Transformers in Reverse

[electrofelon]

But the nameplate shown does have the primary/secondary labeling matching the current usage.

Of course, if the transformer appears to have originally been a step-down transformer, that suggests the current nameplate is not the OEM nameplate, which raises other questions.

Cheers, Wayne

Yes it is a rather odd transformer, and I'm curious what it's original use was for. I also question whether the data plate is the original or not. It seems rather oddly generic how it just states "transformer" and not the manufacturer. It does State the date of manufacturer however, and in my experience transformer remanufacturers hardly ever put the date of manufacturer on the new label for some reason.

 

[jim dungar]

But the nameplate shown does have the primary/secondary labeling matching the current usage.

Of course, if the transformer appears to have originally been a step-down transformer, that suggests the current nameplate is not the OEM nameplate, which raises other questions.

Cheers, Wayne

In which case it is okay, except for the NEC violation of rebounding the neutral at the transformer primary. Then you have your voltage issues of having the neutral connected.

The transformer may have been rewound at some point in time.

 

[W@ttson]

Remove it! Run!

That’s exactly how I feel. Should have never asked to go see the upstream distribution!

 

[W@ttson]

But the nameplate shown does have the primary/secondary labeling matching the current usage.

Of course, if the transformer appears to have originally been a step-down transformer, that suggests the current nameplate is not the OEM nameplate, which raises other questions.

Cheers, Wayne

Yes, that was my point. It actually calls the primary “208/120v LV”.

 

[W@ttson]

In which case it is okay, except for the NEC violation of rebounding the neutral at the transformer primary. Then you have your voltage issues of having the neutral connected.

The transformer may have been rewound at some point in time.

So this transformer was installed in 1990. I don’t know of it’s life before then but if taking the 1989 date as correct looks like it’s first life was at this installation.

This transformer has not been touched since 1990 except for the time it was partially submerged during a storm event.

The electricians at the time cleaned off the conductors and reterminated. This was around 2012. Whether they reterminated the way it used to be on the primary I am not sure but the secondary hasn’t changed at all for sure since the building I was originally investigating was always 480VAC without a neutral.

 

[W@ttson]

In which case it is okay, except for the NEC violation of rebounding the neutral at the transformer primary. Then you have your voltage issues of having the neutral connected.

The transformer may have been rewound at some point in time.

Ok so I want to get my ducks in a row here:

1. The NEC violation of multiple neutral - Gnd bonds is a violation of 250.24(5) correct?
2. The voltage anomalies that are sometimes false triggering the generator to kick on is likely the cause of the primary Neutral being landed on the XO. Is there any literature I can reference to read up on this phenomenon more?
3. I read somewhere else in this forum that such a case with the neutral connected can lead to large loads on the neutral conductor being fed back to the source. A conductor that typically does not have OCPD. Is that part of 2 above's issues?
4. I will eventually recommend replacement of the transformer, especially considering it beign partially submerged in the past, its time is due. In the mean time action items are:
   1. Disconnect the sources neutral from the XO terminal. terminate it on an insulated lug within the xfmr.
   2. Disconnect that XO bonding strap?
   3. Try to recommend some GFP monitoring until the new transformer arrives? What is the most cost effective solution for this one considering that a new transformer is probably on the order of 50 week lead time?

 

[jim dungar]

Back in the 90s, it was permissible to install 480V ungrounded deltas, like yours, without any Ground detection. The only requirement was qualified/trained maintenance personnel on site.

There are several Ground detection devices available, in days gone past it was fairly common to use two 240V incandescent light in series connected from each leg into a Wye configuration. One problem with this scheme was complacency, after years of staring at the lights people began to ignore them.

 

[W@ttson]

Back in the 90s, it was permissible to install 480V ungrounded deltas, like yours, without any Ground detection. The only requirement was qualified/trained maintenance personnel on site.

There are several Ground detection devices available, in days gone past it was fairly common to use two 240V incandescent light in series connected from each leg into a Wye configuration. One problem with this scheme was complacency, after years of staring at the lights people began to ignore them.

This is a place that is restricted, but I don't think anyone realized what they have. The people back in 1990 are no longer around that knew this was an ungrounded delta.

I have seen those lights only 1 time in my life.

Thank you for the info.

 

[GoldDigger]

Simply put, the connected Wye point will force the primary windings to remain at 120V which prevents the 480V from truly accommodating an unbalance loading, like you may experience when starting large motors.

I do not know of a single transformer manufacturer that does not warn against connecting the Wye when a transformer is run in reverse.

There is only a major problem when the load side winding is delta or there is a delta tertiary winding. And it results from input voltage imbalance, not necessarily related to output loading.
Utilities often use wye-delta transformers, in part because of ferro-resonance problems with wye-wye, but they have complete control over the balance of the input phase voltages on the wye side and can thus avoid the problem.
The delta winding, being a closed loop, forces the sum of the phase voltage phasors to be zero. Anything else will cause large circulating currents in the winding and corresponding unwanted currents on the line side. But this phasor sum condition is not automatically met in the windings on the wye side. If the X0 is left floating (not connected to either neutral or ground), the voltage at the wye point will be free to vary, allowing the input side constraint equation to be met. The current sum condition (no current through X0 when left floating) is automatically met because of the delta winding, regardless of phase load imbalance on that side.

 

[LarryFine]

1. Disconnect the sources neutral from the XO terminal. terminate it on an insulated lug within the xfmr.
2. Disconnect that XO bonding strap?

Both. The X-0 lug should have nothing connected to it. Leave it floating as if it didn't exist.

The neutral conductor serves no purpose, but can be left for possible future use.

 

[Tulsa Electrician]

I would also recommend you look at remote building grounding prior to making any change.

I read:
A 3-wire 480 volt secondary ungrounded Delta to remote building.
A back up generator at remote building 480 volt 4- wire wye with netural bonded at Generator.
Netural from Generator isolated at building ATS.

You were put in charge of adding a GEC at remote building.


There is an EGC from transformer to separate building ATS.

Questions.
Is there an existing grounding electrode at the separate building.
If so, where how connected.
Where is the feeder EGC terminated in the transformer.
Does the transformer have a GEC with a rod driven or other electrode.
Where is the EGC of the feeder connected in the ATS.
Does the 4- wire feeder primary have an EGC.

BE sure grounding requirements are satisfy prior to removing that X0 bond at transformer.

Draw up a grounding detail and post for review. That way your sure your plan is NEC compliant. Or consult an engineer.

Weird thing happen with a system when a wire is unhooked that has worked for years.

Proper grounding of the remote building is a must expecilly with an ungrounded secondary.

Last does the ATS have an OCPD since it is the point of entry for the remote building. Hopefully I put this together correctly in my head.

 

[electrofelon]

considering that a new transformer is probably on the order of 50 week lead time?
[/LIST]
[/LIST]

That is a very common transformer and is an off the shelf item available to ship Monday (assuming you can use a ventilated dry type and we aren't talking about replacing it with another oil filled).

 

[W@ttson]

Here is the connection diagram below the name plate. I am sketching up the existing grounding condition to answer one of the other posts. Will post shortly:

 

[W@ttson]

That is a very common transformer and is an off the shelf item available to ship Monday (assuming you can use a ventilated dry type and we aren't talking about replacing it with another oil filled).

I would recommend a more robust transformer. I like the Hammond HPS Titan series encapsulated transformers.

 

[W@ttson]

I would also recommend you look at remote building grounding prior to making any change.

I read:
A 3-wire 480 volt secondary ungrounded Delta to remote building.
A back up generator at remote building 480 volt 4- wire wye with netural bonded at Generator.
Netural from Generator isolated at building ATS.

You were put in charge of adding a GEC at remote building.


There is an EGC from transformer to separate building ATS.

Questions.
Is there an existing grounding electrode at the separate building.
If so, where how connected.
Where is the feeder EGC terminated in the transformer.
Does the transformer have a GEC with a rod driven or other electrode.
Where is the EGC of the feeder connected in the ATS.
Does the 4- wire feeder primary have an EGC.

BE sure grounding requirements are satisfy prior to removing that X0 bond at transformer.

Draw up a grounding detail and post for review. That way your sure your plan is NEC compliant. Or consult an engineer.

Weird thing happen with a system when a wire is unhooked that has worked for years.

Proper grounding of the remote building is a must especially with an ungrounded secondary.

Last does the ATS have an OCPD since it is the point of entry for the remote building. Hopefully I put this together correctly in my head.

Going to answer your questions in order:

1. There is no existing ground electrode at the separate building.
2. There should be an electrode at least for the generator since it is a SDS (see commentary in sketch)
3. The feeder from the service disconnect to the pad mount transformer has 3 phase conductors and the neutral. The EGC is the rigid conduit. An insulated throat grounding bushing on the feeder conduit has a ground connection to it and that is terminated to the case ground lug. The case ground lug is connected to a ground electrode and the XO. (see sketch)
4. Yes the transformer has a ground electrode and it is connected to the ground lug that is attached to the case and subsequently connected to the XO lug.
5. The feeder from the disconnect switch at the remote building to the ATS has an EGC in it and it terminates on a ground lug within the cabinet. The neutral from the generator goes to the ATS but it is landed on an Isolated neutral lug.
6. The 4 wire feeder to the pad mount does have an EGC but via the rigid.
7. Attached is the grounding detail of the whole system as-is. I also placed some notes on what I want to do for the remote building grounding.
8. Worked for years is tough one to say. They have been having a bunch of voltage anomalies along the way. Phase rotation issues. etc.
9. Yes agree. Proper grounding everywhere is important.
10. The point of entry is the disconnect switch. The ATS does not have an OCPD. However, the disconnect switch also does not have an OCPD.

 

[Tulsa Electrician]

Thanks I have a full picture now.

 

[synchro]

... I will eventually recommend replacement of the transformer, especially considering it being partially submerged in the past, its time is due. In the mean time action items are:

1. Disconnect the sources neutral from the XO terminal. terminate it on an insulated lug within the xfmr.
2. Disconnect that XO bonding strap? ...

For the reasons that GoldDigger, Jim and others have mentioned, the feeder neutral should be disconnected from Xo. But it could make matters worse if you don't also disconnect the Xo bonding strap. If the strap is left connected, then the possible high current through the neutral that we are concerned about might instead flow only through the bonding strap and feeder EGC.

The connection diagram on the transformer shows taps on the delta side to accommodate different source voltages. The taps allow a consistent and proper volts-per-turn on the windings to be attained over a range of input voltages. Clearly this shows that the transformer was designed to be energized from the delta side, even though the low voltage side is labeled as the primary.

 

[Tulsa Electrician]

I do have a question for board.
Would the generator in this case be considered an SDS.

 

[Tulsa Electrician]

Sorry one more question does the generator have an OCPD built in.

 

[W@ttson]

I do have a question for board.
Would the generator in this case be considered an SDS.

Yes, by definition, none of the conductors of that system are ever touching the other systems conductors EXCEPT for ground conductors. It has to be SDS. Phases conductors between systems usually never are possible to touch because even old school transfer switches either have you connected to the one source phases or the other. The neutral is the item that usually is the thing that causes the SDS or not SDS determination. Since one of the systems doesn't have a neutral, there is no chance for the generators neutral to touch the primary source in anyway. It has to be SDS.