Separately Derived Systems

[NickParker]

Separately Derived Systems:
This is defined as “an electrical source having no direct connection to circuit conductors of any other electrical source other than those established by grounding and bonding connections”.

Does this mean a Wye-Wye transformer can be considered as a separately derived source?

Can somebody provide examples of separately derived sources?

 

[waross]

Yes. A conventional transformer is a separately derived source.
An auto-transformer is not a separately derived source.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jghrist]

No, a grd wye - grd wye transformer is not a separately derive system. The neutral conductors of the primary and secondary have a direct connection. The connection of the neutrals is not dependent on incidental ground connections. There is direct zero-sequence connection between systems.

 

[jghrist]

The main concern is whether or not there is a source of ground current. With a wye-wye transformer, there can be no ground current on the secondary unless there is a source on the primary system. A delta-wye transformer is a ground source, so there can be ground current on the secondary even if there is no source on the primary.

 

[LionelHutz]

By the quoted definition, yes a Y-Y transformer is because the only common connection is the neutral bonding which is allowed as per the statement wording.

 

[jghrist]

Pad-mounted wye-wye transformer per IEEE C57.12.22 have internal connections between the high voltage and low voltage neutrals, independent of grounding and bonding connections, therefore do not meet the definition of separately derived systems.

 

[stevenal]

Seems to me the key words are "circuit conductors". If the neutral is carried through to feed unbalanced loads, it is a circuit conductor. If not, it is only a ground bond.

 

[davidbeach]

jghrist and stevenal are correct.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[LionelHutz]

So, to summarize. If the Y-Y transformer comes with an internal connection or if a neutral conductor from the primary is connected to the secondary then the answer is no. The answer would be yes if the transformer secondary neutral point has no internal connection and gets a grounding or bonding wire (only) connected to an external terminal when the transformer is installed.

The sources of current or types of current don't matter since that is not part of the quoted definition.

 

[stevenal]

Consider a grdwye-grdwye transformer feeding three phase loads with three wires. I'm suggesting this could be considered separately derived. Am I wrong?

 

[waross]

I am in complete agreement with you stevenal.
The secondary of a wye:wye transformer is a separately derived system.
This is common in industrial plants where 480 Volt to 120/208 volt lighting transformers are common.
The secondary of each lighting transformer is a separate system.
The codes must be followed in regards to system grounding.
A connecting link between the primary neutral and the secondary neutral is one method of grounding the secondary system, but this method may not be acceptable in all instances.
Utility owned transformers with a connection link between the primary and secondary neutrals are at the crossroads of two codes.
As such they are not an example that may be taken as always valid in regards to NEC questions.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[bacon4life]

I had assumed the main concern with separately derived systems was avoiding flow on the grounding conductor during steady state conditions, rather than controlling the flow of ground current during a fault.

The setup Lionel described appears to be possible using dry-type transformers since IEEE C57.12.01 section 5.10.2 allows for dry-type transformers to have separate neutral connections. Without a bonding jumper between the secondary transformer neutral grounded conductor and the system grounding conductor, there would not be an effective fault current path.

stevenal- The OP was about sources, so I do not see how the connection to the load (3 wires vs 4 wires) would change the properties of the source transformer.

 

[waross]

The OP was about sources, so I do not see how the connection to the load (3 wires vs 4 wires) would change the properties of the source transformer.

There has been a major code change in the Canadian code. I don't have access to a copy of the code just now, but one change that I ran across recently was for a sub panel in a separate building.
It is no longer allowed to ground the neutral anywhere in a system regardless of the number of buildings served.
In the last code that I have access too:
In Canada for a service entrance, the grounded circuit conductor must be grounded at the transformer and at the main panel.
In Canada for a local transformer, the grounded circuit conductor must be grounded at the transformer or at the first downstream switch or panel.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

The word derived does not appear in the Canadian Code in this regard.
But; examples of system sources that are NOT separate:
Sub panels fed directly with no transformer.
Systems fed by an auto-transformer.
Any system fed by a conventional transformer.
Systems fed from a standby generator.
Systems fed from an alternate source such as batteries, solar, wind power, and the associated inverters.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[LionelHutz]

I had assumed the main concern with separately derived systems was avoiding flow on the grounding conductor during steady state conditions, rather than controlling the flow of ground current during a fault.

The quoted definition says nothing about currents, it just says by conductor. Generally, the bonding conductor is never supposed to carry operating current, but it carrying current or not isn't part of the definition.

 

[jghrist]

What is the purpose of treating a separately derived system differently from one that is not? It is to determine whether or not to bond the grounded circuit conductor (neutral) to ground at the system source. For a separately derived system, you do bond the neutral. This is the only ground connection to the neutral and there are no parallel paths for neutral current that could affect ground protection.

If the system is not separately derived, then you don't ground the neutral at the source because it would provide parallel paths for neutral current and would affect ground protection. You don't ground the neutral at two points.

A wye-wye transformer provides a path for neutral current to the primary system source. If the neutral is grounded, there will be parallel paths to the primary source and ground protection would be affected. This is why it is not considered a separately derived system and the neutral is not bonded to ground.

 

[waross]

You may be a little off base here.
The utility owned installation including the primary distribution circuit, the transformer(s), and the conductors to the customer's service do not fall under the NEC or CEC that most of us work with.
The utilities often use multiple earthing of their neutral conductors including bonding of both their H0 and X0 connections.
In normal operation no neutral current flows through the H0-X0 link.
In the event of a ground fault, secondary current flows to ground.
That fault current returns through the ground electrodes.
A ground faulted secondary conductor will pass current to ground.
The ground fault current will return to the neutral trough the system ground electrodes in parallel and if there is an H0-X0 connection, through the primary ground electrode and through the H0-X0 link.

A standby generator will often have a common neutral connection with the system it is supplying.
Despite the common neutral connection the generator is a separately derived system.

Pad mount wye:wye transformers with a factory installed H0-X0 link are most likely intended for the utility market.
If such a transformer is installed in a plant with a primary voltage service and is owned by the plant the local AHJ may or may not require or allow the link to be removed.
In some instances removing the H0-X0 link will make ground fault detection simpler.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

Think of a separately derived system as a system that will not be inherently grounded without an intentional ground connection.
A system fed from an auto-transformer is grounded by the system that it is supplied from.
A secondary of a conventional transformer is not inherently grounded and is a separate system
The H0-X0 link that seems to be causing confusion is a feature of the individual transformer, not a feature of the wye-wye system.
By the way, the Canadian code is basically the same except that it doesn't use the word "derived".
The Canadian Code refers to Separate Systems. It may make it easier to think of separate systems rather than system derivation.


Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jghrist]

The concern is with the NEC, not with utility systems that don't have to comply with NEC. In the case of a generator with a common neutral connection, it is definitely not a separately derived system. See the attached pages from the NEC 2017 Handbook. It is separately derived only if the neutral is switched with a 4-pole transfer switch.

 

[waross]

I apologize.
You are correct under the example that you show.
I don't have any objection to the grounding methods.
I have planned and installed a lot of standby generators with a common neutral and single ground point as shown in your diagram.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!