Single house isolation transformer and GFCI protection?

[_SAK_]

I was reading this thread - which I thought was phenomenal in its exploration of some big questions about power distribution and safety...

https://forums.mikeholt.com/showthread.php?t=162981

Unfortunately the thread is closed, but it got me discussing this with some colleagues:

If a single-family home were to use an isolation transformer - no neutral-ground bond on the load/secondary side - would capacitive coupling be an issue? Could GFCI breakers be used and provide good protection in the highly-unlikely chance one of the conductors became grounded and a second fault (or person shock path) occurred?

 

[drcampbell]

I think you might be underestimating capacitive coupling.

In factory wiring, parasitic capacitive coupling can be sufficient to prevent a plain old ordinary motor starter from turning off if there's a three-wire, two-pushbutton control station located remotely. I don't remember the details but it wasn't an absurd, extreme case -- a few hundred feet of control wire could have enough capacitance to pass a few dozen mA through the contactor solenoid and hold it closed.

 

[electrofelon]

What exactly do you mean by "an issue"? Ungrounded systems are fairly common in industry. Everything is still bonded together, and probably unintentionally earthed anyway thru inevitable attachment to masonry, etc.... Grounded and ungrounded systems have certain design characteristics for OCPD ratings and where they are placed in the circuit so making a system intended to be grounded ungrounded may apply voltages to equipment they were not designed for (think slash rated breakers).

 

[_SAK_]

So the thinking that we (another electrician and myself) are having, is use a shielded isolation transformer for this residence that has a bit of space from neighbors (hundreds of feet), and has its own transformer out on a pole. The client is interested in cleaning up the signal as well as not having the poco's primary neutral running all the way up to their residence due to emf/health/ground current concerns. So the theory is to reduce high frequency (1.5-150kHz) current sometimes called "stray voltage" around the property.

The house would derive its own ground, but we were thinking of using a shielded isolation transformer 80 feet away from the house. Then we were thinking of NOT having (and so far it sounds like inspector is OK with it) a bond established between the secondary center tap neutral of the iso transformer and the house's ground. Referring back to the thread linked in the OP, as is sometimes done some places, the system could be megged and ensured there are no faults to ground and therefore no current paths.

Frankly I'm surprised the inspector said he's OK with this, but this is something new to me so it's interesting and fascinating, but I'm skeptical too. A fault of one of the lines/legs A/B to the equipment grounding conductor...wouldn't be cleared by breakers. But then again a single fault shouldn't pose an electrocution risk. There would have to be more than one fault to ground (one being through a person) between 2 different conductors A/B/neutral in order for an electrocution risk to be in place.

I don't know about all the capacitive coupling stuff or leaking. I know it doesn't take much current (some sources say even less than 100 microamps) to stop a human heart. So I was posting here to see what everything thinks about code and safety.

 

[ptonsparky]

I believe remote or end of line.dairies.could install Ronk Blockers to isolate the farm from.the primary neutral.

 

[texie]

This whole discussion is moot as the NEC requires this service to be bonded and grounded.

 

[kwired]

The client is interested in cleaning up the signal as well as not having the poco's primary neutral running all the way up to their residence due to emf/health/ground current concerns. So the theory is to reduce high frequency (1.5-150kHz) current sometimes called "stray voltage" around the property.

If you eliminate the primary neutral, are you running two ungrounded primary conductors? For example 12.47kV primary instead of 7.2 kV? If so you really want to have that setup with no low impedance fault return path should that transformer primary fault out? Might just raise everything in the vicinity several kV when that happens.

Or if you still use a 7.2kV primary but don't run a neutral conductor then that means you are using earth return for one side of the primary winding - that can also make many voltage gradients in the vicinity, and even worse if you don't use conventional bonding/grounding methods so that at least all those objects are all at same potential.

Where is all this high frequency you are concerned about coming from?

Harmonic distortion that may already be on the lines won't go away by not grounding/bonding things.


If you ask me you are going to create more problems than you solve if you try something like this, especially in a dwelling where nobody is going to know how these "non customary" methods work or how to maintain it, and when the HO or handyman that doesn't know any better adds things they won't understand these special circumstances and will unintentionally contribute to making things worse, and nobody will know any better, until someone is injured or killed and someone that does know their stuff investigates and points out what went wrong and why the present conditions masked a problem.

 

[_SAK_]

You raise some really good points and concerns. So just to clarify we were never talking about using earth as primary neutral - that would cause the exact problem (ground current) we are trying to reduce on this property. We meant eliminating the bond between primary neutral and secondary neutral. Sometimes this is done by power companies where they will drive a separate ground rod quite a distance (50 or more) feet away. This helps prevent neutral current from neighbors (both primary and secondary) from traveling in/out on ground rods in the building(s). It isn't a complete fix since ground rods are still a path and current takes ANY/ALL/EVERY path at all times.

I suppose another option that would be to code would be to have the ground rods out by the pole where the transformer, meter, and main service disconnect are. Then somewhere in between we could install a choke (inductive) type low-pass filter on the secondary neutral running to the house and perhaps one of the legs too to stop high frequencies from external sources from traveling to the house. That would eliminate all those concerns about transformer failures and non-customary or non-code-compliant (even though the inspect has OKed this) practices. I do see your point about potentials for danger if there is a failure.

I've heard of those "Ronk Blockers" before - but it has been some time. I don't remember their design and limitations. I have to talk to Dave Stetzer who is an expert in power quality and ground current aka "stray voltage" (what an erroneous term). Ground or earth current is the term that should be used. It's a real thing and affects livestock and people as well - particularly when it is frequencies in the kilohertz ranges. Energy is proportionate to frequency (Planck's Law). Tighter restrictions and standards need to be put into place (https://www.emfscientist.org/)- but the opposite seems to be occurring as more and more electronic, non-linear loads are being placed onto the grid. THD (total harmonic distortion) limitations were just raised, for example. Current is going everywhere including through animals and humans. It's a mess and people are wanting their own environments as clean and free of this stuff as possible.