AC Coupling

[Mustang65]

I did a couple searches with no finds, so I guess I will ask the question... I was looking at the Enphase website and came across a document regarding "AC Coupling" using their 250 micro inverters and power grid power failures.

I was wondering if anyone has implemented this in their systems. I know that some of the inverter mfr have incorporated small AC systems in their inverters with very limited output. I was under the impression that you could not fool the micro inverters by using an inverter's output to simulate the power grid allowing the micro inverters AC output to be used in your private system. I understand that there is a lot of design that needs to be included to protect the grid from the micro inverters output and limiting what can be powered by the micro inverters, overcharging.....

The link to the document is below.

Just wondering...

Don

AC Coupling of Enphase Micro inverters to Battery Based Systems

[URL="http://enphase.com/global/files/Enphase_Application-Note_AC-Coupled-Battery-Based-Systems.pdf"]

[ButchDeal]

I did a couple searches with no finds, so I guess I will ask the question... I was looking at the Enphase website and came across a document regarding "AC Coupling" using their 250 micro inverters and power grid power failures.

I was wondering if anyone has implemented this in their systems. I know that some of the inverter mfr have incorporated small AC systems in their inverters with very limited output. I was under the impression that you could not fool the micro inverters by using an inverter's output to simulate the power grid allowing the micro inverters AC output to be used in your private system. I understand that there is a lot of design that needs to be included to protect the grid from the micro inverters output and limiting what can be powered by the micro inverters, overcharging.....

The link to the document is below.

Just wondering...

Don

AC Coupling of Enphase Micro inverters to Battery Based Systems

[URL="http://enphase.com/global/files/Enphase_Application-Note_AC-Coupled-Battery-Based-Systems.pdf"]

you do AC coupling but not with a small system and it needs to have support for back feeding the power and know how to control the micros like the OutBack systems http://www.outbackpower.com/index.ph...pled-solutions

[inetdog]

I did a couple searches with no finds, so I guess I will ask the question... I was looking at the Enphase website and came across a document regarding "AC Coupling" using their 250 micro inverters and power grid power failures.

I was wondering if anyone has implemented this in their systems. I know that some of the inverter mfr have incorporated small AC systems in their inverters with very limited output. I was under the impression that you could not fool the micro inverters by using an inverter's output to simulate the power grid allowing the micro inverters AC output to be used in your private system. I understand that there is a lot of design that needs to be included to protect the grid from the micro inverters output and limiting what can be powered by the micro inverters, overcharging.....

The link to the document is below.

Just wondering...

Don

AC Coupling of Enphase Micro inverters to Battery Based Systems

[URL="http://enphase.com/global/files/Enphase_Application-Note_AC-Coupled-Battery-Based-Systems.pdf"]

I find the Enphase PDF dangerously incomplete and potentially misleading.
There are a limited number of off-grid inverters which are bidirectional and can either provide AC from the battery bank or tap the AC for battery charging with no additional external hardware (such as an external CC).
The PDF gives the impression that all off-grid inverters work this way.

I also find that turning the GTI micros on and off based on the battery voltage is a crude and unreliable way to avoid back feeding the off grid inverter AC output.

The mention of frequency shift regulation (by the off-grid inverter) as done within the SMA SunnyBoy/SunnyIsland family leaves out what I believe to be a critical feature of the system, namely that the AC frequency shift provides a proportional analog control of the SB output rather than simply taking the frequency out of the acceptance limits of the SB to shut them off entirely.

I see that there may be some situations in which AC coupling of Enphase micros may be practical and desirable, but I think that the PDF is just too general to be useful in evaluating a particular proposed installation.

Looking at the Outback documents that ButchDeal linked, I wonder just what the effect of high frequency micro-cycling (by their estimate as often as every six minutes) will have on battery life and health.) The Outback document does at least make clear that the GTI modulation is a strictly on-off rather than proportional control method.

It is also not clear to me how the Outback magically manages to choose how much power to pull from the AC bus for battery charging based on the excess of GTI output over load consumption.

[billvon]

It is also not clear to me how the Outback magically manages to choose how much power to pull from the AC bus for battery charging based on the excess of GTI output over load consumption.

It doesn't. It pulls power based on the power that its charging profile wants, and based on the sell and AC input limit settings. In no case will it back off charging because the GTI is supplying less power than expected (or vice versa.) Thus the requirement for the crude "bang-bang" relay to stop GTI inverter operation. Beyond that the system remains "in balance" because the batteries source/sink power based on the demand at the AC output. It does this by measuring voltage; it uses the inverter/charger to regulate the voltage to around 120 volts.

[inetdog]

It doesn't. It pulls power based on the power that its charging profile wants, and based on the sell and AC input limit settings. In no case will it back off charging because the GTI is supplying less power than expected (or vice versa.) Thus the requirement for the crude "bang-bang" relay to stop GTI inverter operation. Beyond that the system remains "in balance" because the batteries source/sink power based on the demand at the AC output. It does this by measuring voltage; it uses the inverter/charger to regulate the voltage to around 120 volts.

But if the Outback draws only what the batteries need, the DC voltage on the batteries will not rise just because the GTI is providing more power than the combination of the loads and charging.
Assuming that the Outback can both provide a reference AC voltage, contributing power, and suck AC for charging at the same time, it would also have to have some way of sensing that the GTI output is greater than the sum of load and charging. I suppose it could do that by monitoring for negative power transfer at its own AC output, but that has its own set of vulnerabilities and was never mentioned in any of the descriptions.

[ButchDeal]

But if the Outback draws only what the batteries need, the DC voltage on the batteries will not rise just because the GTI is providing more power than the combination of the loads and charging.
Assuming that the Outback can both provide a reference AC voltage, contributing power, and suck AC for charging at the same time, it would also have to have some way of sensing that the GTI output is greater than the sum of load and charging. I suppose it could do that by monitoring for negative power transfer at its own AC output, but that has its own set of vulnerabilities and was never mentioned in any of the descriptions.

This document seems to have more detail: http://sunwize.com/documents/OutBack...ril%202013.pdf

and a webinar: https://www.youtube.com/watch?v=gOy2oZMA_gw

[billvon]

But if the Outback draws only what the batteries need, the DC voltage on the batteries will not rise just because the GTI is providing more power than the combination of the loads and charging.

Right. The inverter won't take into account the status of external GT generation. All it senses is the AC output voltage, and will keep that within regulation when operating independently. (When grid is present, of course, it just passes power to the grid and this isn't an issue.)

Thus, in independent operation, if sources balance loads then nominal voltage is maintained and the Outback will do very little. If sources are significantly lower than loads then the Outback will use battery power to maintain the voltage (in other words, be a regular inverter.) If sources are higher than loads then the inverter will draw AC power and send that power to the batteries to regulate the (now too high) AC voltage. This is the "unusual thing" that a bidirectional inverter will do.

The relay is needed because batteries are not infinite sinks. At some point the battery absorbs all the power it can. At that point the battery voltage rises and the relay opens, ending the grid tie operation.

An interesting question that I don't know the answer to is - what if the relay fails to open and excess GT power is still available? I suspect the result would be that the Outback limits voltage to the batteries to prevent battery damage, the AC voltage rises out of range, the Outback trips off-line due to an overvoltage fault followed quickly by the GT inverters. However I have never seen this described explicitly.

[inetdog]

Right. The inverter won't take into account the status of external GT generation. All it senses is the AC output voltage, and will keep that within regulation when operating independently. (When grid is present, of course, it just passes power to the grid and this isn't an issue.)

Thus, in independent operation, if sources balance loads then nominal voltage is maintained and the Outback will do very little. If sources are significantly lower than loads then the Outback will use battery power to maintain the voltage (in other words, be a regular inverter.) If sources are higher than loads then the inverter will draw AC power and send that power to the batteries to regulate the (now too high) AC voltage. This is the "unusual thing" that a bidirectional inverter will do.

Which does not necessarily work if the Outback has to maintain the AC output voltage so that the SB or other GTI will think that it is syncing to the grid.
The mind boggles, and intuition may fail to give a correct answer.

[ButchDeal]

Which does not necessarily work if the Outback has to maintain the AC output voltage so that the SB or other GTI will think that it is syncing to the grid.
The mind boggles, and intuition may fail to give a correct answer.

stumbled upon this which gives some more details

https://www.youtube.com/watch?v=gOy2oZMA_gw

[billvon]

Which does not necessarily work if the Outback has to maintain the AC output voltage so that the SB or other GTI will think that it is syncing to the grid.

Well, sure it does. Think of the Outback as a bidirectional voltage regulator. Voltage too high due to GT operation? It draws power. Voltage too low due to loads? It supplies power. The other GT inverters like this just fine since they see a valid AC signal. As long as the batteries have headroom to accept or source power (i.e. they are neither fully charged nor fully discharged) the inverter will be able to do this to some extent.

[russ]

the inverter will be able to do this to some extent.

Key words are there

[billvon]

Key words are there

Yes. The limits are:
-The inverter's power rating. GT generation must be less than the inverter's rating.
-Ability of batteries to accept charge. A large bank at 50% will be able to support the inverter's full power rating. As the battery moves from bulk (constant current) to absorb (constant voltage) then its ability to absorb power will decline. Hence the additional relay needed, to disconnect GT generation before the battery reaches the point at which it can no longer accept full power.

[Dag Johansen]

So . . . is anyone building a unit like that Outback system that will work with the Tesla Powerwall so that Enphase system owners can easily add a Tesla Powerwall and critical load panel their systems?

[Naptown]

I am not a fan of A/C coupled systems
Particularly in a back up situation.
Batteries get too low and the entire system shurs down and will not restart.
In this instance i prefer a system like the Schneider Context system. In this case the inverter may not start but the batteries will charge.

[jflorey2]

So . . . is anyone building a unit like that Outback system that will work with the Tesla Powerwall so that Enphase system owners can easily add a Tesla Powerwall and critical load panel their systems?

I don't get the whole Tesla Powerwall thing. It's expensive and not all that useful, since it operates at ~400 volts. It is shiny though.

If you want battery backup, you can do that with an Outback of your choice and batteries of your choice. Like lead-acid? No problem. Like LiFePO4? They are available too.

If you want to do AC-coupled, the options I know of are the SMA Sunny Island system (fairly well integrated) or a Rube Goldberg-ish system with relays that turns banks of microinverters on and off to match battery state. There are a few app notes out on how to do this, but nothing very clean.