Stereophile electrical system improvement article

[Ingenieur]

View attachment 161704

Can’t make it more clear.
Wit a broken PEN a RCD/CGFI/FI won’t trip because IL=IN
Ipe = IL - "earth rod current"

In other words Vtouch = Vline * (R ground rod / (R ground rod + R Appliance Load) )

Or the lower the earth rod resistance the lower Vtouch

It will trip a 20 A CB without the person
since G < 2 Ohm whole system.
120/2 ~ 60 A

If isolated
With the person being the path, 1 Ohm or 25 no difference since <<<< 1000
1001 or 1025, a few mA delta.
The GFCI will trip
I will pass thru the X0 CT but not return thru it, it will return to the source via its rod.

You realize there are ungrounded delta 3 ph systems?
And you can touch a phase and not be shocked.

Facts
Irrefutable

 

[Lambda]

It will trip a 20 A CB without the person
since G < 2 Ohm whole system.
120/2 ~ 60 A

At this point i’m not even sure what your trying to tell me or if you even listening.

If you referring to the picture i posted:
No fuse (CB) wont trip because current is limited by Load/Appliance impedance
Ground Impedance in the picture is not <2Ohms it is what ever your local ground rod impedance to earth is (of cause not the rod itself)
So assume worst case 25 Ohm for ground impedance.

Assume a ground rod to ground impedance to ground of 25 Ohms and a load impedance of 6ohm ~2,2kW
So Vtoutsh is 110V*(25/(25+6))=~88V
vs 5ohm ground rod impedance 110V*(5/(5+6))=~50V

It is not 2 phase.
I would not call it that since it is incorrect.
They are referred to as Line, not phase.
It is single split phase wired with a coil in opposition. The source is single phase.

And how would 2 pahse with 180° phase angle look like?

 

[Ingenieur]

Just to refute misinformation for the sake of safety:

A GFCI will work with NO ground, only a H and N.
Obviously if the service N is disconnected no 120 will work properly, V will sag.
Assume a 120 W load at 120 V, 1 A 120 Ohm. If the earth path is 20 Ohm the load will see 120/140 ~ 0.85 A and 103 V.

If you cut the N to it, drops out
In fact for a 2W ckt. It may be more important.

This assumes a 25 Ohm ground
120 V 10 Ohm (12 A load)
With utility and <25, usually <3 total if code compliant, the frame will rise 28 V.
A 1000 Ohm person would see 28 mA
He could pull away from that in most cases.
The fact that there are <100 residential fatalities out of 330 million, most from people working on stuff, no GFI (code non-compliance) , ladders near power lines, is a testament to the safety of the system.

 

[Ingenieur]

At this point i’m not even sure what your trying to tell me or if you even listening.

If you referring to the picture i posted:
No fuse (CB) wont trip because current is limited by Load/Appliance impedance
Ground Impedance in the picture is not <2Ohms it is what ever your local ground rod impedance to earth is (of cause not the rod itself)
So assume worst case 25 Ohm for ground impedance.

Assume a ground rod to ground impedance to ground of 25 Ohms and a load impedance of 6ohm ~2,2kW
So Vtoutsh is 110V*(25/(25+6))=~88V
vs 5ohm ground rod impedance 110V*(5/(5+6))=~50V



And how would 2 pahse with 180° phase angle look like?

You are making little sense.
You are saying if you remove the ground safety from the system it is less safe.

The earth is NOT an allowable fault path.
The code clearly state that.
Lower is not better.
Both are hazards.

I guess you could argue to run 3 grounds to the xfmr.
Monitor them
Etc.

But the only no risk is no power.
Then physical injuries would increase

An open N or G on any system will do that.
Your G is no lower than ours.
A 1000 Ohm person on 220 V with a 1 Ohm bed is worse of than a person on 120 V with a 25 Ohm bed.
It is that simple.

If you are counting on the earth ground for circuit protection you are making a mistake.

 

[Lambda]

A 1000 Ohm person would see 28 mA
He could pull away from that in most cases.

Most cases is good enough?

The fact that there are <100 residential fatalities out of 330 million,

So now your the one saying <100 residential fatalities out of 330 million is good enough?
I think there is no good enough and we should do everything to make it zero fatalities?

The earth is NOT an allowable fault path.

So if its not allowable the fault can’t happen? This sounds very German to me?

A GFCI will work with NO ground, only a H and N.

Never said the opposite.
CFGI will Trip if the current over live different to current over neutral (lets call call them il and in)
Now look at the pictur:

https://www.audiosciencereview.com/forum/index.php?attachments/1635351476978-png

il=in
So the current over live and neutral is the same in the CFGI. therefore CFGI wont trip.

 

[Ingenieur]

Most cases is good enough?

So now your the one saying <100 residential fatalities out of 330 million is good enough?
I think there is no good enough and we should do everything to make it zero fatalities?

So if its not allowable the fault can’t happen? This sounds very German to me?


Never said the opposite.
CFGI will Trip if the current over live different to current over neutral (lets call call them il and in)
Now look at the pictur:

https://www.audiosciencereview.com/forum/index.php?attachments/1635351476978-png

il=in
So the current over live and neutral is the same in the CFGI. therefore CFGI wont trip.

The code has no control over a persons weight, health, clothing, standing on a wet floor, etc.

No, that is not the goal, that is the metric to analyze and implement measures.
The goal is 0, the code is revised to reduce that number. Identify the cause, implement a solution. In 1980 they did not say 500 is ok, now below 100.
In context
1980 500/230 mil ~ 2.2 per mil
2020 <100/330 mil ~ 0.3
A 700% reduction

Faults happen
You focus on the most common first
Broken N, rare
Contact with live wire, common
Use 120 instead of 220, better odds of survival
Shock in wet areas, GFCI

 

[Lambda]

120 V 10 Ohm (12 A load)
With utility and <25, usually <3 total if code compliant, the frame will rise 28 V.

So your using "usually" Values instead of worst case values?!
If calculating safety measures here we always use worst case plus a safety margin.

So why would you assume the whole house is only a 12A lode?
If calculating emergency braking system for an elevator we don’t use an usually only halve full elevator wight...

The code has no control over a persons weight, health, clothing, standing on a wet floor, etc.

So? The saver way would be to assume the worst case. but they don’t because the chose the "Save enough" for them compromise.
Becuase its normal and ok to makes compromise even if it comes to Safety.

Use 120 instead of 220, better odds of survival
Shock in wet areas, GFCI

And you have GFCI on every circuits?

Also i was only talking about the special case of a PEN fault (you keep calling it N but in this case its PE combined with N )
You don't Acknowledge that GFCI wont help in this case?!
Looking at "the picture" it should be clear net current in the GFCI sense coil is 0 and therefor it won’t trip.

But you keep avoiding this and instead you make assumptions under which the system proposed by you or your local code (barely) is good enough.
Again what kind of industry uses normal or usual values when calculating for safety measures?

A lower "Additional earth electrode" impedance (to ground) means lower voltage on PE in case of A PEN fault.
and is therefore among other things more save.

Also bigger area and deeper earth electrode means lower step voltage on this area because the are becomes a equipotential area.

 

[BlackTalon]

No, that is not the goal, that is the metric to analyze and implement measures.
The goal is 0, the code is revised to reduce that number. Identify the cause, implement a solution. In 1980 they did not say 500 is ok, now below 100.
In context
1980 500/230 mil ~ 2.2 per mil
2020 <100/330 mil ~ 0.3
A 700% reduction

87% reduction (sorry for being petty)

 

[Ingenieur]

So your using "usually" Values instead of worst case values?!
If calculating safety measures here we always use worst case plus a safety margin.

So why would you assume the whole house is only a 12A lode?
If calculating emergency braking system for an elevator we don’t use an usually only halve full elevator wight...


So? The saver way would be to assume the worst case. but they don’t because the chose the "Save enough" for them compromise.
Becuase its normal and ok to makes compromise even if it comes to Safety.



And you have GFCI on every circuits?

Also i was only talking about the special case of a PEN fault (you keep calling it N but in this case its PE combined with N )
You don't Acknowledge that GFCI wont help in this case?!
Looking at "the picture" it should be clear net current in the GFCI sense coil is 0 and therefor it won’t trip.

But you keep avoiding this and instead you make assumptions under which the system proposed by you or your local code (barely) is good enough.
Again what kind of industry uses normal or usual values when calculating for safety measures?

A lower "Additional earth electrode" impedance (to ground) means lower voltage on PE in case of A PEN fault.
and is therefore among other things more save.

Also bigger area and deeper earth electrode means lower step voltage on this area because the are becomes a equipotential area.

Those are worse case values
With the utility bonded <2 Ohm
If tied to a metal water piping system and slab, ~1 Ohm

You have never calculated a safety value, lol. That is obvious with your 'good enough' some deaths are OK philosophy.
No, strive for 0. A 700% reduction, it will go lower. 0 is the target.

We calculate on measured values using statistical methods. We would not use 1 MOhm or 0 Ohm.

Why would you assume things will run without a N?
I am a member of the A17 committee.
Do you know what OCW is.
The elevator will not move at 50% load 50% OCW.
We test at 125% descending
0% ascending

No such thing as 'emergency braking'
There the electrical motor brake
Machine brake
Safeties on the car

A lower ground does not increase safety from a line shock. May make it worse.
Lost N vs line shock
A million times more likely.

If you disconnect a safety system you increase risk
Changing something unrelated does not reduce or mitigate it.

Again, do you know what step potential is? Our grounds are the same order of magnitude as yours, perhaps less.
Out V is 1/2 which greatly reduces risk.
That is why out systems are safer if maintained.
<100 out of 330,000,000

You are much safer with a 25 Ohm bed and 120 V than a 1 Ohm and 231.94 V

You are misinterpreting our code.
One rod can't exceed 25 Ohm
You bond to other points plus the utility.
<3 Ohm

 

[Ingenieur]

87% reduction (sorry for being petty)

Let me clarify, thanks

2.2/0.3 x 100 a factor of 733%

 

[Lambda]

I am a member of the A17 committee.

I hope the other Member's are better at following a discussion about a hypothetical fault scenario without getting all emotionally.
So seems like you don't understand how GFCIs work since you think it would trip in the case of a broken PEN?!

Why would you assume things will run without a N?

Because there is current going to your utility's and the ground rod?!
The line voltage would drop (because Neutral voltage would rise because the return part trout ground is not as low impedance)
So depending on Load an depending on your Ground resistance your devices might not work.

But a situation where you have no lights and apparently no power but device enclosures are at potentially Dangerous voltages
is not save because you keep repeating this.

Can’t make it more clear.
Wit a broken PEN a RCD/CGFI/FI won’t trip because IL=IN
Ipe = IL - "earth rod current"

In other words Vtouch = Vline * (R ground rod / (R ground rod + R Appliance Load) )

Again, do you know what step potential is?

Yes.
Step potential is the voltage difference between
the feets of a person near an energized, grounded
Object.
occurs when a current flows to earth through an earth impedance, the potential relative to a distant point on the Earth is highest at the point where current enters the ground, and declines with distance from the source.

i.a. If the ground rod is right at the surface the potential per step is highest right if you stand right next to the ground rod.

Don’t you think if the grounding electrode is bigger, lower impedance and buried 3m deep underground the step potential over it would be lower (for a given current)
compared to the step potential over or right next to a rod?

You bond to other points plus the utility.

So do we? but with thicker gauge wire.

Do you know what OCW is.

You see there is obviously kind of language barrier and its not getting Better that you keep using halve sentences and abbreviations.

 

[Ingenieur]

It's got the point where the systems and rules are safer than the people.

I have researched 100's of electrical accidents from the last 30 years.
Over 90% have one thing in commom
Violation of work rules
Lock out/tag out/verify
Clearances
PPE
Training
Misuse of equipment

If observed you could count accidents due to equipment fault/failure on one hand.

But people are smarter than the system.
Faster than electricity

When we analyze an accident
Did they kill themselves unintentionally
Was there a violation of rules
Would these rules have prevented it
Can we do something more

Usually no. If someone intentionally does wrong we can't always protect them from themselves

You build a better mousetrap and along comes a smarter mouse

 

[Ingenieur]

I hope the other Member's are better at following a discussion about a hypothetical fault scenario without getting all emotionally.
So seems like you don't understand how GFCIs work since you think it would trip in the case of a broken PEN?!


Because there is current going to your utility's and the ground rod?!
The line voltage would drop (because Neutral voltage would rise because the return part trout ground is not as low impedance)
So depending on Load an depending on your Ground resistance your devices might not work.

But a situation where you have no lights and apparently no power but device enclosures are at potentially Dangerous voltages
is not save because you keep repeating this.




Yes.
Step potential is the voltage difference between
the feets of a person near an energized, grounded
Object.
occurs when a current flows to earth through an earth impedance, the potential relative to a distant point on the Earth is highest at the point where current enters the ground, and declines with distance from the source.

i.a. If the ground rod is right at the surface the potential per step is highest right if you stand right next to the ground rod.

Don’t you think if the grounding electrode is bigger, lower impedance and buried 3m deep underground the step potential over it would be lower (for a given current)
compared to the step potential over or right next to a rod?


So do we? but with thicker gauge wire.


You see there is obviously kind of language barrier and its not getting Better that you keep using halve sentences and abbreviations.

No snark necessary cupcake.

Emotional? You're feverishly posting BS and your fingers are bleeding. Lol

No emotional content on my end, hope so on yours. You do not exist to me.
Just another internet expert.

I know one thing: they are smarter that you. Emergency brake loll
I followed your 'imaginary' scenario.
We call it mental masturbation.
Lower grounding will not help personnel shock.

No language barrier, only a shortage of understanding, education and intellect.

It will trip with no ground. The person completes the fault path. A ground is not required will it drop out on UV?

So if you pull the N off a lamp it will light?
You are saying the ground path is 25 Ohm
If the load is 25 Ohm the device will run on 60 V? And a fault will not be noticed?

OK what does step potential have to do with a person touching a frame at elevated V? You are all over the place.
Crazy town.

Step potential at a 120 V rod may be a few V, a few mA, a person would likely not even sense it. You are reaching now.

No, the rod it driven 10' and in wet soil. Likely no gradient at the surface.
That is why rods are used rather than shallow plates.

What gauge do we use?
What is the ratio of its R to the electrode?
100:1, 1000:1, lol

A lower ground bed R offers no personal protection from line contact shock.
240 V is more dangerous than 120.
Fact

You system offers no more protection against an open N than ours, likely less due to the higher V.
In addition our ground beds on average are lower.

 

[Ingenieur]

I hope the other Member's are better at following a discussion about a hypothetical fault scenario without getting all emotionally.
So seems like you don't understand how GFCIs work since you think it would trip in the case of a broken PEN?!


Because there is current going to your utility's and the ground rod?!
The line voltage would drop (because Neutral voltage would rise because the return part trout ground is not as low impedance)
So depending on Load an depending on your Ground resistance your devices might not work.

But a situation where you have no lights and apparently no power but device enclosures are at potentially Dangerous voltages
is not save because you keep repeating this.




Yes.
Step potential is the voltage difference between
the feets of a person near an energized, grounded
Object.
occurs when a current flows to earth through an earth impedance, the potential relative to a distant point on the Earth is highest at the point where current enters the ground, and declines with distance from the source.

i.a. If the ground rod is right at the surface the potential per step is highest right if you stand right next to the ground rod.

Don’t you think if the grounding electrode is bigger, lower impedance and buried 3m deep underground the step potential over it would be lower (for a given current)
compared to the step potential over or right next to a rod?


So do we? but with thicker gauge wire.


You see there is obviously kind of language barrier and its not getting Better that you keep using halve sentences and abbreviations.

You need to relax.
It's ok to be wrong.
It's not ok to think you are right when blatantly wrong.

No emotion.
You have 0 influence over my life.
I can only hope you have none over that of others. At least as far as personal safety based on your 'good enough' kill a few is OK mindset.
So unwind yourself lol

 

[Blumlein 88]

The only way to decide who is right is a blind test.

Build a home which meets the German code and one the USA code. Release 100 blind folded people to wander around touching electrical stuff. See which version injures or kills the most people.

 

[Ingenieur]

Here's another reason ours is safer:
We have 2 lines 180 deg out of phase, opposite polarity. The N only carries the imbalance. So if loads are equally balanced the ground I will be 0.
All heavy loads are 240: dryer, AC, stove, etc. They do not use the N.
Lights only contribute a few amps and at 2-3 Ohm raise the frame <10V or <10 mA shock, barely perceptible.
In the rare instance that it happens.
Maybe a few times a year nationwide.

Even the 120 that flows thru ground will be of mixed poles and cancel.

That is why it is not an issue. You generally will have none or very small N current.

2 phases of 3 phase do not add to 0.
All 3 must be balanced. 400/230/3 wye is dangerous for a home. It only benefits the utility with load balancing.
In the US this is only allowed for commercial/industrial use due to complication and risk.

 

[Lambda]

No emotional content on my end, hope so on yours. You do not exist to me.

Don’t downplay yourself fellow internet expert. your sure Abel to amuse me and some other.

Emergency brake lol

The detailed regulations for elevator brakes in elevator systems are documented in EN 81. The standard stipulates that two brakes or brake circuits that work independently of one another must be present in an elevator drive ( redundancy ). This ensures that the car is held safely even in the event of a total brake circuit failure.

But lets not get ahead of ourselves i’m sure your an elevator expert to. I’m not but this is beside my point.

It will trip with no ground
The person completes the fault path
A ground is not required

I never said the RCD needs a ground
As reminder how a RCD works:

Now go back to the picture i posted and trace the current with your fingers if you need to...

Current goes in the RCD over Live (IL) the same current goes to the appliance
From the Appliance the current flows over Neutral (IN) back to in the RCD
(Notice current on line and neutral are the same but with opposite direction therefore they chancel out and the RCD is operating normal without triping)
This is alle the RCD "sees" and cars about.
The current Coning out of the RCD over N splits most of it goes to the Local Earth electrode and over earth back to the transformer.
But some of the current is going over the "link" from N to PE (IPE) this current IPE goes to the person into ground and from there into the transformer.

Can you follow me? have i layed out the fault scenario clearly? and if so can you explain how the RCD can detect current going over IPE and the person?

We have 2 lines 180 deg out of phase, opposite polarity. The N only carries the imbalance.

And this is diffrent in a 3 phase system?

 

[Blumlein 88]

What about the newer AFCI breakers? An arc from a broken circuit can set them off as well as too many things that aren't due to wiring.

 

[DonH56]

AFCI breakers were a new code change when we finished our basement about ten years ago. They were not ready for prime time (hopefully better now) and would trip on almost anything. We went through three AFCI breakers because whenever we turned on the treadmill it would trip and send my son's bedroom into darkness. Bad planning by the electrician, but as for the AFCIs he hated them and said all new construction required them and he was constantly having to deal with irate customers. He said most vacuum cleaners would set them off (we tried ours; one did, one didn't). After the third time I bought an old-fashioned breaker and replaced it, put the treadmill on a GFCI outlet, and called it a day.

This thread should probably just be ditched, it has become such a train wreck of misunderstanding/mistranslation and insults.

 

[Ingenieur]

Don’t downplay yourself fellow internet expert. your sure Abel to amuse me and some other.




But lets not get ahead of ourselves i’m sure your an elevator expert to. I’m not but this is beside my point.


I never said the RCD needs a ground
As reminder how a RCD works:
View attachment 161757
Now go back to the picture i posted and trace the current with your fingers if you need to...

View attachment 161756

Current goes in the RCD over Live (IL) the same current goes to the appliance
From the Appliance the current flows over Neutral (IN) back to in the RCD
(Notice current on line and neutral are the same but with opposite direction therefore they chancel out and the RCD is operating normal without triping)
This is alle the RCD "sees" and cars about.
The current Coning out of the RCD over N splits most of it goes to the Local Earth electrode and over earth back to the transformer.
But some of the current is going over the "link" from N to PE (IPE) this current IPE goes to the person into ground and from there into the transformer.

Can you follow me? have i layed out the fault scenario clearly? and if so can you explain how the RCD can detect current going over IPE and the person?



And this is diffrent in a 3 phase system?

The difference between you and I, I know I'm an idiot, you don't realize you are.

err, yea. I said that without a picture and fewer words.
The GFCI does not need G, only L and N

In that scenario with our 2 Ohm ground it would trip the 20 A branch.

How much current can a 25 Ohm ground carry before the lights go out?
120/25? 4.8A so 0 V on line?
So in other words any imbalance between lines would drop V to 60 at a sum current of 2.4 A. That would be noticeable.

Some GFIC will drop out on low V,
Do smart meters detect loss of N?

Yes, 2 legs of 3 phase do not cancel, even if of equal magnitude. 'phasors'.
That is why your system is much more dangerous.

With our system the N under worse case only carries a few amps x 2 Ohm frame rise.

Our grounded appliances are not isolated from ground.
That is the flaw in your scenario.
The box should be grounded, then no V rise.
It is a contrived scenario.