480/208V transformer question. Neutral/ground bond

[fastline]

I must be having mental fade. I am not at the equipment currently but I wanted to confirm something. I was restarting some equipment yesterday so I was checking voltages and such. Equipment runs off a 480 to 208 transformer. All loads are 3P, no single phase loads. However, for whatever reason I checked all legs to ground. Two were 120, one was 102. All phase to phase was find, and 480 checked out fine.

It dawned on me that even if that isolation transformer has a neutral tap for L-N, we did not carry it out because it wasn't needed. I cannot remember if isolation transformers normally have a bonded N/G? This might explain 'some' things, but I am focused right now on ground inspections. I know we installed a ground rod for the transformer as well as bonded to the steel in the metal building through unistrut and all switch gear is grounded through that.

I guess more than anything, I am vetting grounds right now but also trying to figure out why I would have L-G voltage without a neutral bond?

 

[LarryFine]

The secondary neutral should be bonded, earthed, and brought to the first OCPD enclosure just like a utility neutral.

You're likely measuring voltages because you're using a high-impedance voltmeter, so you're providing no real load.

A low-impedance meter, a solenoid tester, or an incandescent bulb in parallel with your meter will read almost zilch.

 

[don_resqcapt19]

The secondary neutral should be bonded, earthed, and brought to the first OCPD enclosure just like a utility neutral.

You're likely measuring voltages because you're using a high-impedance voltmeter, so you're providing no real load.

A low-impedance meter, a solenoid tester, or an incandescent bulb in parallel with your meter will read almost zilch.

While there is a requirement for XO to be bonded, there is no requirement that the grounded conductor be run to the first OCPD, like there is for a service. However, no matter if there is a grounded conductor or not, a supply side bonding jumper is required between the transformer and the first OCPD.

 

[LarryFine]

While there is a requirement for XO to be bonded, there is no requirement that the grounded conductor be run to the first OCPD, like there is for a service. However, no matter if there is a grounded conductor or not, a supply side bonding jumper is required between the transformer and the first OCPD.

Would that jumper not still effectively tie the (in this case) secondary neutral to the OCPD enclosure?

 

[augie47]

IMO, although a little uncommon, you could have the SSBJ without having the SBJ to XO. You would be operating it as an ungrounded system and would need fault indicators.

 

[fastline]

I am going to have to review the grounding of this system and code. I will reinspect this installation. There is an outdoor ped that houses the meter can and separate service panel. Conductors were run from this panel to a building. At the moment, I do not recall a ground being carried from that, and I hope I am wrong!

Those conductors were run to an OCPD inside the building, then back outside to the Xfmr. There is a ground rod driven outdoors with solid copper connecting back inside to the building and Unistrut.

I guess in some way, my sniffing around may have turned up a grounding concern. Not cool with 480 and I hope I just missed something during inspection.

 

[fastline]

OK, I got a few things inspected today and it certainly is not 'right'. Basically 480+G was run into building. 3 conductors landed in the OCPD, and ground is an oversized Aluminum, and landed on the Unistrut that mounts the OCPD. I suspect because it would not fit in the lug. Ground is run from that exact same location to the operating equipment. As well, a ground was run on the secondary of the transformer back to the unistrut as well, so transformer 'technically' is grounded, but not proper.

It appears the lug in the 480 switcher needs upsized to accept the larger ground conductor. Easy fix. I did notice in the 208 switcher that connects to the load that it has no lug for ground in it, thus why ground was just stabbed on the unistrut that the switcher mounts to. Wondering how I might handle this? It's looking like I need to drill a hole in the 208 box for that ground.

As far as the transformer bonding, I will have to get the cover off but a safe bet it isn't bonded due to the odd voltages. Further, the tranformer data tag is sun bleached off so now I have no idea what I am looking at. I need to get inside it.

 

[fastline]

Update, just found the transformer data. Acme 480 prim, 208/120 sec. Apparently not bonded. Issue?

 

[augie47]

By far most of the installs I see would have a SBJ at the transformer bonding XO to grounding and a SSBJ to the 208 disconnecting means.
A few, where the neutral was not needed, would not have the SBJ and be operated as an ungrounded system bringing 240.85 and 250.21(B) into play.
With your having a 208/120 secondary the probability of a neutral being utilized is high and I would see a grounded conductor + a SSBJ to the 208 disconnect/panel

 

[paulengr]

While there is a requirement for XO to be bonded, there is no requirement that the grounded conductor be run to the first OCPD, like there is for a service. However, no matter if there is a grounded conductor or not, a supply side bonding jumper is required between the transformer and the first OCPD.

1. Last I knew OCPD is a three pole device. If you are saying that the bonding should be carried through as long as it’s not a separately derived system, I’d agree. But OP said “isolation transformer” which generally means a delta-wye with a separately derived system. This isolates ground faults on both sides of the transformer since zero sequence is open circuit. The last thing you want to do is tie the bonding together. The only issue remaining is what to do about case grounds. Practice is to tie them (and surge arresters) into the primary side system. This is always 4 or 6 wire but could be 3P+N or 3P+G or 3P+N+G depending on how you arrange things. It is common to see ground rods next to or under a transformer only because often installs involve some excavation anyway.
2. Only reason you carry bonding through is if it’s not a separately derived system.
3. It is more common than you think for the X0 jumper to be missing. Once a transformer is installed nobody ever looks for it again until something happens that makes you question grounding.

 

[fastline]

Hmm, there seems to be some controversy regarding grounding methods. I might need to review here. On this application, I want to be clear that even though secondary is 208/120, the neutral was not carried and landed on it's OCDP. It is not needed as the 75kva transformer is a dedicated supply for a machine that only requires 3P.

I know the transformer has a ground "path" but is not directly tied to the ground rod next to it. Easy fix.

I tend to look at this like utility power. Primary has ground reference, then secondary is technically 'isolated' but ground is reestablished on the secondary as a 'separately derived system'.

It would seem the right approach (in my head) would be to bond or ensure jumper is there between N-G, and carry that to both the ground rod and first OCPD. However, if the Xfmr did not have a N, it still needs grounded in any case.

I try to look at this from the safety side and what if a primary wire contacts the Xfmr case? It would seem the right way is land BOTH primary and secondary grounds on the case, ensuring any potential fault has a nice ground path.

 

[electrofelon]

bonding should be carried through as long as it’s not a separately derived system

The last thing you want to do is tie the bonding together. The only issue remaining is what to do about case grounds. Practice is to tie them (and surge arresters) into the primary side system.

Only reason you carry bonding through is if it’s not a SDS.

I'm confused here. Perhaps you can be more specific about what you mean by "carried through" or "bonding together". The EGC run with the primary will be bonded to the transformer case along with the XO and ultimately to the secondary EGC 's. There is no way around it.

 

[Frank DuVal]

that even if that isolation transformer

Is it an isolation transformer or just a 480 to 208 transformer. Of course the same transformer can be used for both cases, this is asking if the install is supposed to be an isolation transformer system.

 

[don_resqcapt19]

1. Last I knew OCPD is a three pole device. If you are saying that the bonding should be carried through as long as it’s not a separately derived system, I’d agree. But OP said “isolation transformer” which generally means a delta-wye with a separately derived system. This isolates ground faults on both sides of the transformer since zero sequence is open circuit. The last thing you want to do is tie the bonding together. The only issue remaining is what to do about case grounds. Practice is to tie them (and surge arresters) into the primary side system. This is always 4 or 6 wire but could be 3P+N or 3P+G or 3P+N+G depending on how you arrange things. It is common to see ground rods next to or under a transformer only because often installs involve some excavation anyway.
2. Only reason you carry bonding through is if it’s not a separately derived system.
3. It is more common than you think for the X0 jumper to be missing. Once a transformer is installed nobody ever looks for it again until something happens that makes you question grounding.

The is no code legal method to install a transformer, other than those specifically permitted in Article 517, that does not have a bond between the primary and secondary grounding system. Both the primary and secondary EGC are required to be connected to the metal transformer frame and enclosure.

 

[jim dungar]

The is no code legal method to install a transformer, other than those specifically permitted in Article 517, that does not have a bond between the primary and secondary grounding system. Both the primary and secondary EGC are required to be connected to the metal transformer frame and enclosure.

Yep.
For the most part general purpose complete isolation transformer installations have been prohibited by the NEC for some 40 years now.

 

[LarryFine]

I know the transformer has a ground "path" but is not directly tied to the ground rod next to it. Easy fix.

It would seem the right approach (in my head) would be to bond or ensure jumper is there between N-G, and carry that to both the ground rod and first OCPD.

Make sure you're using the existing building GEC system, and not only an isolated individual rod.

 

[wwhitney]

The is no code legal method to install a transformer, other than those specifically permitted in Article 517, that does not have a bond between the primary and secondary grounding system.

So given that, and given that the primary grounding system is connected to the GES, what function is served by the requirement to connect the secondary grounding system to the GES an additional time?

Cheers, Wayne

 

[don_resqcapt19]

So given that, and given that the primary grounding system is connected to the GES, what function is served by the requirement to connect the secondary grounding system to the GES an additional time?

Cheers, Wayne

Its function is so that you get a green tag and not a red one from the inspection authority. It serves no technical function.

A number of cycles ago, I submitted a proposal that would permit the primary EGC to serve as the grounding electrode for a separately derived system that is in the same building and where the feeder originates at the service equipment, but it was rejected.

 

[electrofelon]

A number of cycles ago, I submitted a proposal that would permit the primary EGC to serve as the grounding electrode for a separately derived system that is in the same building and where the feeder originates at the service equipment, but it was rejected.

Let me guess, there was no technical substantiation for the change.....even though there is no technical substantiation for the additional conductor in the first place.

 

[don_resqcapt19]

Let me guess, there was no technical substantiation for the change.....even though there is no technical substantiation for the additional conductor in the first place.

Report on Proposals – May 2004
5-99 Log #2716 NEC-P05
(250-30(A)(4) Exception No. 2 (New) )
Final Action: Reject
Submitter: Donald A. Ganiere Ottawa, IL
Recommendation:
Renumber existing Exception as Exception No. 1 and add new Exception No. 2 as follows:
"Exception No. 2: Where a transformer is used as a separately derived system and where the transformer primary feeder circuit
originates in the same building or structure, a grounding electrode shall not be required."
Substantiation:
There is no electrical or safety related reason to require a grounding electrode for a transformer used as a separately derived system when
the primary power source for the transformer is located in the same building or structure. The requirements of 250.4(A)(1) are met
without the use of a grounding electrode at the secondary side of the transformer. Lighting is not a problem within the building or
structure. The only possible contact with a higher voltage system is a fault on the primary feeder and this fault will be cleared by the
primary feeder equipment grounding conductor. The bonding required by 250.30(A)(1) will stabilize the voltage to earth under normal
operating conditions. The same bonding in combination with the primary feeder equipment grounding conductor will limit the effect of
line surges. The addition of a direct connection from the secondary side of a separately derived system to a grounding electrode
provides no additional protection over that which is already provided by the primary feeder equipment grounding conductor.
Panel Meeting Action: Reject
Panel Statement:
The present requirements for separately derived systems to be grounded to a reference earth by a dedicated grounding electrode
conductor apply to all systems without regard of installation inside or outside the building or structure served. The primary reason is to
establish an earth ground reference and stabilize the system voltage around this reference. The equipment grounding conductor from the
source to the separately derived system does not meet the requirements for size, not having a choke effect when installed in metal
raceways, multiple terminations, etc. The equipment grounding conductor's primary purpose is to provide a low impedance path for
fault current in the event of a ground-fault on the system up to and including the primary of the transformer, not to act as the low
impedance earth reference conductor.
Number Eligible to Vote: 16
Ballot Results: Affirmative: 15 Negative: 1
Explanation of Negative:
RAPPAPORT: The purpose of the grounding electrode conductor in the proposed Exception is to stabilize voltage. See my Comment
on Affirmative on Proposal 5-91. An equipment grounding conductor, run with the feeder to the separately derived system and sized for
the separately derived system, should be adequate.