Hi Forum

would someone be able to explain and or point me to some good literature about the difference between an inverter that works with a transformer vs a switching transformer. are there any pros and cons? I realise that this might be a slightly controversial topic, but in the interest of learning i would really appreciate some help.

Many thanks in advance.

It's largely marketing. Transformerless is ostensibly "better". Mostly it is used as a synonym for "high frequency", and it is considered transformerless because there is no big heavy block of iron (or a toroid, which is the same thing, but in a round ring with a different core material) bolted to the back of it. Instead of having the big transformer, a smaller one is mounted on the PCB. You can get away with a smaller transformer because you're using a higher frequency.

But let me take a step back and get into a bit of theory. In power electronics there are two things you usually need to do: buck, and boost. Buck means to reduce the voltage, and boost means to increase the voltage. You want to do this as efficiently as possible, so that if you multiply the voltage and current on the input and output, you get more than 90% of the power back on the other side. In order to do this, just about every design on the planet will use a magnetic component of some kind. By converting the input energy into magnetic flux, and then converting it back to electrical energy, you can swap volts for amps.

A transformer is not the only kind of magnetic component. You can of course do this with plain inductors as well (this is how an MPPT works by the way). The transformer does however give you isolation between input and output, and that is actually a desirable thing.

Now... every battery inverter on the market, except maybe for some models with a high voltage battery, needs a boost stage. Therefore it will have magnetic components to get the job done, and it more than likely WILL have a transformer. There are some exceptions, eg some PV-inverters (SMA etc) might be truely transformerless, but they will still have magnetic components in order to buck/boost the voltage to the right level... they just do away with the isolation (it is more efficient, and cheaper, because you save all the copper/space of the secondary windings).

So, to get back to my opening statement: It is mostly marketing. The real difference is whether it's a low frequency or high frequency design.

So now let me get into the HF vs LF design difference.

Low Frequency designs work like this. You take your 48VDC, and you convert it to 48VAC at 50Hz (a little bit less really, there's some losses in this part). Then you feed this 50Hz low voltage into a big old conventional transformer and on the other side pops out 230VAC. The transformer needs to be big, because the time period t = 1/f is relatively long when f = 50Hz, so you need a nice big store of magnetic energy.

A high frequency design works similar, but it has an extra stage at the end. You again start with your 48VDC, and convert it to 48VAC... but at a MUCH higher frequency (typically 40Khz and above). This also doesn't have to be a sine wave. You then feed this into a transformer again, and convert it to a higher voltage, and then you rectify it back to DC, so that you end up with around 350VDC. This is the so-called high-voltage DC bus that we sometimes talk about, and there is a reason why it needs to be higher than the expected 230V.

You then have a final stage that takes this 350VDC and switches/slices it into a sine wave, and voila, you have 230VAC (RMS).

Because your frequency is much higher, the time constant t = 1 / f is much smaller, and hence a smaller magnetic store is needed.

Also, why 350VDC? Because the 230VAC we are used to is actually an average, an "RMS" value. It's the equivalent DC voltage if you will. Visually, you could think of taking the peaks of the sine wave, slicing them off, and dumping them into the valleys, and it will then level out at 230VDC. The peaks of the sine wave is actually around 325V... and this is why the high voltage DC bus must be at a higher voltage.

OK kids... class dismissed. If I got something wrong, there will be a teacher along to correct me shortly 🙂

 

Thanks plonkster! I’ve been challenged for installing an inverter with an “old technology” transformer, and since I’m not an electrical engineer, I was not quite sure how to answer. There must be some reason why this particular brand uses the transformer - I just don’t know why. I’m happy with my install and don’t plan on changing it. I did read somewhere in my electronic tinkering that a bench power supply that uses a transformer gives a much cleaner signal vs a switching one. Does that relate or is relevant to solar inverters too?

Many thanks again

2 hours ago, plonkster said:

There are some exceptions, eg some PV-inverters (SMA etc) might be truely transformerless, but they will still have magnetic components in order to buck/boost the voltage to the right level... they just do away with the isolation (it is more efficient, and cheaper, because you save all the copper/space of the secondary windings).

There must also be the claim that 'transformerless' is more efficient? (i.e. no losses primary/secondary windings)

1 hour ago, jasonvanwyk said:

I did read somewhere in my electronic tinkering that a bench power supply that uses a transformer gives a much cleaner signal vs a switching one. Does that relate or is relevant to solar inverters too?

People who sell low-frequency inverters will tell you they are much better at starting large loads. Though that is somewhat true, modern HF inverters are much better than they used to be, so it isn't as "obviously true" anymore. It will depend on the brand.

Re clean signal, again it will depend on the brand, but the transformer definitely has a bit of a filtering effect and should get you a slightly cleaner sine wave. Neither inverter will get close to the signal of a good grid connection or a large Diesel generator though.

HF inverters are typically more efficient, because the boost stage is much more efficient (93% to 95%). It's typically as good as the best switch mode power supplies (because it is a switch mode power supply). They are more complex though.

HF inverters can be made cheaper (smaller transformer). LF inverters typically cost more.

Here I might get a bit of flack though, but most of the top inverters (re warranty, support, track record) are LF designs. If you need robustness, go LF.

And with all that said, many makers of LF designs are branching into HF too.

Thanks again Plonk - this definitely helps me understand it a bit more.

1 hour ago, plonkster said:

And with all that said, many makers of LF designs are branching into HF too.

I see.. the Smart RS 6000, looks compact

1 hour ago, plonkster said:

And with all that said, many makers of LF designs are branching into HF too.

I can't find any pricing/specs on the WWW, about the RS smart 6000.

I did see that it is not parallel capable "yet".

I would be quite interested in this if it became parallel capable.

10 minutes ago, phil.g00 said:

I can't find any pricing/specs on the WWW, about the RS smart 6000.

It was introduced at InterSolar last year. I don't think it is in full production yet. An inverter/charger option will likely follow later.

7 hours ago, jasonvanwyk said:

 I did read somewhere in my electronic tinkering that a bench power supply that uses a transformer gives a much cleaner signal vs a switching one. Does that relate or is relevant to solar inverters too?

I doubt this. When a 50Hz transformer is used in a power supply (I presume a DC power supply) the bridge rectifier that rectifies the voltage on the secondary winding only produces a voltage pulse every every 10 ms. This is used to charge an electrolytic capacitor which has to be large enough to store the voltage until the next pulse.

Switched mode PSUs rectify the mains voltage and this produces a  DC rail of 320V approx. This is the supply to the dc-dc converter that reduces this voltage way down to the DC voltage required. As stated this happens at a high frequency so the charge pulses are arriving every x micro seconds. Also due to this speed the voltage regulation is also operating at this speed which allows for better voltage regulation.

Transformers do have their place. The substations that transmit power to you are all regular 50Hz transformers. They have the edge with being able to handle power.

Small laminated core transformers that we used to see have vanished having been replaced by switched mode PSUs.

However every now and then you come across a beauty like this one: I salvaged it from a wrecked (having been 'repaired' ) old school battery charger. I'm sure these are destined to become a collectors prized item!

Edited February 18, 20205 yr by Richard Mackay

2 hours ago, Richard Mackay said:

Transformers do have their place. The substations that transmit power to you are all regular 50Hz transformers. They have the edge with being able to handle power.

Small laminated core transformers that we used to see have vanished having been replaced by switched mode PSUs.

However every now and then you come across a beauty like this one: I salvaged it from a wrecked (having been 'repaired' ) old school battery charger. I'm sure these are destined to become a collectors prized item!

@Richard Mackay that still small. If you get the opportunity to check inside n MLT Powerstar inverter, then you will see a really big one. I wish you guys can see it. Pictures can't do it justes. The complete unit weight is 82kg where I think 70 kg is just transformer in side. 

3 hours ago, Richard Mackay said:

a beauty like this one

Here is another one from my old Multiplus Compact. the 3KVA models had two, one on either side. The Multiplus-II has only one (the main reason for the lower no-load consumption).

Hi @jasonvanwyk

just to give you some more info on the differences between HF and LF inverter, this is how HF (aka transformer-less) inverter creates 50Hz sinewave:

• The IGBTs in a full-bridge configuration are producing ON/OFF PWM signal.
• Since there are just 2 states (ON or OFF), there's a very little resistance to the current flow on the semiconductors themselves. Therefore, even for several kilowatts of switched power, there's not much heat generated.
• The 50Hz sinewave peaks, valleys and edges are being reconstructed based on the variable width of pulses, as you can see in the picture.
• Since PWM operates at roughly 22kHz of frequency, the produced sinewave is very smooth. Each digital pulse is precisely generated by the CPU and there's a loopback too, so the CPU can adjust width of the pulses based on the actual AC load.

For example, just compare this PWM approach with a conventional audio amplifier based on the transistors, producing analog 50Hz sinewave of 10kW: The transistors would be super-hot and smoking.

Since there's not much heat wasted and you don't have to magnetize a huge LF transformer, the HF inverters are very energy efficient, have low self-consumption and some of them even have no fans at all, therefore are super-silent.

Also, in the HF architecture, it's very easy to implement a low-consumption eco-mode:

• Basically, PWM will omit majority of pulses and will generate a waveform with narrow peaks, just to "probe" the connected circuit for any connected loads.
• Once the loads are connected and detected, the CPU will start to produce full PWM, which will turn into a pure sinewave.

 

On the other hand, since CPU is switching at the rapid frequency of 22kHz, if one of the IGBTs will fail, everything will blow-up in a second without a warning. Also, not every HF machine is really built for effeciency. For example, mine InfiniSolars are HF, but they have the self-consumption higher that a typical LF inverter.

Anyway, HF inverters are more and more common and almost every modern 3-phase hybrid inverter is internally a HF machine.
The next step in technology is the capacitor-less inverter. A machine that has no electrolytic capacitors inside, therefore is much more reliable in the long run.

Edited February 18, 20205 yr by Youda

Here is some images of mine. 

Front tronsformer:

And in the rear a monster of a Toroidal Transformer: 

 

Edited February 19, 20205 yr by Gerlach

The new MLT units. 

7 hours ago, Youda said:

just to give you some more info on the differences between HF and LF inverter

@Youda Wow! thank you! This is really great. So my next question, then, would be why do manufacturers continue making LF inverters? Is there still a place for them? What are there Pros or claim to fame?

Many thanks

56 minutes ago, Gerlach said:

Here is some images of mine. 

wow - some serious transformers in that machine!

I reckon this will be like the great audio debates:

Valves vs transistors.

Class A vs class D

8 hours ago, Youda said:

just to give you some more info on the differences between HF and LF inverter, this is how HF (aka transformer-less) inverter creates 50Hz sinewave:

The same technique is used to create a sine wave in LF inverters. Just on the other side of the boost stage (aka transformer). In the Multi, the PWM frequency is 20khz. With a good scope you can even see the steps in the waveform.

Johannes also explains it here.

38 minutes ago, jasonvanwyk said:

What are there Pros or claim to fame?

Well, just like @Crankshaft said, as of today, it's a similar debate like in the HiFi audio world...

IMHO, the @plonkster's comparison of HF vs LF in the above post listed all the main differences and benefits of each design.

A similarly-sized LF inverter will be much better at starting spinning loads, like pumps, fridges, etc. And you can overload it for couple of minutes without killing it. A huge transformer acts like a sort of "flywheel". But this largely depend on brand name and the quality of the build. As many chinese LF inverters are formally rated around 10.000W, while technically their guts are sized for 5.000W or less. And the makers are exploiting the overload capability of the LF architecture. This is the main reason why there's so many cheap LF inverters available on the market.

So, this is something that I would be really careful: what's the REAL power capacity of the machine.

 

As semiconductors evolved over the years, we are now able to build an efficient LF inverter, which combines great reliability together with a low energy consumption in idle mode. Therefore you can see that many manufacturers are sticking to modern versions of LF for their high-end product lines. Especially when a superb reliability is needed and the price is not the main decision factor. A typical use-case for such a modern LF is a sailing yacht or commercial vessel, as you definitelly don't want to look for a repair center in middle of the Pacific ocean.

For the on-grid solar, the HF are used mostly. Because they are efficient, cheap, small, lightweight and the flywheel for starting large loads is provided by the power distribution grid.

7 minutes ago, Youda said:

's comparison of HF vs LF in the above post listed all the main differences and benefits of each design.

There are two more reasons for favouring LF:

1. It is much easier to AC-couple PV to it.

2. It is much easier to put units in parallel. You're essentially wiring identical transformers in parallel, a well known and easy to balance setup.

These two might be the very reasons why the Goodwe cannot parallel or AC-couple with a PV inverter. It has a high frequency design.

There is one other "trick" that is available with an HF design: You can inject your PV on the high voltage DC bus. So instead of bucking the PV voltage down to battery voltage, you instead buck/boost it to the 350VDC high voltage rail. The PV therefore skips the battery boost stage and you get an increase in efficiency for direct use, but a decrease in efficiency for charging batteries. Again, this is how the Goodwe does it, which is why it has a higher voltage PV input.

13 minutes ago, Youda said:

A similarly-sized LF inverter will be much better at starting spinning loads, like pumps, fridges, etc

I remember Goodwe had something about certain loads is not recommend.. 

6 minutes ago, plonkster said:

You can inject your PV on the high voltage DC bus

And while on this line of reasoning... that is how the Axpert King does its "double conversion" thing too. It doesn't really double convert, at least not all the way down to battery voltage and back up again. They use an HF design (just like in their other inverters), but the incoming AC is rectified and pushed directly onto the high voltage DC bus, and then chopped back into AC. The buck/boost stage from the battery can add power to the HV bus (to assist), or take power from the HV bus (to charge). If the grid fails... the battery buck/boost powers it all.

The downside to that topology is efficiency (a bit of loss because of the rectification and switching), and because all the power has to pass through the final stage of the inverter, you are limited to the size of those switches (usually IGBTs).

Edited February 19, 20205 yr by plonkster