12kv transformer isolated ground 6

Safety violation, code violation, common sense violation; I could go on and on. You should stay far, far, away.

Nice catch garetoo.

Welcome to Eng-Tips.

Keith Cress
kcress -

Boy, if a brass monkey was standing with one foot near the ground rod and the other foot near the steel, a good ground fault may emasculate him. A star for picking that up.

Bill
--------------------
"Why not the best?"
Jimmy Carter

I am not looking for any commendation for noting this potential issue (thank you), I was hoping to get some input on the proper method to ground this, or does it even need one? It just gets stepped down to 480 at the end of its run.

An excerpt of what I asked/pointed out to the engineer of record, with no answer is (quoted below):

Part of what I am referencing is at NEC 250.184 Solidly Grounded Neutral Systems

(B) Single-Point Grounded Neutral System. Where a single-point grounded neutral system is used, the following shall apply:
(1) A single-point grounded neutral system shall be permitted to be supplied from (a) or (b):
a. A separately derived system
b. A multigrounded neutral system with an equipment grounding conductor connected to the multigrounded neutral conductor at the source of the single-point grounded neutral system
(2) A grounding electrode shall be provided for the system.
(3) A grounding electrode conductor shall connect the grounding electrode to the system neutral conductor.
(4) A bonding jumper shall connect the equipment grounding conductor to the grounding electrode conductor.
(5) An equipment grounding conductor shall be provided to each building, structure, and equipment enclosure.
(6) A neutral conductor shall only be required where phase-to-neutral loads are supplied.
(7) The neutral conductor, where provided, shall be insulated and isolated from earth except at one location.
(8) An equipment grounding conductor shall be run with the phase conductors and shall comply with (a), (b), and (c):
a. Shall not carry continuous load
b. May be bare or insulated
c. Shall have sufficient ampacity for fault current duty

By not bonding the equipment grounding conductor (4) and not running that with each group of conductors (5) & (8) the installation seems non-compliant. The lack of a bond actually appears to make the system a low-impedance grounded system, if there is such a thing? (Taken from NEC handbook). If there is going to be a deliberately introduced impedance between the equipment ground and the insulated neutral conductor it appears it must be a high impedance, I am not sure the 2’ earth separation is such, but I also do not think the earth unto itself is an actual bond as provided by proper materials and connections.

END QUOTE


gare

There are times when authority and hubris overwhelm knowledge and common sence.
Your problem may be more a people problem than a technical problem. Watch your liability and cover your assets.

Bill
--------------------
"Why not the best?"
Jimmy Carter

The simplest solution would be to bond the EGC to the local electrode and transformer neutral point.

You could use a low resistance ground if you need to reduce available ground current, although this is not common at 480V. Using the earth as the low resistance is not acceptable. A large voltage can develop across the resistor during a fault, so the resistor has to be located where it someone cannot get shocked. This would be difficult if the resistor were part of the earth. Designing for a particular fault current would also be difficult if the earth is the resistor.

I suspect the reason for the original design is not meant to reduce ground fault currents, but a misguided attempt to isolate the grounds for power quality reasons.

Thanks JG. The ground is on the 12kv side and I am merely installing - not designing. I do think the bond needs to be made from the EGC to the transformer neutral point. You are also correct in that the engineer has stated that they put it there for power quality. Without the bond you have suggested, the the distance between the two rods becomes an impedance path and is not safe. Therefore, I have refused to complete the install, a tough thing for an electrician versus approved drawings.

gare

You are a better person for it. Plus! You won't be defending yourself in a death through negligence trial.

Keith Cress
kcress -

 
Unless the transformer was being used for some kind of well-defined research/temporary load {one that comes to mind for me years ago was for large-area soil heating for environmental remediation—but it was quite the opposite—three graphite ground electrodes were used connected to their respective bushings} I can’t envision any reasonable justification for a floating system.

This might be of some help, and it is a little bit dated but this NIST document may give you some ammunition.

Busbar,
That is a great article! Especially page 30 diagram 4. I appreciate the comments with relative content toward a solution.

Thanks!
gary

gare

A less confrontational solution may be to just install an appropriately sized jumper from the ground rod to the adjacent steel or ground grid. Then the installation may be code compliant. I see this quite often with the "Magic Triangle" (Three ground rods in a 10 foot triangle for a dedicated ground.)
The magic triangle by itself is not code compliant due to the issues stated here. A large (often 4/0 AWG) jumper between the magic triangle and the power ground grid assures compliance and is common.
You will have made a safe compliant installation and if the jumper is then removed, responsibility will devolve elsewhere.
Size the jumper from the code as part of the equipment grounding grid.

Bill
--------------------
"Why not the best?"
Jimmy Carter

I whined enough they have asked us as the installer to present how we think it should be done, an electrician second guessing a PE..... not a good feeling. This is the first residence I have ever done with a 12kv primary. I will look up the "magic triangle". I will try and incorporate what seems valid from everyone's input and let you know what happens.
Is bonding a leg of this delta 12kv even necessary? Should an equipment grounding conductor be run with the feeders to the step-down transformer 1,700 feet away?

gare

Talk to the utility in order to understand their supply system. They may not want a center-tap 12kV ground on their system and you may not need it.

I just noticed that the ground is on the 12 kV winding of the transformer. I've never heard of center tapping a 12 kV delta winding. That gives you a 10.4 kV high leg.

The drawing notes say that there is no ground run with the three 12kV conductors between the MV switchgear and the utility yard padmounted transformer. The cable 12 kV will be shielded. If the shield is grounded at both ends, then the shield acts as a ground wire. If there is a ground fault on the 12 kV cable at the utility yard padmounted transformer, fault current will flow through the cable shield and it has to be sized for the current.

If the shield in not grounded at both ends, then you need to run a ground wire. Otherwise, a ground fault at the utility yard end will have to flow through earth and the fault current may not be enough to blow the 65E fuse at the MV switchgear. High step- and touch-potentials will also occur during a ground fault.

Without a metallic return path, a ground fault at the utility yard on the high leg of the 12 kV system could result in touch-potentials at the padmounted transformer as high as 10 kV. This high voltage could also be transferred through the secondary EGC to the utility building.

jghirst: You are absolutely correct. Even if the shield is landed at both ends, could it handle a fault? It is landed in every vault along the way, to a ground rod, but there is no detail for the ends.

apowrengr: This is not owned by the utility, I wish it was. The utility is taking their 12kv, running it about 75 ft to a 12kv/480 trans, then about 30 feet into the main switchgear and metering it. After that, the client is bumping it back up to get it to the next location where it is knocked down again.

Interestingly, there is a detail at the other end showing a bond between equipment without explaining how it is accomplished.

gare

I can’t cite specific code references, but I think you’ll find that for an ungrounded 12kV system {id est, 12kV-delta windings back-to-back} the insulation level of the interconnecting cables should be 173% at 15kV, or possibly 100% at 25kV.

If the intent of this work is to provide a ground on the separately derived 12kV system it should be done at the source as is proposed and then a suitable grounding conductor needs to be run with the circuit.

QUOTE:
apowerengr (Electrical)
15 Mar 09 21:08
If the intent of this work is to provide a ground on the separately derived 12kV system it should be done at the source as is proposed and then a suitable grounding conductor needs to be run with the circuit.
END QUOTE

Bond H1 or H2 or H3 to ground and run an EGC with the feeders?

gare

I would not corner ground or ground the midpoint of one winding of a 12 kV transformer. I can't find offhand where the NEC prohibits this, but IEEE Std 142-1991 recommends against midpoint grounding systems over 240V. Ø-ground voltage will be 12 kV on two phases if corner grounded, 10.4 kV on one phase if mid-point grounded. Insulation, connectors, and arresters must be rated appropriately.

If you want a grounded system, I recommend installing a zig-zag or grdY-delta grounding tranformer.