480V 3 phase to 120/240V single phase Xfmr

[jim dungar]

If you really want both 3 phase in _and_ 120/240V, I can describe the transformer connection. But do you really want to use a non-standard transformer from some random dude on the internet.

-Jon

A transformer with 3 phase in and single phase 120/240V out will not present as a balanced three phase load on the primary so it offers very little advantage. I last saw one called a Hoppy-Hookup from a manufacturer from 50 years ago.

 

[petersonra]

1000 A at 120 V is about 120 kva.

If you got a three phase 480 service, that would be about 150 Amps. You could run 2 sets of 350 kcm aluminum 1500 feet and end up with about 3% VD. Chances are if the 1000 Amps of 120 V load came from a standard load calculation, it will never come close to ever seeing that much actual load.

At the building you could install a 3 phase 208/120 3 phase wye transformer for all the 120 V loads, and a 240 V delta transformer for the 240 V loads.

 

[winnie]

Panels are rated for Maximum voltage.

Panels are also rated for maximum current.

While unlikely I could imagine a panel where the neutral current rating makes it unsuitable for 120/208.

Jon

 

[winnie]

A transformer with 3 phase in and single phase 120/240V out will not present as a balanced three phase load on the primary so it offers very little advantage. I last saw one called a Hoppy-Hookup from a manufacturer from 50 years ago.

Agreed. You would essentially need 3 separate 120/240V secondaries in a non standard 3 phase transformer, or a bank of 3 single phase transformers.

Jon

 

[jim dungar]

Panels are also rated for maximum current.

While unlikely I could imagine a panel where the neutral current rating makes it unsuitable for 120/208.

Jon

The neutral is always fully rated, if not it could not handle a 100% imbalance.

 

[wwhitney]

The neutral is always fully rated, if not it could not handle a 100% imbalance.

I think the question is whether the panel would still pass the temperature rise tests with full current on the neutral and both ungrounded conductors simultaneously. Versus a 120/240V system where that is not possible. Do you know if the UL listing tests cover that scenario automatically?

Cheers, Wayne

 

[jim dungar]

Agreed. You would essentially need 3 separate 120/240V secondaries in a non standard 3 phase transformer, or a bank of 3 single phase transformers.

Jon

You would have to ground the center point of each 120/240V winding which would not be possible with a single transformer core.

 

[jim dungar]

I think the question is whether the panel would still pass the temperature rise tests with full current on the neutral and both ungrounded conductors simultaneously. Versus a 120/240V system where that is not possible. Do you know if the UL listing tests cover that scenario automatically?

Cheers, Wayne

Yes it would.

 

[kwired]

Will POCO not bring their primary closer to the service or do they just want to charge an arm and a leg to do so?

Single MV coaxial type cable run most of that 1500 feet should cost less for materials than needing conductors for 1000 amps plus whatever might need increased for voltage drop - even at 480 volts that VD is a lot more significant than it would be at say 7.2 kV.

 

[winnie]

You would have to ground the center point of each 120/240V winding which would not be possible with a single transformer core.

I don't see why one couldn't have a standard 3 leg core, each leg with a primary coil and a center tapped 120/240V secondary coil.

There might be a bit of awkward routing to get the center tap out from behind the primary coils, but nothing too insane.

-Jon

 

[jim dungar]

I don't see why one couldn't have a standard 3 leg core, each leg with a primary coil and a center tapped 120/240V secondary coil.

There might be a bit of awkward routing to get the center tap out from behind the primary coils, but nothing too insane.

-Jon

You have to look at the magnetic flux distribution in the core during period of significant unbalance. This is partly why center tapped Delta wound transformers often have limitations on the amount of single phase loading. I have seen the calculations that show you can overload a three winding single core center tapped 240/120 delta simply by turning off the 120V load on one leg.

 

[winnie]

You have to look at the magnetic flux distribution in the core during period of significant unbalance. This is partly why center tapped Delta wound transformers often have limitations on the amount of single phase loading. I have seen the calculations that show you can overload a three winding single core center tapped 240/120 delta simply by turning off the 120V load on one leg.

Ahh, I see what you are saying. The transformer I described would develop the voltages I described (3 120/240V single phase systems distributed across all three phases) but the sort of unbalanced loading normally _expected_ for such single phase systems could potentially cause huge magnetic unbalance in the core and serious problems.

I see this as a design issue for the transformer, but clearly makes my simplistic approach (use a standard core) likely wrong. My _hunch_ is that you would need a larger core for a given kVA or possibly a 5 leg core. If this approach were being seriously considered it would need to go through a custom transformer house with a good engineer on board.

-Jon

 

[jim dungar]

If this approach were being seriously considered it would need to go through a custom transformer house with a good engineer on board.

-Jon

Why bother. there are standard configurations which fill this need, either 3 single 120/240V units or a 1leg center tapped configuration.

When I was fresh out of college, with a transformer design course under my belt, it was brought to my attention that almost every practical transformer design was already being manufactured. I guess they were right because in the 45 years since then all I have seen are tweaks to existing designs, primarily involving materials.

 

[wwhitney]

The transformer I described would develop the voltages I described (3 120/240V single phase systems distributed across all three phases) but the sort of unbalanced loading normally _expected_ for such single phase systems could potentially cause huge magnetic unbalance in the core and serious problems.

Can you expand on that?

Say we have a transformer wound as in the diagram below, with 480V delta primary, and 120/240V wye/hexa(hemi)phase secondary. Say windings 1-3 are the primary windings, with 4 times the turns ratio as windings 4-9. And windings 4 and 7 are connected in series to provide a 120V/240V center-tapped secondary phase; likewise with windings 5 and 8 and with windings 6 and 9.

If the windings are rated for the current, what difference does it make if the secondary load on say windings 4 and 7 are full current on 4 only (120V load), full current on 7 only (120V load), or full current on 4 and 7 (240V load or balanced 120V loads)?

Thanks, Wayne

 

[tortuga]

TBH, clients don't want MV.

Then I would look into a 600V or 1000V service.

The total load is approximately 500amp@240V / 1000 amp@120V. Mostly load is 1 phase 120v, except the AC unit, which is 240V.

I would be interested to see the load calc, I tend to convert everything to VA like the examples in the back of the book.

 

[petersonra]

I'm curious why the client doesn't want medium voltage. Did someone tell them it was a bad idea? Maybe the best answer is to have a chat with them and explain the reality of their situation and try to allay any fear somebody may have put into them about medium voltage.

It might be something as simple as an electrician told him 50 years ago medium voltage is bad for some reason and he just remembers what he was told.

 

[tortuga]

Can you expand on that?

Say we have a transformer wound as in the diagram below, with 480V delta primary, and 120/240V wye/hexa(hemi)phase secondary. Say windings 1-3 are the primary windings, with 4 times the turns ratio as windings 4-9. And windings 4 and 7 are connected in series to provide a 120V/240V center-tapped secondary phase; likewise with windings 5 and 8 and with windings 6 and 9.

If the windings are rated for the current, what difference does it make if the secondary load on say windings 4 and 7 are full current on 4 only (120V load), full current on 7 only (120V load), or full current on 4 and 7 (240V load or balanced 120V loads)?

Thanks, Wayne

View attachment 2561455

So to make standard 120/240 single phase would you tie
I4,I5 an I6 together for L1
V4-I7, V5-I8 -V6-I9 tied together for N
and
V7, V8, V9 tie together for L2?

 

[wwhitney]

So to make standard 120/240 single phase would you tie
I4,I5 an I6 together for L1
V4-I7, V5-I8 -V6-I9 tied together for N
and
V7, V8, V9 tie together for L2?

No, you don't get just one single phase system on the secondary, you get 3 different ones.

If the transformer in the diagram brought out all 18 ends of the 9 coils, then call those coils 1 through 9 and the ends +/-. [The Ix / Vx labeling is apparently for current and voltage, just something that was already on the diagram I grabbed.]

Then on the delta primary side you'd connect 1-/2+ = H1 ; 2-/3+ = H2; 3-/1+ = H3.

On the secondary side, you'd connect 4-/7+/5-/8+/6-/9+ = X0. That's your common neutral on all 3 single phase systems, which can be grounded. Then one 120/240V system is 4+ / X0 / 7- ; the second system is 5+ / X0 / 8- ; and the third system is 6+ / X0 / 9-. These 3 systems are all 120 degrees pair-wise out of phase. You'd have to apportion your 120/240V loads among the 3 systems.

Cheers, Wayne

 

[synchro]

The transformer I described would develop the voltages I described (3 120/240V single phase systems distributed across all three phases) but the sort of unbalanced loading normally _expected_ for such single phase systems could potentially cause huge magnetic unbalance in the core and serious problems.

Can you expand on that?

Say we have a transformer wound as in the diagram below, with 480V delta primary, and 120/240V wye/hexa(hemi)phase secondary. Say windings 1-3 are the primary windings, with 4 times the turns ratio as windings 4-9. And windings 4 and 7 are connected in series to provide a 120V/240V center-tapped secondary phase; likewise with windings 5 and 8 and with windings 6 and 9.

If the windings are rated for the current, what difference does it make if the secondary load on say windings 4 and 7 are full current on 4 only (120V load), full current on 7 only (120V load), or full current on 4 and 7 (240V load or balanced 120V loads)?

The flux in the core induced by a primary winding will be proportional to the integral of the applied voltage waveform across that primary winding. If the primary and secondary windings are tightly coupled and so we can ignore leakage flux, then the flux produced by a load current on a secondary winding will be cancelled by the corresponding current that flows through the primary, via Lenz's law. And so loading on a transformer secondary will not have a major effect on the flux within the core.

If there is a zero sequence current through the three secondary windings of a 3-leg core (in other words, a vector component of current that is the same in each winding) then that could cause problems because there is no return path for the zero sequence flux in a 3-leg core. A 5-leg core would help in this case. The amount of zero sequence flux there would be in practice when there are 3 independent single phase secondary windings is unknown. It would require some different reactive loads on the secondary windings to produce a zero sequence flux in the core. These comments assume that there is a delta primary, because a delta primary cannot supply a zero sequence current to cancel out any zero sequence flux produced by the secondary.

The situation that winnie proposed and Wayne described has three secondary windings that are not connected to each other and will behave independently, and differently than a delta connected secondary. And so the load current on the center tap will be limited by just the windings themselves. With the proposed configuration, a wye-grounded primary could be used because there would be not be the problem of circulating currents that a delta secondary would have with imperfectly balanced wye primary voltages. And importantly, the wye primary could supply zero sequence currents that might be drawn on the secondary, eliminating the zero sequence flux issue in the core. But then triplen harmonics would also be conducted through the neutral. However, when considering all of these factors there's not really much to be gained over just using 3 commonly available single-phase units as Jim has recommended.

 

[tortuga]

Nice explanations thanks fellas.
What about using a Scott T tansformer set up? then you could get two larger single phase outputs instead of 3 smaller ones?
Example for electric rail:

Scott T Transformer – Voltage Disturbance