bobby ocampo
Senior Member
What is the purpose of equipment grounding for Ungrounded System? Is Equipment Grounding required in UNGROUNDED system if the Overcurrent protectind ve device will not trip on a single line to ground Fault?
Equipment Grounding UNGROUNDED SYSTEM
- Thread starter bobby ocampo
- Start date
- Status
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
It is still required and you are right in thinking it does not provide a ground fault path that will open the OCPD if a GF occurs.
- Location
- Chapel Hill, NC
- Occupation
- Retired Electrical Contractor
Why is it required? Is it in case the system ever goes to a grounded one?
zbang
Senior Member
- Location
- Roughly 5346 miles from Earls Court
How do you detect an unintended ground fault if there's isn't a ground to fault onto? (Not for the OCPD for the required fault detection.)
texie
Senior Member
- Location
- Fort Collins, Colorado
- Occupation
- Electrician, Contractor, Inspector
As you know, the first ground fault will not impose fault current on the metallic components. That is why you ground detectors are required as otherwise it would not be known. But consider what happens if you have an unknown GF in an ungrounded system and then there becomes another GF from another phase somewhere in the system-then the metallic system has to carry fault current current.
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
I wonder how common it is to get GF's on two separate phases on ungrounded 3 phase systems.
hillbilly1
Senior Member
- Location
- North Georgia mountains
- Occupation
- Owner/electrical contractor
Probably not as uncommon as you would think. Take say a 120/208 volt transformer that XO was not bonded, Fault happens on a circuit, now the other two phases are 208 volts to ground, but still 120 volts to neutral. A fault happens on another circuit elsewhere on a different phase, now one or the other, or both breakers trip, but it would be a phase to phase fault, at the higher voltage, not a phase to neutral fault.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I think it depends on how rapidly the first ground fault gets repaired.
In how many places does one light go out and the other two go bright, and no one bothers to fix anything because the machines are still happily running.....
If you don't have a reliable EGC, then in the case of a ground fault some random bit of metal will become energized, but only in that one location. The GF detection system might not even notice this and people just keep using the equipment. Then you get another fault, but because you don't have a reliable EGC this simply means that yet another random bit of metal gets energized. Nothing gets noticed until something or someone completes the path between the two.
-Jon
texie
Senior Member
- Location
- Fort Collins, Colorado
- Occupation
- Electrician, Contractor, Inspector
Yes, and I should have mentioned that in my earlier comments.
- Location
- Illinois
- Occupation
- retired electrician
As far as the code requirements, the only difference in the rules for bonding and grounding of ungrounded and grounded systems is that the ungrounded system does not have a main or system bonding jumper. Everything else is identical.
1. The metal frame can be capacitively or inductively energized WITHOUT contact.
2. If there is a fault it is directly energized. If it is grounded it becomes corner grounding.
Just because currents other than faults don’t normally flow on grounded wiring doesn’t mean there is no electrical activity.
Do not forget that there is no such thing as “ungrounded”. An “ungrounded” system if nothing else is insulated from Earth (even if you didn’t ground anything). The definition of a capacitor is two conductors separated by an insulator. It is still grounded, just that it is grounded through capacitance. Which brings up the major issue with ungrounded systems: transients. When you do get a ground fault, it causes damage to other healthy equipment unrelated to the fault.
It's in this document:
If the equipment grounding is not contiguous, MSHA calls that a local "Peg Grounded" system, and cites it as significant in fatal accident investigations. Metal parts or isolated building parts, equipment frames, can become energized shorting to any line conductor. It does not matter if the supply is floating or grounded, isolated non contiguous metal can have a voltage on it. Throw in the Earth reference, people standing on the Earth to climb on equipment, mine dewatering pump and a guy in a metal boat servicing it ...
If all equipment and metal frames are bonded continguous, which MSHA may require in a mine site, voltage potential difference is greatly reduced, sitewide. Then when the supply conductors fault to ground ,which can be anywhere, the floating system will just reference to ground and the grounded system will trip the breaker.
It also requires a ground fault alarm or at least “phase lights” in an ungrounded system as far as a power system goes. There are limited exceptions for instrumentation like PTs in switchgear but that’s it.
The insurance carriers take this further. At one time Factory Mutual stated that alarming has to be in a 24/7 continuously monitored location like a control room or guard shack, or else tripping (high resistance grounding) was required. The latest standard just says convert to high resistance grounding, no exceptions.
Bender promotes the idea of ungrounded systems with monitoring since they sell ground fault monitors. But after dealing with one for years I’ve become convinced Factory Mutual is right...get rid of this stuff.
- Location
- Illinois
- Occupation
- retired electrician
I forgot about the requirement for the ground detection found in 250.21(B). An important point.
If one Ground Fault has already occurred it is more likely that a second one will follow. Something caused that first ground fault. That first Ground Fault did not trip the circuit conductor Over Current Protective Device. In the absence of ground fault detection no one may yet know that the first fault has occurred. That means the underlying cause of the ground fault is probably ongoing. Whatever it is it is more than a little likely that the problem will cause one of the other phases to fault.
--
Tom Horne
When I was still in the craft I asked for permission to seek a change order to ground one phase or make up a simple fault detection circuit that would include a dry contact relay for the building alarm system. Since modern alarm systems use addressable devices the alarm monitoring company would be instructed to treat an alarm from the contact opening detector module as a building trouble signal for which they would contact a "responsible person." I was always refused permission. Not because the "Extra;" as in work that is not already paid for within the contracted price, would not be profitable but rather because no one was willing to let the electricians, no matter how well qualified in control circuit installation, create a detector for which there would be some liability for any failure. The legal system has made cowards of us all. I was particularly concerned about installations such as medical research laboratory refrigeration which was powered from 3 phase delta transformers. Since you could not hold the phase to ground voltage under the prescribed voltage the US NEC does not require the installation of a Main Bonding Jumper to one of the current carrying conductors. Yet if those refrigerators fail years of research may be lost or worse still some pathogen might grow when the temperature rises and escape it's container thus contaminating the entire inside of the refrigerator. The laboratory staff will find out about that when they open the door and contaminate the lab. If they then flee the laboratory they may cause the escape of a communicable disease organism. I know. I worry too much.
Likelihood of two different phase GFs is 67% of the time!
Start with the first GF. Which phase? Does not matter. On the second GF two out of three phases are the “wrong” phase so the odds are 2/3 or 67%.
As to whether it happens...how often do you get ground faults, and how quickly do you clear them? That has more to do with maintenance practices. If the practice is to treat ground faults like a fire alarm(work quickly and efficiently towards locating and eliminating the issue) then it’s pretty unlikely. If practices are to get around to it some day, it’s common.
Start with the first GF. Which phase? Does not matter. On the second GF two out of three phases are the “wrong” phase so the odds are 2/3 or 67%.
As to whether it happens...how often do you get ground faults, and how quickly do you clear them? That has more to do with maintenance practices. If the practice is to treat ground faults like a fire alarm(work quickly and efficiently towards locating and eliminating the issue) then it’s pretty unlikely. If practices are to get around to it some day, it’s common.
My concern is that, at least in some of the biological and pharmaceutical laboratories I helped to build or serviced, that they did not have an in house electrician. So what I was trying to do was give them the means to know when they needed to call in the electrical service contractor that they normally turned to. In the research laboratory they had some nasty pathogens under refrigeration. The reason that the Electrical Engineer had specified Delta supply was to avoid interruption of the power to the refrigeration from a single fault. To me the only way that makes sense is to have ground fault detection so that they can act on the fault very promptly. Since they did not have a contract for a regular electrical testing routine as you indicated the first fault could be around long enough for the second fault to occur. To me that doesn't make any sense whatsoever. Which is why I brought that up in the first place. I'm at a loss to figure out why the NEC does not require fault monitoring when the installation is taking advantage of the exception for continuity of service.
MTW
Senior Member
- Location
- SE Michigan
It does, and the section was mentioned earlier. I just means that the installation you mentioned was not compliant.
- Status