Sol-Ark 12K + Enphase IQ8 microinverters

[timselectric]

This implies that when you have power exceeding the loads on the critical loads panel that power will flow into the Sol-Ark. this must mean the load out connection also allows power in. This implies it will flow back into the main service panel and off into the grid. OR it is used to charge the battery. What happens when The solArk is set to not export power and the battery is full. In this case the frequency gets shifted to turn down/off the microinverters. Is there a table to correlate frequency shift to %power generation on the microinverters or is this just an on/off situation?

I believe it's just an on and off. Not a variable control.
It's based on battery SOC.
If I'm not mistaken. It's off, at 100%. And on, at 90% or lower.

 

[Johndo]

It changes the frequency of its output, to control the micr inverters. They only output, if they see a clean power, already on the line.

 

[offthehook]

I have dual Sol-Ark 12Ks with 4 strings of IQ7 microinverters in an AC coupled install. Once the battery reaches a certain voltage, the sol-arks begin ramping up the frequency to curtail IQ7 output. Using the right Enphase configuration and the IQ7s will adjust linearly based on frequency. It works well and adjusts quickly for changing loads. Of course, this is an off-grid solution otherwise the excess power will be fed into grid. Can't shift frequency while grid-tied! I'm sure the IQ8 hase a similar configuration although, it's big claim to fame is grid-forming without the need of another inverter.

 

[Johndo]

I have dual Sol-Ark 12Ks with 4 strings of IQ7 microinverters in an AC coupled install. Once the battery reaches a certain voltage, the sol-arks begin ramping up the frequency to curtail IQ7 output. Using the right Enphase configuration and the IQ7s will adjust linearly based on frequency. It works well and adjusts quickly for changing loads. Of course, this is an off-grid solution otherwise the excess power will be fed into grid. Can't shift frequency while grid-tied! I'm sure the IQ8 hase a similar configuration although, it's big claim to fame is grid-forming without the need of another inverter.

That is interesting that you can not shift frequency when grid tied. They (SolArk engineering support) told me they could and would. Have you verified this? Considering both scenarios in their manual (tied into the sub panel or tied into their box through the generator in) I see no reason technically they could not do this. What they did say is they could not throttle it. They would just turn off the string if it could not be balanced out with the DC tied PV running through their box. Their recommendation was to have at least an equal amount of DC tied PV.

 

[Ampster]

That is interesting that you can not shift frequency when grid tied.

The grid is a pretty massive force. I suppose the question turns on what is meant by grid tied? The only way that I know that a hybrid inverter would generate its own AC frequency different than the grid, would be to disconnect from the grid. There is no need for frequency shift when grid tied since any excess solar would go to the grid which presents an infinite load to the inverter.

 

[Johndo]

The grid is a pretty massive force. I suppose the question turns on what is meant by grid tied? The only way that I know that a hybrid inverter would generate its own AC frequency different than the grid, would be to disconnect from the grid. There is no need for frequency shift when grid tied since any excess solar would go to the grid which presents an infinite load to the inverter.

 

[Johndo]

You do not need to shift the frequency of the entire grid. You only have to shift that to your sub panel or to the line connected to the generator in connection whichever you have your enphase microverters tied into. You are still grid tied on the main service panel and have access to flow through energy for your sub panel. If your Micro-inverters are tied into your main service panel then your SolArk's have no interaction with your microinverters.
Have you talked to SolArk on this? You really should. I could have gotten the wrong information but they were pretty specific on how this would work.

 

[Ampster]

You are still grid tied on the main service panel and have access to flow through energy for your sub panel.

How does that flow work when the frequencies are different? I still believe there is no need for frequency shift because the grid can take all the power from the AC coupled solar panels.

 

[Hedges]

That is interesting that you can not shift frequency when grid tied. They (SolArk engineering support) told me they could and would. Have you verified this?

If they said that, chances are they don't know what they are talking about.

You do not need to shift the frequency of the entire grid. You only have to shift that to your sub panel or to the line connected to the generator in connection whichever you have your enphase microverters tied into. You are still grid tied on the main service panel and have access to flow through energy for your sub panel.

If Sol-Ark contains a single bi-directional inverter, it could then run the grid-tied inverters at 60.75 Hz to charge battery while letting sub-panel get 60 Hz from grid. But it could not provide the full functionality you want unless it had two separate inverters. (Not to be confused with two inverters for two phases of split-phase. It could have two 120V inverters for grid interactivity plus one 240V inverter for frequency-shift of GT PV inverters on generator input. But it does not have that.)

As Ampster says, it can take all GT PV when on-grid. That just flows through wires to grid.

Off-grid, it may in fact use frequency shift to manage GT PV inverters, but may have trouble sucking down excess power during "load dump" during the time it slowly shifts frequency.
By having less power AC coupled and more power DC coupled, with GT PV on the "generator" input with relay disconnect, it can simply dump GT PV when desired and substitute DC coupled PV.

AC coupled does work, if battery can accept the surge for a few seconds. SMA has it working. What I don't know is how well that works with lithium battery.

 

[Ampster]

AC coupled does work, if battery can accept the surge for a few seconds. SMA has it working. What I don't know is how well that works with lithium battery.

Apparently my Skybox reports that my micros are AC coupled when on the grid. I guess that refers to the fact that the micro output flows through the Skybox on it way to the grid. The Skybox only needs to frequency shift when the grid is down if the batteries cannot accept the output or surge as you stated.
I also agree with your conclusion that a hybrid would need two internal inverters to be able to concurrently run two distinct frequencies.That makes the reported statement by SolArk suspect without further clarification.

 

[Shimmy]

Thanks for all your comments, and I did see that diagram also. I'm referring to this section in the Sol-Ark install manual:
View attachment 87404
I'm assuming, but can someone confirm, that this is talking about using Panels+microinverters connected to the "Gen input"

NO. AC coupling is on the load side, per the diagram.

 

[Shimmy]

So, I still don't feel like I have gotten my question answered - can you connect a panel+IQ8 directly to the Sol-Ark 12K "gen input" and have it work. Has anyone actually done this?.

NO. The IQ8s require an external voltage reference and the input side of the inverter would not have one. That is why the diagram shows it connected to the load side, where there is always a voltage reference-- either grid on pass-through or inverter from the SolArk.

 

[Johndo]

So this is not what SolArk told me. They may be feeding me bull or made a mistake or maybe I just did not understand what they said but I asked multiple times about this. They were clear that in a 0 grid conditions if the microinverters over produced the load and the batteries were limited by SOC they could turn the inverters off. They made it very clear that they could not throttle them up or down but they could turn them off by frequency shifting.
I agree with sgturner that it would be nice is someone with real life experiences could illuminate us on how it actually works.

Let me qualify what I was originally after. I simply wanted a device that I could place on my main service panel which my enphase micro-inverters also fed into. Using a set of CT's it would read the in and out flow of energy. It would then absorb the energy being sent out so there would be 0 export and it would supplement the energy input so there would be 0 energy used. Basically a true 0 grid situation. In both cases pending the SOC of the batteries (agnostic LiFePo4) and limitations of the charger and inverter. It appears there is NOTHING on the market that can do this without some custom programming. If someone disagrees PLEASE let me know because I am a little incredulous that something like this does not exists.


For me SolArk is just a very expensive compromise

 

[solardad]

@Johndo nice recap on a AC coupled setup that he runs 'off-grid'

 

[Shimmy]

They were clear that in a 0 grid conditions if the microinverters over produced the load and the batteries were limited by SOC they could turn the inverters off. They made it very clear that they could not throttle them up or down but they could turn them off by frequency shifting.

Yes, this is going to be true for any grid forming inverter and modern grid-tied inverter.

The key item is that if you want it to work the microinverters need to be electrically downstream of the grid forming inverter-- on the load side.

 

[Ampster]

For me SolArk is just a very expensive compromise

There are less expensive hybrid inverters that will do what the SolArk will do. It is not a matter of compromise but rather understanding the physics of AC coupling and how to use it to leverage GT inverters and hybrid inverters with batteries. Optimization is the word that comes to mind.
To clarify the response of @Shimmy and the answer you received from SolArk;
Any modern micro inverter should be able to throttle down or modulate as a modern hybrid inverter shifts frequency. The latest UL1741SA is the specification to look for. If you have older UL1741 micros (without the SA suffix) then SolArk is correct, they can only react with on or off.

 

[Shimmy]

There are less expensive hybrid inverters that will do what the SolArk will do. It is not a matter of compromise but rather understanding the physics of AC coupling and how to use it to leverage GT inverters and hybrid inverters with batteries. Optimization is the word that comes to mind.
Any modern micro inverter should be able to throttle down or modulate as a modern hybrid inverter shifts frequency. The latest UL1741SA is the specification to look for. If you have older UL1741 micros (without the SA suffix) then SolArk is correct, they can only react with on or off.

Is throttling capability in the hybrid inverter guaranteed by UL1741SA?

 

[Ampster]

Is throttling capability in the hybrid inverter guaranteed by UL1741SA?

AC coupling is not garanteed. You have to look at each hybrid inverter to see if and how they handle it and the relationship between the AC coupled inverter capacity and battery capacity in the hybrid. UL 1741SA requires inverters conforming to that spec which includes modulation based on frequency shift or throttling as you refer to it. So yes it would be garanteed in the GT inverter being controlled by a hybrid that is AC coupling to it. I do not know the proper terminology to describe the hybrid inverter which controls the GT inverter that is AC coupled to the hybrid, but typically the controlling inverter is a hybrid with batteries and the inverter being controlled is a GT inverter. I think the term you used, "grid forming inverter" is descriptive of part of the AC coupling process.

 

[Hedges]

NO. The IQ8s require an external voltage reference and the input side of the inverter would not have one. That is why the diagram shows it connected to the load side, where there is always a voltage reference-- either grid on pass-through or inverter from the SolArk.

What do you mean by "NO"?

We understand SolArk can support frequency shift AC coupling of inverters on the load size, but may have problems.
We also understand it supports that on the generator input. That is same as load connection, except there is a relay in between, which can be opened to instantly shut off the power.

Yes, this is going to be true for any grid forming inverter and modern grid-tied inverter.

The key item is that if you want it to work the microinverters need to be electrically downstream of the grid forming inverter-- on the load side.

And the generator input is downstream of the grid-forming inverter.

If a generator is connected, SolArk could treat it like grid. If GT PV inverters are connected, SolArk can just close the relay. Different configuration options.

 

[Shimmy]

What do you mean by "NO"

Every inverter I have researched treats a generator input as unstable voltage reference and utility input as stable voltage reference. The inverter will not "stabilize" the generator voltage as that is impossible... it will simply isolate the load from it, using it to charge batteries. If the inverter is not providing a voltage reference to the microinverters, what is?