Shortstub
Electrical
- Dec 30, 2002
- 268
I am admitted technocrat... one, who upon learning that 4,378,573 people were married last year, asks the question "why is it not an even number!" In other words, I'm a detail nut... requiring factual explanation of electrical failures.
Now I would like to respond to the 3 M's... misconception, misunderstanding, and malarkey... existing in past threads covering "electrical faults!" Initially I would like to address: a) ungrounded systems; b) arcing ground-faults; c) unwarranted fuse blowing; and any others that posters agree will fit the thread. This post will start with the dreaded "ungrounded systems. Others will follow:
a) Ungrounded Systems.
First of all it's a misnomer. While it is true that such systems are not deliberately connected to a grounded reference point, their phases are still capacitively connected to "ground." All electrical components have ground capacitance... motors, heaters, power-factor correction capacitors, surge arrestors, bus duct, transformers, and the most significant, cables. And, as long as capacitance is equal on all three phases, then the system ground becomes the electrical-neutral-point. As a result, all three phase-to-ground voltages are equal, i.e., Vpp/sqrt3.
A little history. I cut my engineering teeth in one of those industries that employed an ungrounded 480V system. Justification was based on the "false belief" (quotes are mine) that "an inadvertent ground-fault won't cause an undsirable outage!" See NEC for exact verbage. It was anr old plant, 50+ years. Yes, all substations were equipped with ground-detectors in the form of wye-connected 480-V bulbs connected between phase and ground! One day, I noticed all bulbs were out in one substation, but only two fuses blown. While the ph-to-ph voltages were the same, the grounded phase, say 'A' showed 0-V, but the "sound" phases 'B' and 'C' showed 800-V, and 1,200-V, respectively. Furthermore, I learned that everytime this happened, 3, 4, sometimes more, motors would fail within 1 to 3 weeks.
Now about the intermittent arcing phenomena... While I agree that intermittent phase-to-ground exacerbates the problem, I assure you, arcing is not involved, at least not by my definition. The root-cause is transient-overvoltage, not current. In short, the 'sound' phases are exposed to extreme overvoltages, resulting in insulation breakdown. As a consequence, a phase-to-phase fault occurs (now current is involved), blowing fuses or tripping breakers.
Several observations were made in the thread regarding ungrounded systems that didn't exhibit such failure. It has nothing to do with luck, nor good-maintenance. The parameters that lead to this phenomena are very large cable systems, hence large ground-capacitance and old systems having degraded system insulation. One mention was made about a Navy system being OK, but I have two observations by Navy personnel with just the opposite observation. Also, I would venture an educated guess that such a fault receives very prompt response, compared to land-based systems.
Regarding the plant I was in. A change was made to a grounded system and motor failures disappeared.
BTW, this phenomena also explains most of the mysterious MOV failures on VFD systems.
Now I would like to respond to the 3 M's... misconception, misunderstanding, and malarkey... existing in past threads covering "electrical faults!" Initially I would like to address: a) ungrounded systems; b) arcing ground-faults; c) unwarranted fuse blowing; and any others that posters agree will fit the thread. This post will start with the dreaded "ungrounded systems. Others will follow:
a) Ungrounded Systems.
First of all it's a misnomer. While it is true that such systems are not deliberately connected to a grounded reference point, their phases are still capacitively connected to "ground." All electrical components have ground capacitance... motors, heaters, power-factor correction capacitors, surge arrestors, bus duct, transformers, and the most significant, cables. And, as long as capacitance is equal on all three phases, then the system ground becomes the electrical-neutral-point. As a result, all three phase-to-ground voltages are equal, i.e., Vpp/sqrt3.
A little history. I cut my engineering teeth in one of those industries that employed an ungrounded 480V system. Justification was based on the "false belief" (quotes are mine) that "an inadvertent ground-fault won't cause an undsirable outage!" See NEC for exact verbage. It was anr old plant, 50+ years. Yes, all substations were equipped with ground-detectors in the form of wye-connected 480-V bulbs connected between phase and ground! One day, I noticed all bulbs were out in one substation, but only two fuses blown. While the ph-to-ph voltages were the same, the grounded phase, say 'A' showed 0-V, but the "sound" phases 'B' and 'C' showed 800-V, and 1,200-V, respectively. Furthermore, I learned that everytime this happened, 3, 4, sometimes more, motors would fail within 1 to 3 weeks.
Now about the intermittent arcing phenomena... While I agree that intermittent phase-to-ground exacerbates the problem, I assure you, arcing is not involved, at least not by my definition. The root-cause is transient-overvoltage, not current. In short, the 'sound' phases are exposed to extreme overvoltages, resulting in insulation breakdown. As a consequence, a phase-to-phase fault occurs (now current is involved), blowing fuses or tripping breakers.
Several observations were made in the thread regarding ungrounded systems that didn't exhibit such failure. It has nothing to do with luck, nor good-maintenance. The parameters that lead to this phenomena are very large cable systems, hence large ground-capacitance and old systems having degraded system insulation. One mention was made about a Navy system being OK, but I have two observations by Navy personnel with just the opposite observation. Also, I would venture an educated guess that such a fault receives very prompt response, compared to land-based systems.
Regarding the plant I was in. A change was made to a grounded system and motor failures disappeared.
BTW, this phenomena also explains most of the mysterious MOV failures on VFD systems.