Why is North American residential power called single phase?

Question

I am under the belief that North American residential power is a special case of two phase power adequately described by the term split phase. Clearly I am in the minority and want to be convinced otherwise but have not found any convincing explanation.

I always assumed there was a functionally or historical reason but I'm always met with an inadequate theoretical explanation. Best I can figure it's a contextual reason to help the people who work on motors or it's to differentiate from systems with two phases that are 90deg out of sync. A theoretical explanation seems to require a unique definition of phasors or to define our electrical system by the loads we attach rather than the voltage source supplying the current with a zero point reference.

Here is my theoretical argument. Pretty simple. Transformers transform. With the secondary having a center tap, 3 wires, each pair of wires has either a unique phase angle from the other pairs or a unique magnitude, even accounting for direction. Now arbitrarily calling the center tap 0v. The other two legs are 180deg out of phase. (Note, grounding an edge would give the same phase angle but different magnitudes, still a two-phase system.)

Answer 1

Because it is a single phase, with a center tap.

That's all it is. There's nothing more exciting than that. It is a 240V transformer secondary with 3 taps at 120V spacing.

Most installations have a safety earthing system (not to be confused with electronics GND/Vss). Leakage in the transformer might float all 3 taps thousands of volts above earth. We don't want that, so we pick 1 tap and bond it to the safety earthing. Does that make sense?

The one we bond to earth, we call "neutral".

Let's imagine they picked the tap at the end of the transformer winding. Neutral would be at the "bottom" and there would be a 120V tap and a 240V tap. * In that case, you would unquestionably call it single-phase with taps - right?

Using the center tap for neutral is marginally less dangerous, since it means neither hot wire will be more than 120V to ground. The transformer is exactly the same. Nothing has changed but the bonding.

You can call it 2-phase if you really want to; I see where you're not wrong, in a principled sort of way. But the whole point of language is to communicate with others. And others think 2-phase is a 4-wire system with two loops 90 degrees out of phase.

---

---

---

* Once I found a 120V service like this, a fault had caused "the white wire formerly known as neutral" to be 120V from ground. Some circuits had a 0V hot, others had a 240V hot. Everything worked fine. Nothing cared. It was found during routine maintenance.

Answer 2

Why is North American residential power called single phase?

Because that's exactly what it is.

Utility power in North America is distributed via single-phase transformers with their primaries being fed 12.47 kV using 2 lines of a three-phase system or 7.2 kV using a line and neutral.

The distribution transformers, which are pole-mounted, are also known as a 'pole pigs'.

The single secondary winding L1 - L2, with an earthed centre tap designated as the neutral, provides a single-phase 240 V supply between L1 and L2 and two single-phase 120V supplies, which are out-of phase by 180°, between L1 & N and L2 & N.

Hence it is referred to as a 'single-phase, three-wire' system or 'split-phase' system.

I am under the belief that North American residential power is a special case of two phase power adequately described by the term split phase.

A theoretical explanation seems to require a unique definition of phasors or to define our electrical system by the loads we attach rather than the voltage source supplying the current with a zero point reference.

A single-phase alternator, has a single winding that produces a single, continuously alternating voltage.

A two-phase alternator, now obsolete, had two windings spaced 90° (electrical degrees) apart with the voltages generated in them having 90° phase displacement.

A three-phase alternator, has three windings spaced 120° apart with the voltages generated in them having 120° phase displacement.

A single phase induction motor, having two windings spaced 90° (electrical degrees) apart is actually a two phase motor. The single phase supply is split into two phases that are out of phase by 90°, by connecting a capacitor in series with one winding. A single phase induction motor is hence referred to as a split-phase motor.

The 'single-phase, three-wire' system or 'split-phase' system, with its two voltages being out of phase by 180°, does not qualify as a two-phase system.

Answer 3

The "single-phase" terminology is nearly always used in the context of three-phase supplies as this is the standard method of power transmission used throughout the industrialised world.

Image source: Plug Socket Museum.

When you study the diagram of the delta-connected high-leg transformer arrangement it becomes clear that the L1 - N - L3 output is derived from one phase of the three-phase supply. There is no phase rotational direction associated with them so, on their own, they cannot drive an induction motor without additional help to start them in a particular direction.

I think calling it a two-phase system is likely to imply a phase angle other than 180° and lead to some confusion. "Split single-phase" would give the least risk of confusion.

Answer 4

This is perhaps opinion based, but I believe the reason we avoid the term "two-phase" is historical. A long time ago, two phase power with a 90 degree phase angle was somewhat commonly used. The term "split phase" differentiates our domestic electrical system from two-phase 90 degree power.

I am not sure if this is relevant to your question or not, but the power on the utility poles in North America (as well as buried distribution wires) is truly single phase in many places. Split phase power is created by the center tap on the transformer secondary as you say in your question. So when people refer to North American power distribution, in many places, that power is correctly called single phase. The actual power inside the residence and at the service entrance could be called two phase or split phase. So in some specific instances, it may be perfectly accurate to say "single phase."

Also, just for completeness, in many places in north america, power distribution is three phase. For example at my house, the power on the utility poles is three phase power. The transformer that supplies my house converts one of those phases to split phase residential power. But a second transformer on my property supplies a second meter with three-phase power from all three phases of the utility wires. That second meter is for a well pump. The reason for that is definitely off-topic. But basically there is a three-phase powered well pump connected to that meter, and the electricity used by that pump has to be paid for separately from my utility bill.

Answer 5

The nomenclature is largely historical, and indeed there was (briefly) a 2 phase 90 degree system used as this could run an induction motor (Which 180 degree two phase power cannot without help).

Of course two phase (of either variety) takes the same number of conductors as three phase delta which produces way less torque ripple and is generally superior, so it took over rather rapidly.

You could call 120-0-120 two phase, but you will just confuse folk, which is a good enough reason not to.

Answer 6

The two split phases do not meet any of the usual constraints usually implied by the term "separate phases"

In a true N phase system the voltages in each isolated phase are at 360/N degrees electrically. In the North American system there are 3 such phases, each 220 V and each at 360/3 = 120 degrees to the other.

It is customary to split each of these phases into 2 x 110V non-isolated portions. Each of these is constrained to be at the same electrical rotation by virtue and constraint) of being on the same core of a 3 phase transformer and are unable to be isolated electrically without the provision of an additional isolating transformer.

It would be possible to physically split the winding at the centre point and to provide two voltages, each still in 0 degrees or 180 degrees isolation to each other - but this is not done. Doing this would require that each of the 100V windings be connected to Neutral at one end, and they would then be unable to be connected together to allow the full 220V phase to be used in eg heating applications.

Answer 7

The answer about the pole top transformer is correct as far as it goes, but the primary of the pole top transformer (usually in the 11000-33000 volts) is really one leg or a high voltage 3 phase coming from the local distribution substation. If primary of the pole top transformer has one high voltage line and a ground (which which is connecting to an earthing ground at the base of the pole, then it came from a 3-phase "Y" distribution from the substation (if the transformer primary has two HOT wires, then from a 3-phase delta) transformer. That, in turn was driven by a primary transmission line of 69,000 to 200,000 volts.). Thus, what runs down the street is ONE (single) Phase of a three phase power system. Thus, refered to as a SINGLE PHASE final distrubution.

Just follow the single phase back toward the substation, and you will eventually come to a pole which has all three phases (feeding different neighborhoods or industrial complexes)

The local/area substation transformer primary voltage is usually anywhere between 69,000 volts and 240,000 volts which were probably an intermediate voltage stepped down from the long distance transmission lines ("the Grid") which could be up to 2 million volts. Next time you drive under one, count the wires...3 HOT lines for each phase.

Dan, BSEE, 40 years experience, starting as Asst. Operator of a power station for BVG&E

Answer 8

Because it just is single phase

Octave code:

x = 0:0.1:(2*pi+0.5);
plot (x, (x-x), "-k;reference/neutral;", x, sin(x), "-r;Line1 [=sin(x)];", x, 2*sin(x), "-m;Line2 [=2*sin(x)];", x+pi, sin(x), ":^g;Line1+180deg;", x, -sin(x), "--vb;Line1*(-1) [=-sin(x)];")

#1 - ground at one outer tap

• Black reference/neutral is the grounded outer tap (=GND henceforth in this post).
• Red Line1 is the middle tap with 120Vrms to GND.
• and Magenta Line2 the other outer one with 240Vrms to GND.

Voltage between Line1 and Line2 is 120Vrms too. -> 120V+120V=240V -> these two "120Vs" must be in phase and not 180° out of phase or you wouldn't get 240V.

As shown/stated by @vu2nan the pole transformer has just one primary/input winding and only one core and one output winding (with a center tap). Even if different phases were to arrive at the primary winding (which is electrically impossible) the winding itself would only swing with the sum of those phases (replace phases with frequencies in this sentence and it makes sense...).
Thus everything on the secondary winding must be in one phase too (there may be an exception if the transformer isn't magnetically/mechanically balanced mirror-able).

#2 - GND at center tap eg. "split phase"

• Black is the GNDed center tap.
• Red Line1 is the 1st outer tap L1 with 120Vrms to GND.

And you have two options for the 2nd outer tap L2:

1. it has exactly the same voltage profile as L1 / Red Line1 when you measure from GND to L1 and from L2 to GND. (use a scope where inputs 1+2 have completely isolated inputs including there grounds or two separated scopes with synced triggers(?). Basically if you measure exactly as in #1 but with a shifted GND.)
2. or it's Blue/Line1*(-1), the inverse of L1 when you measure from GND to L1 and from GND to L2 (eg. flipped #1).

This doesn't change anything with the phase. It's still just one phase and not out of phase by 180° (or any other degree).

To mathematically get 240Vrms here between L1 and L2 you have two options again (with 1 on the y-axis being equivalent to sqrt(2) * 120V):

1. sin(x) + sin(x) = 2*sin(x)
2. sin(x) - sin(x+pi) = sin(x) - -sin(x) = 2*sin(x)

but sin(x) + sin(x+pi) = sin(x) - sin(x) = 0 (one 180° out of phase to the other)

Now if you want to add two times 120V together they need to be in phase. If their phases are pi-shifted you need to subtract(!) both potentials from each other to get 240V.
Mathematically the result is the same but one is is a bit more counter intuitive (from this perspective).

Why the term "phase" is out of place (here in my opinion)

Short: If you first learn this split phase out of phase you need to re-learn it when you want to work in electrical power distribution and so on (in the US you have multi-phase AC there too?).

Much longer....

Phase is usually only used with n-phase systems (with n>1) in relation to the generators.
Imagine a simplified/idealized generator with just three stator coils in star configuration 120° apart from each other and one bar-magnet as the rotor. Now if you start to turn this generator with 60rpm = 1Hz there is a 'real' phase difference between the three outputs when you compare the sinus waves between any two random but different pairs of measuring points (L1/2/3). The V-peaks on each pair follow each other 1/3 *s apart (or 120°).

Example: L1~L2 has it's V-peak 1/3s before L2~L3 but 1/3s after L1~L3. I used "~" instead of "-" here because the polarity is irrelevant for the phase. If you measure one pair in the opposite direction to the other one it's up to you to use the inverse or you'll falsely think the rotating field suddenly switched direction which can quickly lead to catastrophes (as if a big gas turbine would suddenly switch directions).
In other words: You need to decide on one rotation direction and measure along the Ls that way top get a correct result. If you switch directions in the same set of measurements the result cant be trusted.

When you you have only one phase you have no rotating field at all and cant have any 'true' out of phase "phases"...

2nd example: If you connect a new power plant to the grid and the generator is not exactly in phase with the grid it will either be sped up or slowed down by the grid to get it in phase (the grid will do the opposite but it's rotational momentum is far greater so this effect is negligible). If it's too far out of phase (eg. by 180°) it will probably destroy itself, some transformers, the turbine and more...
This out of phase doesn't really relate to the split phase system's out of phase.

Another example: modern eg. DUSPOL voltage testers and equivalent can measure the rotation direction of/between just two phases of a three phase system by - I'm not actually sure how - maybe measuring the time between V-peaks (or with a capacitively connected third pole through the handlers hands - need to test that)?
The problem with that is: Under ideal conditions you get a perfect sinus wave between two phases regardless of variations in phase difference or if it's L1~L2, L2~L3 or L3~L1. Reason is that the sum of any two sine waves with the same frequency (eg. 60Hz) and any phase shift of f<>pi is another sine wave, just with a different amplitude. (If f=pi=180° the sum is 0)
But If one measures between GND and any of the three phases the voltage tester randomly flips between right and left rotation. It would be the same in a single phase center taped system - no or randomly flipping rotation direction but between any pair of measurement points (not just between GND and L* but between Lx and Ly too).

The phase offset I know of is in direct relation to the mechanical construction of a multi-phase generator (degree between sets of coils and if two sets are 180° apart they are connected so they are in the same set...).

Why Wikipedia says "Two 120 V AC lines are supplied to the premises which are out of phase by 180 degrees with each other (when both measured with respect to the neutral)" is beyond me. As if split phase alone wouldn't be the source of enough confusion, let's throw in an out of phase.

Much better would be (imho): Two 120V AC to ground lines are supplied to the premises with 240V between the two
And maybe add (which means with respect to the grounded center tap both AC lines are inverted relative to each other)
but leave phase out if it...

RE your kinda vague questions

The common North American residential power is not a 2-phase system by any reasonably accurate technical definition. Why the confusing term split phase is used instead of eg. center tapped single phase - which would be exact, accurate and technically correct - I don't know.

The reason is purely technical but kinda depends on how you define phase in this regard. As the mathematical/physical time offset of different sine waves of the same frequency with and/or ;-) without regards to polarity.
For time offset take look at the green wave from the graph.

Calling them out of phase is very misleading imho. Maybe phase shifted? Don't like that much more so maybe phase inverted?!

Answer 9

I always assumed there was a functionally or historical reason

Historically the purpose of the electricity grid was to connect motors and lights to generators.

A "split phase" system is closer to a single phase system than a true polyphase system in two important ways.

1. It cannot directly create a rotating magnetic field. This means that hacks are required to make AC motors work on single phase or split phase.
2. While the voltages and currents are 180 degrees out of phase, instantaneous power is defined by voltage * current. So the power waveforms of the two "phases" are in phase with each other. Power drops to zero twice per cycle. That means that your motors and generators need more inertia than they would in a true polyphase system which can deliver power continuously.

Answer 10

Spit-phase: Keeping track of any arbitrary direction say L1 to N to L2. Mind, you can't start at N and switch which way you are "looking" without switching the sin (+/-) of the voltage. L1-N would add linearly with N-L2 signifying they are in-phase. Since there is no phase angle difference between L1-N and N-L2 we say the total system is single phase.

180deg two phase: If L1-N and out N-L2 where 180deg out of phase then N-L2 would subtract from L1-N and the voltage from L1 to L2 would be 0. This would be physically accomplished by winding N-L2 in the opposite direction. Then you could say there is a phase angle difference between L1-N and N-L2 and thus could say there are two phases.

Answer 11

The voltage between 2 phase wires that have a phase difference of 180° is 0. If the phase difference is 120° as in a three-phase system it is √3 x the N-phase voltage, so the US system is clearly not 2-phase. Only if they are in the same phase the voltage will be twice as high.

Answer 12

To my knowledge, North American residential power uses three systems.

120/240 single phase is straightforward, one transformer winding with the centre tap connected to a grounded conductor. The two live conductors have phase angles of 0 and 180° with respect to the grounded conductor.

120/208 wye is also straghtforward. 3 windings are each connected between the grounded conductor and a 'live' conductor. The three live conductors are 120v and 120° in phase apart with respect to the grounded conductor, and 208v between any two live conductors. Residential properties typically use two live conductors and the grounded conductor, with high powered appliances connected between the two live conductors to take advantage of the 208 volts.

120/208/240 delta however is an oddball system. They use 3 transformer windings in a 240v 3 wire delta connection, but connect the centre tap of one winding to the grounded conductor. This creates two live conductors at 120 volts to the grounded conductor, and 0° and 180° phase angles, and a third "high leg" conductor at 208v and 90° phase angle.

Single phase resedential connection on that system bring in the two '120v' live conductors and the grounded conductor. It is not possible to use the voltage difference between these three conductors to create a rotating magnetic field. They are electrically equivalent to the "split-phase" system, and so are called "single phase".