assume i'm not too bright.. i know its a stretch, but can someone briefly explain the difference between the series fault and the parallel fault and why the afci is lacking in the former? i copied this from a previous thread "we monitor arcs to ground for the parallel arcs in NM-B cable at a 30mA level, and MOV's on the circuit board protect from surges", i dont get that? do you? also do we use the same smoke detectors with the backup battery on afci circuits, the battery doesnt make any difference, right?
If the smokes are on the same ckt as the AFCI,which I guess the ones in the bedroom would have to be(NEC2002)and the AFCI tripped the battery back up operation would power the smoke. It is more difficult to determine/detect a series arc because the load is still in series with the arc and may be drawing a significant amount of the total current leaving a small amount of relative current for Arc detection for the micro-processor located in the AFCI.
Chris; you'll find that even Zlan's site beats around the bush concerning series arcs Cindy, actually your brighter than the manufacturer's consider the average sparky. A 'series' arc is simply a heated termination, a 'parrallel' arc is one that goes to ground, the contention of the manufacturer's is that all series arc's are not dangerous, and progress to ground eventually. There is some good discussion Here Basically an arc-fault is a glorified GFPE. Never has the NEC been so buffalo'd......
In addition to the Arc Fault sensing circuitry, all 4 manufacturers of AFCI devices have ground fault protection in their design. This is not 5mA personnel protection, but equipment level 30mA protection. By adding ground fault protection, we are able to greatly expand the protection that these devices provide.
In the AFCI, we look for short duration sporadic high-energy spikes of current. I've gone into detail in some previous posts if you are interested in reading how we analyze the current waveform. High-energy parallel arcs (arcs between conductors at different potentials such as line and neutral) conduct an amount of current determined by the available short circuit at the fault and the magnitude of the arc voltage. UL has determined that the minimum available short current at a residential receptacle outlet to be 75-amperes RMS or about 100-amperes peak. When an arcing condition develops instead of a "short circuit" the arcing current can be as low as 50-amperes peak, much lower than the peak inrush current of normal household appliances. The sporadic and small peak arcing current result in an RMS current that can be less than the handle rating of the breaker or fuse and the small peak current is below the magnetic trip level of a circuit breaker. We demonstrate parallel arcs by cutting across an energized SPT (lamp) cord. The result is a dramatic spark and expulsion of molten copper. If done in the presence of a combustible material, the result is ignition of the material and is clearly an unsafe condition.
It should be noted here that a "series arc", defined as current that would jump from a break in only one wire, would only carry the amount of current that the load on that circuit is pulling. This is a normal, safe arcing condition that occurs whenever a switch turns an electrical load such as a light off or on. I’ve had people in the industry tell me that they have "tested" AFCI's by wiring them up, and taking one of the wires and cutting it, and making and breaking that wire repeatedly, noticed an arc, and stated that the breaker did not trip. By making and breaking one wire, all that these folks have done is created a switch in the circuit similar to a light switch. If you were to measure the current and voltage of making and breaking the wire, and do the same with teasing a light switch, they would look the same. Clearly, we do not want circuit breakers to react to the operation of light switches. The amount of current that jumps across this opening is equal to the amount of current drawn by the load. At 120 volts, and 300 watts of load, the amount of current is 2.5 amps, which is not enough to cause a fire. If you want to test this, surround the broken wire with cotton, and do the test...nothing will happen.
Ok, so parallel arcs are dangerous and series arcs are not...but, high resistance series FAULTS are another matter. If you want to create one of these, simply plug a 1500 watt space heater in to a receptacle, loosen the line or neutral terminal at the receptacle and carefully jiggle the wire creating sparks. After five or ten minutes a glow will develop at the copper wire-to-receptacle screw interface then let of the wire. The repeated sparking causes a build-up of copper oxide, which is a semiconductor. Once formed this connection unfortunately is mechanically stable and will remain even if the space heater is turned off. You have now created the hazardous "glowing contact". Due to the characteristics of this connection the result is that all of the current will flow through this very fine strand of material. The resulting high resistance produces a voltage drop of about 2-volts which when multiplied by the current results in a connection power dissipation of over 20-watts causing the wire to glow like a toaster.
The resulting heat generated by this glowing contact will cause melting of the wire insulation, and the plastic material of the receptacle device. Since there is not an increase in current, the standard thermal magnetic device will not recognize this unsafe condition. However, by protecting this circuit with an AFCI with equipment ground fault protection, we can respond by either sensing ground current caused by the melting and shifting of components within the device, or by sensing a parallel arc when one of the damaged conductors comes in contact with another. If you try this with an AFCI device, you will find that it will respond and de-energize the circuit. If you try it with a standard thermal magnetic circuit breaker or fuse, it would be advisable to have a fire extinguisher handy. The bottom line is that AFCI responds to connection issues like loose terminals or loose wire nuts, loose aluminum wire connections, where standard thermal-magnetic breakers or fuses do not. This is the next step in circuit breaker technology.
Your reasoning is sound. Series faults are a problem. Series arcs are not. When you break a single conductor, the circuit is interrupted which turns off the load. It is self-announcing to the homeowner...the light flickers, or goes off completely. Series faults, as I described, are the issue with wire nuts, loose connections, aluminum wire issues, etc. I don't think I will be able to convince you, so please try the tests as I recommended with an AFCI and with a standard breaker or fuse.