Technical Power System Definition - 640.3

Status
Not open for further replies.

dereckbc

Moderator
Staff member
Location
Plano, TX
I think we agree that the EGC normally never carries current unless there is a fault condition. (Except for "leakage" or "touch" current.)
In a Single Phase 120 volt circuit there is always leakage current in the EGC. That is the problem with unbalanced load supplies. In a Balanced 120/60 there is no leakage current and no Neutral Conductor is ran, just an EGC

I am saying that since the illustrations show that middle wire as connecting to the midpoint of the secondary, that it is by definition a Neutral as stated in Article 100. What I believe should be added to that illustration is an EGC.
The EGC is shown in both drawings, you are mistaken it for a neutral conductor

The same with 240 V single phase, 3-wire plug (plus Ground) for a Range or Dryer circuit. I suppose you could call the 240 V single phase a balanced circuit but I haven't heard anyone use that term before in the context of a range or dryer circuit. The current on L1 and L2 are in differential mode with one another - a difference circuit as the waveforms are in opposite polarity from one another. Any unbalance returns via the Neutral.

A 240 3-wire circuit is a Balanced Circuit. There is no imbalance because L1 Current = L2 Current, thus there is no neutral current and no neutral conductor is ran with it. It is impossible for any neutral current to flow. There is only an EGC in which there is never any current flowing on the EGC except in the event of a fault.

As Don and I have tried to explain you do not seem to understand the definition of a Grounded Circuit Conductor as it IS INTENDED TO CARRY NORMAL LOAD CURRENT. In a Single Phase 240/120 the neutral is needed to derive 120 volts. In a 3-Phase 208/120 Wye service the neutral is required to derive 120 volts. In a Balanced 120/60 no neutral is required because no 60 volt loads exist. Same for a 240 volt single phase 3-wire for a dryer or cook top as no 120 volt is needed. Now with that said many dryers, ovens, and cook tops electronics and fan motors operate on 120 volts. In that case you run a 4 wire circuit to derive 120 volts and you run L1, L2, N, and G.
 

Smart $

Esteemed Member
Location
Ohio
...

A 240 3-wire circuit is a Balanced Circuit. There is no imbalance because L1 Current = L2 Current, thus there is no neutral current and no neutral conductor is ran with it. It is impossible for any neutral current to flow. There is only an EGC in which there is never any current flowing on the EGC except in the event of a fault.

...
Please quit calling this a 3-wire circuit. It has no neutral conductor. The EGC is not counted as a circuit conductor.

L1, L2, and EGC is a 2-wire circuit.

L1, L2, N, and EGC is a 3-wire circuit.


A 120VAC balanced power source supplies only 120VAC 2-wire circuits: L1, L2, and EGC. There is no neutral circuit conductor.
 

smalltime

Member
Location
Roanoke, VA
<duncecap>
First, thanks for your patience. I probably would have been name calling by now. Second, my apologies for being so thickheaded.

I wasn't thinking beyond a dryer or range circuit and applied that same thought to the 120 V line-to-line (Art 647).

For 240 V, where I don't need to derive 120 V (welder, HW, kiln circuits, etc.) and for the 120 V line-to-line, I don't need a Neutral. The return current from the two line conductors cancel each other out.

I got seriously tripped up with the Equitech illustration as the diagram looks like what would be a Neutral at the source. It's just the EGC and since there's no Neutral, there's no bond with a Neutral, thus the EGC has to be connected back to the source for OCP.

Both of the above examples are 2-wire plus Ground, balanced circuits.

</duncecap> Maybe.

So back to my original issue with the TPS definition in Art. 640.2. A TPS is for "reduction of electromagnetic noise". Art. 647 is 120 V balanced power and that's for "sensitive electronic equipment".

I don't see where 640.2 defines a TPS as being a balanced power system. Couldn't this be using 120 V line-to-neutral branch circuits? Am I missing this?

If a TPS is using 120 V line-to-neutral branch circuits, I would expect to find: Line, Neutral, EGC (for grounding the boxes, conduit, panels, etc.) and an IG conductor. The IG would be isolated from the EGC until the bond back at the source.

This is my issue: 640.2 defines the TPS and says, "...where the equipment grounding conductor is isolated from the premises grounded conductor...". In a normal 120 V line-to-neutral, this would be the case anyway.

In a TPS branch circuit, as I am assuming above - a 120 V L-to-N, with EGC for grounding the boxes, conduit, panels, etc., I would also have an IG conductor. In other words, I should have... the IG equipment grounding conductor isolated from the premises (equipment) grounding.

Is that making sense?
 

dereckbc

Moderator
Staff member
Location
Plano, TX
In a TPS branch circuit, as I am assuming above - a 120 V L-to-N, with EGC for grounding the boxes, conduit, panels, etc., I would also have an IG conductor. In other words, I should have... the IG equipment grounding conductor isolated from the premises (equipment) grounding. making sense?
Yes I understand you and have the answer but short on time right now. Wife's birthday and she is Hungary. For now it cannot be Isolated from the premises ground. This trips a lot of people up and when I get back will walk you through it. Once done you will understand why recording studios went with Balanced power systems.
 

smalltime

Member
Location
Roanoke, VA
That's really weird... It's my wife's birthday today as well. But she's at choir practice tonight. We'll celebrate tomorrow.

In any event, I didn't want to so much get into the merits of balanced power as I did want to properly get a handle on the definition of a TPS. Is it really L + N + EGC (for boxes, conduit, etc.)+ IG and if so, is the definition as worded in 640.2 incorrect?

I know TPS and Art.647 are a bit esoteric so I know we're not discussing everyday stuff here. There are a lot of venues that want a TPS but not to the extent of going with Balanced Power. (Large churches, Performing Arts Centers, etc.) In other words, they want the IG to limit noise currents and potential ground loops from flowing on their isolated equipment ground conductor and that has merit as well.

I know AV people are using a definition out of a book called Audio Systems Design and Installation by Philip Giddings. It says,
?The terms technical power, technical power panel (or panelboard), and technical power system, and, as discussed in a following subsection, technical ground, are used to denote dedicated panelboards, equipment, and systems used for the purposes of supplying AC power to technical systems such as audio, video, and computers.?

And it goes on to say, "Technical power may be implemented in systems of high reliability which may not require a technical ground, such as an emergency evacuation system having a signal-to-noise requirement of about 50 dB. Technical grounding is necessary only where electrical noise (EMI) is an anticipated problem. Technical ground systems may be implemented in systems without technical power, such as a home studio."

"Technical Ground" equals IG to Giddings.

So Giddings is much more loose in his definition of what constitutes "Technical Power" but his book was published 9 years before the NEC defined a TPS in 1999. No wonder there's confusion when I see Technical Power in a specification.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
... In other words, they want the IG to limit noise currents and potential ground loops from flowing on their isolated equipment ground conductor and that has merit as well. ...
I have never seen any evidence that IG systems do much of anything, other than drive up the cost of the installation.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
OK first it is important you understand how Balanced Power Systems operate and induce noise into ground circuits. So that was not a waste of your time and I am not sure you really understand the point of how a 120 volt asymmetrical system induces noise currents into grounding circuits.

First define NOISE as any unwanted signal. I do not care what that signal is, if it is unwanted, it is a noise. For Sparkies, no offense intended guys, grounding conductors can and do serve other purposes other than safety. But Sparkies are only concerned and have been taught safety and code. The code recognizes ground circuits can and do serve other purposes, and makes allowances for modifications to grounding systems to mitigate NOISE. In the communications world ground also serves as a SIGNAL REFERENCE POINT in addition of power. Problem is the 120 volt single phase world is inheritable noisy by injecting unwanted currents into grounding conductors.

The unwanted currents, or NOISE, is injected by two primary sources:

1. L-G and N-G cable capacitance. As the circuit conductors get longer, the coupling capacitance increases, and thus injects or couple 60 Hz line power and harmonic frequencies into the ECG conductors.
2. At the utilization equipment most electronic devices like amplifiers or really anything with a semi-conductors have RFI and surge protections devices installed between L-G and N-G on the incoming line power conductors. Although these devices have fairly high impedance to power line frequencies they still inject power line frequencies into the EGC. The L-G common mode is th emost offensive of the two.

OK number 1 & 2 are additive and compound the problem. Result is you get some 60 Hz line current flowing in ground conductors. Not enough to generate a high enough voltage to be a safety concern, but more than enough to drive signal circuits crazy. All of us know one of the most common problems that annoying hum in PA speakers, stereos, ect.

With that said let me say Isolated Ground Receptacles are antiquated and really serve no purpose other than to pad a Sparkies check and give a warm fuzzy feeling to the AV nerd. In the old days of stereos, PA systems etc used unbalance signal transmission meaning they used ground as a signal path and in the COMMON MODE. Today most all communications circuit used either balanced signal transmission, Wireless RF, or completely isolated with optical or digital balanced Pulse Code Modulation.

So how did that annoying hum get into the audio stream? Well it came in at low signal levels called Line Level. Line Level would be the output of a microphone, electric guitar, mixer, pre-amps etc... All those analog line levels used ground as a signal path and reference level. Remember line level is very low power at 600 Ohms @ 0 dB or about 1 volt RMS at full output level. So here is what happens you have a guitar on stage through a long line cord to a stage mixer Powered on a AC circuit. From there to the main mixer board and EQ powered by another AC circuit. From there to a rack of Power Amps powered again by another AC circuit. See where this is going. Each of the AC circuit receptacles are referenced by the EGC. Along the length of th eEGC noise current is flowing. All the mics, pre-amps, and gizmos are connected the EGC at different points along its length. There is impedance between those points of connections. If there is a current flowing through an impedance then there is a Voltage Drop. Th emixers and mostly the Power Amps are picking up those very small signals and amplifying them and you hear it as that dang annoying HUM.

So some 50 years ago a bunch of engineers came up with the idea of a Isolated Ground Receptacle, or the forth wire in a standard 120 VAC circuit. It was magic so most thought. In reality an IGR has 3 outcomes, and what follows is liste din order of what happens.

1. Nothing
2. Makes the problem worse.
3. Desired effect.

The idea or principle is sound but corrupted by either installation or the connected equipment itself. That principle is a SINGLE GROUND POINT of power supplies, equipment racks, and signal point. In a true SPG there can be NO current flowing. If NO currents flow, there is no noise or voltage. For current to flow there has to be a complete circuit or a loop for current to enter and exit. It works if you can pull it off, but very easy to corrupt and just about all the signal ground reference equipment corrupts the SPG by the interconnect cable carrying the signal you are trying to clean up.

In reality it is an easily solved problem especially in a home environment or even commercial. Just pug all the equipment into the same receptacle. OK so you are saying that only has two ports. Well install a quad or even better by a decent Power Strip with 10 ports. That way all the EGC are physically connected to a SINGLE POINT, thus no difference of potential between them. No potential difference, no NOISE

OK Smalltime now that is out of the way and hopefully you understand what I just said lets answer your question. Don't get offended. But your understanding of 640.2 is incorrect. No ground can be isolated from the premises ground. To do so would not only be extremely dangerous, but you give you all kinds of technical problems like NOISE.

What 640.2 is telling you is the Isolated Ground is allowed to pass through all panels and junctions without bonding to them like you are required to do with standard EGC. It allows you to go all the way back to the service entrance point where the holly grail of ground is right at the Main Bonding Jumper point where N & G are bonded to each other. Th eN-G bond is where the real SINGLE GROUND POINT originates or ZERO POTENTIAL reference point. It does not have to go all the way back to the service entrance, but allowed if you wish to do so and most do prefer that point.

However someone who really knows what they are doing will not use any IGR as we have a much more effective means at our disposal. We create a new reference ground point near where we need and use it. How you ask? Simple we install a Isolation Transformer with a new N-G bond point. A Isolation transformer even a dry one gives us at least 60 dB of common mode noise rejection isolating the output from everything. Use a true isolation transformer like farromagnetic resonator and we can get 140 dB of common mode noise rejection. FWIW 60 dB = 10,000,000 : 1 or reduces 10 million volts to 1 volt, 140 dB is 1,000,000,000,000,000 : 1 rejection. Only God can comprehend that number)

Ok today none of that is needed. Modern engineering of electronics has eliminated all those headaches and problems. On stage they use Wireless mics and equipment pick ups, optical, digital, or balanced analog signal transmission and whatever power is available.

See how that word "balanced" keeps popping up? ;)

Two last points to blow your mind but very true.

1. In any stick built house that uses NM cable and plastic junction and receptacle boxes is an Isolated Ground Circuit. Well that assumes you only have one Main Panel and no sub panels.
2. Any 2-wire plus ground in a stick built house as above is a Balanced Isolated Ground Circuit.
 
Last edited:

smalltime

Member
Location
Roanoke, VA
So that was not a waste of your time...

Learning and understanding is never a waste of time.

First define NOISE as any unwanted signal.

I always have defined it that way.

OK Smalltime now that is out of the way and hopefully you understand what I just said lets answer your question. Don't get offended. But your understanding of 640.2 is incorrect. No ground can be isolated from the premises ground. To do so would not only be extremely dangerous...

No offense taken. I get this. It was drilled into me many years ago. The 3-to-2 prong "grounding adapter" is a most dangerous device. If you recall my original post, I said, "In other words the IG grounding conductor is isolated (insulated) from the EGC until a single grounded termination point." That's back at the Service Entrance or Separately Derived Source just as you describe.

What 640.2 is telling you is the Isolated Ground is allowed to pass through all panels and junctions without bonding to them like you are required to do with standard EGC. It allows you to go all the way back to the service entrance point where the holly grail of ground is right at the Main Bonding Jumper point where N & G are bonded to each other. Th eN-G bond is where the real SINGLE GROUND POINT originates or ZERO POTENTIAL reference point. It does not have to go all the way back to the service entrance, but allowed if you wish to do so and most do prefer that point.

Really, I do get this. Absolutely. I also teach this to others. The use of the word "Isolated" in itself causes misunderstanding, I think.

Unfortunately, this doesn't fix the potential issue I see with definition of TPS in 640.2:

If a TPS is using 120 V line-to-neutral branch circuits, I would expect to find: Line, Neutral, EGC (for grounding the boxes, conduit, panels, etc.) and an IG conductor. The IG would be isolated from the EGC until the bond back at the source (as I've said before).

This is my issue: 640.2 defines the TPS and says, "...where the equipment grounding conductor is isolated from the premises grounded conductor...". In a normal 120 V line-to-neutral, this would be the case anyway.

Shouldn't it say, "...where the equipment grounding conductor is isolated from the premises groundING conductor..." The Neutral would be insulated (isolated) anyway until the MBJ. Shouldn't the definition of a TPS, an IG system, be saying the the normal EGC be isolated IG until the MBJ.

See how that word "balanced" keeps popping up? ;)

Balanced circuits have been popping up since their origin in the early days of telephony. I have a copy of a Jan 1935 Bell System Technical Journal called Wide Band Transmission Over Balanced Circuits. If you read it you'll find that it's really about an impedance balance between the two conductors. It just so happens that with electrical power sources you also have signal symmetry. So you get both.

Two last points to blow your mind but very true.
1. In any stick built house that uses NM cable and plastic junction and receptacle boxes is an Isolated Ground Circuit. Well that assumes you only have one Main Panel and no sub panels.

Agreed.

2. Any 2-wire plus ground in a stick built house as above is a Balanced Isolated Ground Circuit.

Agreed. As long as it's a 240 V 2-wire. A 120 V derived from 240 V would not be balanced as the impedances are different on the two conductors because one of them is connected to ground. (Referencing the Jan '35 BSTJ)

As far as noise being injected onto the EGC, NM has another benefit - its physical construction. The EGC lies perfectly in between the L and N and since the magnetic fields that are generated by the L and N are opposite, the net result is that they cancel each other out in the EGC. Again, because it lies perfectly in between the two current carry conductors generating opposing magnetic fields. I know people that have been measuring this. Another method that has been successful in keeping noise out of the EGC is twisting the L and N together before pulling it into the conduit. The physical twisting of the two conductors around each and generating opposing magnetic fields generates little energy to be induced into the EGC (raceway or copper wire) or any other surrounding circuits.

It doesn't fix problematic ground loops, but that's a different issue.

This has all been great, but I'm still not finding an answer or resolution to my original issue in regards to the definition of a TPS.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
This is my issue: 640.2 defines the TPS and says, "...where the equipment grounding conductor is isolated from the premises grounded conductor...". In a normal 120 V line-to-neutral, this would be the case anyway.

Shouldn't it say, "...where the equipment grounding conductor is isolated from the premises groundING conductor..." The Neutral would be insulated (isolated) anyway until the MBJ. Shouldn't the definition of a TPS, an IG system, be saying the the normal EGC be isolated IG until the MBJ.

OK to me it sounds like your are cherry picking. The definition is:

Technical Power System. An electrical distribution system
with grounding in accordance with 250.146(D), where the
equipment grounding conductor is isolated from the premises
grounded conductor except at a single grounded termination
point within a branch-circuit panelboard, at the
originating (main breaker) branch-circuit panelboard, or at
the premises grounding electrode.


Now we have to reference 250.146 (D) and put the two together.

250.146 (D) Isolated Receptacles. Where installed for the reduction
of electrical noise (electromagnetic interference) on the
grounding circuit, a receptacle in which the grounding terminal
is purposely insulated from the receptacle mounting
means shall be permitted. The receptacle grounding terminal
shall be connected to an insulated equipment grounding conductor
run with the circuit conductors. This equipment
grounding conductor shall be permitted to pass through one
or more panelboards without a connection to the panelboard
grounding terminal bar as permitted in 408.40, Exception,
so as to terminate within the same building or structure directly
at an equipment grounding conductor terminal of the
applicable derived system or service. Where installed in accordance
with the provisions of this section, this equipment
grounding conductor shall also be permitted to pass through
boxes, wireways, or other enclosures without being connected
to such enclosures.


To me all that is saying is reinforcing the fact that the Grounded Circuit Conductor can only be bonded once as a warning not to bootleg a ground at a receptacle (bonding N-G again downstream of entrance). Bootleg grounds are quite common with IT and AV morons who do not understand what they are doing. Personally 640 is a waste of paper to me as it is already covered elsewhere many times. It appears to be written just for AV and IT weenies to teach them they have been doing it wrong for decades. That is my best guess anyway. Otherwise it sounds like you understand.

FWIW I am a Telecom Power Protection Engineer.
 
Last edited:

smalltime

Member
Location
Roanoke, VA
OK to me it sounds like your are cherry picking.

I'm really not trying to cherry pick. If a TPS is specified, I just want to make sure everyone understands exactly what is being specified and how its expected to be installed.

To me all that is saying is reinforcing the fact that the Grounded Circuit Conductor can only be bonded once as a warning not to bootleg a ground at a receptacle (bonding N-G again downstream of entrance).

I don't know why even a moron would look at the white and green/yellow and go, "Hey, I have a good idea, let's connect these together." Interesting perspective, though.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
I don't know why even a moron would look at the white and green/yellow and go, "Hey, I have a good idea, let's connect these together." Interesting perspective, though.
Because it works in some cases, and done frequently. If you have a 2 wire with no ground amp, signal ground is chassis inside the amp which is bonded to Neutral. Connect a preamp or a device with a equipment ground and now you have significant potential difference at the amp signal ground and preamp.

Down stream from the service, the grounded circuit conductor develops a voltage drop with respect to ground. The further you get away from the entrance and the more current, the higher the voltage is. Most sensitive equipment specifies a maximum N-G voltage drop of 2 volts or less. Lower to zero is better. That equates to a 3% voltage drop, 2 volts on neutral,and 2 volts on L. So now you have a plant technician or a EC looking to make a quick buck and you get a call stating you have a 3 volt N-G voltage on your gizmo. Fix it. Bootleg the ground by bonding the N-G at the receptacle. Problem solved in 5 minutes and a $100 check in your pocket.

The right solution is pulling new over sized circuit conductors to minimize the voltage drop to acceptable levels. but that takes a lot of money and time to do it right. Or use an Isolation Transformer like they sell at high end AV shops.
 
Last edited:
Status
Not open for further replies.
Top