Low THD split phase 240/120v high wattage inverters?

[FlyFishn]

All,

I was intrigued by the Sigineer inverters for a long time - they had a 15kw for years then they made an upgrade with an 18kw. However, I got to looking at them and the THD on them is pretty high - I want to say up close to 20%?

Are there any similar inverters that have clean power? Preferably under 4-5% THD?

Or am I going to have to go to single 120v units at much lower wattage?

The idea was to be able to tie in to a conventional residential power system that is split phase - hopefully at high enough wattage to start/run an AC unit.

My low THD requirement is I have some electronics I want to be able to run off of said inverter that are sensitive to THD.

[Warpspeed]

Twenty percent THD sounds like a "modified sine wave" inverter, which is really just a square wave inverter.

THD on the grid might be in the 4% to 5% THD range these days, and you can equal or better that with a 27 step inverter waveform which is about 3% THD direct and unfiltered.  That would be the lowest cost and most robust way to do it at fairly higher power levels, and its perfectly adequate for most loads.  Although the voltage waveform looks a bit jagged, the current in any inductive load is a surprisingly smooth sine wave.

That kind of waveform is generated by combining the outputs of thee separate square wave inverters directly. Its a kind of digital to analog conversion, where the combining is done at high power and full output voltage.

If you want to do better than that, high frequency pwm will get you down to 2% THD or less, but it would be a pretty expensive inverter at that kind of power level.

Where to buy one, I have no idea.
I always build my own.

[joestue]

the switching losses are so low these days, there is no longer much reason to use complicated multilevel converters.. except once you exceed 480 volts or somewhere in that range, perhaps higher now that there are 1700 volt rated igbts that are pretty good.

for example, you can use a buck or boost 3 phase autotransformer topology to make a 12 pulse rectifier or inverter, and almost meet your 5% THD specification.
but the transformer will need to be rated for minimum 22% of the load. and in addition you need the inter phase reactors.

the expense and losses of the transformer and the interphase reactors would exceed the losses of a simple controlled rectifier, medium frequency pwm and 3 small high frequency inductors.
as for OP..

here you go
https://www.amazon.com/Sigineer-Power-Inverter-Charger-Frequency/dp/B093KZH31P?th=1

45kW, 20 seconds.

should be enough to start a 5 ton heat pump.

now that i think about it, if you want to mix square waves together using 3 or 4 transformers (which gets you a 27 or 81 level inverter).. the minimum kva of the transformers is about 1.3  Not actually that bad, and basically equivalent to the minimum kva needed for a 24 pulse transformer.

the question is: does 3 or 4 switches (they could be mosfets igbts or scr's ) in series actually have lower losses than a single switch driving a pwm duty cycle into a small inductor instead of 3 or 4 switches in series driving multiple transformer coils. -- almost certainly not.

another issue with the 3 or 4 transformer multi level inverter is that the transformers are varying sizes.. smaller transformer = less efficient.

[FlyFishn]

@joestue - the inverter you linked to is the 15kw Sigineer I've seen for years. It has some really nifty features that make it an excellent off-grid unit.

However, the THD is the killer in my application. I want to be able to run 240v inverter welders and my current tig/stick machine will absolutely not take the higher THD. AC and other household loads I am not as concerned with, though its always a good idea to keep the THD down.

@Warpspeed - homebrewing an inverter sounds like an interesting project. I like that idea - if I make it then I can fix it - and have backup parts on-hand, if not another unit there ready just-in-case. I'll have to do some digging on schematics and maybe some examples of others that have done just that and see what I come up with.

[Warpspeed]

the question is: does 3 or 4 switches (they could be mosfets igbts or scr's ) in series actually have lower losses than a single switch driving a pwm duty cycle into a small inductor instead of 3 or 4 switches in series driving multiple transformer coils. -- almost certainly not.

The switches are not all in series, the separate inverters run in parallel across the incoming dc.  Only the transformer secondaries are in series. A very important distinction.

You are quite right though, about the total kVA rating of three or four transformers being excessive, but the combined magnetizing current need be no higher than a single transformer using high frequency pwm. So no load idling power is very comparable for both methods.

My own four transformer 5kVA inverter has a zero load idling power of 34 watts, and less than 1% measured THD.
Third harmonic is by far the largest and its down by -41db.

What you might be missing is that the four inverters diminish drastically in size with each successive stage. Only the largest 50/60 Hz square wave inverter runs at significantly high power, and switching losses will be negligible.  Second stage only runs at a third of that, so additional losses of successive stage when combined are correspondingly less.

Conduction losses are pretty comparable, pwm requires a series choke, but a hard switched square wave inverter does not.

Multistep only has one very serious disadvantage, the cost of winding all those transformers. Technically its more than competitive for idling power, efficiency, and THD with high frequency pulse width modulation.
It gains most in advantage as the power level risess, but loses out in all the extra components required at low power levels.

Its always been a commercial failure, but only because nobody could manufacture multistep and be price competitive in the power ranges we are mostly interested in around here.

But for home construction its a winner if you use recycled material for your transformers, and the required power level is perhaps 5kW or more.   This is not armchair theory, many of these have now been built and are all running successfully without any serious problems. 

On the other hand many novice home constructors have had continuing ongoing blow ups with high power home built inverter projects. Nothing wrong with pwm in principle, but it all becomes a lot more critical in layout and details when trying to drive a large number of parallel mosfets at 20 Khz plus.

[Warpspeed]

@Warpspeed - homebrewing an inverter sounds like an interesting project. I like that idea - if I make it then I can fix it - and have backup parts on-hand, if not another unit there ready just-in-case. I'll have to do some digging on schematics and maybe some examples of others that have done just that and see what I come up with.

My Warpverter design is public domain and all the details are available to anyone that is interested.
Here is the complete overall wiring diagram:

On the left is the main control board that just connects to the incoming battery supply, for internal power and to measure battery voltage.  Note there is no voltage feedback required from the final ac output.

There are four inverters consisting of eight half bridge gate driver boards. These are all optically isolated so that a blowup cannot propagate back into the control board.  These just plug into the driver board and receive the appropriate drive waveforms, all eight driver boards are identical and interchangeable.

Each half bridge inverter uses either single individual IGBTs or mosfets for the smaller two inverters, or  high power IGBT half bridge power block for the larger two inverters.  Everything is modular and just plugs in, it can be repaired without requiring a soldering iron in very short time.  The modularity and isolation also prevents a cascade failure, where one point of failure can rapidly lead to total destruction of just about everything. A common problem with pwm inverters. When they blow its usually total devastation.

[joestue]

the transformer windings are in series, which puts the switches on the battery side in series as well, as the current has to flow somewhere.


anyhow i wonder if you can augment the transformers with an  ac bi directional switch on the secondary to short out the transformer when the voltage is zero. they turn on after the switches turn off, and they turn off before the switches turn on. this reduces the "resistance" of the transformer during the off state.

[joestue]

that 15kw inverter says its 5% thd, is that a lie?

have you tried adding an LC choke to the output?

i would buy a line load reactor off ebay and then add 50uf of capacitors after the choke.

[Warpspeed]

the transformer windings are in series, which puts the switches on the battery side in series as well, as the current has to flow somewhere.


anyhow i wonder if you can augment the transformers with an  ac bi directional switch on the secondary to short out the transformer when the voltage is zero. they turn on after the switches turn off, and they turn off before the switches turn on. this reduces the "resistance" of the transformer during the off state.

If you have two separate off grid inverters connected to the same battery, each providing 120 volts.
Now lets assume they are phase locked and you connect the secondaries in series to get 240 volts.

Does that mean the mosfets in each inverter are now operating in series  ?
I don't think so.

Each transformer can generate three voltage steps on its secondary. Positive, negative and zero volts.
As the secondaries are in series, zero volts can only be achieved by shorting the primary, not by open circuiting the primary.
So turning off all four mosfets in a bridge does not get you zero volts.
What must be done is either the upper pair OR the lower pair of mosfets must turn on simultaneously to actively short out the primary winding.

Another thought experiment for you, something else you probably have not considered.
The combined secondary waveforms create an almost perfect low THD voltage waveform.
With a resistive, and even more so with an inductive load, the secondary current waveform will also be sinusoidal.

In any transformer the secondary current reflects back directly into the primary,allowing for turns ratio of course.
So the current in all of the windings of all four transformers will be a perfect unbroken 50/60Hz sinusoid.

The transformer voltages are however all rectangular waves on both primaries and secondaries.

Its a rather difficult thing to visualize, square wave voltages and sinusoidal current, but that is the way it is.

[Warpspeed]

Its mind numbing stuff trying to imagine how this all works, but it certainly does work, and very effectively too.

Another aspect that many people have great difficulty with is how can it possibly operate over a 2:1 dc input voltage range, provide very stiff output voltage regulation, without having any voltage feedback.....

Well consider this. 
Assume you have a suitably large transformer connected up to the grid and you measure the no load output voltage. 
You then switch a load onto the secondary and again measure the output voltage, and find its hardly changed.
No need for secondary voltage regulation, because the transformer is supplied from a voltage source that hardly changes. There will be a slight voltage drop of course, but for practical purposes we can ignore that.

So if we have an inverter and we can insure that the ac drive voltage on the primary is pretty much fixed, we don't really need to regulate the secondary voltage. It should be fine without any further correction being necessary.

So how can we rigidly fix the primary voltage when our incoming dc is all over the place ?
Easy, measure the incoming dc voltage and use that to control the output amplitude of whatever drives the primary.
If the incoming voltage falls by a certain percentage, we just increase the drive amplitude by that exact same amount to compensate.

Its dead easy to do either with pwm or this multistep method.  Measure the dc voltage with an analog to digital converter, then use that to select an appropriate lookup table that contains data to generate our inverter drive waveform.
Its very fast, because a dc voltage can be measured in an instant and amplitude correction applied.

An ac voltage is always changing sinusoidaly. Its difficult to measure because it needs to be rectified and the rectifier output averaged.  That is slow.  Feedback must also be applied gradually with a PID control loop or it can over correct and become unstable.

Feedforwad is super fast and cannot ever become unstable, Feedback is slow, can be unstable, but at least long term it will settle at an exact output voltage. 

To correct the only disadvantage of this, the slight voltage sag under load, current feed forward can be added.
This works similar to correcting the voltage, except we measure the dc input current to the inverter with a Hall current sensor.

The measured dc input current can provide just that little bit of additional amplitude to compensate for the voltage drop under load due to conduction losses in the inverter.  In fact its possible to tweak it so the inverter output voltage actually rises slightly with increasing load.  That too is completely stable and operates very quickly.

The result is very good output voltage regulation and minimal light flicker with massive surge loads on the inverter.
And all without requiring any voltage feedback.

Some people have great difficulty with this whole concept, but it does actually work really well.