Subpanel Grounding
#1
05-25-04, 04:40 AM
Subpanel Grounding
A friend of mine has hooked up a subpanel in his garage. It's a 60amp running from his main panel in the house. The service from the house to the garage is all underground. My questions are these:
1) He didn't install a separate ground rod for the subpanel. How should we go about installing one, if this is needed? (I haven't looked at it yet, but I am guessing that he probably ran 4 wire cable [Hot, Hot, Neutral, Ground] from the house to the subpanel - the ground wire having a continuity between the house panel and the garage). I know we can drive a ground rod, and run a conductor, but should we do something with the (probable) incomoing ground wire from the house to the subpanel?
2). In the subpanel, is it absolutely required/encouraged to separate the neutral and grounding busbars. I read, in one of Rex Caldwell's books, I think, that in some cases these two should/could be connected. What are the safety concerns of the two being connected?
3). If the busbars are to be separate, and if they are not isolated by an insulator of some sort on the neutral busbar (I think some panels aren't meant to be subpanels, and I don't know yet which type he has), what do I look for at the Home Depot in the way of an insulator.
My plan is to go to his home this weekend (can't get there any earlier). So, I'm sorry that I don't have all the specifics of the way it is currently setup. I'd appreciate any prep advise you can give me.
Thanks,
dim
1) He didn't install a separate ground rod for the subpanel. How should we go about installing one, if this is needed? (I haven't looked at it yet, but I am guessing that he probably ran 4 wire cable [Hot, Hot, Neutral, Ground] from the house to the subpanel - the ground wire having a continuity between the house panel and the garage). I know we can drive a ground rod, and run a conductor, but should we do something with the (probable) incomoing ground wire from the house to the subpanel?
2). In the subpanel, is it absolutely required/encouraged to separate the neutral and grounding busbars. I read, in one of Rex Caldwell's books, I think, that in some cases these two should/could be connected. What are the safety concerns of the two being connected?
3). If the busbars are to be separate, and if they are not isolated by an insulator of some sort on the neutral busbar (I think some panels aren't meant to be subpanels, and I don't know yet which type he has), what do I look for at the Home Depot in the way of an insulator.
My plan is to go to his home this weekend (can't get there any earlier). So, I'm sorry that I don't have all the specifics of the way it is currently setup. I'd appreciate any prep advise you can give me.
Thanks,
dim
#2
05-25-04, 07:33 AM
If this is a detached garage (which probably is since the service is underground) then a 4 wire feeder has to be used (2 hots, neutral and ground). At the main panel the hots go to the breaker, and the ground and neutral go to the shared neutral/ground bar. At the subpanel the hots go to the main breaker, the neutral goes to the neutral bus bar and the ground goes to the ground bus bar. The neutral and ground bars are not to be bonded (don't use the green bonding screw). You'll also most likely need another bus bar for the neutral unless your sub is pretty large.
You'll also need to run a ground rod (I generally run 2 5/8" rods, 8' apart) and run a grounding wire (probably #4) to those rods. So the subpanel will now have 2 ground wires.
You'll also need to run a ground rod (I generally run 2 5/8" rods, 8' apart) and run a grounding wire (probably #4) to those rods. So the subpanel will now have 2 ground wires.
#3
05-25-04, 07:37 AM
Member
1) Yes the ground rod is needed. The ground rod serves a completely different purpose than the grounding wire to the house. They both connect to the same place in the subpanel, but are part of different grounding systems.
2) There are different conditions. But since you ran a grounding wire from the house, you absolutely must keep the grounding and neutral electrically isolated in the subpanel. Otherwise you would have neutral current running on your grounding wire. There is no controversy here. It's a very clear cut situation.
3) The neutral bar that comes with the subpanel is already electrically isolated from the case, as long as you don't install the green screw. You don't need an insulator. Many people think the neutral bar is not isolated after a visual inspection, but test it with an ohmmeter (with all the wires disconnected) to convince yourself. You usually need to buy a grounding bar kit, although some panels come with one already installed.
2) There are different conditions. But since you ran a grounding wire from the house, you absolutely must keep the grounding and neutral electrically isolated in the subpanel. Otherwise you would have neutral current running on your grounding wire. There is no controversy here. It's a very clear cut situation.
3) The neutral bar that comes with the subpanel is already electrically isolated from the case, as long as you don't install the green screw. You don't need an insulator. Many people think the neutral bar is not isolated after a visual inspection, but test it with an ohmmeter (with all the wires disconnected) to convince yourself. You usually need to buy a grounding bar kit, although some panels come with one already installed.
#4
05-25-04, 02:34 PM
convincing
Thanks for those replies. Now comes the convincing part, for BOTH my friend AND myself. I, of course will, and do, take you at your word on this ground separation, and I know it's been discussed over and over here. I've read previous posts regarding the grounding issues, and I've read about it elsewhere, but it's still not clear to me why, if, as John says, "Otherwise you would have neutral current running on your grounding wire", the same situation is not happening in the main panel, where the neutral and ground do not float. If there's the possibility that current will flow out onto nonconductive (i.e. metal appliance frames) parts in a bonded neutral/ground subpanel, why isn't the same thing possible in the bonded main panel? I know you folks are probably tired of rehashing all this, but I need to convince my friend that there is a problem, but first I have to understand this myself.
Why isn't there "neutral current running on your grounding wire" in the main panel, and therefore, onto the metal frames of, say washers and dryers?
Thanks for your time and patience.
-dim
Why isn't there "neutral current running on your grounding wire" in the main panel, and therefore, onto the metal frames of, say washers and dryers?
Thanks for your time and patience.
-dim
#5
05-25-04, 03:18 PM
Member
The difference is that you only have three wires running back to the utility transformer, not four. If you only had three wires running to the garage (only acceptable if there is no metallic path between the buildings), then the same rules would apply.
Remember, current doesn't just run around randomly. It only flows towards a point of lower potential.
Also note that the earth has about a billion times more resistance than copper does.
Remember, current doesn't just run around randomly. It only flows towards a point of lower potential.
Also note that the earth has about a billion times more resistance than copper does.
#6
05-25-04, 03:38 PM
The answer is that the same thing sometimes _does_ happen at the main panel, and it _is_ a problem. But when all of your house interior is properly bonded with 'single point' grounding then you don't get current flow _inside_ of the house, and there isn't a shock risk between the various metal surfaces inside the house.
One point of confusion: you are confusing voltage and current. Voltage is measured _between_ two points, and is a measure of the 'push' available to move electrons. When the electrons start moving, that is current.
When the neutral and ground are bonded at a _single_ point in your home, even if you do have some ground current, all of the metal frames connected to the ground system will be at the _same_ potential, and there will be no voltage difference between them. No voltage difference means no shock. It's the old 'bird on a wire' thing; the bird is sitting on a bare wire at thousands of volts, but doesn't get shocked because it is on _one_ wire, and there is no circuit path through the bird.
In situations where there is neutral current flowing on a home's ground wire, the problems occur between that building and other buildings. Neutral currents flowing on grounding conductors are known to cause problems with metal piping systems; plumbers have gotten shocks when cutting the pipes going into houses. If you do a google search on 'stray voltage', you will see many references to a rather ill defined subject that IMHO has as its primary cause neutral to ground currents.
-Jon
One point of confusion: you are confusing voltage and current. Voltage is measured _between_ two points, and is a measure of the 'push' available to move electrons. When the electrons start moving, that is current.
When the neutral and ground are bonded at a _single_ point in your home, even if you do have some ground current, all of the metal frames connected to the ground system will be at the _same_ potential, and there will be no voltage difference between them. No voltage difference means no shock. It's the old 'bird on a wire' thing; the bird is sitting on a bare wire at thousands of volts, but doesn't get shocked because it is on _one_ wire, and there is no circuit path through the bird.
In situations where there is neutral current flowing on a home's ground wire, the problems occur between that building and other buildings. Neutral currents flowing on grounding conductors are known to cause problems with metal piping systems; plumbers have gotten shocks when cutting the pipes going into houses. If you do a google search on 'stray voltage', you will see many references to a rather ill defined subject that IMHO has as its primary cause neutral to ground currents.
-Jon
#7
05-25-04, 06:27 PM
Originally Posted by John Nelson
The difference is that you only have three wires running back to the utility transformer, not four. If you only had three wires running to the garage (only acceptable if there is no metallic path between the buildings), then the same rules would apply.
Remember, current doesn't just run around randomly. It only flows towards a point of lower potential.
Also note that the earth has about a billion times more resistance than copper does.
Remember, current doesn't just run around randomly. It only flows towards a point of lower potential.
Also note that the earth has about a billion times more resistance than copper does.
John, only three wires to the main panel? Is there not also a grounding conductor coming in with the two hots and grounded/neutral conductor? Even if there is no grounding conductor (bare wire) how would this change things?
I am still at a loss as to why current would flow to metal appliance cabinets in a bonded neutral/grounded subpanel, but not at the main.
And I don't believe I'm confusing I with E. I realize that current flow is a result of a load (resistance) and that voltage is the EMF supplied to the main lugs at the panels.
I hope I'm not being too much of a pain.
#9
05-25-04, 07:27 PM
Ahh, but I do believe that you are confusing I and E. Current does not simply flow to a bonded component. For current to flow you need to have a closed circuit. What does 'flow' to all bonded elements is _voltage_.
If you have a bonded subpanel along with your bonded main panel, then the equipment ground conductor is sitting in parallel with the neutral conductor. Because current will flow in parallel paths, you will see current flowing in _both_ of these conductors. You won't see current flowing out to the appliances, because these are not closed circuit current paths. But any equipment bonded to the ground/neutral in the subpanel will be at the same _voltage_ as the subpanel neutral.
Similarly, any equipment bonded to the ground/neutral in the main panel will be at the same voltage as the main panel neutral. So in a sense, with the main ground/neutral bond, the neutral _voltage_ does 'leak out' to all of the bonded equipment in the building.
The problem is that the voltage of the ground/neutral bond in the sub panel will be different from that in the main panel.
You cannot have _current_ leaking until you have _two_ ground/neutral bonds, in order to form a closed circuit path for the current. So if you just have the single main ground/neutral bond, then all the bonded metal in the house is tied to the _voltage_ of the main bond. If you add another ground/neutral bond, you will then see _current_ leaking and following whatever paths are available between the two bond points.
-Jon
If you have a bonded subpanel along with your bonded main panel, then the equipment ground conductor is sitting in parallel with the neutral conductor. Because current will flow in parallel paths, you will see current flowing in _both_ of these conductors. You won't see current flowing out to the appliances, because these are not closed circuit current paths. But any equipment bonded to the ground/neutral in the subpanel will be at the same _voltage_ as the subpanel neutral.
Similarly, any equipment bonded to the ground/neutral in the main panel will be at the same voltage as the main panel neutral. So in a sense, with the main ground/neutral bond, the neutral _voltage_ does 'leak out' to all of the bonded equipment in the building.
The problem is that the voltage of the ground/neutral bond in the sub panel will be different from that in the main panel.
You cannot have _current_ leaking until you have _two_ ground/neutral bonds, in order to form a closed circuit path for the current. So if you just have the single main ground/neutral bond, then all the bonded metal in the house is tied to the _voltage_ of the main bond. If you add another ground/neutral bond, you will then see _current_ leaking and following whatever paths are available between the two bond points.
-Jon
#10
05-26-04, 01:01 AM
#11
05-26-04, 10:58 AM
"Why isn't there a Neutral-current ( conducted thru the Ground-wire) onto the metal frames---"
I'll attempt an explanation with an "Hydralic Anology" but first----
The basic 2-wire 120-volt Branch-Circuit consists of a Black conductor, a White conductor ( the Neutral) and a bare Equiptment Grounding Conductor (EGC)>.At the "source" (the panel) the Neutral and EGC and "Bonded" together. There is 120 volts Black-to-White, 120 volts Black -to -EGC, and Zero volts White-to-EGC.
The Neutral and the EGC are "semi-parallel" conductors in the Branch-Circuit cable to the connected load,but AT the connected load they are seperate, the Neutral connected to the load and insulated from any metallic surface and the EGC "Bonded" to the metallic surface.
We have a water-line with a pressure of 120 lbs.-per-inch above atmospheric pressure.The purpose of the water-line is to fill a vessel with a "constant" supply of "pure" water at a rate, of say, 10 gallons-per-minute.The "impure" water flows out of the vessel thru a drain-line.
Should the vessel burst,or over-flow, it's imperative that the released water flow into an "emergency" drain-line to prevent a "flood".This requires an "emergency" drain-line that "parallels" the "normal" drain-line back to the "source" where it connects to the same point as the "normal" drain-line. In "normal" operation, there is 10 gallons-per-minute thru the water-line, the vessel, and the "normal" drain-line.Should the vessel burst, there is a "surge" of water thru the "emergency" drain-line".
This is similiar to a "Ground-fault" in a connected load-- a "surge" of current thru the Black wire connected to a circuit-beaker, thru the connected load, and thru the EGC back to the "source". The "surge" will trip the breaker to open the circuit.
I'll attempt an explanation with an "Hydralic Anology" but first----
The basic 2-wire 120-volt Branch-Circuit consists of a Black conductor, a White conductor ( the Neutral) and a bare Equiptment Grounding Conductor (EGC)>.At the "source" (the panel) the Neutral and EGC and "Bonded" together. There is 120 volts Black-to-White, 120 volts Black -to -EGC, and Zero volts White-to-EGC.
The Neutral and the EGC are "semi-parallel" conductors in the Branch-Circuit cable to the connected load,but AT the connected load they are seperate, the Neutral connected to the load and insulated from any metallic surface and the EGC "Bonded" to the metallic surface.
We have a water-line with a pressure of 120 lbs.-per-inch above atmospheric pressure.The purpose of the water-line is to fill a vessel with a "constant" supply of "pure" water at a rate, of say, 10 gallons-per-minute.The "impure" water flows out of the vessel thru a drain-line.
Should the vessel burst,or over-flow, it's imperative that the released water flow into an "emergency" drain-line to prevent a "flood".This requires an "emergency" drain-line that "parallels" the "normal" drain-line back to the "source" where it connects to the same point as the "normal" drain-line. In "normal" operation, there is 10 gallons-per-minute thru the water-line, the vessel, and the "normal" drain-line.Should the vessel burst, there is a "surge" of water thru the "emergency" drain-line".
This is similiar to a "Ground-fault" in a connected load-- a "surge" of current thru the Black wire connected to a circuit-beaker, thru the connected load, and thru the EGC back to the "source". The "surge" will trip the breaker to open the circuit.
#12
05-26-04, 02:20 PM
PATTBAA et al
I do understand somewhat the needs and the basics for grounding and how current gets back to the load. That's not really my question here. My question is how bonding the neutral and ground at the main is any different at the subpanel. I DO understand the danger of jumpering the neutral and ground at a receptacle, and how that causes a parallel circuit, as was discussed in previous threads.
It just seems to me that the same situation is available at a bonded subpanel as is at the main. That is, that if we can bond the neutral in the main why not in the sub? How is it that fault current will flow back to the source in the main panel which will hopefully trip the CB, but it will flow to the appliaces at the subpanel. Wait a minute, I'm having some brain activity ---- Nope, that wasn't it
racraft said, in his reply to the other post http://forum.doityourself.com/showthread.php?t=167510, that "The neutral and ground are tied together so that if a fault occurs at an outlet or at an appliance and the hot wire connects to the grounded metal junction box ...., the current will have a path back to the source, which will hpoefully cause breaker to trip ...." "In a sub panel the ground and neutral are kept separated so that no current flows on the ground side of the circuit. .... If the wiring to the subpanel had the ground and neutral the same size, and if they were connected at the subpanel, then the current across each wire would be roughly equal."
So, if what racraft said is true at the subpanel, why isn't it also true at the main? That is, if the current would flow in the ground wire at the subpanel, why wouldn't it also flow at the main?
I do appreciate the fact that what you are telling me is gospel and according to code and safety and all, but I just can't seem to see the difference, and explaining to my friend that "well it's just code" may not be enough to convince him to make the necessasry changes. And if there's a risk to his well being, I would like this to be corrected.
Dave: I read that very detailed explanation you gave in the other post as well as a 32+ page thread that occured back in 2001 (I had to print it). I speant about 2 hours on all this, and your explanation was very detailed and good and it made me understand alot more. Other's, in 2001, made very good points as well. I'm going to reread your posts cause I think there's alot of info in there if I can just concentrate on it enough.
PATTBAA: Your explanation was also good and I appreciate the time you spent.
Winnie: After reading your replies again, I think I was, in fact, confusing I and E. But maybe so was some of the others? If I've got this right now, if the hot wire disconnects at an appliance and touches the chassis, there will be voltage on that chassis, but no current flow. Is this correct? If so, I would assume that the current flow would begin the moment I decide to touch this chassis, and that I would then be the unlucky neutral carrying the fault back to the source. Gee, I would sure hope the breaker trips
Again to all, thanks for the replies and time.
dim
It just seems to me that the same situation is available at a bonded subpanel as is at the main. That is, that if we can bond the neutral in the main why not in the sub? How is it that fault current will flow back to the source in the main panel which will hopefully trip the CB, but it will flow to the appliaces at the subpanel. Wait a minute, I'm having some brain activity ---- Nope, that wasn't it
racraft said, in his reply to the other post http://forum.doityourself.com/showthread.php?t=167510, that "The neutral and ground are tied together so that if a fault occurs at an outlet or at an appliance and the hot wire connects to the grounded metal junction box ...., the current will have a path back to the source, which will hpoefully cause breaker to trip ...." "In a sub panel the ground and neutral are kept separated so that no current flows on the ground side of the circuit. .... If the wiring to the subpanel had the ground and neutral the same size, and if they were connected at the subpanel, then the current across each wire would be roughly equal."
So, if what racraft said is true at the subpanel, why isn't it also true at the main? That is, if the current would flow in the ground wire at the subpanel, why wouldn't it also flow at the main?
I do appreciate the fact that what you are telling me is gospel and according to code and safety and all, but I just can't seem to see the difference, and explaining to my friend that "well it's just code" may not be enough to convince him to make the necessasry changes. And if there's a risk to his well being, I would like this to be corrected.
Dave: I read that very detailed explanation you gave in the other post as well as a 32+ page thread that occured back in 2001 (I had to print it). I speant about 2 hours on all this, and your explanation was very detailed and good and it made me understand alot more. Other's, in 2001, made very good points as well. I'm going to reread your posts cause I think there's alot of info in there if I can just concentrate on it enough.
PATTBAA: Your explanation was also good and I appreciate the time you spent.
Winnie: After reading your replies again, I think I was, in fact, confusing I and E. But maybe so was some of the others? If I've got this right now, if the hot wire disconnects at an appliance and touches the chassis, there will be voltage on that chassis, but no current flow. Is this correct? If so, I would assume that the current flow would begin the moment I decide to touch this chassis, and that I would then be the unlucky neutral carrying the fault back to the source. Gee, I would sure hope the breaker trips
Again to all, thanks for the replies and time.
dim
#13
05-26-04, 02:49 PM
Winnie: After reading your replies again, I think I was, in fact, confusing I and E. But maybe so was some of the others? If I've got this right now, if the hot wire disconnects at an appliance and touches the chassis, there will be voltage on that chassis, but no current flow. Is this correct? If so, I would assume that the current flow would begin the moment I decide to touch this chassis, and that I would then be the unlucky neutral carrying the fault back to the source. Gee, I would sure hope the breaker trips
However, by design, appliances are _not_ isolated. We electrically connect them to the EGC, 'grounding' the appliance chassis. In this case, when the hot wire touches the chassis, it is electrically connected through the EGC system back to the main neutral-ground bond. Ideally, so much current flows that the breaker trips.
The important point to remember is that current does not flow until you have a complete circuit.
For the EGC system to be able to provide the function of tripping a breaker when a hot conductor touches the chassis somewhere, you _must_ have at least one connection between the EGC system and neutral. This particular function works just fine if you have more than one EGC to neutral bond, but you need at least one bond.
But the EGC system provides a second function, which is to make sure that all of the metal bonded to the EGC is at the same voltage, so that you can't get a shock between any two bonded items. This aspect of the EGC system will only work when there is no current flowing, because if there is current flow, there _must_ be a voltage difference because of the conductor resistance. In order to be certain that no current flows through the EGC system, you need to make sure that there are no energized _circuits_ which include the EGC system.
You reconcile these two requirements by having one, but only one, bond between the energized system and the EGC system. A circuit requires a closed loop, and you can't have a circuit if your conductors are joined at only one location.
-Jon
#14
05-26-04, 04:10 PM
Member
Unfortunately, the current required to trip the breaker is more than a thousand times the current required to kill you. So even if the breaker does trip when you touch that ungrounded but live chassis, you'll be dead long before the breaker trips. That's why we need GFCI and EGCs to save our lives. The breaker just won't do that by itself.
#15
05-26-04, 08:24 PM
I have been going thru this with one of my neighbors and it takes time. I get some questions about the same thing and have been working with my mechaic over this,, he is getting it. I see many people with this train of thought. "I dont understand it so I am going to do it this way" I have to explain,,, Because you dont understand it is all the more reason to follow the instructions, not a reason not to. Explain it to your friend in those same terms.
#16
05-27-04, 09:13 AM
Member
Indeed. Following rules you understand is intuitive. But it is often even more important to follow the rules you do not understand.
#17
05-27-04, 11:36 AM
And, I agree. I do plan to attempt to get this corrected, and up to code for him. Am still reading over posts. Are there any books on the market which exclusively address Residential Electrical Grounding?
Thanks for all your help, and I hope more will reply to this thread before the weekend when I plan to go.
I'll ask once more and then leave this alone:
That is, that if we can bond the neutral in the main why not in the sub? How is it that fault current will flow back to the source in the main panel which will hopefully trip the CB, but it will flow to the appliaces at the subpanel
Thanks,
Terry
Thanks for all your help, and I hope more will reply to this thread before the weekend when I plan to go.
I'll ask once more and then leave this alone:
That is, that if we can bond the neutral in the main why not in the sub? How is it that fault current will flow back to the source in the main panel which will hopefully trip the CB, but it will flow to the appliaces at the subpanel
Thanks,
Terry
#18
05-27-04, 11:54 AM
I'll take a stab at your question.
If you connect the neutral and the ground together at the main panel and at the sub panel, you will have return current flowing from the sub panel back to the main panel across both the neutral wire and across the ground wire. This is because electricity flows across all available paths back to the source.
Because there is separation between the main panel and the subpanel, the voltages of the current carrying conductors at the subpanel is different than the voltages of these same conductors at the main panel. If the voltages at the main panel are exactly 120 volts for each hot leg and exactly 0 volts for the neutral, the the voltages at the sub panel would be something less than 120 volts for each hot leg and something more than 0 volts for the neutral. The exact voltages depend on the length and material of the conductors, among other things.
If your ground is bonded to the neutral at the subpanel, then your ground is not really 0 volts, but rather whatever the voltage is on the neutral. This means that the metal shell of a grounded tool or appliance is not really grounded, but sitting at some voltage. Somebody touching the metal shell of a device and a true ground would truly feel a shock.
If you connect the neutral and the ground together at the main panel and at the sub panel, you will have return current flowing from the sub panel back to the main panel across both the neutral wire and across the ground wire. This is because electricity flows across all available paths back to the source.
Because there is separation between the main panel and the subpanel, the voltages of the current carrying conductors at the subpanel is different than the voltages of these same conductors at the main panel. If the voltages at the main panel are exactly 120 volts for each hot leg and exactly 0 volts for the neutral, the the voltages at the sub panel would be something less than 120 volts for each hot leg and something more than 0 volts for the neutral. The exact voltages depend on the length and material of the conductors, among other things.
If your ground is bonded to the neutral at the subpanel, then your ground is not really 0 volts, but rather whatever the voltage is on the neutral. This means that the metal shell of a grounded tool or appliance is not really grounded, but sitting at some voltage. Somebody touching the metal shell of a device and a true ground would truly feel a shock.
#19
05-27-04, 01:56 PM
racraft
OK, I'm gonna have to beat this horse 
This, I understand completely
This is interesting. Am learning something
And this is where the confusion comes. Why isn't it the same way at the main? For example, I am visualizing the following current flow:
From the main panel CB, current flows from the Black Hot wire through a load (lamp, for example), and returns to the source via the neutral. However, at the main panel, the neutral and the ground are in continuity, and therefore, I am visualizing the neutral current spreading out to the ground wire and returning to the grounded receptacle via the ground wire which is bonded to the neutral. As you said. "This is because electricity flows across all available paths back to the source".
As a result, I am visualizing the amount of current on the neutral (or more precisely the "grounded conductor"), being available to the metal chassis of the receptical (could also be a dryer, washer or whatever, as opposed to the lamp I suggested as the "load), because the neutral and ground are in continuity at the main panel.
If I were to kill the circuit, and check for continuity with an ohmmeter between the white grounded conductor and the bare/green grounding conductor at the receptacle, would I not get a reading of continuity? And if so (and I believe I would since the neutral and ground are bonded at the main), how could there not be return current on this ground? I guess I'm gonna have to clamp an ammeter on this to find out.
Despite my still growing confusion and stubborness on this issue, I do appreciate your reply and hope you can help me further. I don't think I'm really too far away from understanding this, as many people have offered good knowledge, and for that, I'm grateful.
Thanks,
dim
Originally Posted by racraft
.... If you connect the neutral and the ground together at the main panel and at the sub panel, you will have return current flowing from the sub panel back to the main panel across both the neutral wire and across the ground wire. This is because electricity flows across all available paths back to the source.
This, I understand completely
Because there is separation between the main panel and the subpanel, the voltages of the current carrying conductors at the subpanel is different than the voltages of these same conductors at the main panel. If the voltages at the main panel are exactly 120 volts for each hot leg and exactly 0 volts for the neutral, the the voltages at the sub panel would be something less than 120 volts for each hot leg and something more than 0 volts for the neutral. The exact voltages depend on the length and material of the conductors, among other things..
This is interesting. Am learning something
If your ground is bonded to the neutral at the subpanel, then your ground is not really 0 volts, but rather whatever the voltage is on the neutral. This means that the metal shell of a grounded tool or appliance is not really grounded, but sitting at some voltage. Somebody touching the metal shell of a device and a true ground would truly feel a shock.
From the main panel CB, current flows from the Black Hot wire through a load (lamp, for example), and returns to the source via the neutral. However, at the main panel, the neutral and the ground are in continuity, and therefore, I am visualizing the neutral current spreading out to the ground wire and returning to the grounded receptacle via the ground wire which is bonded to the neutral. As you said. "This is because electricity flows across all available paths back to the source".
As a result, I am visualizing the amount of current on the neutral (or more precisely the "grounded conductor"), being available to the metal chassis of the receptical (could also be a dryer, washer or whatever, as opposed to the lamp I suggested as the "load), because the neutral and ground are in continuity at the main panel.
If I were to kill the circuit, and check for continuity with an ohmmeter between the white grounded conductor and the bare/green grounding conductor at the receptacle, would I not get a reading of continuity? And if so (and I believe I would since the neutral and ground are bonded at the main), how could there not be return current on this ground? I guess I'm gonna have to clamp an ammeter on this to find out.
Despite my still growing confusion and stubborness on this issue, I do appreciate your reply and hope you can help me further. I don't think I'm really too far away from understanding this, as many people have offered good knowledge, and for that, I'm grateful.
Thanks,
dim
#21
05-27-04, 02:09 PM
Terry,
A slight misunderstanding on 'our' part: you are asking about _fault_ current, but we are giving explanations about _normal_ current.
When you have a conductor to ground _fault_, current will flow in the grounding system. When current flows in the grounding system, the various parts of it become energized to different voltages, depending upon the resistance of the grounding system. Anything connected to the grounding system will become energized to a greater or lessor extent during ground faults. This is true even with exactly proper grounding and proper separation of ground and neutral at the subpanels.
The point is that during a fault condition, a _circuit_ exists that includes the grounding system conductors. Current flows through this circuit, and the various parts of the circuit become energized. Hopefully a circuit breaker or ground fault detector trips and cuts off this current flow.
During these fault conditions, it doesn't matter if the subpanel neutral is bonded to ground or not. There is lots of current flowing through the grounding conductors, and significant potentials. In fact, during these fault conditions, it is quite likely that having a grounded subpanel would actually _reduce_ these potentials. But don't do it
The reason that you don't want to ground a subpanel is because of current flow during _normal_ operation. In normal operation, current is supposed to flow in the neutral conductor, but is not supposed to flow in the grounding conductors. If you connect the grounding conductors to neutral at more than one location, then you will _always_ have the grounding conductors in a circuit, and you will _always_ have some current in the grounding system.
Again, the point of single point grounding is to make sure that the grounded appliance chassis are electrically connected and at the same _voltage_ as the ground reference point, but that none of these elements are in a circuit, so that there is no _current_ flow.
-Jon
A slight misunderstanding on 'our' part: you are asking about _fault_ current, but we are giving explanations about _normal_ current.
When you have a conductor to ground _fault_, current will flow in the grounding system. When current flows in the grounding system, the various parts of it become energized to different voltages, depending upon the resistance of the grounding system. Anything connected to the grounding system will become energized to a greater or lessor extent during ground faults. This is true even with exactly proper grounding and proper separation of ground and neutral at the subpanels.
The point is that during a fault condition, a _circuit_ exists that includes the grounding system conductors. Current flows through this circuit, and the various parts of the circuit become energized. Hopefully a circuit breaker or ground fault detector trips and cuts off this current flow.
During these fault conditions, it doesn't matter if the subpanel neutral is bonded to ground or not. There is lots of current flowing through the grounding conductors, and significant potentials. In fact, during these fault conditions, it is quite likely that having a grounded subpanel would actually _reduce_ these potentials. But don't do it
The reason that you don't want to ground a subpanel is because of current flow during _normal_ operation. In normal operation, current is supposed to flow in the neutral conductor, but is not supposed to flow in the grounding conductors. If you connect the grounding conductors to neutral at more than one location, then you will _always_ have the grounding conductors in a circuit, and you will _always_ have some current in the grounding system.
Again, the point of single point grounding is to make sure that the grounded appliance chassis are electrically connected and at the same _voltage_ as the ground reference point, but that none of these elements are in a circuit, so that there is no _current_ flow.
-Jon
#22
05-27-04, 02:19 PM
Terry,
First, the above is the crux of your misunderstanding. A ground wire that is connected at one end only is _not_ a path back to the source. It is a potential return path, if a fault occurs, but without a fault, it is an open circuit. _Current_ does not _flow_ on an open circuit. _Voltage_ spreads out to the entire electrically connected object, but _current_ does not flow unless there is actually an electrically conductive path that it can follow. When the grounding conductors are connected at one point only, then there cannot be a current flow path.
Second, as I mentioned way back at the beginning of this thread, _outside_ of your house, multipoint grounding is used, and current flows on the ground electrodes are often found. Especially in urban areas with metal piping systems and several houses sharing a single transformer, considerable current can flow through the ground electrodes, and different houses will have different 'ground' potentials.
But, ideally, inside your house, you will have _single_ point grounding, no circuits which include your ground conductors, and no current flow on your ground system except during faults.
-Jon
First, the above is the crux of your misunderstanding. A ground wire that is connected at one end only is _not_ a path back to the source. It is a potential return path, if a fault occurs, but without a fault, it is an open circuit. _Current_ does not _flow_ on an open circuit. _Voltage_ spreads out to the entire electrically connected object, but _current_ does not flow unless there is actually an electrically conductive path that it can follow. When the grounding conductors are connected at one point only, then there cannot be a current flow path.
Second, as I mentioned way back at the beginning of this thread, _outside_ of your house, multipoint grounding is used, and current flows on the ground electrodes are often found. Especially in urban areas with metal piping systems and several houses sharing a single transformer, considerable current can flow through the ground electrodes, and different houses will have different 'ground' potentials.
But, ideally, inside your house, you will have _single_ point grounding, no circuits which include your ground conductors, and no current flow on your ground system except during faults.
-Jon
#23
05-27-04, 02:33 PM
Winnie
OK, the key here, then, is that since there is a bond at the main panel, but not at the other end, for example, a recepatcle, then there is no closed circuit? There is no parallel circuit? This just seems like a parallel circuit to me, but I guess not since the bond is at one end only. Do I have that right?
Thanks, Jon! I'm getting there, I think
Thanks, Jon! I'm getting there, I think
#24
05-27-04, 03:55 PM
OK, the key here, then, is that since there is a bond at the main panel, but not at the other end, for example, a recepatcle, then there is no closed circuit? There is no parallel circuit? This just seems like a parallel circuit to me, but I guess not since the bond is at one end only. Do I have that right?
-Jon
#25
05-27-04, 05:23 PM
" ( Why does) a Fault-current flow back to the source, but will (not?) flow to the appliances"
To return to my "Analogy"---- The "source" is a connection-point for 2 drain-lines--a "Normal" drain (N-D)for conducting "normal" water-flow ( Neutral current), and an "Emergency-drain" (E-D) for conducting an "abnormal" water-flow ( fault-current).
Both drain-lines extend from the "source", at Elevation 0, in parallel to the "appliance",at Elevation 10, where they make SEPERATE connections---the N-D to a connection inside the "appliance", and the E-D to a connection outside the "appliance" where the E-C will "collect" any "abnormal" water-flow.
There is a Elevation-difference of Zero beween the N-D and E-D at any point between the "source" and the "appliance"--- similiar to a "potential-difference" of Zero volts between the Neutral and EC in a Branch-Circuit cable.
In "normal" operation the water will flow by gravity thru the N-D from Elevation 10 to Elevation Zero ( the "source")---- this is the "Neutral current". It's obvious that no water will flow from the N-D into the E-D at the "source" connection and back to the "appliance" because of the Zero difference in Elevation between the two drain-lines. This should answer the question "Why isn't there a Neutral current thru the Ground wire (back) to the appliance?"
Also obvious is the fact that any water-leakage from the "appliance", at Elevation 10, will flow into the E-D from Elevation 10 to Elevation Zero.This should answer the question that preceeds this explanation.
With "Leakage", there could be water-flow (current) in both the N-D (Neutral) and E-D ( EGC).If the water-vessel were to burst or rupture "open", there would be an large volume of "abnormal"water-flow ( Fault-current) directly into the E-D and back to the "source".
To return to my "Analogy"---- The "source" is a connection-point for 2 drain-lines--a "Normal" drain (N-D)for conducting "normal" water-flow ( Neutral current), and an "Emergency-drain" (E-D) for conducting an "abnormal" water-flow ( fault-current).
Both drain-lines extend from the "source", at Elevation 0, in parallel to the "appliance",at Elevation 10, where they make SEPERATE connections---the N-D to a connection inside the "appliance", and the E-D to a connection outside the "appliance" where the E-C will "collect" any "abnormal" water-flow.
There is a Elevation-difference of Zero beween the N-D and E-D at any point between the "source" and the "appliance"--- similiar to a "potential-difference" of Zero volts between the Neutral and EC in a Branch-Circuit cable.
In "normal" operation the water will flow by gravity thru the N-D from Elevation 10 to Elevation Zero ( the "source")---- this is the "Neutral current". It's obvious that no water will flow from the N-D into the E-D at the "source" connection and back to the "appliance" because of the Zero difference in Elevation between the two drain-lines. This should answer the question "Why isn't there a Neutral current thru the Ground wire (back) to the appliance?"
Also obvious is the fact that any water-leakage from the "appliance", at Elevation 10, will flow into the E-D from Elevation 10 to Elevation Zero.This should answer the question that preceeds this explanation.
With "Leakage", there could be water-flow (current) in both the N-D (Neutral) and E-D ( EGC).If the water-vessel were to burst or rupture "open", there would be an large volume of "abnormal"water-flow ( Fault-current) directly into the E-D and back to the "source".
#26
05-27-04, 06:01 PM
Member
I must go back to two points I made earlier:
You often hear that electricity takes all available paths. That is true. But if 6 electrons go one way, and 6,000,000,000,000 go some other way, we normally ignore those 6 and pretend like all of the current went the other way. If you put your hand on your washing machine, a few electrons returning to the panel from your television will flow out of the panel to the washing machine receptacle, through the grounding wire on the washing machine, through your body, through the structure of your house, through the earth, and out to the transformer. The number, however, is so incredibly small, we pretend that it is zero.
- Current doesn't just run around randomly. It only flows towards a point of lower potential.
- The earth has about a billion times more resistance than copper does.
You often hear that electricity takes all available paths. That is true. But if 6 electrons go one way, and 6,000,000,000,000 go some other way, we normally ignore those 6 and pretend like all of the current went the other way. If you put your hand on your washing machine, a few electrons returning to the panel from your television will flow out of the panel to the washing machine receptacle, through the grounding wire on the washing machine, through your body, through the structure of your house, through the earth, and out to the transformer. The number, however, is so incredibly small, we pretend that it is zero.
#27
05-28-04, 12:01 AM
thanks all
Some very good, and understandable final posts from Winnie, PATTBAA and John N. Of course, again, I appreciate everyone's replies. I am beginning to see the light
Jon (Winnie): Although my question may have sounded as if I was referring to fault current, I was, in fact, thinking about normal current flow. But your explanations were very clear. Thank you.
PATTBAA: Although your post first struck me as being borderline technically over my head, your explanation was very good, and I thank you, and
John N: A ball does indeed have a rough time going uphill without some help. The analogy to the roughly 0 potential at the panel helped me understand a lot better. Thank you for your time.
Thanks to everyone. Will post back when things are hopefully safe.
-Terry
Jon (Winnie): Although my question may have sounded as if I was referring to fault current, I was, in fact, thinking about normal current flow. But your explanations were very clear. Thank you.
PATTBAA: Although your post first struck me as being borderline technically over my head, your explanation was very good, and I thank you, and
John N: A ball does indeed have a rough time going uphill without some help. The analogy to the roughly 0 potential at the panel helped me understand a lot better. Thank you for your time.
Thanks to everyone. Will post back when things are hopefully safe.
-Terry
