Securing Power from a 120VAC Generator: H3 and H8 Taps

[jimboz]

TL;DR

Can I use a multitap transformer to step up 120 to split phase 240? Transformer schematic attached

My plan is to power the secondary with my 120VAC generator and use the 240VAC primary connection configuration and use the H3 and H8 taps as the neutral.

 

[berkeman]

Welcome to PF.

How much power are we talking about here? How many kVA is the transformer that you want to use? How clean is the output from your generator? What size fuses are you planning on using for the primary and secondary sides?

 

[jimboz]

7.5 kva (on hand), 'Pretty clean' honda eu2200i (bout 2kw) generator, probably 15 amp fuses on 120 side and 8 amp on 240 side. The load is a submersible well pump 6 amps running at 240VAC.

 

[DaveE]

Summary:: Can I use a multitap transformer to step up 120 to split phase 240? Transformer schematic attached

View attachment 278130

My plan is to power the secondary with my 120VAC generator and use the 240VAC primary connection configuration and use the H3 and H8 taps as the neutral.

Yes, but not in the normal application. You want to basically drive a 120V winding, and get a center tapped 240V output, which would be two 120V, out of phase, and connected at the output neutral, correct?

The problem is this transformer only has the X1-X2 and X3-X4 windings rated at 120V. The other windings are much larger in voltage. As below:

H1-H2 = H9-H10 = 216V
H1-H3 = H8-H10 = 228V
H1-H4 = H7-H10 = 246V
H1-H5 = H6-H10 = 252V

So you can make ~240V in the H windings, but you can't get a 120V center tap. What you are looking for is a winding ratio of 1:1:1.

It can be done if you connect the X windings in series, the 240V configuration. But you would only drive it at 120V. Then connect the H windings in a 480V CT (or 2x 240V) configuration. Since you are only driving the X windings at half voltage, you only get half of the voltage out of the H windings. So, if you use the H3 and H8 taps, as you suggest, you'll get 228V CT out. If you really want 240V CT out, I would connect it as suggested for 480V with the neutral at H4 and H7.

Of course you will have to derate the power (VA) rating by 2x to avoid melting the copper windings. You don't get to double the current because you cut the voltage in half.

 

[Tom.G]

If you do not need the isolation between the generator and pump, there is an easier way:

Connect the transformer Primary for 240V (X2 to X3)
Connect the generator to one half of the primary (for instance to X4 and X2/X3)
Connect the pump to X1 and X4

You can get away with this because the transformer is way oversized (5 times) for the pump. (It would need a minimum of 3X oversized)

edit: If one power lead from the generator is low side ("Ground"), connect it to the X2/X3 terminals. That gives you the conventional 2-phase 120V/240V connection.

Cheers,
Tom

 

[jimboz]

If you do not need the isolation between the generator and pump, there is an easier way:

Connect the transformer Primary for 240V (X2 to X3)
Connect the generator to one half of the primary (for instance to X4 and X2/X3)
Connect the pump to X1 and X4

You can get away with this because the transformer is way oversized (5 times) for the pump. (It would need a minimum of 3X oversized)

edit: If one power lead from the generator is low side ("Ground"), connect it to the X2/X3 terminals. That gives you the conventional 2-phase 120V/240V connection.

Cheers,
Tom

Whoa! I wouldn't have thought of that in a million years and I only kinda understand it now. If you've got a few minutes would you mind explaining how it works and how the derating factor is arrived at.

Thanks

 

[DaveE]

An autotransformer works just like a normal transformer, except that there isn't any isolation between the primary and secondary windings. This means that the primary and secondary currents are flowing in the same wires and will partially cancel.

I like to explain it as in the picture below: Imagine that when you wind a normal transformer you use a bifilar winging. That means that the primary and secondary windings are identical and are physically laid side by side. Because they are identical, the (instantaneous) voltage at any point along the primary winding is equal to the voltage on the secondary winding next to it. So, there is no need for insulation between the primary and secondary windings. No voltage across the insulation means no current flow. Of course, insulation is required between the adjacent turns of each winding; but let's say the windings are spaced apart (turn to turn) so air is the insulator.

If no insulation is necessary between primary and secondary, then we can remove it and have the wires touching along their entire path. But touching wires is practically the same as a single winding. Now the primary and secondary currents are flowing together in a single wire. The polarities mean that they cancel, at least partially. That is the genesis of the derating. Some of the current can then bypass the transformer "primary" winding and flow directly through the load.

 

[Tom.G]

For a hopefully simpler explanation, refer to the transformer wiring diagram in the first post.

In particular the Secondary Volts table connections of X1, X2, X3, X4.

Now remember that a transformer doesn't care which windings you use as a primary, secondary, tertiary, or whatever. The relative voltages are determined by the turns ratios between the windings.

1) For 120V on the secondary, the two windings are wired in parallel. This implies that the windings are identical; if they were NOT identical there would much current from one winding flowing thru the other one. Not very desirable.

2) Now for 240V, the windings connect in series; two identical 120V windings in series for 240V. If this is an output it would be considered 120/240V two-phase.

3) Since a transformer doesn't care where the power comes in or goes out, you can apply 120V INPUT from the X2/X3 connection (the center tap) to either end of the winding, X1 or X4.

4) The 'other secondary' (other half) winding is still connected in series with the winding where you supplied power.

5) Since they are in series (with the correct phasing), you get 240V across the winding ends (X1 to X4).

As @DaveE pointed out, this configuration is called an 'autotransformer', or rarely an 'autoformer'. There are commercial products used as test equipment that can supply a continuously variable voltage from zero to above line voltage. This is done with a sliding contact that can be positioned at each turn of the winding. Two common brands in the U.S. are 'Variac' and 'Staco'. Try an internet search for them for more info.

As for the derating, I chose 3 because I was (and am) too lazy to look it up/investigate in detail. It is based on the fact that the wire used for the powered part of the winding (and the Iron core) must support the total power withdrawn, plus losses. Perhaps the derating is closer to 2, but I chose to be conservative.

This ended up a longer post than I expected! Hope it helps.

Cheers,
Tom

 

[DaveE]

The increased power handling ability of autotransformers vs. isolation transformers is due to the elimination of the primary winding. The size of transformers is a combination of the core size required plus the amount of copper wire required. The core size is determined primarily by the applied voltage and frequency. Then the geometry of that core allows a limited amount of space for the windings. The power is determined by the voltage applied times the current capability. The current capability is determined by the I²R losses allowable in all of the windings.

In my previous example, the isolation transformer needs primary and secondary windings: 2 windings that can carry 10A each and one winding that can carry 20A. It would be common practice to split the Cu losses equally between the primary and secondary windings (equal power dissipation density). This means that the primary winding resistance will need to to be 1/2 of each secondary winding resistance (let's call that Rs). So the total resistive heat is 20²⋅(Rs/2) + 2⋅(10²⋅Rs) = 400⋅Rs. The unused primary winding would be left out to allow more copper in the other windings.

In the case of the autotransformer, the core can be the same (since the voltages on the windings are unchanged), but there isn't a primary winding any more. This will allow you to increase the diameter of the wire in the secondary windings to allow more current. Twice the wire area means half of the winding resistance. so, for the same resistive losses the secondary can now carry 20A ( 400⋅Rs = 2⋅20²⋅(Rs/2) ).

The elimination of the need for a primary winding allows more space and thus more current in the secondary windings for the same amount of copper losses.

But, this applies to a design optimized as an autotransformer. Since you have a transformer that was intended to operate as a normal transformer, with isolation, you won't get this benefit. If you wire it as an autotransformer, it will work as originally rated, but some of the winding space will be filled up with windings that you aren't using. Transformers like this give extra windings to allow flexibility at the expense of optimization of efficiency. Your transformer is only working at peak efficiency when all of the windings have the maximum allowable current flowing through them. Otherwise, they should be derated to some extent.

It is common in autotransformers to see dramatically reduced wire diameters in some windings because the application doesn't require much current flow through them. This can't be done in general purpose transformer designs.

 

[Tom.G]

@jimboz,

If you want to learn more about transformers, there is an older thread,
https://www.physicsforums.com/threads/how-to-calculate-toroidal-core-maximum-va-capacity.876425/,
that digs into details. See especially the links in posts #5 and #12 there.

Thanks to @Charles Link for finding and pointing out that thread.

Cheers,
Tom

 

[jimboz]

Well, given those in depth descriptions I'll leave transformer design to you guys! I do plan on using the 'h' terminals only as this seems the simplest approach. Really appreciate y'alls help.

 

[berkeman]

(Apologies if this has already been discussed earlier in the thread)

honda eu2200i

Does anybody know if this generator ties Neutral and Earth ground at its output? At first I was not concerned with Earth ground issues, since it's a floating generator that is making the AC Mains 120Vac waveform. But now that I think about it, if the generator ties its Neutral and Earth ground connections together at its outlets, and you use an autotransformer to make the split center tap 120V-120V to get 240Vac, the placement of the "Neutral" output of the generator may matter.

It may matter if the pump is grounded, or if another grounded appliance is also plugged into the generator. I guess if the Neutral output of the generator is connected to the "center tap"/Neutral connection of the autotransformer it might be okay. But it would be good to see a diagram of the whole system to be sure. Kind of like the last sketch in Post #7 by @DaveE but with explicit Earth ground connections shown and the pump and another grounded appliance connected to the autotransformer...

 

[hutchphd]

I got interested: see page 18:
http://cdn.powerequipment.honda.com/pe/pdf/manuals/00X31Z446010.pdf

Looks like it floats

 

[Averagesupernova]

Looks like it floats

I think a lot of them do. I have mixed feelings about that.

 

[anorlunda]

I think a lot of them do. I have mixed feelings about that.

I understand your unease. But remember those things can be used in places like flying hot air balloons or on a boat where no actual Earth is nearby.

 

[DaveE]

I think a lot of them do. I have mixed feelings about that.

But it has a ground terminal (output grounds and case) and the user should use it. It must have double/reinforced insulation also, or it would never get safety certifications. But the "distribution system" it may be connected to should be grounded. The wiring in your house isn't designed for floating sources. OTOH, the wiring in your house should have it's own grounding too.

Does anybody know if this generator ties Neutral and Earth ground at its output?

Be careful when discussing this with others. There are two common interpretations:
1) Neutral is the electrical "midpoint" of the AC sources (sorry, I'm too lazy to really define this better).
2) Neutral is the grounded conductor.

IMO, it's best to speak of grounding separately. For example, there are configurations, like a true 3Φ delta secondary (which is stupid, BTW) that have to be grounded, so they ground one of the phases. While there is no physical neutral connection, there is a neutral point in a phasor diagram, that is sometimes a useful thing to discuss. That fictitious neutral can't be grounded.

 

[jimboz]

Currently (no pun intended) the well pump is ungrounded. I'm pretty sure the generator has a bonding screw that is in the grounded position. My thinking was to unbond the generator and hook to the house wiring which has the neutral bonded to ground.

 

[berkeman]

Currently (no pun intended) the well pump is ungrounded.

Interesting. So it must be double-insulated (between the AC Mains input and its metal bits), so that a single fault can't energize the metal enclosure? Can you post the model number? Does it have a UL number on its information label/plate?

 

[jimboz]

Sorry, don't know any of those details, it was here when I bought the place. Only two wires going to it and it's 220V

 

[DaveE]

it must be double-insulated

I think it would have to be for an underwater motor. There just isn't a great way to ground water; or it's already grounded, what with being in the ground. I think it would be essentially for reliability to keep the electrons inside the wires and the water out.

 

[jimboz]

Since it's in the ground and assuming submersed it should go to ground and trip the breaker through the skin of the pump. Don't know it that's code but seems like it would work. Additionally, the pump itself might have a ground terminal which the installer did not use... what I meant is, there is not a ground wire connected to the pump.

 

[Tom.G]

The main reason that end-use centralized power (the stuff you get in-home from the power company) has a physical Ground is for safety when there is leakage from the 16kV primary to the secondary.

The three main ways leakage occurs is by capacitive coupling or leakage from the primary, a transformer failure (when it goes up in a ball of flame), and a lightning strike on either the primary or secondary side.

Here in the U.S., the transformer housing, and presumably the core, is also grounded. If you see a transformer on a pole you will see a Ground wire running down the pole.

Since there are no higher voltages available to an independent and isolated power source, such as a local single-phase generator without aerial transmissions lines, there is little to no reason for an Earth Ground.

This is covered in the National Electrical Code; unfortunately my copy was loaned out and not returned, so I can't quote the applicable section.

The one exception I can think of is if you are in an isolated area where lightning strikes are expected to occur. Even a ground strike nearby can induce some pretty good voltages.

I once was visiting a friend at his desert cabin when a thunderstorm came thru. He had commercial power but there was a ground strike perhaps 100 feet away, after which the well pump stopped working. The induced surge destroyed the circuit breaker but not the pump. We figured we could do without Hot water better than with no water, so we repurposed the water heater breaker to the well pump; and stayed a few more days.

Cheers,
Tom

 

[Shane Kennedy]

Summary:: Can I use a multitap transformer to step up 120 to split phase 240? Transformer schematic attached

View attachment 278130

My plan is to power the secondary with my 120VAC generator and use the 240VAC primary connection configuration and use the H3 and H8 taps as the neutral.

There are readily available 240 -> 120V transformers available for using power tools on building sites. They are generally "autotransformers" in that the secondary is simply half way through the primary i.e. centre tap, no isolation. You CAN use any twin-primary transformer by just connecting the primaries in series, connect 240 to the "outer" ends, and take 120V from the one winding.
On the generator, there are two identical windings. A switch connects them either in parallel, or in series. With the switch connecting them in series, you can tap into the bridge that links one winding to the other and use that as 120V BUT, only at half the usual 120V power. If you are using 240V at the same time, then the total power will be reduced too, otherwise you will cook the winding that is providing your 120V.