Ungrounded-Wye/Grounded-Wye Transformers and Providing Single Phase (L-N) Loads

[laurenm]

I am working on a system that is installed with an ungrounded-wye (480V) to grounded-wye (13.8kV) step-up transformer (T1). After the voltage is stepped-up the system goes through a series of switches, and eventually gets stepped back down via a grounded-wye (13.8kV) to grounded-wye (480/277V) transformer (T2). The 480/277V feeds a series of single phase, line to neutral loads.

I understand that I do not have an effective ground on this system, due to the network sequence of the original step-up transformer (T1). Can T2 provide a 277V (L-N) load without an effective ground on the system? Or will issues arise due to T1 being ungrounded (e.g. floating voltage issues, ground fault problems, etc)?

I've tried to search for some information on this topic but haven't been successful in finding anything useful. Thanks.

 

[davidbeach]

If neither of those transformers includes a third, delta, winding you won't have a stable phase-neutral voltage on the low-side of T2. It's worth a check of the nameplates, manufacturers have been known to include a buried delta in wye-wye transformers to save the cost of additional core limbs.

If they're existing and there isn't a delta in there somewhere you need one more transformer. It could be a grounding bank on the low-side of T2 or it could be a 480V delta-480/277V grounded wye to carry the loads that need the neutral.


When one this sentence into the German to translate wanted, would one the fact exploit, that the word order and the punctuation already with the German conventions agree.

-- Douglas Hofstadter, Jan 1982

 

[laurenm]

That makes perfect sense, and quite similar to what I was thinking. Interesting about the buried delta. What do you look for on the nameplate that will show if you have one or not?

 

[waross]

As David pointed out, the voltages will not be stable as described.
(Actually missing a neutral connection. An ungrounded neutral connection back to the source will stabilize the voltages grounded or not.
However there are a couple of mitigating factors.
Saturation. The voltage will tend to rise on the lightly loaded, ungrounded phase.
When the voltage rises 10% or 20% saturation will tend to limit further rises.
If either of the transformers has a three legged core, that will create a phantom delta which will tend to stabilize the voltages.
That said, I would not be comfortable depending on either of these factors.
A four wire connection (neutral connected) wye:delta transformer connected at any point in the system.
The delta voltage is unimportant and is not used.
Sizing: The transformer shall be sized to support the maximum possible single phase load.
Faults: The wye delta bank will back feed into any upstream faults and to that end, the transformer withstand capability shall be coordinated with the protection clearing time.
With fault withstand in mind, I would suggest installing the transformer on the secondary of T2. It will be most effective there and any fault current will be mitigated by the impedance of T2.

How does it work?
If wye voltages and the phase angles are not equal, the delta will not close.
If the unequal wye voltages are reflected in the sides of the delta, there will be a gap or overlap at the last corner drawn.
That represents a voltage difference.
But that is a bolted connection, there cannot be a voltage difference.
Instead there is generated a circulating current.
That circulating current acts to oppose the force creating it.
The circulating current tends to load the higher voltage phases and to lessen the load on the lower voltage phases.
The circuit will transfer power between phases, so that the lighter loaded phases tend to support the higher loaded phases.


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

 

[davidbeach]

That winding will show up on the diagram if present, may not have terminals.

Got some good project work once by describing how a wye-wye with buried delta would have fouled up a simple source loss detection scheme not knowing that that the original engineer could never figure out why their design failed repeated tests.

When one this sentence into the German to translate wanted, would one the fact exploit, that the word order and the punctuation already with the German conventions agree.

-- Douglas Hofstadter, Jan 1982

 

[laurenm]

Thanks all for the responses. It certainly confirms what I was already thinking and gives me additional food for thought. I was hoping there was a way to fix this system without adding a grounding bank. But it certainly seems like this is the best way to go. Thanks again!

 

[Kiribanda]

Since your T2 is solidly grounded and single phase loads are served, any GF current will return back to the NEUTRAL of
T2. Therefore, your 480V system fed by T2 is eff. grounded irrespective of T1 grounding method.
If you perform a SC study you will find that COG of T2 system is less than 85%.

 

[davidbeach]

Nope, draw out the sequence diagram and you won’t find a path that allows zero sequence current on the low-side of T2.

When one this sentence into the German to translate wanted, would one the fact exploit, that the word order and the punctuation already with the German conventions agree.

-- Douglas Hofstadter, Jan 1982

 

[waross]

Since your T2 is solidly grounded and single phase loads are served, any GF current will return back to the NEUTRAL of
T2.

This assumes that the voltages supplied to T2 are equal and stable.
May I attempt to moderate here;
First, we must make a distinction between grounding and connecting the neutral to the source.
The distribution folk use the term Grounded to refer to both a connection to ground and to a connection of the neutral to the source.
This system is not grounded (Neutral connected to source) according to universal distribution practice.
This may be the result of a feed from a delta connected generator, with no neutral available.
A line to ground/neutral fault on an ungrounded wye system causes the voltages on the ungrounded phases to rise to line to line voltage to ground (Subject to saturation mitigation).
A line to neutral fault on T2 secondary will reflect as a line to ground/neutral fault on T2 primary, on T1 secondary and on Ti primary.
T1 secondary voltages will be very low on the faulted A phase and will rise on the unfaulted B and C phases.
A fault on A phase at T2 will result in very low EMF and very low fault current.
A fault on A phase at T2 will result in voltages approaching line to line voltage on B and C phases, mitigated somewhat by saturation.
As the voltage rises on B and C phases, saturation will cause increased current.
This will increase the current in B and C phases and mitigate against further voltage rise.
This increased current has the effect of B and C phases supporting a portion of the already low fault current.
(I am using the power flow convention for transformer primary and secondary.)

This is easy to visualize if you consider a single phase, 120:240 circuit with unbalanced loads and an open neutral
The single phase calculation is fairly straightforward, the three phase effect is more difficult to calculate.
So: Unbalanced line to neutral currents transferred to the ungrounded primary of T1 will result in unbalanced voltages in the secondary of T1. These unbalances will be transferred through to the secondary of T2.
A grounding transformer anywhere in the circuit will stabilize the voltages.
My first choice for a clean sheet install would be to install a grounding transformer on the secondary of T2, but not my only choice.
If there are other loads on the system, I would install a grounding transformer at T1.
If small distribution transformers are readily available I would consider a wye:delta bank of small distribution transformers on the 13.8 kV line. Subject to cost and availability.

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

 

[prc]

Waross, thank you for the explanation. I have a question: Why is the primary neutral of T1 kept ungrounded? If it is grounded, is there no longer a need for a stabilizing delta winding?

 

[waross]

Why is the primary neutral of T1 kept ungrounded?

I am going to repeat something here that you already know, prc.
You are not the only one reading this and I want to be clear for the sake of others who come here for learning.
Please be patient.
Thanks.
The primary neutral may easily be grounded but grounding is not the issue.
The lack of a neutral connection to the source is the issue.
Imagine a television set that draws 1.2 Amps at 120 Volts, connected in series with an electric kettle that draws 12 Amps at 120 Volts.
That series connection is now supplied with 240 Volts.
Do the math.
Simple math tells us that the voltage division will be about 10:1,
The kettle will see about 21.8 Volts and the TV will see about 218 Volts.
I have seen this several times.
The TV set dies.
A universal connection in North America is both 120 Volt loads and 240 Volt loads fed from center tapped 120/240 Volt transformers.
Now in our example, if the transformer center tap is grounded, it makes no difference to the voltages to the loads.
And in our example, if the series connection point between the two loads is grounded, the voltages across the loads become voltages to ground, but it makes no difference to the voltages to the loads.
What is needed is a connection from the source, that is the transformer center tap, to the point of connection between both loads.
If the load neutral is properly connected to the source neutral, the voltages at the loads will be stable, regardless of whether the neutral is grounded or ungrounded.
Why this lengthy discourse on a tangent subject?
We are assuming that the term "Ungrounded" is distribution jargon and implies that the neutral is not connected to the source.

Back on subject:
Why is the neutral not grounded (connected to the source neutral)?
It is very possible that the source has no neutral to connect to.
How is that possible?
The source may be a delta connected generator or transformer secondary.
Case in point, the power plant in the micro-grid that I used to oversee:
The number of generators online depended on the load.
One of our generators was delta connected.
When the loading was very light that may be the only generator online.
Fortunately we had two industrial customers with wye:delta banks that stabilized our voltages when no source neutral was available.
If no source neutral connection is possible, the text book solution is a special order zig-zag grounding transformer.
The down and dirty solution is three distribution transformers connected wye:delta.
The cost is often lower.
Initial equipment and possible future replacements are often available "Off the Shelf" and if not, lead times are short.



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