Residential Split Phase 180deg Phase Difference

[Cjmeziere12]

Hi, all.

Got a question regarding residential split-phase systems.
After looking into how the split-phase systems are derived I keep seeing it mentioned that the Lines are 180degrees out of phase from each other.
I understand that in a 3-phase system the Lines are 120deg out of phase (and that is how we end up with 120V/208Y, etc.) but what I'm not understanding is how a 180deg phase difference doesn't result in phase-cancellation of residential power.

Am I missing a piece of information? Is the "180deg. out of phase" an oversimplification?

Just curious, any help appreciated.
Thanks!

 

[infinity]

This should be interesting. There is only one coil so I never understood how they're could be a phase angle. Let's see what others have to say.

 

[garbo]

Hi, all.

Got a question regarding residential split-phase systems.
After looking into how the split-phase systems are derived I keep seeing it mentioned that the Lines are 180degrees out of phase from each other.
I understand that in a 3-phase system the Lines are 120deg out of phase (and that is how we end up with 120V/208Y, etc.) but what I'm not understanding is how a 180deg phase difference doesn't result in phase-cancellation of residential power.

Am I missing a piece of information? Is the "180deg. out of phase" an oversimplification?

Just curious, any help appreciated.
Thanks!

I have been out of school a long time but no such thing as split phase for the typical what we learned as a 3 wire Edison system where the center point of ultility transformer provides the grounded ( neutral ) conductor. The only split phase that I know is single phase motors that have a starting winding with or w/o a capacitor. A 120/208 volt 4 wire system comes from a 3 phase transformer with the "Y" or center point providing the nuetral.the 3 line phases are 120 degrees apart but never considered a split phase system. I worked on old 2 phase systems where the 2 phases are 180 degrees apart.Hope this helps

 

[d0nut]

The voltage and phase are relative to what point you are using for a reference in the system. The 180deg out of phase is a bit of an oversimplification since it is relative to the neutral point L-N loads rather than L-L loads, but since that is what most residential loads are, it works.

If you are using the neutral as your reference and comparing L-N loads, they do cancel out on the neutral. That is why you can reduce the size of the neutral and why multi-wire branch circuits work. The neutral ends up carrying the imbalance of the system, which is the portion that doesn't cancel out.

 

[zbang]

Residential single-phase service is called "split-phase" in at least a few places (I learned that in the DC area 45 years ago).

d0nut covered most of it, but what hasn't been mentioned, maybe it's assumed, is that the source is a center-tapped transformer with the voltage relationships 120 ("hot") <--> 0 (neutral/grounded) <--> 120 ("hot"). If you look from the neutral to each hot, one goes plus while the other goes minus at the same time, hence "180 deg out of phase" (looks like a balanced line, even though it isn't actually one).

 

[wwhitney]

but what I'm not understanding is how a 180deg phase difference doesn't result in phase-cancellation of residential power.

Say you have a transformer like this, where the secondary has two separate coils, each 120VAC:

If you call V(X1,X2) = +120V, then V(X3,X4) = +120V, and of course V(X2,X1) = V(X4,X3) = -120V. Often you'd wire a transformer like that by connecting X2 and X3; that gives you conventional 120/240V split phase, where X2 = X3 is the neutral point. V(X1,X4) = 240V.

You could instead connect X2 and X4. Then V(X1,X2=X4) = 120V, and V(X2=X4,X3) = -120V. So V(X1,X3) = 0V, which is the cancellation you conceived of. Given that, you can also connect X1 and X3. Now you have a 120V only, 2-wire source, with the same available power as the 120/240V split phase wiring.

So I would say that you don't get cancellation because you wire/build the transformer to get addition instead of cancellation, if you want 120/240V split phase.

Cheers, Wayne

 

[gar]

211223-1149 EST

I have never understood why electricians are so confused on this subject.

If you have two waveforms that are of exactly the same frequency, and I do mean really very exactly, then the relation between the two or more waveforms in one aspect can be described as the phase difference.

If the waveform zero crossings are identical, and of the same slope. then the phase difference is 0 degrees between the two waveforms. If the same zero crossings occur, but the slopes are opposite, then the phase shift difference is 180 degrees. If one of the waveforms is shifted only 1/2 as far, then you have a 90 degree phase shift.

This is very simple to understand.

.

 

[Rock86]

211223-1149 EST

I have never understood why electricians are so confused on this subject.

If you have two waveforms that are of exactly the same frequency, and I do mean really very exactly, then the relation between the two or more waveforms in one aspect can be described as the phase difference.

If the waveform zero crossings are identical, and of the same slope. then the phase difference is 0 degrees between the two waveforms. If the same zero crossings occur, but the slopes are opposite, then the phase shift difference is 180 degrees. If one of the waveforms is shifted only 1/2 as far, then you have a 90 degree phase shift.

This is very simple to understand.

.

I was just typing a similar response.

I will add... It is single phase, there is no phase cancellation. Thinking of it in a DC way... L1 to N is like reading (+) to (-) on a meter and getting a (+) result, then taking the meter and reading N to L2 as (+) on the N and (-) on the L2 you would get a (-) reading. Its visibly explained with two sine waves mirroring each other.

(I don't know if that helps... it sounds better in my head I think haha)

 

[LarryFine]

We had a lengthy discussion a while back about this. A center-tapped secondary is effectively two sources wired in series, like a pair of batteries. It's a difference in polarity, not a difference in timing.

 

[wwhitney]

If you have two waveforms that are of exactly the same frequency, and I do mean really very exactly, then the relation between the two or more waveforms in one aspect can be described as the phase difference.

True for pure sinewaves. And in that case the difference between "180 degrees apart" and "negative" degenerates, they look the same.

But consider an AC voltage waveform of period T that's of the form V(t) = A * sin(2π (t/T)) + B * sin(2π (2t/T)). So a mix of a sinewave and its 2nd harmonic. Then it's no longer true that V(t+T/2) = - V(t); "180 degrees apart" and "negative" are now two different things.

If we feed that waveform into a transformer with a center-tapped secondary, L1-N-L2, what is the relationship now between V_N-L1 and V_N-L2? My understanding is that V_N-L1 = - V_N-L2, and they are not 180 degrees apart.

Cheers, Wayne

 

[Carultch]

Am I missing a piece of information? Is the "180deg. out of phase" an oversimplification?

It is mathematically identical to two waveforms that are 180 degrees out of phase, assuming ideal sinusoidal waveforms with no irregularities. You might initially think it should be called 2-phase, because by extending the defining characteristics of 3-phase electricity to the WHAT-IF scenario with just two waveforms that are 120V nominal to neutral, it is identical to the 120/240V split phase systems that we use in 1&2-family dwellings.

However, there are two reasons we don't call it "2-phase":
1. It isn't produced that way. It isn't produced by generator magnet/coil interactions that are delayed in timing, in the same manner as 3-phase is produced by three magnet/coil interactions that are delayed in timing. Instead, it is produced by center-tapping the transformer secondary just one of the phases of a 3-phase system. Split-phase circuits are built based on producing two waveforms that are equal and opposite relative to neutral, rather than two waveforms that are time-delayed.

2. "Two-phase" is a term that is already coined, for an obsolete electrical distribution system, from the early 1900's. It was built with two waveforms that were phase-shifted by 90 degrees, in order to have the advantage of starting motors, which is different from the 180 degree apparent phase shift in question. Internal to some designs of modern single phase motors, the same concept is used for powering the starter coil, but it is no longer used as a distribution system, thanks to the advantages of 3-phase power over the 90 degree phase shift.

 

[Carultch]

True for pure sinewaves. And in that case the difference between "180 degrees apart" and "negative" degenerates, they look the same.

But consider an AC voltage waveform of period T that's of the form V(t) = A * sin(2π (t/T)) + B * sin(2π (2t/T)). So a mix of a sinewave and its 2nd harmonic. Then it's no longer true that V(t+T/2) = - V(t); "180 degrees apart" and "negative" are now two different things.

If we feed that waveform into a transformer with a center-tapped secondary, L1-N-L2, what is the relationship now between V_N-L1 and V_N-L2? My understanding is that V_N-L1 = - V_N-L2, and they are not 180 degrees apart.

Cheers, Wayne

There in-lies the essential difference.

This is what the ideal sine waves look like for a 120/240V split phase system. The amplitude of the line-to-line waveform is 340V, due to the fact that the nominal voltage is 240V. These numbers are related via a factor of sqrt(2).


Now suppose we define VLL(t) = A*sin(w*t) + B*sin(2*w*t), so that we now have a harmonic with a frequency twice the fundamental. I've set w=2*pi, so that the horizontal axis has units of cycles. Continue with A=340V, but now add the second harmonic at half the amplitude, B=170V.

The way split-phase is derived in reality, the line-to-neutral waveforms will look like this. And this is what split phase wpuld look like, if extreme harmonics like this were to exist.


Suppose now that we keep the same waveform for V1 to Neutral, but instead phase-shift it by 180 degrees to define V2 to N. This is what the waveforms will instead look like, in the hypothetical case that it were derived by a phase shift instead of a center-tap:

 

[synchro]

I will add... It is single phase, there is no phase cancellation. Thinking of it in a DC way... L1 to N is like reading (+) to (-) on a meter and getting a (+) result, then taking the meter and reading N to L2 as (+) on the N and (-) on the L2 you would get a (-) reading.

We had a lengthy discussion a while back about this. A center-tapped secondary is effectively two sources wired in series, like a pair of batteries. It's a difference in polarity, not a difference in timing.

The analogy of "spit-phase" to two DC sources of opposite polarity also has some historical basis. Edison's original 3-wire arrangement was for +110VDC and -110VDC in order to lower the voltage drop and save copper when 110VDC loads (such as his light bulbs) were part of the loads that had to be served.
The advantages of such a 3-wire arrangement also apply with a "single phase" AC power source, and in this case the arrangement was named "split-phase."
I think the term split-phase, or any reference to the word "phase" on single-phase circuits or systems occured after the emergence of "polyphase" systems such as 2-phase or 3-phase. After all, phase is a relative measure between two voltages, currents, signals, etc. If there is only one sinusoidal source within a system, the term "phase" becomes somewhat meaningless, unless it is in comparison to some time reference such as GPS. Or if we are dealing with harmonics.

 

[Carultch]

If there is only one sinusoidal source within a system, the term phase becomes relatively meaningless (unless it is in comparison to some time reference such as GPS).

The term phase refers to the constant phi, in the general equation for a sine wave, that indicates where within the cycle the waveform starts at time, t=0.
V = A*sin(w*t + phi) + D

It still is relevant for a single waveform, it just turns out to be something isn't as essential to think about. It is usually an arbitrary choice where t=0, so we often select it the start time so that phi can also equal zero. With three phase, you have two waveforms that will require phi to equal something other than zero, which is why it is a lot more relevant of a concept for 3-phase systems.

 

[synchro]

The term phase refers to the constant phi, in the general equation for a sine wave, that indicates where within the cycle the waveform starts at time, t=0.
V = A*sin(w*t + phi) + D

It still is relevant for a single waveform, it just turns out to be something isn't as essential to think about. It is usually an arbitrary choice where t=0, so we often select it the start time so that phi can also equal zero. With three phase, you have two waveforms that will require phi to equal something other than zero, which is why it is a lot more relevant of a concept for 3-phase systems.

I guess when I said "meaningless" I meant for practical purposes, because as you said, phase can usually be considered relative to an arbitrary timing reference. If the frequency of a sinusoidal waverform varies with time as a function f(t), then the phase will change as the time integral of f(t). In the short term the POCO frequency will have some variation and hence the phase (measured relative to a precision time reference like GPS, WWV, etc.) will also change. There are corrections made for long term drift, but even then I'm sure that many degrees of phase error are tolerated. The main concern is that all of the sources on the grid maintain adequate synchronization with each other, even in the short term.

 

[OldBroadcastTech]

Hi, all.

Got a question regarding residential split-phase systems.
After looking into how the split-phase systems are derived I keep seeing it mentioned that the Lines are 180degrees out of phase from each other.
I understand that in a 3-phase system the Lines are 120deg out of phase (and that is how we end up with 120V/208Y, etc.) but what I'm not understanding is how a 180deg phase difference doesn't result in phase-cancellation of residential power.

Am I missing a piece of information? Is the "180deg. out of phase" an oversimplification?

Just curious, any help appreciated.
Thanks!

Well, I'm just a retired Broadcast Engineer, and I had to re-join just to put in my $ 0.02....but I have heard the '180 degrees out of phase' ever since Electronics School in 68-70.

What you have ( in a single-phase 120-240 volt system) is two voltages of OPPOSITE POLARITY. When voltage on L1 is going positive, voltage on L2 is going negative. Either one, referenced to N, is 120 volts, but measured L-L, you get 240 volts.

With a PURE sine wave, being of opposite polarity is INDISTINGUISHABLE from being '180 degrees out of phase'.

But consider a NON-SINUSOIDAL wave-form, for instance a positive-going half-cycle of triangular shape and a negative-going half-cycle of a square wave.
In that case, '180 degrees out of phase' would be a waveform with the negative-going square wave first, then the triangular positive-going part. Or to put it differently, 'displaced by a half-cycle' (180 degrees).

Soapbox now available, 'You may fire when ready, Gridley !'

 

[Hv&Lv]

No math. hope to keep it simple…


Notice the two coils at the bottom. They are “hot” on the left side and the right side, with the center being grounded. So if we draw a “+” on the left and right sides and a “-“ in the center, it helps give a picture of what Larry Fine was saying about polarity being opposite. As the primary single phase wave rises from “0” to its peak voltage(say 7200volts), a voltage relative to coil windings (60:1) is induced in the secondary, but with different polarity. So the voltage on the right is increasing “positive” (or up on a graph) to 170 volts (peak voltage) at the same time the voltage on the left coil is increasing “negative”(or down on the graph) to 170 volts (again, peak voltage). Look at the peak to peak voltage of the black and the red lines and divide by square root 2. (340/1.414)

so you get a sine wave picture that looks like this:
As Calrutch pointed out there’s a square root 2 relationship to RMS voltage and peak to peak voltage. It’s the DC equivalent.

Now, to make the waters even muddier, when we connect three transformers together for a 120/208, we connect the two coils in parallel like this:
you can see the polarity on both coils is the same and only good for 120 volts whereas the picture above they are opposite in series. We do this internally with a regular ”residential” transformer right before we build the bank. Once three are connected together on a true three phase line the voltages are 120 degrees apart the same as the primary voltages are 120 degrees apart.

 

[LarryFine]

So if we draw a “+” on the left and right sides and a “-“ in the center, it helps give a picture of what Larry Fine was saying about polarity being opposite.

I would instead draw it with one line +, one line -, and the center tap with one of each, like this diagram. Incidentally, in my opinion, the two sine waves should be drawn the same, not opposite like they are here:

 

[LarryFine]

Here's another one showing the polarities. If the two halves of the secondary were actually of opposite "phase", wiring them in series would produce zero volts instead of adding to produce 240v.

 

[zbang]

It's amazing how many ways there are to say the same thing .