I read the post regarding a generator synchronization at http://control.com/thread/1272536595 and I kindly need an answer:

I'm not a HW person but a SW one, and I hold a degree in EE Engineering.

The above post helped me to understand why the voltage difference between the generator and the grid is, in a way, meaningless regarding the current flow in any direction. That is when we talk about synchronous generators, with real motors and stators, and with a real inertia.

But,when we talk about inverters (solar, wind, etc.), and we want to sell the energy our PV systems provide, there is no motor/stator that produces the torque that causes the current to flow from the inverter into the grid, but still current does flow in that direction.

My question is how is this possible, unless there's a voltage difference between the inverter and the grid ( |V_inv| > |V_grid| ) ?
And if there is no voltage difference (synchronization fundamentals), what causes the current to flow into the grid from the inverter?

Please help me to understand this.

Sincerely,
Oryan

 

Oryan,

I've often had the very same thoughts when considering the differences between inverters and generators. I've done no World Wide Web searching on this topic, and I suspect there is some information available somewhere--possibly from inverter manufacturers. I quite often find that companies wishing to sell me something I can buy from several other companies will detail in their adverts and white papers why their products are different from others--and I will quite often learn a LOT about the products and the industry in general.

Write back when you learn more, hopefully you can share some URLs for interesting and informative sights you find!

Finally, the concepts I've presented are pretty basic in nature--just things that power plant operators will observe and experience. Certainly, there is EMF and Counter-EMF to be considered in any power balance, including rotating synchronous generators. However, measuring these is not typically done in a power plant, and trying to explain theoretical principles to people who can't observe them or visualize them can be very difficult. I try to just use measurements and observations that power plant operators can or may use every day, combined with meters and measurements they see every day.

I'm always trying to figure out how things work--many times before I have any idea about how they actually work. I try to use similar equipment and concepts I have encountered, and sometimes I find I was pretty close, if not spot on. Other times, I miss the mark completely--but I so enjoy the process of analyzing the situation and trying to figure out the process and then trying to "anticipate" how things work. It's very invigorating, and very satisfying--at times. I almost never make ass-umptions about how things work, and by that I mean that I never just assume something works or should work or doesn't work in some particular way. I always try to spend at least a few minutes thinking through what I know about the process, or the equipment, what I know--or can discern--about how things might work, and then try to form some reasoned presumptions and "hypothesize" about how something works. I have encountered <b><i>so many</b></i> people who just ass-u-me something will or will not do this or that and are quite adamant they are correct without every having given it a moment's thought. And, quite frankly, that percentage of people is increasing because most people have not learned to practice critical thinking skills--because they're too difficult. I hear that from parents and colleagues all the time, "Why do they make my child think so hard? Why do they want them to question things? Why can't they just memorize things until they can repeat it correctly? Isn't that enough?" And, I just cringe when I hear that, because, that's what separates creative people from non-creative people, and this world needs more creative people--LOTS more creative people who will think about things differently and improve things and invent new things. Not just build the same old things.

Anyway, think about it like this, maybe. You use a converter to charge a battery. To do so, the converter output voltage has to be a little higher than the battery voltage in order for current to flow from the charger converter to the battery. The larger the difference, the more current that flows; the smaller the difference, the less current that flows.

I would think that a good solar inverter would be smart enough to know how to respond to changing system (grid) voltages when trying to "export" power from the inverter. (If you've never taken voltage measurements at the receptacle in your home periodically throughout the day, the week, the month, and the year and recorded those readings--you would be very surprised to find out how much that voltage fluctuates!) And, that when the amount of DC being produced is not sufficient to develop the required AC differential that it would shut itself off.

But, that's just my five-minute analysis for this question today. Right or wrong, that's how I start my research and investigation of a new process or piece of equipment.

And, I want you consider one other very important--and oft-overlooked--aspect of solar inverters: Can they "produce" reactive current? And, if they can't, what is going to be the effect on grids as more and more solar power comes on-line in the coming years. VArs are an important part of "keeping the lights on"--and can't just be overlooked. The refrigerator in your home or office, the air conditioner in your home or office, the cooling air fan in your home or office--most electric motor-driven appliances and most water pumps and industrial motors are all induction motors, and they "require" VArs. So, what part--if any--do solar inverters play in "supplying" VArs to a system? Rotating synchronous generators play a very important part in keeping the light on in an AC power system (grid) by "supplying" VArs, and if they are increasing supplanted by solar where will they come from?

(And, as an added bonus--were you aware that many, if not most, wind turbine-generators are induction machines which "require" VArs to operate? Hmmm....)

 

In most inverters, some sort of controlled rectifier or switching device is used together with inductance. If you do some research into choke input filters, you will find that a full-wave rectifier followed by a large inductance will tend to maintain a steady current flow - if it is a single-phase full-wave rectifier, it will have an output voltage of 0.636 x Vp. The current will be essentially steady although the voltage waveform is quite lumpy. Any ripple in the current will peak about 90 deg after the voltage peak.

With a simple full-wave rectifier, the voltage applied to the inductance is two half-waves of the same polarity starting at 0 deg and continuing to 180 deg. If the incoming rectifier is not allowed to turn on as normal, but is triggered on under some sort of control, the conduction half-cycle will start later and continue until the other rectifier is triggered on. So a rectifier element may start to conduct when the angle is 30 deg, and continue to 210 deg. Even though the voltage applied to the rectifier anode is negative, it will still conduct as the back emf developed by the inductor makes the cathode more negative and maintains the necessary forward voltage across the rectifier. Single-phase is not useful as an inverter, but with multi-phase arrangements it is possible to delay the firing so that the average output voltage is negative so that the DC side effectively absorbs power and transfers it into the AC side.

You can get some really weird effects with rectifier-inductor or switching transistor-inductor combinations - look up voltage boosters for examples. But a simplified version is that the current flow is maintained by the inductor, and it produces an emf that is whatever it takes to maintain conduction in the circuit. So any mismatch between the voltage out of the inverting or switching element and the supply it is connected to is accounted for by the inductor back emf.

 

CSA & BruceD,

Thank you for your replies!

I did a LOT of research through the internet, and read a whole lot of articles the past week. None is talking about how power (P nor Q) is injected by the inverter into the grid.

However, I asked Phil Corso via email and he sent me for this post: http://control.com/thread/1419896241 and I read all the comments.
He says there, in simple words:

- Power (P) is transmitted if the two "Bus-Angles" differ
- VAr (Q) is transmitted if the two "Bus-Voltage" magnitudes differ

These can be managed by a processor with the correct algorithm.

Now here is another question I couldn't find an answer for:
What is the amount of Power and VArs that is usually transmitted into the grid? For Power, is it the maximum power (current) that the PV can produce (with minimum value monitoring) or is there a limit to how much current can flow into the grid?

And what about the VArs? My guess is that VArs is a function of P in order to maintain the correct angle (phi)... right?

Thanks,
Oryan

 

If you take a simple example - a single-phase AC supply from a DC source such as a car battery.

The output AC can be produced using a full-wave system with the positive side of the battery connected to opposing ends of a split transformer winding via switching devices, and the negative side connected to the transformer winding centre-tap.

Now if the switching devices are controlled from a PWM system, it is possible to develop a sine wave output from the transformer secondary. Switching device A is turned ON for a short time, and OFF for a longer time. On subsequent cycles, the ON time is increased and the OFF time is decreased, till A is ON for almost a full cycle(which can be at a high frequency of above 10 kHz). Then reduce the ON time and increase the OFF time. Once A is OFF for a full cycle, repeat the sequence with device B and you have generated a full cycle.

Synchronising can be done by starting the switching sequence at the appropriate part of the other voltage cycle. This is done for instance in a UPS where there is an inverter-generated supply and a bypass supply derived directly from the mains. (This can give rise to some interesting problems. For example, I was once involved with an inverter system which maintained a 50 Hz supply if the mains supply frequency was more than 2.5 % out due to disturbances. When the mains frequency recovered, the synching system pulled the two cycles together again - but over a very short time (I think about 25 cycles). Under extreme cases, this resulted in an undervoltage trip of some connected equipment as the inverter used a ferro-resonant transformer and the rapid change in inverter phase effectively altered the frequency for a short time.)

The reactive power issue is an interesting one. Reactive power is needed because reactive devices store energy during part of the cycle and release it during the other part - a capacitor takes energy from the source while voltage is increasing, and loses it while voltage is decreasing, and an inductor absorbs energy while current is increasing and loses it when current decreases. With the right choice of switching device and control circuitry, it is possible for the inverter to return the reactive energy to the battery during the appropriate part of the cycle. Because the switching device is operating at a frequency much higher than the output AC, the VAr problem changes to one of dealing with reverse power flow through the controlled switching device for a short time in its switching cycle. A very simple example - by using diodes in conjunction with a switching transistor, it is possible to feed energy from a coil back to the battery as part of surge limiting.

At this level, AC power theory goes out the window and you need to start thinking in terms of switched DC electronics.