How did you come to this conclusion? His whole post, including the title, concerns sparks flying. You are correct about GFCI's forte', but if everything was hooked up properly and the GFCI was the source of the stray current, this is exactly the scenario that they should take care of (current being sourced by the GFCI which isn't being returned to the device).
A GFCI outlet monitors for a current imbalance between the hot and neutral wires and breaks the circuit if that condition occurs. That "imbalance" isn't exactly the perfect description though. Because ANY power directed into the neutral or the ground will trip a CGFI. Which is exactly what the "reset button" on the outlet is designed to do: Short directly between the hot and ground terminals inside the outlet. Just like dropping a hair dryer into a bath tub full of water.
A circuit breaker usually will trip if you receive a shock, but it may not act fast enough to protect you from harm. A GFCI outlet is more sensitive and acts faster than a circuit breaker or fuse and is thus an important safety feature. Most will trip in as fast as 1/30th of a second. But CGFI's are NOT thermal interrupters and will seldom trip because of amperage over draws.
A circuit breaker works as a thermal switch and is designed to trip when the heat from power draw climbs higher than the breaker is designed to allow. However electricity, just like water or air, always takes the path of least resistance.
Simply explained, a GFCI measures the amount of current that it is supplying, and compares it to what is coming back on the other leg. If there is a difference (usually in the neighborhood of 5 milliamps), it trips out because the current is taking a different route than what it should be. This would be caused by a ground fault. That is why in my earlier post I was conjecturing that the problem was possibly in the conduit prior to the GFCI's and the conduit was poorly grounded. If the conduit has a high-resistance ground, it can fail to create enough fault current to trip the breaker. If the fault was downstream of the GFCI, the GFCI should have tripped instantly when there was a difference in the outbound vs what was being returned to it.
That is a fair explanation of how a CGFI works, and is pretty close to the description given by manufacturers. But it isn't exactly what a CGFI actually does.
As I understand the situation:
The room is supplied @ the main with a 20 amp breaker which was used in construction to supply power to 1 or 2 rooms with limited outlets/lighting, so it is either 14-2 or 12-2 wire. AND it's run inside of metal conduit.
MikeRoOrganix tapped into the circuit at a junction box with heavy-duty extension cord, which is most likely 12-3 wire, but could possibly be 10-3 if he paid a premium for it.
The design of the room as described, is foil faced foam-board. Which of course is a capacitor in and of it self. (Read the bottom quote to understand how a capacitor works).
Now the reason I gave the previous response is as elementary as the ABC's.
Wires getting hot.
The air pump shorted and fried.
Ballast arcs and smells.
The room is constructed in such a manner as to create a capacitor. AND isn't grounded. Thus the "OZONE odor" mentioned in the original post.
The wiring at the end of the circuit is much smaller and will become the weakest link. Heat=resistance = more heat. When the circuit is under draw it begins heating the wiring all the way back to the source, heating the smallest wires and weakest connections first. Since the majority of the wiring is encased inside of metal conduit, the heat cannot radiate disperse. As the wires continue to heat, and wire-nuts get hot, both the wires and the connectors oxidize, which creates a capacitance gap.
Also: Appliances and electrical equipment that is operated on lower voltage or amperage than it is designed to operate, overheats and wears out faster. But it's the heat that causes problems, as it melts the plastic components inside that hold wrappings and brushes in place. The discharges being described are indicative of appliances operating at below normal or optimum power, creating heat which is causing resistance, resulting in more heat and then discharging either through thin, worn cords or appliances, OR wire connection that are corroded (oxidized) from heat.
A capacitor (formerly known as condenser) is a device for storing electric charge. The forms of practical capacitors vary widely, but all contain at least two conductors separated by a non-conductor. Capacitors used as parts of electrical systems, for example, consist of metal foils separated by a layer of insulating film.
A capacitor is a passive electronic component consisting of a pair of conductors separated by a dielectric (insulator). When there is a potential difference (voltage) across the conductors, a static electric field develops across the dielectric, causing positive charge to collect on one plate and negative charge on the other plate. Energy is stored in the electrostatic field. An ideal capacitor is characterized by a single constant value, capacitance, measured in farads. This is the ratio of the electric charge on each conductor to the potential difference between them.
Capacitors are widely used in electronic circuits for blocking direct current while allowing alternating current to pass, in filter networks, for smoothing the output of power supplies, in the resonant circuits that tune radios to particular frequencies and for many other purposes.
The capacitance is greatest when there is a narrow separation between large areas of conductor, hence capacitor conductors are often called "plates", referring to an early means of construction. In practice the dielectric between the plates passes a small amount of leakage current and also has an electric field strength limit, resulting in a breakdown voltage, while the conductors and leads introduce an undesired inductance and resistance.