Split Phase Output From Full Bridge Inverter

[Wrydog]

How is a split phase output on a full-bridge inverter set-up?

I'm trying to accomplish having 240Vrms and 120Vrms outputs on an inverter

I have currently have a Psim simulation that has two capacitors connected in series after the LC filters of the upper and lower legs (see attached pic). For a load, I have two resistors. With both resistors, the simulated output is fine with Vout1 measuring 240Vrms and Vout & Vout0 both measuring 120Vrms. However, when I disable the lower resistor (simulating a resistive load plugged into the 120V receptacle), the voltage drops to nearly zero.

Thinking it might be simply a control issue, I added a DC offset to the amplitude of the modulating signal to try to boost the voltage of the upper leg across the resistor but didn't have any success. It this just the wrong circuit to do this with?


Ideally, I'd like to have 240V and 120V appliances running simultaneously. Obviously control is a major part of this but I wan to be sure I have a circuit that will support that.

Any suggestions?

Edit: Included basic overall topology.

Thank you.

[drussell]

The easiest way would probably be to do it just like the commercial inverter products do it.  Tie two inverters together with linked oscillator and controls.  Very simple with most 120 volt inverters, you just link the second one in with opposite phase so each handles any imbalanced 120 volt load and the pair produces the full 240 for any load across them.

I have experience in actual use of the Trace (now Xantrex) DR series at a friend's cabin, very, very simple and straightforward....  That kind of set-up should be a breeze to implement, even DIY from scratch, it just makes sense.

The operating manuals etc. for those commercial inverters are readily available and I think I even pulled service data and schematics at one point.

[jbb]

Are you building an inverter from scratch here, or looking to make some existing solution work?

If you're extending an existing solution, a simple 60Hz transformer (while not especially efficient) could be very effective.

If building from scratch, there are a few options, each with pros and cons.  You'll almost certainly need some more semiconductors to get good performance.  Also, I suggest you show an outline of the whole inverter concept including the DCDC stage and H bridge.

[David Hess]

I have no idea how real inverters do it.  Two floating outputs that are in phase could be stacked but I assume real split phase inverters do something more sophisticated.

My solution was to build a phase splitter from some big surplus 60 Hz power transformers that I had.  See the photograph below.

If you're extending an existing solution, a simple 60Hz transformer (while not especially efficient) could be very effective.

The efficiency of a transformer based phase splitter is just fine and better than an inverter.  It is just heavy.

[Zero999]

To get 240V you're using a full bridge with an output filter either side?

Can't you just make a neutral in the middle by adding some large capacitors to either side of the DC bus? Obviously the capacitors need to have a low enough impedance and high enough current rating to pass the desired current, without changing the voltage much. That way, you'll be effectively using a half bridge if you just connect a large load between one of the phases and neutral.

[retrolefty]

Why not just use a centered tapped output transformer, that's how my home is supplied from a street transformer. Center tap becomes the neutral, secondary ends become L1 and L2, ground the neutral at it's service panel?

 

[Zero999]

Why not just use a centered tapped output transformer, that's how my home is supplied from a street transformer. Center tap becomes the neutral, secondary ends become L1 and L2, ground the neutral at it's service panel?

Because there's no mains frequency transformer?

Battery -> DC:DC converter, giving 340VDC -> H-bridge, giving PWM, -> Low pass filter, giving 240VAC pure sine wave.

[T3sl4co1l]

If you have bipolar supplies, then an H bridge gives "+/-" 120V with 240V between the two.

Tim

[David Hess]

Why not just use a centered tapped output transformer, that's how my home is supplied from a street transformer. Center tap becomes the neutral, secondary ends become L1 and L2, ground the neutral at it's service panel?

Inverters with low frequency output transformers do this and easily support split phase operation.  Or you can add an external low frequency transformer to any inverter to do the same thing which is what my photograph shows.

I thought Wrydog was asking about how to do it in a design which lacks low frequency transformers.

[Wrydog]

From scratch.

[Wrydog]

This is for a battery supplied inverter. Has to be comparatively lightweight.

[Wrydog]

Yes. My problem stems from the "what if only one 120V load is plugged in as a load?" My simulation shows nice waveforms with two balanced resistors, but that stops when only one is active.

[Ian.M]

You'd have to actively maintain the neutral point at the mean voltage of the two phases.  You cannot assume that the neutral voltage is half the DC bus voltage unless your inverter drive is fully symmetrical.. Use a potential divider across the two Line outputs to get the neutral reference voltage, and add another leg to the 'H' bridge PWMed in a servo loop to maintain the load neutral at the reference neutral voltage.

[Benta]

If I understand your setup correctly, you have a full-bridge PWM-driving a high-frequency tranformer to provide a sinusoidal output at 60 Hz.
Is that right?
If so, having two independent secondaries on said transformer, each delivering 120 Vrms solves the problem. After filtering, they can be series connected to 240 Vrms centre-tapped.

[Wrydog]

Thanks! This seems to be what I need. Can you explain further or point me to some more references?

[Zero999]

Yes. My problem stems from the "what if only one 120V load is plugged in as a load?" My simulation shows nice waveforms with two balanced resistors, but that stops when only one is active.

That's what you should expect to happen with your current schematic. The impedance of the neutral point is equal to that of C33 and C34 in parallel. C33 & C34 in parallel = 4.4µF, so Z = 1/(2pi×FC) = 1/(2pi×60×4.4×10^-6) 603R.

If I understand your setup correctly, you have a full-bridge PWM-driving a high-frequency tranformer to provide a sinusoidal output at 60 Hz.
Is that right?
If so, having two independent secondaries on said transformer, each delivering 120 Vrms solves the problem. After filtering, they can be series connected to 240 Vrms centre-tapped.

Thanks! This seems to be what I need. Can you explain further or point me to some more references?

All right, there's nothing wrong with that set up. It will work but then you say:

This is for a battery supplied inverter. Has to be comparatively lightweight.

How lightweight does it need to be?

What power rating are you after?

I hope you know that the transformer will need to be capable of passing 60Hz, which means it needs to be an ordinary mains transformer and those aren't lightweight.

The only way to make it more lightweight is to add a DC-DC converter, to boost the voltage to the peak mains voltage, before the h-bridge.

[Benta]

The best I could find at short notice is this:
https://www.onsemi.com/pub/Collateral/SMPSRM-D.PDF
page 12, Fig. 12.
Ignore the bridge rectifier etc., but imagine that the secondary is only filtered. Add an additional secondary with filtering, and you have your two 120 V independent supplies.

I hope you know that the transformer will need to be capable of passing 60Hz, which means it needs to be an ordinary mains transformer and those aren't lightweight.

 

Please explain. A full-bridge with a "12 V primary" and a "120 V secondary" is no problem (quotation sign, because with a switching supply the voltages are a bit different)

The only way to make it more lightweight is to add a DC-DC converter, to boost the voltage to the peak mains voltage, before the h-bridge.

 

Again, why? Running an H-bridge directly from a battery to generate a higher voltage is no magic.

You're confusing me.

[Zero999]

The best I could find at short notice is this:
https://www.onsemi.com/pub/Collateral/SMPSRM-D.PDF
page 12, Fig. 12.
Ignore the bridge rectifier etc., but imagine that the secondary is only filtered. Add an additional secondary with filtering, and you have your two 120 V independent supplies.

Good. That looks like an excellent suggestion for the DC:DC converter part of the inverter, to generate the high voltage DC bus.

I hope you know that the transformer will need to be capable of passing 60Hz, which means it needs to be an ordinary mains transformer and those aren't lightweight.

 

Please explain. A full-bridge with a "12 V primary" and a "120 V secondary" is no problem (quotation sign, because with a switching supply the voltages are a bit different)

That's no problem at all.

The only way to make it more lightweight is to add a DC-DC converter, to boost the voltage to the peak mains voltage, before the h-bridge.

 

Again, why? Running an H-bridge directly from a battery to generate a higher voltage is no magic.

You're confusing me.

I was under the impression that the idea was to use an h-bridge to convert the DC voltage from that battery to PWM AC and step it up to the mains voltage with a transformer. That will work the transformer still needs to be able to pass 60Hz.



Perhaps I misunderstood what you meant? No schematic has been posted of the actual inverter.

[Wrydog]

Yes. My problem stems from the "what if only one 120V load is plugged in as a load?" My simulation shows nice waveforms with two balanced resistors, but that stops when only one is active.

That's what you should expect to happen with your current schematic. The impedance of the neutral point is equal to that of C33 and C34 in parallel. C33 & C34 in parallel = 4.4µF, so Z = 1/(2pi×FC) = 1/(2pi×60×4.4×10^-6) 603R.

If I understand your setup correctly, you have a full-bridge PWM-driving a high-frequency tranformer to provide a sinusoidal output at 60 Hz.
Is that right?
If so, having two independent secondaries on said transformer, each delivering 120 Vrms solves the problem. After filtering, they can be series connected to 240 Vrms centre-tapped.

Thanks! This seems to be what I need. Can you explain further or point me to some more references?

All right, there's nothing wrong with that set up. It will work but then you say:

This is for a battery supplied inverter. Has to be comparatively lightweight.

How lightweight does it need to be?

What power rating are you after?

I hope you know that the transformer will need to be capable of passing 60Hz, which means it needs to be an ordinary mains transformer and those aren't lightweight.

The only way to make it more lightweight is to add a DC-DC converter, to boost the voltage to the peak mains voltage, before the h-bridge.

Sorry, I had meant to reply to IanM saying that was what I was looking for. There is a high frequency full-bridge DC-DC stage on the front end of the inverter to boost the inverter's input voltage in order to negate the need for a  60Hz transformer. I was looking for more references to control strategies for the neutral point.

[Benta]

Hero999, first a big Thank You for providing much better graphics than what I found.

I was under the impression that the idea was to use an h-bridge to convert the DC voltage from that battery to PWM AC and step it up to the mains voltage with a transformer. That will work the transformer still needs to be able to pass 60Hz.

Yes, but why include the extra 60 Hz transformer? Just step it up directly to 2 x 120 VAC with two separate secondaries and filters. Then the outputs can be used freely.

[Benta]

Sorry, I had meant to reply to IanM saying that was what I was looking for. There is a high frequency DC-DC stage on the front end of the inverter in order to negate the need for a  60Hz transformer. I was looking for more references to control strategies for the neutral point.

This is still not clear, but you get back what you offer: a fractional schematic, unclear questions and, quite frankly, no real idea what you want.
The quality of the answers correspond to the quality of the input.

[Wrydog]

Yes. My problem stems from the "what if only one 120V load is plugged in as a load?" My simulation shows nice waveforms with two balanced resistors, but that stops when only one is active.

That's what you should expect to happen with your current schematic. The impedance of the neutral point is equal to that of C33 and C34 in parallel. C33 & C34 in parallel = 4.4µF, so Z = 1/(2pi×FC) = 1/(2pi×60×4.4×10^-6) 603R.

If I understand your setup correctly, you have a full-bridge PWM-driving a high-frequency tranformer to provide a sinusoidal output at 60 Hz.
Is that right?
If so, having two independent secondaries on said transformer, each delivering 120 Vrms solves the problem. After filtering, they can be series connected to 240 Vrms centre-tapped.

Thanks! This seems to be what I need. Can you explain further or point me to some more references?

All right, there's nothing wrong with that set up. It will work but then you say:

This is for a battery supplied inverter. Has to be comparatively lightweight.

How lightweight does it need to be?

What power rating are you after?

I hope you know that the transformer will need to be capable of passing 60Hz, which means it needs to be an ordinary mains transformer and those aren't lightweight.

The only way to make it more lightweight is to add a DC-DC converter, to boost the voltage to the peak mains voltage, before the h-bridge.

Thank you, the low combined impedance was a big factor in the simulated voltage drop. I think I'll have to redesign with higher value caps.

[Wrydog]

You'd have to actively maintain the neutral point at the mean voltage of the two phases.  You cannot assume that the neutral voltage is half the DC bus voltage unless your inverter drive is fully symmetrical.. Use a potential divider across the two Line outputs to get the neutral reference voltage, and add another leg to the 'H' bridge PWMed in a servo loop to maintain the load neutral at the reference neutral voltage.

Thanks Ian, this is helpful.

Can you explain further about adding a third leg or point me to other resources/search terms?

I've added a diagram of the basic overall scheme to the original post.

[Ian.M]

Another leg: another pair of upper and lower MOSFETs like a three phase H bridge, only instead of driving the phases to approximate sine waves  at 120 deg to each other after filtering, you drive two phases as a conventional H bridge output inverter and drive the third phase to maintain the neutral.

[T3sl4co1l]

Some variations on the output network.

Note that the supplies are already shown as bipolar, so I added a ground reference, implying the supplies are not just plus and minus of a single supply, but a dual, symmetrical supply.


Original


Two notes: 1. The transformer must carry the full low frequency content regardless.  2. LF transformers have high core losses at high frequencies.  Might as well move the filter in front.


The supplies and phases are complementary and symmetrical, so we can divide the LPF in half against ground (or, since the ground terminal is just capacitors, it doesn't care what DC is -- either supply would work as well, if you prefer).  This is identical if the phases are perfectly symmetric, so seems like a lot of component duplication, however in practice they will never perfectly match, and this provides equal common mode and differential mode filtering.  Huge EMI win!


Continuing the symmetry theme, the primary can be CT-grounded without problems*.

*Given that the flux walking constraint now applies to both outputs individually, not just their difference.  That is: DC output current must be very close to zero, for both outputs.

Also, as long as we're putting taps on things, the secondary can be tapped anywhere you like (not just half), getting the split phase that was asked about, or any other voltages for that matter.


Since it's symmetrical, we can even eliminate the transformer altogether -- assuming isolation is not required.

Now it is especially important for the LPFs to be independent, rather than balanced, because there's no potential filtering benefit from the transformer.

Tim