Grid tie inverter operation clarification

Hi,

I have to carry out a project for my studies and I choose to have a look into solar PV business in the Middle East.

While most things have now become clear to me, I would like to understand how a grid-tie inverter exactly works. From what I have seen and understood, inverters will be connected in parallel to the grid. If we consider the load to be higher than the energy produced by the solar system, the connected load should draw both from the grid and the inverter. If solar acts as a voltage source, it may cause problem since the two branch will have different potentials and there would be no way of ensuring that energy coming from solar will be totally consummed before using energy from the grid. My understanding is it works more like a current source but once again how do you ensure to draw the maximum possible from the solar system energy and not only a part of it.

So if anyone could enlight me on how an inverter works to ensure that the load will use the totality of the power coming from the solar panels and only the balance from the grid, it would be greatly appreciated.

Thanks in advance.

Comments

  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    Re: Grid tie inverter operation clarification

    The Grid tie inverters are the key here as the convert the DC voltage to match the grid AC voltage. They also sync the freq to match. The solar is supplied on the inside of the meter and if enough loads are present then all is consumed and any additional needed will come from the grid, otherwise the power excess flows to the grid. The meter must be able to run "Backwards" as in old mechanical meters or new smart meters to make a net metering plan work.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Grid tie inverter operation clarification
    Onitoto wrote: »
    My understanding is it works more like a current source but once again how do you ensure to draw the maximum possible from the solar system energy and not only a part of it.

    So if anyone could enlight me on how an inverter works to ensure that the load will use the totality of the power coming from the solar panels and only the balance from the grid, it would be greatly appreciated.

    Thanks in advance.

    Think of it as a tub with two faucets feeding it and one drain and an overflow.
    The drain is your varying load.
    One faucet is the power coming from POCO, and it has a float valve that only opens it when the level goes down. The other is the power coming from the grid tie inverter, and it stays wide open (the full output of the panels). The overflow is the sale back to POCO.

    Now start with a full tub.
    When you open the drain, it does not matter which faucet the water comes from that goes out the drain. The inverter faucet will keep putting out water while the POCO faucet will either supply water or the excess water will flow back out the overflow to POCO. Once the water is in the tub, it no longer matters where it came from.

    Sorry, I am addicted to analogies.
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Grid tie inverter operation clarification

    Specifically, related to Off Grid vs Grid Tied inverters...

    Off Grid Inverters are "voltage sources" -- Basically, think of an "AC Battery". The sine wave voltage source appears to have "near zero" resistance (impedance)--So no matter the load (up to maximum current limits), the output voltage of an OG Inverter is just the Sine Wave Voltage (~0-+/-170 Volts) and the current is a function of the load (V=IR for linear loads, spiky current for non-linear loads like DC power supplies without power factor correction--By the way V=IR are "vectors", so you get into Impedance "Z" or V=I*Z).

    Grid Tied Inverters are something entirely different. They are "Current Sources"--Basically, the inverter output current according to the point in the voltage cycle (I=V/R) and the amount of current/power available from the solar panels. And then I goes between 0% and 100% of available current from the array (cloudy weather, the Ipk may be 5 amps, full sun, Ipk may be 20 amps, etc.).

    You can make an inverter/generator be "synchronous" with the grid (i.e., sort of like paralleling two or more batteries)--and this is what your utility/power companies are doing... The rotational frequency of the generating stations (huge alternators to be precise) are "locked" together and rotating at the same RPM/phase for the entire western grid (as an example). To boost the power output of a single alternator, they apply more "torque" (steam/water pressure/etc.) to push the alternator harder--which forces it to output more current to the grid. If the alternator was out of phase/frequency when connected to the grid--very bad things can happen (over current, possibly even mechanical issues/broken mounting, etc.).

    Instead of going to the trouble of making a GT inverter "synchronous", they can instead make it a "current source" which simply follows the AC wave form. Much easier (and probably "less exciting" if something goes wrong). It simply outputs current based on the AC voltage wave form.

    Note, I do not design inverters--so this is just my limited understanding of what is done in the industry. There are many different ways to accomplish similar results--and I have not addressed those other methods here (and I may have messed up more than one analogy).

    -Bill "ain't pefect" B.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Grid tie inverter operation clarification
    Onitoto wrote: »
    Hi,

    I have to carry out a project for my studies and I choose to have a look into solar PV business in the Middle East.

    While most things have now become clear to me, I would like to understand how a grid-tie inverter exactly works. From what I have seen and understood, inverters will be connected in parallel to the grid. If we consider the load to be higher than the energy produced by the solar system, the connected load should draw both from the grid and the inverter. If solar acts as a voltage source, it may cause problem since the two branch will have different potentials and there would be no way of ensuring that energy coming from solar will be totally consummed before using energy from the grid. My understanding is it works more like a current source but once again how do you ensure to draw the maximum possible from the solar system energy and not only a part of it.

    So if anyone could enlight me on how an inverter works to ensure that the load will use the totality of the power coming from the solar panels and only the balance from the grid, it would be greatly appreciated.

    Thanks in advance.
    Grid tied inverters are current sources. Think of the inverter as being separate from the loads. The inverter feeds X kWh's into the grid, and the loads draw Y kWh's from the grid. The loads don't know or care whether the kWh's they are using originally come from the inverter or the coal fired power plant (or whatever) out on the other end of the grid.
  • jaya
    jaya Registered Users Posts: 3
    Re: Grid tie inverter operation clarification
    ggunn wrote: »
    Grid tied inverters are current sources. Think of the inverter as being separate from the loads. The inverter feeds X kWh's into the grid, and the loads draw Y kWh's from the grid. The loads don't know or care whether the kWh's they are using originally come from the inverter or the coal fired power plant (or whatever) out on the other end of the grid.

    Hi,

    I landed on this thread while looking to try and clear the confusion in my mind about how a PV grid-tie inverter works, but after going thru the discussion, my questions persist, so I would really appreciate more clarity.

    When I read the analogy with the tub, I thought yes, everything's clear now, but then an instant later, the doubts came right back :( because in order for the tub analogy to work, there must be some storage/buffering, which is not the case. The point about the current source makes sense, until I try to understand how it might be implemented, and then I get to this:

    As I have understood how the GTI works:

    a. When instantaneous load ~= instantaneous PV output, GTI feeds all output to local load
    b. When instantaneous load < instantaneous PV output, GTI feeds excess into the grid
    c. When instantaneous load > instantaneous PV output, deficit from GTI flows from the grid to drive local load

    At the same time, isn't the GTI's ability to feed the grid, based on maintaining a voltage level that is slightly higher than the grid?

    If so,

    1. does the GTI maintain a higher-than-grid voltage only when it senses condition (c) above, or all the time?

    2. If the GTI always maintains a higher-than-grid voltage, then shouldn't all or most of the energy flow back into the grid, since grid impedance is lower than impedance of the local loads (or so I'm told)?

    Thanks!
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification

    Grid-tie inverters do not make any differentiation between household and grid. They only see an AC connection which has (hopefully) proper Voltage and frequency. When that is present they take any power available from the PV array and supply it to the AC mains. If the household uses it that's where it goes. If the GTI output is greater than household needs it goes to the grid, which in this case does not act as a supply but as a load. Inevitably this requires a small Voltage difference between the two, so the GTI output will be slightly elevated in respect to grid Voltage. It also uses this as a 'test'; if the Voltage is allowed to rise too high it assumes there is no grid present and shuts down. A similar test for correct frequency is also made.

    The function is very similar to panels + controller feeding batteries and DC load in an off-grid system except for the nature of the current, the tests for connection quality, and the fact that the grid is an "unlimited capacity battery" which can never be fully charged.

    This excludes hybrid GT systems which have the ability to shut down 'SELL' so that no power produced goes anywhere except to connections on the 'LOAD' terminals as needed. In that instance any power available above load requirements is not realized, the same as with an off-grid system.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Grid tie inverter operation clarification

    Replace the tub with Hover Dam... Virtually unlimited available water and any water you use/consume does not change the level of the lake behind the dam.

    Your "loads" (sprinklers, kitchen sink, etc.) and a GT source (a small piston water pump from your well)

    The dam/lake level will keep your water pressure at 60 PSI--No matter what... All the facuets on, still 60 PSI at that 18" water pipe outside your home.

    Now turn on the piston pump, it just starts pumping 5 GPM into your home's water pipes and out to the street pipe--Still at 60 PSI.

    That is the first approximation... No losses.

    Second level of approximation. Now add pipe friction... Your 5GPM flow from the street to your sprinkler is now 55 PSI. Add 5 GPM from your pump and the there is no water pressure drop from the street--Back to 60 PSI in your house.

    Draw 10 GPM of water, and pump 5 GPM back, you are a net 5 GPM from the street/dam... And your current pressure in the house is ~55 PSI. The pump does not care, it will be programmed to pump from 50 PSI to 80 PSI--and turn off if the water pressure it outside of those limits.

    Third level of approximation--momentum of the water in the pipes and the pump output (pulsed, vs continuous stream, variable output based on sunlight, etc.). Add "water hammer" (water pressure spikes from turning off water quickly, momentary pressure drop when valves are first opened, etc...

    Fourth level of approximation--Water evaporation, vapor pressure, type of chlorination, pipe materials, water hardness, leaks, etc... -- I.e., more of teh crazy.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jaya
    jaya Registered Users Posts: 3
    Re: Grid tie inverter operation clarification
    Grid-tie inverters do not make any differentiation between household and grid. They only see an AC connection which has (hopefully) proper Voltage and frequency. When that is present they take any power available from the PV array and supply it to the AC mains. If the household uses it that's where it goes. If the GTI output is greater than household needs it goes to the grid, which in this case does not act as a supply but as a load.

    Thanks for the super quick responses.
    Even when the GTI output is less than the instantaneous household demand, since the GTI is injecting at a higher voltage into grid and load simultaneously, isn't some of the instantaneous PV energy output always "shared" between the grid and the local loads? In other words, does one have to always reckon with some loss to the grid, even when household load is greater than PV output?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification
    jaya wrote: »
    Thanks for the super quick responses.
    Even when the GTI output is less than the instantaneous household demand, since the GTI is injecting at a higher voltage into grid and load simultaneously, isn't some of the instantaneous PV energy output always "shared" between the grid and the local loads? In other words, does one have to always reckon with some loss to the grid, even when household load is greater than PV output?

    No, not really.
    Looking at current flow on an AC system can drive you nuts because it goes back and forth instead of one way, but in essence it flows from source(s) to load(s). With the GTI operating as a source and the grid operating as a source the current flow is to the loads. The only time the grid becomes a load is if the GTI's output exceeds the 'local' (household) load demands.

    Think of it in DC terms:
    Grid "battery", GTI "battery", household loads.
    Grid battery has much higher capacity than GTI battery, but essentially the same Voltage (GTI slightly higher).
    Household loads exceed GTI battery's capacity, extra power required is drawn from the Grid battery.
    Household loads less than GTI battery's capacity, its slightly higher Voltage potential causes all of its current to flow to the lesser Voltage of the grid.

    It doesn't take a large Voltage difference to create the current flow either.
  • jaya
    jaya Registered Users Posts: 3
    Re: Grid tie inverter operation clarification

    Ah..!

    Merging BB's and your explanations, I now see it as a multipoint manifold with two inlets (GTI & grid)
    and multiple outlets (local loads).

    When the cumulative flow rate from all the outlets is equal to inflow rate from the GTI inlet, the
    manifold is in "equilibrium".

    When the cumulative flow rate from all the outlets is higher than the inflow rate from the GTI inlet,
    the deficit in the manifold filled from the grid inlet.

    When the cumulative flow rate from all the outlets is lower than the potrential inflow rate from the
    GTI inlet, the pressure in the manifold rises (higher voltage at the GTI inlet) and is relieved by
    a reverse flow out through the grid inlet.

    Many thanks to you and BB!
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Grid tie inverter operation clarification

    Long live water and plumbing analogies! :D

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Grid tie inverter operation clarification
    jaya wrote: »
    Thanks for the super quick responses.
    Even when the GTI output is less than the instantaneous household demand, since the GTI is injecting at a higher voltage into grid and load simultaneously, isn't some of the instantaneous PV energy output always "shared" between the grid and the local loads? In other words, does one have to always reckon with some loss to the grid, even when household load is greater than PV output?
    Well, yes and no. A load does not know or care where the electrons that flow through it are coming from. If your PV is producing 1000 Watts and your household loads are consuming 1000 Watts, it matters not whether all the loads are getting all their power from the PV or only half and the other half is coming from the grid and half the power from your PV is "escaping". The net result is the same.

    The PV is not injecting current at a higher voltage; there can only be one voltage at a given spot on a given conductor. The higher voltage at the inverter is due to voltage drop in the conductors because of the current flow; if the conductors were superconducting the voltage would be exactly the same at both ends and the current would still flow at the same rate. That's the thing about current sources, the voltage is higher because of the current flow, not the other way around like it is with a voltage source.
  • SUNUP
    SUNUP Registered Users Posts: 24 ✭✭
    Re: Grid tie inverter operation clarification
    ggunn wrote: »
    Well, yes and no. A load does not know or care where the electrons that flow through it are coming from. If your PV is producing 1000 Watts and your household loads are consuming 1000 Watts, it matters not whether all the loads are getting all their power from the PV or only half and the other half is coming from the grid and half the power from your PV is "escaping". The net result is the same.

    The PV is not injecting current at a higher voltage; there can only be one voltage at a given spot on a given conductor. The higher voltage at the inverter is due to voltage drop in the conductors because of the current flow; if the conductors were superconducting the voltage would be exactly the same at both ends and the current would still flow at the same rate. That's the thing about current sources, the voltage is higher because of the current flow, not the other way around like it is with a voltage source.

    The PV produces power? I thought it produced a voltage, power or watts is the results of the load applied across the voltage. Still confused on how the grid will sink the current from the GTI? and why the user loads will only use the current from the GTI until their requirements are met and then draw from the grid if they are not? I have been looking for the answer for over a year and called a few MFG of inverters and still don't have an answer. looks like I might have to buy an inverter and do some testing to find the answer.
    Charlie
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification
    SUNUP wrote: »
    The PV produces power? I thought it produced a voltage, power or watts is the results of the load applied across the voltage. Still confused on how the grid will sink the current from the GTI? and why the user loads will only use the current from the GTI until their requirements are met and then draw from the grid if they are not? I have been looking for the answer for over a year and called a few MFG of inverters and still don't have an answer. looks like I might have to buy an inverter and do some testing to find the answer.
    Charlie

    PV is a current source. The current is not created by Voltage differential. It will 'try' to produce Imp at any Voltage from zero on up to Vmp. The output of the GTI is the same, just converted to AC. Once Voltage is matched, current flows to any loads no matter where they are. This includes the grid, which can be considered "loads in other peoples' houses".

    Neither the GTI nor the household nor the utility know where the power is coming from or going to, nor do they care. It's all just one big circuit. If the GTI could be kept active and all loads removed including the grid it's output current would be zero and its Voltage would be the Voc equivalent for '240' (this does not actually happen due to the nature of the anti-islanding circuitry which keeps the GTI in sync with the grid, just like an additional generating plant).

    For functional purposes:
    If GTI power > household loads then difference goes to grid.
    If household loads > GTI output then difference comes from the grid.

    There's no "here first, there second".
  • SUNUP
    SUNUP Registered Users Posts: 24 ✭✭
    Re: Grid tie inverter operation clarification

    Thanks for the reply,
    Neither the GTI nor the household nor the utility know where the power (current ) is coming from or going to, nor do they care.
    Ok
    Now if we remove all the loads in the house, the grid is still up 240v the inverter is at 240v and can supply 40 amps, how or why will the current flow into the grid? Impedance of grid =240v divided 40 amps = 6 ohms or less? acting as the load? are we still on the same page?
    Don't hang up on me I'm a slow learner.
    Charlie.
  • gww1
    gww1 Solar Expert Posts: 963 ✭✭
    Re: Grid tie inverter operation clarification

    I don't know but my basic assumtion was it was like a river. it flows down hill. So if the meter only allows 120 volts that is going to flow down hill to the loads. if an inverter puts out 121 volts that stops the flow and the new down hill is from the inverter to things that are less then what it is putting out. Again, only my assumtion.
    gww
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification

    It is a little hard to wrap your head around.

    Two different types of power source are involved here: Voltage based (the utility grid) and current based (the GTI).

    The grid tries to maintain a fixed Voltage against a varying current demand. As far as it is concerned that current demand can even be locally negative.

    The GTI doesn't so much "put out" Voltage as match it and supply current. As far as it is concerned everything connected to it is a load, including the grid.

    I think it's the concept of the grid as load that most people have trouble with. Think of it as a gigantic AC 'battery'; no matter what it's 'SOC' you can draw power from it or push power to it.

    Maybe you have trouble with the idea that current can flow without a Voltage differential, but it can. Those photons just slam into that junction and kick the electrons forward; if there's a circular path they go 'round. If there's not, they just bounce back.
  • SUNUP
    SUNUP Registered Users Posts: 24 ✭✭
    Re: Grid tie inverter operation clarification
    It is a little hard to wrap your head around.

    Two different types of power source are involved here: Voltage based (the utility grid) and current based (the GTI).

    The grid tries to maintain a fixed Voltage against a varying current demand. As far as it is concerned that current demand can even be locally negative.

    The GTI doesn't so much "put out" Voltage as match it and supply current. As far as it is concerned everything connected to it is a load, including the grid.

    I think it's the concept of the grid as load that most people have trouble with. Think of it as a gigantic AC 'battery'; no matter what it's 'SOC' you can draw power from it or push power to it.

    Maybe you have trouble with the idea that current can flow without a Voltage differential, but it can. Those photons just slam into that junction and kick the electrons forward; if there's a circular path they go 'round. If there's not, they just bounce back.

    Current needs a complete circuit from the source thru the load and back to the source.
    The grid or load Impedance will determine how much current will flow based on the source voltage up to the capacity of the source.
    The way I see it now, the grid is low impedance compared to the GTI so it is sinking the current from the GTI and will sink current as the other loads in the home at the same time.
    I thought photons had to do with light.
    I am still learning and with each reply, I think I am getting closer to the answer.
    Charlie
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification

    Photons do have to do with light: they strike the photovoltaic cells and cause the electrons to jump.

    The GTI, in oversimplified form, converts this DC action to an AC one; electron (current) flow.

    How much depends on the intensity of the light on the PV (as well as all the regulating factors). By coincidence the old term for Amperes was Intensity, hence the use of the letter 'I' in Ohm's law.
  • SUNUP
    SUNUP Registered Users Posts: 24 ✭✭
    Re: Grid tie inverter operation clarification
    Photons do have to do with light: they strike the photovoltaic cells and cause the electrons to jump.

    The GTI, in oversimplified form, converts this DC action to an AC one; electron (current) flow.

    How much depends on the intensity of the light on the PV (as well as all the regulating factors). By coincidence the old term for Amperes was Intensity, hence the use of the letter 'I' in Ohm's law.

    I am only interested in the current flow from the Inverter to grid, one thing at a time.
    Charlie
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie inverter operation clarification
    SUNUP wrote: »
    I am only interested in the current flow from the Inverter to grid, one thing at a time.
    Charlie

    It's the same thing, only instead of going in one direction it alternates.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Grid tie inverter operation clarification

    The impedence of the grid--You can measure it pretty easily if you what a Utility Meter that gives you Watts being used/supplied at the time and a meter to read voltage (and/or an AC/DC current clamp DMM). But the math would look like:

    Say you have a 5 kWatt electric drier:

    5,000 Watts / 240 VAC = 21 Amps

    And say that your voltage at the meter goes from 244 volts to 240 volts when the drier is turned on and running--And you have 21 amps more draw...

    Resistance is:

    V=I*R
    R=V/I=change in voltage / change in current
    R=4 volts / 21 amps = 0.19 Ohm resistance to grid (roughly correct)

    Remember that AC voltage and current are actually "vectors" with both amplitude an direction (phase angle). And the AC Utility power, like Marc/Cariboocoot said looks for all the world like a giant AC Battery bank--Its job is to clamp the voltage at ~240 VAC... It will sink power if the current generated by the solar array goes to 21 amps/248 VAC, or it will source power at 240 VAC/21 amps if the drier is running. If both GT solar and drier are sourcing/sinking 5,000 watts, then the line voltage will be at ~244 volts nominal line.

    The easiest analogy to AC power flow I can think of giving is -- Imagine you are riding a bicycle with no "freewheel" (aka like a tricycle from when we were kids).

    If you "pump" the pedals correctly, the bicycle will speed up--You are pumping the pedals in phase with the there rotation. If you want to slow down, you resist the motion of the pedals with your feet, and you slow down. And if you pedal randomly (sometimes restrict, sometimes help pedal motion), then, on average, you have no/zero power transfer.

    A home with Grid Tied Solar, local AC loads, and connected to the grid... Operates very much like a car with a very large battery bank and an alternator with no voltage regulator. The battery is "clamping" the battery bus voltage at ~13.8 volts or so--The loads are whatever the loads are (lights, ignition, radio, etc.). And the alternator is outputting current at whatever RPM the engine is running at (engine turns faster, more current; engine stops, no alternator current and all power is supplied by the battery).

    Put an Amp*Hour/Watt*Hour meter on the battery and bill the driver of the car for the power used (credit for power generated).

    Vehicle system works well until battery is fully charged--Then battery bus voltage rises to >14.4 volts--At that point, the alternator turns off completely (for sake of this example). Just like a GT Inverter "shuts down" if grid voltage is outside ~212 to 264 VAC range and/or outside 60+/- 1 Hz frequency range (grid out of regulation).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • SUNUP
    SUNUP Registered Users Posts: 24 ✭✭
    Re: Grid tie inverter operation clarification

    A home with Grid Tied Solar, local AC loads, and connected to the grid... Operates very much like a car with a very large battery bank and an alternator with no voltage regulator. The battery is "clamping" the battery bus voltage at ~13.8 volts or so--The loads are whatever the loads are (lights, ignition, radio, etc.). And the alternator is outputting current at whatever RPM the engine is running at (engine turns faster, more current; engine stops, no alternator current and all power is supplied by the battery).

    The alternator produces a voltage determined by engine RPM that's why they use a voltage regulator and not a current regulator.