diy solar

diy solar

Hybrid inverters supporting Zero Export to External CT on AC input

I think ‘Zero Export to CT’ is pretty clear.

Yes, Zero Export to CT implies that you are grid-tied and outputting energy that is matching phase and frequency with grid (on both L1 and L2).

In addition, ‘Zero Export to CT’ implies you are exporting energy towards the grid but not enough export and power past the CT sensor that is limiting export.

The analogy I find more helpful is that if a DC-coupled array charging a battery through an SCC.

As the battery approaches full charge, the MPPT will throttle-back PV power to exactly match constant-voltage battery charge power.

If additional loads draw more DC energy from the battery, the MPPT will draw more power from the solar panels to supply that additional DC load.

Zero Export to CT is similar except that the approach of consumption where the CT is placed approaching Zero Amps plays the role of a battery approaching full charge state in the DC-coupled analogy and the inverters ability to throttle-back input to avoid Export past the CT sensor plays the role of the MPPT in the DC-coupled analogy.


No, the solar and the charging are not required. What is required is the ability of a DC-powered (battery and/or SCC) inverter to throttle-back grid-tied power generation to avoid export past an external CT sensor clamped to the same leg.

Not understanding how that is relevant, but if you want yo provide a link, I’m happy to have a look…

The Zero Export to CT capability I’m discussing is no different than what any Solark Hybrid inverter can deliver (as well as the Schneider Conext XW Pro).
The functionality you’re describing for AC coupling is frequency shift power curtailment and you’re wanting the hybrid inverter to waggle the frequency of the grid to signal the grid-following device to reduce output. I’m very supportive of the suggest to just go with DC coupled approach instead. More efficient, less limited and better performing.
 
SW will not export but it does have a generator assist function that will effectively work as a peak load shaver if you’re running from a grid source vs. the generator. I’ll say that’s as close as you’ll ever come to getting the functionality you‘re looking for in 24V. Curious: why are you putting a 24VDC condition on the system? You’ll benefit from going to 48V in choice, performance and flexibility.
24V allows 1S charging while 48V requires 2S minimum string length.

I’ve got severe shifting shade in the morning and a 1SxP string of half-cut panels handles it well.

So if I can find an option to stay at 24V, I will (and at the moment, the Victron Multiplus is looking promising).
 
The functionality you’re describing for AC coupling is frequency shift power curtailment and you’re wanting the hybrid inverter to waggle the frequency of the grid to signal the grid-following device to reduce output. I’m very supportive of the suggest to just go with DC coupled approach instead. More efficient, less limited and better performing.
I’ve come to the same conclusion - curtailing DC-coupled is just so much more natural and less complicated than frequency-shift or other communication-based approaches to curtailing AC-coupled power.
 
Most do not have the ability to send power out of the AC input in the other direction to act as a grid tie inverter and directly supply the main panels loads without going thru a critical loads/offgrid panel..
My Outback Skybox has a bidirectional AC input so I export to the grid through that port. I would say that a hybrid without a bidirectional port barely qualifies to be called a hybrid in my book
 
My Outback Skybox has a bidirectional AC input so I export to the grid through that port. I would say that a hybrid without a bidirectional port barely qualifies to be called a hybrid in my book
I absolutely agree. There’s all these so called hybrid inverters that are only chargers and offgrid inverter combos..to me a hybrid is a grid tie inverter and an offgrid inverter combination. Mppt Charger being optional
 
curtailing DC-coupled is just so much more natural and less complicated than frequency-shift or other communication-based approaches to curtailing AC-coupled power.
It is more natural and does not have the constraints that AC coupling does. They can both coexist in a well designed system. AC coupling is cost effective to add battery functionality to an existing GT inverter system.
 
All hybrid inverter capable of export can at least prevent back feeding power to grid. The difference is some hybrid inverters have only internal CT sensors which only allows them to prevent export beyond their AC input terminals. Xantrex XW is like this.

Some units have remote CT sensors, like SolArk, that plug into inverter control that can be located outside of inverter, like right on main grid lines coming into your main panel. This allows them to ensure zero export current from the point where remote current sensors are place. This would allow you to backfeed power from inverter into your main panel and supplement your loads in the main panel circuits and still prevent any current backfeed into grid utility input lines to main panel.

When grid goes downs you will not be able to supplement your main panel loads as inverter will disconnect its AC input from main panel feed. Without grid power present, all loads in main panel will not be powered.

Appropriate description would be 'Hybrid inverter with export capability that has remote CT current sensing capability'. Having export capability means they are UL1741 listed, or they will not be approved. Just telling utility you will be blocking export to their grid is not good enough to get around the UL1741 requirement. It has to meet the anti-islanding requirement of UL1741. California has additional requirements that allows them to block exports when their grid is getting too much backfeed or when they don't need the power, and again, just saying you will never backfeed will not get you exempt from this requirement.

Utility power plants cannot turn on a dime. Electric utilities have to anticipate demand for the next few hours. Different power plants have longer or shorter response time. Nuclear is slowest to react, natural gas turbines are the quickest to react. Oil/coal is in between.

You will see your AC line voltage rise or fall during the day depending how well the utility predicted power demand. Utilities usually error on the side of overproduction to ensure they can meet unexpected demand without too much voltage slump.

I once tried to explain how this works to a marketing person in the large company we worked. I explained the power companies use 'turbine peakers' that are quickly brought online to meet unexpected peak load demand. She smiled and replied, 'that is a great name for an alcoholic bar drink',
 
When grid goes downs you will not be able to supplement your main panel loads as inverter will disconnect its AC input from main panel feed. Without grid power present, all loads in main panel will not be powered.
Just a few critical loads can be moved to a critical loads panel to be powered by the inverters AC output in the event of a rare (in my area and rare for most areas) power outage. The critical panel can still get its power from the main panel when the inverter is powering the home as a grid tie inverter. If there’s a grid connection to the home there’s no reason to have all the start up surge strain on the batteries and inverters shortening their life. The grid can cover the start up surges when the inverter is powering the main panel as a grid tie inverter via the bi-directional AC input. Since power outages are usually rare for most people it would be ok to occasionally use a manual transfer switch to switch between powering the critical panel from the main panel/bidirectional AC input or the inverters AC offgrid output.
All this isn’t needed if using an inverter with a 200amp pass thru (like the SolArk 15k) or enough pass thru to power the home. Then the inverter can simply be installed between the meter and the main panel and done.. Although I am still very confused as to how the main panel loads would utilize the 15k SolArks bi-directional input if the inverter is installed between the meter and the main panel...
 
Last edited:
I am still very confused as to how the main panel loads would utilize the 15k SolArks bi-directional input if the inverter is installed between the meter and the main panel...
The way I understand it, the bidirectional flow allows the inverter to feed the grid. It is the same as a hybrid with a critical loads panel except in this case, the main panel is wired to AC output just like a critical loads panel would be wired. There are no loads between the meter and the SolArk.
 
I’m thinking the SolArk 15k doesn’t have a bi-directional input because I can’t see how a bidirectional input would properly power the main panel with grid tie power with the inverter between the meter and main panel. I’m thinking the 15k powers the main panel with grid tie thru the AC output instead of a bidirectional AC input.
 
The way I understand it, the bidirectional flow allows the inverter to feed the grid. It is the same as a hybrid with a critical loads panel except in this case, the main panel is wired to AC output just like a critical loads panel would be wired. There are no loads between the meter and the SolArk.

if the inverter is feeding the grid with a bidirectional AC input then how can the inverter feed grid tie power to the main panel at the same time to cover the main panels loads if the inverter. I mean how can power go out the AC input to the grid but not pass the meter to achieve zero export and at the same time power go out the AC output to the main panel
 
I’m thinking the 15k powers the main panel with grid tie thru the AC output instead of a bidirectional AC input.
It has to be bidiectional on AC input or it could not sell to the grid. You are correct that the main panel is powered from the AC output. That output may not be bidirectional like my Skybox hybrid because SolArk AC couples through the generator input.
 
I mean how can power go out the AC input to the grid but not pass the meter to achieve zero export and at the same time power go out the AC output to the main panel
It is physics depending on the balance of power. It will cover the main panel loads first before exporting. I assume it does it by controlling voltage. The grid presents like a load but with a more constant voltage than the main panel which probably has a lower voltage than the grid. Presumably the SolArk raises it's voltage so the loads on the main panel get served before anything gets exported.
 
How can the inverter sell to the grid via the bidirectional AC input and supply grid tie power to cover the main panel loads via the AC output all at the same time ..
 
Ya know with the 8k grid tie power was all supplied via the AC put to a breaker in the main panel and all is good. The AC output isn’t utilized when the SolArk 8k was supplying gridtie power
 
How can the inverter sell to the grid via the bidirectional AC input and supply grid tie power to cover the main panel loads via the AC output all at the same time ..
See my answer above, It is physics based on voltage that the SolArk directs the current flow where the loads are. If no loads on the main panel the current will flow to the grid. The same concept happens at the transformer. The current from exported power flows to your neighbors on that transformer first before any excess goes through the transformer to the grid. Electricity follows the path of least resistance.
 
I suppose the 15k does just that. It supplies grid tie power via the Ac output to cover the main panel loads. When they are fully covered the inverter can then also export via the bidirectional AC input if export is allowed.. this is different than the 8k which only supplied grid tie power out the AC input. AC output was only to supply offgrid power
 
Last edited:
The way I understand it, the bidirectional flow allows the inverter to feed the grid. It is the same as a hybrid with a critical loads panel except in this case, the main panel is wired to AC output just like a critical loads panel would be wired. There are no loads between the meter and the SolArk.
The backfeed from SolArk to main panel is effectively acting as a PV GT inverter. It is good to supplement non-inverter backed up loads but, like a GT inverter, if grid goes down, so do all the main panel loads.

The only thing the remote CT current sensor is doing is moving the zero current control reference to the utility input lines if you want to avoid sending any power to the grid. As long as some amount grid current is flowing into main panel the SolArc will push as much power it can to main panel. This will supplement any non-subpanel loads in the main panel.

There are zero export options on some PV GT inverters with remote CT current sensors on grid input lines that do the same thing.

Moving more main panel loads over to subpanel is not a realistic option. Any loads you put in subpanel is forced to be supported by inverter when grid goes down. The way I do this is with a control panel with about six heavy duty 30A DPDT toggle switches that I can take selected loads like water heater, washer, drier, and few lesser loads, at my discretion, so I can switch these individual loads between main panel and subpanel. This also allows me to individually select loads to run on battery backup during outages to manage battery consumption.
 
Last edited:
The backfeed from SolArk to main panel is effectively acting as a PV GT inverter. It is good to supplement non-inverter backed up loads but, like a GT inverter, if grid goes down, so do all the main panel loads.
If the grid goes down the 15k Solark will automatically cover the main panel loads offgrid.
The 15k is installed differently than the 8k SolArk. The 8k AC bidirectional input gets wired to a breaker in the main panel. The 12k AC bidirectional input gets wired to the meter. The 12k AC output (not the AC input) is wired to the main panel, which enables the 12k to easily switch between powering the main panel grid tied or offgrid or directly from the grid via the 200amp pass thru
 
Last edited:
The backfeed from SolArk to main panel is effectively acting as a PV GT inverter. It is good to supplement non-inverter backed up loads but, like a GT inverter, if grid goes down, so do all the main panel loads
I agree that that any hybrid inverter behaves like a GT inverter when the grid is up. However, if the hybrid is connected to a battery it gan form a grid and power loads and AC couple with a GT inverter. This has been mentioned previously. The wiring of a SolArk 15k may be different than the earlier models which left some loads stranded as you mention.
 
Back
Top