEEE 1584.

Just had a few phone meetings this morning as we move towards addressing comments, cleaning up the next draft of the next edition etc. Trying to clean up some of the text. Arcing time / duration was selected as the IEEE 1584 term to mean how long the arc flash lasts as JBD pointed out. It is usually based on how long it takes an overcurrent device to trip and clear the arc flash (or 2 seconds)

From IEEE C37.100 and IEEE 100:

arcing time (of a fuse): The time elapsing from the severance of the current-responsive element to the final
interruption of the circuit.

arcing time (of a mechanical switching device): The interval of time between the instant of the first initiation of the
arc and the instant of final arc extinction in all poles.

Since arc flash is linked to breakers and fuses, it would seem appropriate (and safer) to use the same definitions. Since arcing duration remains undefined in the two dictionaries, I suggest 1584 use that term exclusively for arc flash exposure time.

For a lot of breakers arcing time varies from 0 to 1/2 cycle which is entirely dependent on the phase angle at which opening completes and depends on when the next current (not voltage) zero occurs at which point if restriking has been eliminated, the arc fully extinguishes, so the arcing time is usually given as either a worst case or an average over this time range. This obviously applies almost as an absolute when vacuum is the breaker medium. In air the arc has to elongate within the arc chute to the point where it quenches and although it's pretty fast this time is usually longer than it is in a vacuum interrupter but there can also be a puffer device that mechanically blows the arc out shortening the time to at least theoretically possibly even faster than vacuum. Similar arguments apply with alternative gas media such as SF6 and oil (which is really extinguishing inside hydrogen gas bubbles). There are also some breakers on the market particularly for utilities that attempt to minimize contact wear and other negative effects such as transient recovery voltage by timing the opening such that opening time almost coincides with a current zero so that the arcing time is almost eliminated. In this case the arcing time uncertainty is traded off for opening time uncertainty since the breaker electronics vary the timing for coil energizing to control when contact opening actually completes but obviously in this case the uncertainty is still there...it just shifted to the contact opening time part of the estimate. From speaking to an interrupter engineer personally when discussing the Federal Pacific Autojet medium voltage disconnect switch, their design uses a mechanical bellows for the puffer. It takes about 100 milliseconds to inflate the bellows as the main cam is moving and the disconnect is opening, then the bellows rapidly accelerates right at the point of opening of the main contacts and blows the arc out in about 1 cycle. This also seems to coincide well with similar designs for air break based breakers that I've worked on and may explain why the 80 millisecond delay pretty much has to exist in order for this design to work, although there is no reason that the bellows can't be charged during the closing or even charging cycle. This is also an inspection point as many times the puffer tubes or bellows are damaged or missing due to age or mechanical damage.

Another major factor in terms of contact opening time is whether the breaker is fired via a DC control signal or AC control signal. Although substations generally across the board operate on DC, particularly 125 VDC systems, the same is not true of say class E1/E2 medium voltage motor starters or even some low voltage starters on larger motors, synchronous motors, and wound rotor motors, where typically relaying is operated off of power derived from a PT or CPT. In this case the control signal and breaker/contactor coil is AC rather than DC so depending on the phase angle at which the interrupter coil is fired again the delay can be 0-1/2 cycle in theory but is almost always stated as a worst case 1 cycle. With DC systems switching can be arbitrarily fast but the specifications almost always give a 4 millisecond delay, roughly twice to 4 times as fast as the AC equivalents.

Opening time with breakers is mostly dictated by inertia. The required motion is to move the relatively heavy contacts and frame over a certain amount of travel distance including smooth acceleration as well as deceleration (to avoid contact bounce and restriking) in as short of a time as possible. As the power contacts increase in size and/or the travel distances increase, obviously the time involved increases substantially. At the low end ignoring some very exotic devices such as triggered current limiters that use explosives to blow a busbar in half with opening times of around 1-2 milliseconds, almost all small miniature and even low current (<100 A) molded case breakers mechanically break arcs in 1 to 1.5 cycles. Oil and larger air-break circuit breakers in particular gradually slow down to as long as 5-10 cycles. Vacuum breakers only need about 1/4" of movement and are relatively consistent across the board in terms of speed with about 1-2 cycles for contact opening plus a worst case 1 cycle for arc quenching giving either a 2 or 3 cycle breaker design almost universally.

So this may help explain the inner workings of circuit breakers and why the opening and arcing times are what they are. Designs have been slowly moving towards lighter and thus faster designs over time. Roughly 20-30 years ago we saw the rise of reliable vacuum interrupters and a gradual increase in their current handling capability. Between this and the rise of SF6 designs it effectively ended oil breakers across the board. Air breakers have been relegated almost exclusively to small or low voltage designs with vacuum (below 40 kV) or SF6 (above 40 kV) displacing virtually all other technologies. The most recent "big thing" at least in medium voltage breakers is the rise of the magnetic actuator which has only a single moving part (ignoring the ABB Magvac version).

The next big thing may be totally eliminating moving parts.There have been a very small number of experimental and commercial AC solid state breakers produced but so far market penetration has been almost nonexistent. These breakers "trip" in submillisecond speeds. In fact they don't just trip...they can also soft start/soft stop (current limit/modulate) and they can reclose, and even test fire pulses of voltage to actively monitor line conditions. What makes this particularly interesting in an arc flash context is that the thyristors simply CANNOT conduct a typical bolted or even arcing fault level of current and must be modulated to maintain close to name plate current. Thus 35k or 100k AIC solid state breakers are infeasible in the first place. They will simply current limit to something more along the lines of 1-2 kA irrespective of the available fault current. Needless to say this means that between the speed and current limitations, arc flash simply ceases to be a major concern. There are three downsides. The first one is that a solid state breaker in and of itself simply cannot provide mechanical separation of the power conductors so it is not allowed as a LOTO device per OSHA requirements just as a vacuum interrupter isn't either and on top of that it does not provide a visible break so it does not meet utility standards. Thus just as with vacuum breakers a separate (potentially non-load break) disconnect is required for LOTO purposes. The second downside if you want to call it that is that semiconductors frequently fail in a dead shorted state. This would seem to be a nonstarter and the reason that we usually have to protect semiconductors with high speed fuses or breakers in the first place. However snubber circuits are very effective at avoiding very fast dv/dt rise times which gives the semiconductor time to switch and when comparing semiconductor interruption to air/SF6/oil/vacuum the other interruption technologies are also subject to contact welding so as long as the semiconductor switches can be designed to provide equivalent reliability this concern should not be an automatic death sentence. We are after all talking about a circuit breaker vs. a drive which should be two entirely different designs. Finally the third concern is self-commutation. Generally the transient overvoltage limits of SF6 and air breakers are extremely high. With vacuum breakers surge protection (MOV's) are a necessity due to the very short vacuum gap, and this becomes even more critical with solid state breakers because at fairly low voltages all semiconductor switches undergo a phenomena called self-commutation which is just as bad as it sounds...it goes into conduction without any gate power at all. Worse still they can also do it at much lower voltages with fast rise times...hence again the reason that fast rise time protection is also a necessity.

As unlikely as this sounds, solid state breakers are probably not just a scientific curiosity. I believe they are coming relatively soon and might be just as disruptive as vacuum interrupter technology if not more so. They have already begun making inroads in aircraft and military applications particularly with small and medium size DC breakers which have traditionally been very challenging for mechanical designs and AC designs are coming down quickly in price. Two other aspects of it is that since it is solid state, there is no "contact wear" and thus the need for two different components in a motor starter (breaker and motor starter) goes away. On top of that there is the soft start/soft stop aspect, and the breaker itself is already solid state and electronic with on board microprocessor and potentially I/O or networking. Thus every component of both magnetic and soft starters can be replaced by a single component.