buck33k
Senior Member
- Location
- Washington state and North Carolina
What is meant by "stabilize the voltage to earth during normal operation" in 250.4(A)(1)?
stabilize the voltage to earth
- Thread starter buck33k
- Start date
- Status
- Location
- Lockport, IL
- Occupation
- Semi-Retired Electrical Engineer
If the neutral is not attached to planet Earth at the source of power, then the voltage from any hot conductor to planet Earth will vary from point to point within the building, and from moment to moment throughout the day. That's not considered a good thing. But with the N-G bonded and with that point connected to a ground rod, then the voltage from any point in the system to planet Earth will be (approximately) the same at all points in the system (meaning all with the same "nominal voltage") and at all times.
Re: stabilize the voltage to earth
Thirty years ago I clearly understood what they meant. About 15 years ago I started to wonder (or wander). Now, I have no clue.
I can't tell that an NEC specified residential ground rod(s) connected to the neutral does anything for the safety or stability of the US standard residential system - except maybe for some lightning protection.
One thing residential ground rods do is to get the utility a lot of free grounds. And they use/depend on them alot.
carl
buck33k said:
Thirty years ago I clearly understood what they meant. About 15 years ago I started to wonder (or wander). Now, I have no clue.
I can't tell that an NEC specified residential ground rod(s) connected to the neutral does anything for the safety or stability of the US standard residential system - except maybe for some lightning protection.
One thing residential ground rods do is to get the utility a lot of free grounds. And they use/depend on them alot.
carl
- Location
- Illinois
- Occupation
- retired electrician
Charlie,
Don
I have never understood why the equipment would care what the voltage to earth would be. The equipment only sees the voltage between the circuit conductors. It doesn't care what the voltage to "earth" is.
Don
sandsnow
Senior Member
- Location
- Southern California
I'm with Don.
I have heard of floating voltages (different between phases to neutral) and this "grounding" is supposed to magically stop that.
Please advise 8)
I have heard of floating voltages (different between phases to neutral) and this "grounding" is supposed to magically stop that.
Please advise 8)
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
As far as the current flowing through the conductors, or the power delivered to the load, the voltage of the various conductors with respect to Earth is absolutely irrelevant. Electrical systems work on airplanes, and satellites, after all.
One can design electrical systems with quite substantial 'common mode' voltages to Earth. While the NEC requires a wye secondary to be 'grounded' via the neutral tap, a 'corner grounded wye' connection would function 'just fine'. A helicopter bonded to a multi-hundred-KV transmission line continues to function normally.
But where voltage between the conductors and Earth actually makes a difference is for protection of the insulation system between the conductors and Earth. Imagine the following scenario: 13.2KV delta to 277/480V wye transformer. X0 of the transformer, rather than being bonded to ground, is bonded to supply phase A. A 480V motor connected to the secondary would function _normally_ if the insulation to ground could tolerate 13.2KV/1.732....but with a normal motor the insulation system would probably last a half cycle and arc between a conductor and ground.
Where 'stabilizing voltage relative to ground' becomes important is that there are quite a few sources of high voltage, sources that could momentarily push the electrical system to damaging voltages relative to ground. If you simply had an ungrounded electrical system, then any of these voltage sources could damage the insulation system; simply walking across the carpet could produce enough voltage to punch through common building wire insulation.
Sources for 'common mode' voltage to Earth could be leakage from the transformer primary, local static electricity, lightning events, or a ground fault on one phase raising the voltage on the other phases. In the event of a 'restriking ground fault' the interrupting fault current can combine with the natural capacitance between the wiring system and ground to actually 'pump' common mode voltage up to several times the system line voltage.
Solidly bonding the source (transformer, generator, battery, etc.) provides a good low impedance path for the above circuits, preventing them from significantly changing the voltage of any conductor. This is not bonding back to the neutral to provide a fault current path; this is electrically coupling to Earth (as well as all bonded metal) to make it impossible for a high voltage, high impedance, low current source to energize an electrical system to insulation damaging voltages.
-Jon
One can design electrical systems with quite substantial 'common mode' voltages to Earth. While the NEC requires a wye secondary to be 'grounded' via the neutral tap, a 'corner grounded wye' connection would function 'just fine'. A helicopter bonded to a multi-hundred-KV transmission line continues to function normally.
But where voltage between the conductors and Earth actually makes a difference is for protection of the insulation system between the conductors and Earth. Imagine the following scenario: 13.2KV delta to 277/480V wye transformer. X0 of the transformer, rather than being bonded to ground, is bonded to supply phase A. A 480V motor connected to the secondary would function _normally_ if the insulation to ground could tolerate 13.2KV/1.732....but with a normal motor the insulation system would probably last a half cycle and arc between a conductor and ground.
Where 'stabilizing voltage relative to ground' becomes important is that there are quite a few sources of high voltage, sources that could momentarily push the electrical system to damaging voltages relative to ground. If you simply had an ungrounded electrical system, then any of these voltage sources could damage the insulation system; simply walking across the carpet could produce enough voltage to punch through common building wire insulation.
Sources for 'common mode' voltage to Earth could be leakage from the transformer primary, local static electricity, lightning events, or a ground fault on one phase raising the voltage on the other phases. In the event of a 'restriking ground fault' the interrupting fault current can combine with the natural capacitance between the wiring system and ground to actually 'pump' common mode voltage up to several times the system line voltage.
Solidly bonding the source (transformer, generator, battery, etc.) provides a good low impedance path for the above circuits, preventing them from significantly changing the voltage of any conductor. This is not bonding back to the neutral to provide a fault current path; this is electrically coupling to Earth (as well as all bonded metal) to make it impossible for a high voltage, high impedance, low current source to energize an electrical system to insulation damaging voltages.
-Jon
bphgravity
Senior Member
- Location
- Florida
At one point in electrical history, it was very important to be able to measure the voltage to earth from any one line conductor of a telegraph system. This is when grounding became a popular method for "stabilizing" the reading at various points along the line.
The NEC's reference to 25-ohms is an old cross-over from telegraph installations when a man named Samuel Varley determine that 25-ohms or 1-siemens was sufficient ground resistance for establishing a reference point when troubleshooting telegraph systems. So this unit is nothing more than a performance issue for antique telegraph systems and has no real basis for today's electrical systems.
The NEC's reference to 25-ohms is an old cross-over from telegraph installations when a man named Samuel Varley determine that 25-ohms or 1-siemens was sufficient ground resistance for establishing a reference point when troubleshooting telegraph systems. So this unit is nothing more than a performance issue for antique telegraph systems and has no real basis for today's electrical systems.
- Location
- Illinois
- Occupation
- retired electrician
winnie,
Don
How does a connection to earth prevent this? The spikes are are still there, the current is just shunted to ground. The grounding does not change the peek voltage of the transients.
Don
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I agree that the grounding does not change the _source_ voltage of the transients. But these sources that I listed (excluding lightning) are _high impedance_ sources, meaning that there might be lots of voltage, but there is no current to back it up.
Rather than thinking of these sources as 'voltages', think of them as voltages to ground in series with internal resistance. For example, primary to secondary leakage in a transformer is perhaps 8kV in series with 10 megohms, or our human walking across a carpet is a 100pf capacitor charged up to 50kV, in series with 1.5 kilohms (one standard model; the reality is different).
The transient voltage imposed on the electrical system will _not_ be the source voltage. Instead it will be the result of the voltage divider formed by the source impedance and the impedance of the electrical system to ground. The higher the impedance to ground, the greater the fraction of the source voltage that shows up between the electrical system and ground.
If a primary conductor falls into your secondary conductors, so that you have 8kV in series with a couple of ohms, then grounding will do squat for stabilizing voltages, because the source impedance will be quite low. But for the high impedance 'everyday' transient sources, grounding will make a significant difference.
-Jon
Rather than thinking of these sources as 'voltages', think of them as voltages to ground in series with internal resistance. For example, primary to secondary leakage in a transformer is perhaps 8kV in series with 10 megohms, or our human walking across a carpet is a 100pf capacitor charged up to 50kV, in series with 1.5 kilohms (one standard model; the reality is different).
The transient voltage imposed on the electrical system will _not_ be the source voltage. Instead it will be the result of the voltage divider formed by the source impedance and the impedance of the electrical system to ground. The higher the impedance to ground, the greater the fraction of the source voltage that shows up between the electrical system and ground.
If a primary conductor falls into your secondary conductors, so that you have 8kV in series with a couple of ohms, then grounding will do squat for stabilizing voltages, because the source impedance will be quite low. But for the high impedance 'everyday' transient sources, grounding will make a significant difference.
-Jon
winnie -
You are dead on with everything I clearly understood thirty years ago.
Lets take a look at a few different systems. (yeah, I'm pretty sure you already know all of this.)
Impedance grounded 480Y: Probably my favorite for any industrial installation. Gets rid of any restriking ground fault issues. Highly reliable - takes two faults to take it out. Absolutely can't understand what the science or physics is behind the NEC not allowing 480V HID lighting.
Ungrounded 480V, D or Y: Highly reliable. Used on all Navy shipboard systems. Cable is 600V insulation, good, tough, outer jackets, and mil-spec, but not particularly any better than say MC-HL as far as insulation resistance/robustness goes. Lots of lightning strikes to the ships (they are the tallest thing out there) But the ships have a steel hull and shunt the current around the conductors. That's true. And for unground systems used outside , one should highly consider surge arrestors/TVSS. Also, as I understand, surge arrestors/TVSS also mitigates issues with restriking ground faults (I'm out of my area of expertise here. Hopefully someone else will pitch in on this issue)
As for high impedance, high voltage sources causing insulation damage, I have not seen any evidence that it is an issue. Plenty of 24VDC and 125VDC insustrial controls are run ungrounded. Another common ungrounded system is 480V, 3P through a Drive Isolation Transformer to a VFD. One I am currently working is a 13.8kV /6.6kV through a DIT to a 6500hp VFD. The first two are know to work great, the last looks like it will.
I'm just not seeing any evidence of rampant damage from high impedance HV sources to ungrounded systems.
Just so we are clear, I absolutely think the non-current carrying metalic (conductive) parts of electric fed machinery should be bonded to a grounding electrode. This is a lightning protection issue, a fault protection issue.
You are dead on with everything I clearly understood thirty years ago.
Lets take a look at a few different systems. (yeah, I'm pretty sure you already know all of this.)
Impedance grounded 480Y: Probably my favorite for any industrial installation. Gets rid of any restriking ground fault issues. Highly reliable - takes two faults to take it out. Absolutely can't understand what the science or physics is behind the NEC not allowing 480V HID lighting.
Ungrounded 480V, D or Y: Highly reliable. Used on all Navy shipboard systems. Cable is 600V insulation, good, tough, outer jackets, and mil-spec, but not particularly any better than say MC-HL as far as insulation resistance/robustness goes. Lots of lightning strikes to the ships (they are the tallest thing out there) But the ships have a steel hull and shunt the current around the conductors. That's true. And for unground systems used outside , one should highly consider surge arrestors/TVSS. Also, as I understand, surge arrestors/TVSS also mitigates issues with restriking ground faults (I'm out of my area of expertise here. Hopefully someone else will pitch in on this issue)
As for high impedance, high voltage sources causing insulation damage, I have not seen any evidence that it is an issue. Plenty of 24VDC and 125VDC insustrial controls are run ungrounded. Another common ungrounded system is 480V, 3P through a Drive Isolation Transformer to a VFD. One I am currently working is a 13.8kV /6.6kV through a DIT to a 6500hp VFD. The first two are know to work great, the last looks like it will.
I'm just not seeing any evidence of rampant damage from high impedance HV sources to ungrounded systems.
Just so we are clear, I absolutely think the non-current carrying metalic (conductive) parts of electric fed machinery should be bonded to a grounding electrode. This is a lightning protection issue, a fault protection issue.
sandsnow
Senior Member
- Location
- Southern California
winnie
I never heard that point before. Seems to make sense to me.
I will think on it some more.
Thanks
I never heard that point before. Seems to make sense to me.
I will think on it some more.
Thanks
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
coulter:
I appreciate the experience; I only understood this from theory I presume that theory is what you understood 30 years ago, and now your experience gives you pause.
If I understand you correctly, the only issue (from the list above) that you have found actually causes enough of a voltage excursion to damage insulation is something like a re-striking ground fault.
It may well be the case that as insulation systems have improved, HV leakage from transformers has become a non-issue. At the same time, small static discharges (from people) have also become less of an issue (these discharges get shunted directly through the nearest computer chip), and direct lightning hits are too much for anything to deal with
Did you know that theoretically, there is no difference between theory and practise
-Jon
I appreciate the experience; I only understood this from theory
I presume that theory is what you understood 30 years ago, and now your experience gives you pause.If I understand you correctly, the only issue (from the list above) that you have found actually causes enough of a voltage excursion to damage insulation is something like a re-striking ground fault.
It may well be the case that as insulation systems have improved, HV leakage from transformers has become a non-issue. At the same time, small static discharges (from people) have also become less of an issue (these discharges get shunted directly through the nearest computer chip), and direct lightning hits are too much for anything to deal with
Did you know that theoretically, there is no difference between theory and practise
-Jon
- Location
- Illinois
- Occupation
- retired electrician
Jon,
Don
If there is very little current there is also little energy so how does this low energy event become a problem?
Don
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
My understanding is that the risk is insulation damage. The high voltage low current source cannot deliver much total energy, and thus cannot create a sustained fault causing lots of damage. But the voltage is high enough to cause insulation to ionize and break down.
I suspect that these high voltage, low energy 'events' do damage the insulation system, but that they generate small amount of damage that simply leads to reduced life, not instant failure.
The only situation where I've actually read a report of insulation failure caused by voltage excursion in an ungrounded system was a description of a 'restriking ground fault'. In this case, the high voltage source was the low voltage system itself, and while there may not have been much energy behind each impulse, there were probably quite a few high voltage impulses, all backed up by the low voltage system as soon as a fault path was opened up.
-Jon
I suspect that these high voltage, low energy 'events' do damage the insulation system, but that they generate small amount of damage that simply leads to reduced life, not instant failure.
The only situation where I've actually read a report of insulation failure caused by voltage excursion in an ungrounded system was a description of a 'restriking ground fault'. In this case, the high voltage source was the low voltage system itself, and while there may not have been much energy behind each impulse, there were probably quite a few high voltage impulses, all backed up by the low voltage system as soon as a fault path was opened up.
-Jon
winnie said:
Well, there is that of course, but that makes it sound like I understand all I know about this - and I don't :? Examining the individual jobs I've been involved with, has caused me to rethink the theory, trying to expain what I was seeing with the accepted knowledge base - and I can't alyways. Maybe the theory is wrong. Oh No!! The NEC may not be absolute truth. AAARRRGGGG I can feel the brain waves of some in our profession, "Blasphemy. Phillistine. Un-washed Dog."
winnie said:
I think so, that and lightning. I've seen the effects of dropping an HV transmission line on a distribution MV line in a substation (once). It ruined a lot of equipment, but as far I know, it didn't damage the MV xfmrs or distribution conductors.
I don't clearly understand restriking GFs. Probably cause I work with few ungrounded systems - most are impedance grounded - except for the 208/120 stuff and they are of course grounded.
winnie said:
That would be my understanding. Where I live, static in the middle of winter is brutal - there is very little moisture in the air at -50F. Dry air is an excellent insulator. Definitly tough on chips.
winnie said:
Practically speaking, yes. Theoretically, I'm sot so sure :?
One of the issues with ungrounded or impedance grounded 480 has been that one could not use 277V lighting. That was always okay with me, I never liked it anyway. I liked 480V lighting w/contactors - gets rid of the harmonics. Current code says 480V lighting is no longer available. I think I finally figured out why impedance grounded or un-grounded 277 was considered unsafe - that only took ten years. But I still don't understand why the NEC considers 480V HID lighting bad or unsafe. If we have some members out there that understand the physics or science, I'm interested in learning.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
I am curious as to your understanding of why the NEC prohibits line to neutral loads in ungrounded and impedance grounded systems.
I always presumed that it was because in a ground fault the 'neutral' will be at a considerably elevated voltage.
However rather than prohibiting such loads, why not a requirement that all conductors be treated as ungrounded. Eg. double pole switches or contactors controlling the loads, breakers open _all_ conductors to the load, etc.
Is there another reason that I am not seeing?
We've also been using small isolated impedance grounded systems for our motor test stands. This is not for continuity of service, but instead to reduce fault current. We measure ground fault current and open the supply contactor if it exceeds what we expect for capacitive loading. I rather like the idea that the step between a ground fault forming and the supply opening is something other than a shower of sparks This sort of use, impedance grounding with ground fault circuit interruption, does not seem to be anticipated by current code.
-Jon
I always presumed that it was because in a ground fault the 'neutral' will be at a considerably elevated voltage.
However rather than prohibiting such loads, why not a requirement that all conductors be treated as ungrounded. Eg. double pole switches or contactors controlling the loads, breakers open _all_ conductors to the load, etc.
Is there another reason that I am not seeing?
We've also been using small isolated impedance grounded systems for our motor test stands. This is not for continuity of service, but instead to reduce fault current. We measure ground fault current and open the supply contactor if it exceeds what we expect for capacitive loading. I rather like the idea that the step between a ground fault forming and the supply opening is something other than a shower of sparks
This sort of use, impedance grounding with ground fault circuit interruption, does not seem to be anticipated by current code.-Jon
I think you pretty well got it.winnie said:
If I'm understanding, that would take a 4 buss panel so you could put in CBs on the neutral. Never thought of that - interesting idea.winnie said:
I like it - good engineering.winnie said:
Yeah, me too.winnie said:
No, it doesn't.winnie said:
I think you have pretty well hit on why the NEC doesn't like single phase ckts on un-grounded or impedance grounded 3p systems. And, it is pretty complicated to use/design.
Now, why don't they like 480V HID lighting?
carl
- Location
- Illinois
- Occupation
- retired electrician
Carl,
Don
It is fine as long as it is supplied from a grounded wye system.
Don
- Status