• We will be performing upgrades on the forums and server over the weekend. The forums may be unavailable multiple times for up to an hour each. Thank you for your patience and understanding as we work to make the forums even better.

Transformer sizing for deriving a neutral

304sparky

Member
Location
West Virginia
Occupation
Electrician
So, do you want to consider the harmonics of those chargers? Do you have the specs of the chargers? Is the manufacturer of the charger claiming it emits xx% y harmonics level?
Since you have measured a 6 A neutral current on the charger, 6 x 8 = 48A. If that's all you have, then size a transformer that can handle ~48A! For good measure, give it a 150% fudge factor should you happen to buy cheaper and more harmonics-laden ones, say you design for a total neutral current of 75A for the 8 units. That would be like a smaller grounding transformer of 75/3 x 480 x 1.732/1,000 = 20 kVA ~ 30 kVA, 3-phase, 480Y.
So, even though my OCPD on my distribution panel is 200 amps, in this case I can size the transformer off the neutral load because all I’m doing is deriving a neutral?
 

topgone

Senior Member
So, even though my OCPD on my distribution panel is 200 amps, in this case I can size the transformer off the neutral load because all I’m doing is deriving a neutral?
You have options. If you want to consider the full neutral current (equal to the phase currents =200A), you'll need to have at least a 75 kVA grounding transformer; if you only provide for the neutral current requirement of your chargers, then you can choose a transformer that will provide a lesser neutral current of 6 x 8 = 48A, that could be a 15 kVA and for a 75A neutral, a 30 kVA will be fine.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
Picture this: you are using a wye-grounded primary, delta secondary transformer with the delta secondary having no loading at all. The neutral is derived from the grounded wye-point of the primary. The possible full neutral current of 200A on your load side passing the neutral conductor and entering at the neutral point of the transformer will be shared equally by the three phases of the transformer
Thank you for this comment, it took some thinking to believe. It is interesting to contrast the two different ways to use a 480D : 480Y/277V transformer to supply a 277V load from a 480D 3-wire source.

One way is to use it as an isolation transformer. Then if you have a 1A load on the secondary L-N, it draws 1/sqrt(3) A on two of the primary conductors. You need to size your transformer for all the VA of your L-N loads, or rather 3 times the VA of the most loaded phase.

Another way to use it is as you described. The key here is that with the delta side of the transformer unconnected, any current in one of its coils must equal the current in each of the other coils; the current can only go in a loop. That in turn means that the current in any coil on the wye side must equal the current in each of the other coils on the wye side. I.e. as you say, any neutral current that comes into the center point divides evenly among the 3 coils.

In this configuration, if you have 1 A connected L-N, that 1A from the neutral divides into 1/3A on each of the 3 coils, and then on the supply side of the transformer, you end up with currents of 2/3A, 1/3A, and 1/3A on the three line conductors.

As you say, in this configuration you only have to size the transformer for the worst case unbalanced current on the neutral. I.e. if you have (8) identical 1A L-N loads, and you distribute them as balanced as possible, the worst case neutral current is 3A, or 1A on each coil. So you'd only need a 277V * 1A * 3 = 831 VA transformer. Versus the isolation transformer sizing for the same load arrangement of 277V * 3A * 3, or 3 times larger.

Moreover, if you had an arbitrary number of 1A L-N loads, and you are able to arrange that as they turn on and off, the active loads are always as balanced as possible, now the worst case unbalance current is 1A, and you only need a 277 VA transformer. Whereas in the isolation transformer arrangement, the required size of the transformer scales linearly with the number of loads supplied.

Cheers, Wayne
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
So, even though my OCPD on my distribution panel is 200 amps, in this case I can size the transformer off the neutral load because all I’m doing is deriving a neutral?
NO NO NO!

My apologies for shouting. Welcome to the forum. We like providing good information but we _love_ our tangents.

For your application you are not just deriving a neutral. You are deriving an entire system which has a neutral.

All of the power to the chargers is passing through the transformer, not just the neutral current. Calculate the kVA load of all the chargers combined, and size your transformer to supply that load.
A 200A circuit seems a bit small for supplying a transformer for 8 30A loads. But there may be diversity factors involved. Ideally you would first calculate the load kVA, then select your transformer to supply that, then pick OCPD to protect the transformer, then pick conductor sizes to match load and OCPD. Simply saying '200A' will likely leave something undersized.

The idea of simply deriving the neutral is interesting, and has its place, but not for supplying your chargers. Below I'm continuing that tangent discussion.

If the OP has an ungrounded distribution system, then they could use a grounding autotransformer to derive a neutral. But if you have a system with a neutral, you are required (by code) to ground that neutral. This would have the effect of grounding the entire 480V system, not just the 200A panel being fed. Presumably if the plant has an ungrounded delta system, they want to keep it that way.

The other possibility is that the OP has some flavor of grounded 480V distribution. If they use an autotransformer to derive a neutral in this case, they end up with a neutral that isn't grounded. If they do ground this derived neutral then it can cause large circulating currents trying to equilibrate with the system neutral. (Or cause a short circuit leve current if the source is corner grounded.)

That 'ungrounded neutral' is probably not code compliant, and would need yo be treated as an ungrounded conductor if code permitted it.

Yes, a Wye : Delta transformer can function to derive a neutral used as an autotransformer. The OP doesn't need this. They need a Delta:Wye transformer sized for their full load.

Jon
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
That 'ungrounded neutral' is probably not code compliant, and would need yo be treated as an ungrounded conductor if code permitted it.
Thanks for keeping us on track. The NEC limits the use of autotransformers, e.g.

215.11 Circuits Derived from Autotransformers. Feeders shall not be derived from autotransformers unless the system supplied has a grounded conductor that is electrically connected to a grounded conductor of the system supplying the autotransformer.

That would rule out this 'ungrounded neutral' option. 215.11 has an exception for 208V <--> 240V autotransformers.

Cheers, Wayne
 
Top