Allowing Gird-Tie system to function off grid?

I am working on a project to allow a grid-tie system to continue to function when the grid goes down. (Southern California)
From what I have read, this involves "tricking" the grid-tie inverters into thinking they are connected to the power grid by providing a 240V/60Hz signal. I am using a pure sine wave generator (powered by a 12V battery), and connecting it in place of the grid to the solar inverter (Note the grid is disconnected for all of these experiments). This generator provides a clean 60Hz signal at the voltage required by the inverter, but so far I have not been able to get it functioning (brand is Sunpower, but I believe this process to be the same for all inverters).
Has anyone done something like this successfully or have any ideas on the matter? I have been able to get the inverter to the "flashing green light" stage in which it senses a grid connection and is attempting to sych with it, but no farther.

My thoughts are that maybe the inverter needs to "see" more power from the sine-wave generator (my generator can produce 600W) or that there is the possibility that the generator is not producing a signal of 60Hz +- 1Hz required for inverter operation.

Any advice is appreciated!
Thank you

Comments

  • AceNZ
    AceNZ Solar Expert Posts: 104 ✭✭✭✭
    Re: Allowing Gird-Tie system to function off grid?

    What are you using as a load on your test setup?

    If the load is significant, your small inverter will probably have a large voltage drop, which could put it out of range for your GT inverter. Have you measured the voltage during the sync phase?

    Keep in mind that connecting a system like this to the grid during a failure is dangerous for electrical workers; you should have a transfer switch -- assuming islanding of some type is even allowed by your power company, of course.
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Allowing Gird-Tie system to function off grid?
    awls wrote: »
    My thoughts are that maybe the inverter needs to "see" more power from the sine-wave generator (my generator can produce 600W) or that there is the possibility that the generator is not producing a signal of 60Hz +- 1Hz required for inverter operation.
    Welcome to the forum,

    If your grid tie inverter makes power, where do you think the power will go? Can your sine wave generator reverse itself and pump the solar energy back in to the battery? If not, I don't see how the scheme can work.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Allowing Gird-Tie system to function off grid?

    Welcome to the forum Awls,

    First question, what is the size of the GT inverter and solar array?

    In general, the Off Grid AC inverter (I guess you are calling this your sine wave "generator"--We usually use generator/genset to mean a fuel driven internal combustion engine driving an alternator/generator) needs to be at rated at the highest wattage you expect to receive from the GT inverter (i.e., a 600 watt array/GT inverter would need a 600 Watt minimum Off Grid inverter).

    If the GT inverter is >> wattage rating of the Off Grid inverter (or you have a small/fully charged battery bank), you may kick the Off Grid inverter off-line which would knock the GT inverter off line.

    Next--getting a 12 VDC 240 VAC 60 Hz AC off grid inverter is not easy in the US/North America--Normally the off the shelf small pure sine wave AC inverters are ~120 VAC @ 60 Hz...

    And most "UL/NRTL LISTED" GT inverter in the US are rated for 240 VAC (actually 120/240 VAC split phase)--So you may not be matching the 240 VAC GT inverter with the 120 VAC off grid inverter...

    Lastly, some GT inverters may do a "grid impedance test". Checking that there is a solid/low resistance path from the GT inverter to the Utility Power--I believe this test was a requirement for German installed systems--Not sure if it made it over to the US or not.

    Another issue may be line frequency... The GT Inverter needs +/- 0.5 Hz accuracy (or better). Your off grid inverter should be OK, but some are not very accurate/can wander.

    Note that the GT inverter needs stable AC voltage (pure sine wave) and frequency for ~5 minutes before it turns on. If there are any power hick-ups during that 5 minute period, the timer will be reset back to zero.

    Later on, there will be other issues to watch for--A standard Off Grid inverter has no feedback/control when it is being "back driven" by a GT inverter--You can easily overcharge the attached battery bank.

    Anyway--Probably need more information to figure out if you have the basics right first.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • AceNZ
    AceNZ Solar Expert Posts: 104 ✭✭✭✭
    Re: Allowing Gird-Tie system to function off grid?
    vtmaps wrote: »
    If your grid tie inverter makes power, where do you think the power will go? Can your sine wave generator reverse itself and pump the solar energy back in to the battery? If not, I don't see how the scheme can work.

    I think he's saying he wants to power his house load from the GT inverter.
  • awls
    awls Registered Users Posts: 3
    Re: Allowing Gird-Tie system to function off grid?
    BB. wrote: »
    Welcome to the forum Awls,

    First question, what is the size of the GT inverter and solar array?

    In general, the Off Grid AC inverter (I guess you are calling this your sine wave "generator"--We usually use generator/genset to mean a fuel driven internal combustion engine driving an alternator/generator) needs to be at rated at the highest wattage you expect to receive from the GT inverter (i.e., a 600 watt array/GT inverter would need a 600 Watt minimum Off Grid inverter).

    If the GT inverter is >> wattage rating of the Off Grid inverter (or you have a small/fully charged battery bank), you may kick the Off Grid inverter off-line which would knock the GT inverter off line.

    Next--getting a 12 VDC 240 VAC 60 Hz AC off grid inverter is not easy in the US/North America--Normally the off the shelf small pure sine wave AC inverters are ~120 VAC @ 60 Hz...

    And most "UL/NRTL LISTED" GT inverter in the US are rated for 240 VAC (actually 120/240 VAC split phase)--So you may not be matching the 240 VAC GT inverter with the 120 VAC off grid inverter...

    Lastly, some GT inverters may do a "grid impedance test". Checking that there is a solid/low resistance path from the GT inverter to the Utility Power--I believe this test was a requirement for German installed systems--Not sure if it made it over to the US or not.

    Another issue may be line frequency... The GT Inverter needs +/- 0.5 Hz accuracy (or better). Your off grid inverter should be OK, but some are not very accurate/can wander.

    Note that the GT inverter needs stable AC voltage (pure sine wave) and frequency for ~5 minutes before it turns on. If there are any power hick-ups during that 5 minute period, the timer will be reset back to zero.

    Later on, there will be other issues to watch for--A standard Off Grid inverter has no feedback/control when it is being "back driven" by a GT inverter--You can easily overcharge the attached battery bank.

    Anyway--Probably need more information to figure out if you have the basics right first.

    -Bill


    You are correct, I am using a 120V sine-wave inverter that is connected to a transformer to step it up to 240V (http://www.amazon.com/gp/product/B000MX564U/ref=oh_details_o03_s01_i00?ie=UTF8&psc=1), do you think this could cause problems? We have measured 240V so the voltage is fine, we are going to get an oscilloscope to make sure we're within the acceptable frequency. The array size is significant (over 10kW), and we were hoping to have a scenario in which we did not use a large battery bank but only relied on the power being instantaneously generated from the panels (we just want the battery to power the sine-wave inverter to kick the solar inverters on). We would prefer to not need a 10kW battery bank. That grid impedance test is concerning me as well...
    Thanks for the input!
  • awls
    awls Registered Users Posts: 3
    Re: Allowing Gird-Tie system to function off grid?
    AceNZ wrote: »
    What are you using as a load on your test setup?

    If the load is significant, your small inverter will probably have a large voltage drop, which could put it out of range for your GT inverter. Have you measured the voltage during the sync phase?

    Keep in mind that connecting a system like this to the grid during a failure is dangerous for electrical workers; you should have a transfer switch -- assuming islanding of some type is even allowed by your power company, of course.

    Our load was minimal (about 500 watts - a work light and a fan were running), the voltage was measured to be 240.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Allowing Gird-Tie system to function off grid?
    awls wrote: »
    You are correct, I am using a 120V sine-wave inverter that is connected to a transformer to step it up to 240V (http://www.amazon.com/gp/product/B000MX564U/ref=oh_details_o03_s01_i00?ie=UTF8&psc=1), do you think this could cause problems? We have measured 240V so the voltage is fine, we are going to get an oscilloscope to make sure we're within the acceptable frequency. The array size is significant (over 10kW), and we were hoping to have a scenario in which we did not use a large battery bank but only relied on the power being instantaneously generated from the panels (we just want the battery to power the sine-wave inverter to kick the solar inverters on). We would prefer to not need a 10kW battery bank. That grid impedance test is concerning me as well...
    Thanks for the input!

    An off the self GT inverter is not going to work this way...

    For a lot of reasons--It could be done, but since solar panels do not have a surge capability--You would need a 2-4x or larger array than the load just to run a refrigerator. And if a bird lands on the panel or a few puffy clouds go by, your system AC voltage is going to collapse. Very few AC appliances do well with an "unstable" AC supply.

    We can go further into the details (a typical GT inverter is a current source and cannot really regulate a fixed voltage--And an Off Grid inverter is a voltage source which actually regulates the GT inverter's output voltage, same thing the Utility does).

    A typical recommendation is for ~100 AH battery bank for every 1kWatt of output power (battery draw) or solar array input power (at 48 volt battery bank--A 12 volt battery bank would use ~400 AH for every 1kWatt of load/solar array (just rules of thumbs to start a discussion).

    In general, the battery bank is the "heart" of an off grid system--It must absorb/supply current and keep a relatively constant output voltage (i.e., voltage source). The Off Grid AC inverter does the same thing--It supplies arbitrary current at a fixed voltage/frequency. And with a GT inverter, the Off Grid inverter also absorbs excess power (and sends it to the battery bank) to keep the AC voltage/frequency to spec.

    A GT inverter "slaves" to the AC wave form and outputs current to the grid based on available sunlight/array wattage. There is no other output regulation.

    Backing up a moment... What is it you are trying to do (actual AC voltage and wattage, etc.)? There is no good reason to use a GT inverter to back drive a TSW inverter other than for charging the battery bank/supporting AC power usage during the day. Usually, using a GT inverter increases system cost and complexity--And does not make anything more reliable (and probably makes it less reliable).

    There are perfectly good Off Grid inverters that will give you stable AC power with alternate power inputs (AC mains, AC Generator Backup, plus internal AC battery charger). And there are Hybrid Inverters that can do both Grid Tied (feed power back to the grid and turn the meter backwards) and off grid (supply AC power to protected sub-circuits--AKA a big old UPS) power when the utility power fails.

    There are few applications were the variable output (voltage, current, wattage) of solar arrays is really useful to every day loads--The major loads that can direct connect to solar array are water pumping systems... Basically the pump just uses what every solar energy is available to pump water... Pump a little water in mornings/evenings and pump at capacity during bright mid-day sun.

    Otherwise, most applications need the battery bank to supply minutes/hours/days of backup power for when the sun does not shine.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Allowing Gird-Tie system to function off grid?

    I should add--Another issue you may end up with your 240:120 VAC transformer... Many modern 240 VAC (US/North America) GT inverters also measure line voltage from Hot to Neutral (120 VAC). You may need to take the center tap from the 240 VAC side of the transformer and ground reference (ground the "neutral") and run the neutral to the GT inverter so it can measure 120 VAC between Line A to Neutral and Line B to Neutral as ~120 VAC.

    A floating Neutral would cause a fault with many GT inverters.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Re: Allowing Gird-Tie system to function off grid?
    awls wrote: »
    You are correct, I am using a 120V sine-wave inverter that is connected to a transformer to step it up to 240V (http://www.amazon.com/gp/product/B000MX564U/ref=oh_details_o03_s01_i00?ie=UTF8&psc=1), do you think this could cause problems? We have measured 240V so the voltage is fine, we are going to get an oscilloscope to make sure we're within the acceptable frequency. The array size is significant (over 10kW), and we were hoping to have a scenario in which we did not use a large battery bank but only relied on the power being instantaneously generated from the panels

    Here are the issues.

    1) Your transformer has to be able to transfer the entire load, not just "enough to start the inverter." A 240V string inverter does not reference neutral so all the balancing power will have to come from the transformer. If your loads are a) all resistive and b) perfectly balanced in theory you wouldn't be sending any current through the transformer, but in reality this almost never happens.

    2) Your loads have to take the entire amount of power generated. If you are generating 7456 watts, your loads have to total exactly 7456 watts. If you get a cloud in front of the sun your loads have to drop instantaneously - and recover instantaneously when the sun returns. This is unlikely in the extreme. You could design something to do this (a 10kW load that could adjust sub-cycle to always exactly match demand) but this is not something you'd be able to do over a weekend, and probably not something you'd want to attempt without an EE degree and a lot of experience with power electronics, digital switchmode power supply design experience and some SW skills.

    3) If you want to do this the "right" way it's called AC coupling, and you need a bidirectional inverter equal to the rating of your solar array. In your case you could probably get away with a Radian, but of course they're pricey ($4K or so.) If you want to go that route there are plenty of app notes out there.

    4) There's no such thing as a "10kW" battery bank. (Well, there probably is, but no one rates them that way.) Do you mean 10kWhr?

    5) If you just want some power when you lose the grid, replace one or all of your string inverters with SMA inverters with the "secure power supply" feature. It will give you up to 1500 watts of power per inverter with no batteries required when the grid is out.
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Allowing Gird-Tie system to function off grid?
    awls wrote: »
    I am working on a project to allow a grid-tie system to continue to function when the grid goes down. (Southern California)
    From what I have read, this involves "tricking" the grid-tie inverters into thinking they are connected to the power grid by providing a 240V/60Hz signal. I am using a pure sine wave generator (powered by a 12V battery), and connecting it in place of the grid to the solar inverter (Note the grid is disconnected for all of these experiments). This generator provides a clean 60Hz signal at the voltage required by the inverter, but so far I have not been able to get it functioning (brand is Sunpower, but I believe this process to be the same for all inverters).
    Has anyone done something like this successfully or have any ideas on the matter? I have been able to get the inverter to the "flashing green light" stage in which it senses a grid connection and is attempting to sych with it, but no farther.

    My thoughts are that maybe the inverter needs to "see" more power from the sine-wave generator (my generator can produce 600W) or that there is the possibility that the generator is not producing a signal of 60Hz +- 1Hz required for inverter operation.

    Any advice is appreciated!
    Thank you

    The root problem is that an inverter fed by photovoltaics is a virtual current source and cannot regulate current output at constant voltage in response to demand (exception: some SMA inverters' protected loads outputs, but you do not want to connect a generator to that). That, and the fact that your generator can only source current, not sink it. The short answer is that it will not work. Even if you get it to start up, as soon as your load demand falls below the output of your inverter, the system will fail, quite possibly with catastrophic effect on your generator. You cannot "trick" physics.
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Allowing Gird-Tie system to function off grid?

    Just to add…

    If you consider the economics, unless the grid goes down a lot and/or for long periods of time it's just not worth it to even try to get a GT PV system to run while the grid is down. A GT system is a strategic economic tool, meant to offset your electric bill over years. Its contribution to your bottom line by continuing to operate during infrequent and/or rare grid outages is next to nothing. Yes, it leaves you without power during those outages, but it's the wrong tool to solve that problem unless it's one of the SMA inverters that can give you up to about 1500W on a dedicated outlet when the sun is shining. Get one of those and forget about trying to fool a standard GT inverter into running in an outage.