Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

how would I synchronized the frequency and voltage of the grid to my solar inverter?

Status
Not open for further replies.

dann11

Full Member level 3
Joined
Oct 22, 2015
Messages
166
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
2,742
I wanted to make a solar inverter, but the problem i have is, i dont know how to synchronized my inverters frequency and voltage to the grid. Is there any circuit configuration that will help me to do that? thanks..
 

you need a license to do this too.

PLL for frequency and phase lock with suitable filters
current limiter
lightning transient and thermal protection circuits
6kV isolation for safety
real-time power monitor
 
  • Like
Reactions: dann11

    dann11

    Points: 2
    Helpful Answer Positive Rating
thanks for the reply.. I just have a few questions for you, if you dont mind..

with the PLL circuit, will it have help me control any changes in the grid frequency? what are the possible causes of frequency variation across the inverters output? and how about the output voltage of my inverter? will a voltage sense circuit enough for me to control any changes on it?

thanks again..
 

I presume there's pretty much literature about grid-tied inverters.

A PLL to track the grid frequency is a good idea. A grid-tied inverter is usually controlled to feed the available input power to the grid. A straightforward way is to operate it as sinusoidal current source phase locked to the grid voltage.
 
  • Like
Reactions: DanyR

    DanyR

    Points: 2
    Helpful Answer Positive Rating
thanks for the reply.. I just have a few questions for you, if you dont mind..

with the PLL circuit, will it have help me control any changes in the grid frequency? what are the possible causes of frequency variation across the inverters output? and how about the output voltage of my inverter? will a voltage sense circuit enough for me to control any changes on it?

thanks again..

Keep in mind an impedance transformation occurs with your conversion, whether it is current sourced or voltage sourced, for efficiency the source impedance must be kept as low as economically possible.

There are requirements for THD and power factor that need to be regulated by design.

Technology has improved 6 generations in IGBT's and MOSFET's and state-of-the-art designs now use Silicon Carbide switches. Cost demands drive pressure on these exotic parts towards a solution that is $10/kW in volume production, but R&D costs are high and we aren't there yet. Commercial units are around $100/kW cost to make.

There was a million dollar competition last year for a >1kW inverter. I wonder how they made out? The requirement was only at least 50 Watts per cubic inch, but best design wins ( with no cost limit :) . look for "the little box challenge"

Widespread use of micro-inverters requires communication from the grid for grid-array control for power factor , phase and stability control. As grid control involves phase and frequency shift to compensate for massive rotating machine loads with poor power factor, they are no long always kept in perfect sync with coordinated universal time clocks every cycle, yet averaged frequency and regional phase are kept in a tight range for grid control. A central authority regulates these parameters.

For you
One would treat the grid as a short circuit load to a sine function, compared to your source and use that as the voltage and phase reference into two regulators. One for power limit and one for phase control of your controlled oscillator with a phase detector (PLL) and some frequency reference that can be tuned by analog or digital methods. Other regulators you need compare power available to output and charge control to state-of-charge, by CC,CV,float methods. Feedback to control includes overtemp protection (OTP), OCP,OVP , line fault detection.

- - - Updated - - -

https://googleresearch.blogspot.ca/2014/12/little-box-challenge-academic-awards.html
 
Last edited:
I presume there's pretty much literature about grid-tied inverters.

A PLL to track the grid frequency is a good idea. A grid-tied inverter is usually controlled to feed the available input power to the grid. A straightforward way is to operate it as sinusoidal current source phase locked to the grid voltage.

thanks.. but i am kinda new to grid tie inverter design, does it mean? that i need a feedback from the grid where current must be sensed and then the current sensed will serve as the input to the PLL? where exactly should i put the PLL's output? should i connect the PLL's output to the PWM circuit?

thanks for the help..
 

For you
One would treat the grid as a short circuit load to a sine function, compared to your source and use that as the voltage and phase reference into two regulators. One for power limit and one for phase control of your controlled oscillator with a phase detector (PLL) and some frequency reference that can be tuned by analog or digital methods. Other regulators you need compare power available to output and charge control to state-of-charge, by CC,CV,float methods. Feedback to control includes overtemp protection (OTP), OCP,OVP , line fault detection.

- - - Updated - - -

https://googleresearch.blogspot.ca/2014/12/little-box-challenge-academic-awards.html

thanks.. but does these mean, that I need a separate circuit to control my output's phase and voltage? thanks..
 

Hi,

The assumptions in both your latest post are right.

There may be other solutions, but usually you have to adjust frequency and voltage to the grid.
Else there soon will be overcurrent, because you are fighting against the grid power.. and you will loose

That is the main difference to asynchronous generator machines, where you don't need all this.

Before connected to the grid you regulate voltage, after connection you can not change voltage anymore, if you try you will change phase shift (power factor).
You may need a phase shift regulation loop also.

Klaus
 
i need a feedback from the grid where current must be sensed and then the current sensed will serve as the input to the PLL? where exactly should i put the PLL's output? should i connect the PLL's output to the PWM circuit?
Doesn't make much sense.

The inverter has to be synchronized to the grid voltage, respectively the PLL must lock to the voltage.

The grid tied inverter should feed only real power to the grid, not feed or "consume" reactive power or harmonic power. A sinuisodal inverter input current, 180 degree phase shifted against the grid voltage corresponds to this operation.

A usual pwm inverter with voltage fed DC bus is primarly a voltage source, in so far the imagined current source has to be implemented by combined feedforward and feedback control of a voltage source. In a simplified view, you are controlling the voltage difference between grid and inverter output, which is the voltage across a coupling inductor so that the inductor current equals the intended inverter current.

A different control approach would be to treat the inverter like a synchronous generator with respective inductive impedance. In this scenario, you'll control the real power by varying the phase difference between generator e.m.f. and grid voltage and the reactant power by varying the e.m.f. magnitude.

Sketch a phasor diagram for better understanding.
 

Hi,

two modes:

1) as long as it is not connected you must synchronize (adjust PLL) with respect to mains voltage. (your voltage to grid voltage)

2) as soon as it is connected you should adjust PLL with current/phase shift (your voltage to your current)

Klaus
 

2) as soon as it is connected you should adjust PLL with current/phase shift (your voltage to your current)
Synchronization to the current means the inverter's power factor is undetermined and can't be controlled to near unity, as it would be required.
 

Hi,

Synchronization to the current means the inverter's power factor is undetermined and can't be controlled to near unity, as it would be required.
Maybe a misunderstanding. In my eyes the PLL frequency has to be controlled to get the desired phase angle (current to voltage).

In connected case:
One can not synchronize the inverter voltage to the mains voltage. Because both voltages IS the same, they are connected.
But one must ensure that inverter frequency needs to track with the mains frequency. I can´t think of another way as to keep track with the help of the current.

Klaus
 

If you control the inverter in a synchronous generator approach, the PLL would generate the inverter e.m.f., phase shifted against the grid voltage, with the phase shift setting the transferred power.

I think that the current source approach allows a more direct power control, but that's surely a matter of presonal preference and habit.
 

Hi,
This is a very interesting topic, I asked myself also this type of questions.

in principle the grid tied convertor is a very simple thing: only a voltage controlled current source (as said in one of the earlier posts).

To make the principle clear for myself I made a small LTSpice simulation:
Circuit:
GridTied_Circuit.jpg
V1 is the grid voltage, RLoad is the grid load and "InvertorCurrentSource" is ... well like the name says.
As you can see the latter delivers a current into the grid/load depending on the grid voltage. This means that the current will follow the grid with respect to voltage and phase (always "synchronised").
View attachment GridTied_Circuit.txt

Simulation results:
GridTied_Curves.jpg
As you can see here the load takes 4.5A peak, the convertor deliveres 3.5A peak and the grid only deliveres 1A peak.

Of course this is no electrical circuit. The voltage dependant current source is the main thing to develop, including all safery issues and power grid requirements...
One of the main safety problems can be: what will happen if the grid becomes disconnected (the invertor in this simulation uses it as "base", which is normally unless interrupted...)

My extra question: The commercially available grid tied convertors do they approach the sinusoidal current curve very well or do they provide a "modified" sinus (as it is called sometimes) which is no less than a "staircase" type of sinus?
 
Last edited:

Despite of control method details, I see that sine inverters provide a high power factor (e.g. > 0.95) and a good sine shape.

A "modified sine" can work for stand-alone (island mode) generators, but not for grid tied. Due to the low grid impedance, it would cause large harmonic currents.
 

Some time ago I dealt with a system aimed to sell extra power provided by a local inverter to the public mains. The system was operating with the phase advanced in 30 degrees, and the amplitude increased by 10%. It is probably the setpoint to make a system like that.
 

This thesis "proposes a low complexity grid synchronization method which extracts both the parallel component and the orthogonal component from the grid voltage while sufficiently filtering out grid distortions. The grid synchronizer is easy to implement and provides the inverter the capability of controlling the reactive power generation without the need for dq frame transformation."

This also include details on Notch filter required and Bode Plots of control system, with ref to IEEE-1547 on the required 5% Total Demand Distortion. This approach mitigates the grid transient disturbances propagating thru the inverter (found on primitive V sense only frame GTI's ) without the need for current sensing. It allows PF adjustment in case of peak supply limit voltage during low demand periods, so you can inject power by active PF control without overvoltage.

This is the IEEE version submitted

If you are interested in post graduate work in this field, my son-in-law is the Prof at U of T in this paper, would be happy to help. just tell him Tony sent you. I think this is the student who now works at Tesla in Cali.

Tony Stewart
 
Last edited:

The profound thesis is somehow control theory centered, in parts surprizing for "common engineer" like me. E.g. when it describes a kind of all-digital PLL in the grid synchronization block without ever referring to the well-known term. But the functionality matters.

I also contradict the conclusion that dq-transformation and control in dq coordinates can't be applied to single phase systems. The only difference is that you need other means to derive orthogonal components than simple vector arithmetic.
 

This thesis "proposes a low complexity grid synchronization method which extracts both the parallel component and the orthogonal component from the grid voltage while sufficiently filtering out grid distortions. The grid synchronizer is easy to implement and provides the inverter the capability of controlling the reactive power generation without the need for dq frame transformation."
Perhaps a stupid question: Why would a simple generator (here invertor) provide reactive power to the grid? I can understand it could be interesting to make the overall power factor better, but is this the task of an (again simple) invertor?
Personally I would think that if the invertor delivers its power to the grid with its delivered current in phase with the grid voltage (power factor of 1), and a minimum of harmonics, that would already be the main purpose realised.

One note:
when I learned to use an AC generator (a real AC generator machine, driven by a motor) they tought me that before coupling to the mains:
- I should synchronise the generator with the grid: voltage, frequency and phase the same,
- I should connect the generator to the grid, resulting in no power at all delivered by the generator,
- I should try to make the generator run faster (more power of the driving engine), to make the generator deliver power to the grid. In fact this is trying to give the grid a higher frequency, which is impossible, thus the generator speed stays the same, but it gives power to the grid now.

I do not know if the same principles are also valid for a solid state invertor?
 
Last edited:

Controlling the reactive power means zero it out in normal operation. The inverter is connected to the grid by a LCL filter, you can e.g. compensate the reactive capacitor current by a small inductive inverter current. In some situations it's also wanted that an inverter sources reactive power, please read about low voltage ride-through capability as a means to support the grid in case of low-voltage events. https://en.wikipedia.org/wiki/Low_voltage_ride_through

But generally you're correct, the control of grid-tied inverter is pursuing similar objectives as a synchronous generator control.
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top