diy solar

diy solar

Turbo-AC-coupling - will this work???

Depends on if we ever get consistent negative spot prices during summer days. It's theoretically possible if you have non-dispatchable power that can't be ramped down operating at the same time as the solar glut, but 1741SA is supposed to give utility knobs to turn down solar. And the vast majority of solar at that point in the future should be responsive.

Unless 1741SA was mis-designed. Then it would be a shitshow. But 1741SB's explicit control systems would also have to be mis designed.
It’s an interesting question as to whether PF&E is going to require an equipment upgrade once NEM1/2 systems reach the end of their grandfather period…

Does anyone know whether we have any NEM 1.0 customers who have reached the end if their 20 years and been converted to NEM 3.0 yet?
 
It’s an interesting question as to whether PF&E is going to require an equipment upgrade once NEM1/2 systems reach the end of their grandfather period…

Does anyone know whether we have any NEM 1.0 customers who have reached the end if their 20 years and been converted to NEM 3.0 yet?
I have a neighbor that has ticked past 20 years. They were sniffing at upgrading their equipment but it’s not a priority. Presumably PG&E is not harassing them to upgrade. Also no turnkey company wants to try to give them a new inverter using the existing strings ?‍♂️

1741SB seems designed to support minor customer side upgrades (voluntary based on incentives or mandatory) to give the utility more control. Since the new features are not used by a complete system yet.
 
I am starting my 5th year under NEM 2.0 so I should have 10 more years before they mess with me. The only rule that really bothers me is if they do end up trying to charge just for having solar panel wattage available. A fixed fee seems wrong. If they did charge when exporting at noon, I get it. And shifting some of the electric bill from "energy use" based to a "maintenance fee" does seem fair to a point. The one proposal I saw where they base this fixed maintenance fee on the household income feels blatantly wrong. Of course it will flip in my favor in 10 years when I retire, but still, it should certainly not cost ever me more than I used to pay for electricity.

The plan my system is working under now is trying to flatten my grid energy usage to essentially no import or export. But as long as I am getting some credit for export, I am just letting that happen after my battery is full. Today, the battery hit full charge in the XW-Pro at 12:50 pm and so the Enphase system has been exporting since. My 2 KW of DC coupled panels are set to charge to 0.9 volts higher so they keep charging the battery for another 2 to 3 hours, pushing in another 2 to 4 KWHs.

If they do change the rules and try to charge me for the energy export from the Enphase system, I will look into the electric water heater route. I use a gas water heater now. I may also be adding either a plug in hybrid or full EV car. The extra power I export now is only enough for about 12 miles a day. My typical commute is about 50 miles.
 
I have a neighbor that has ticked past 20 years. They were sniffing at upgrading their equipment but it’s not a priority. Presumably PG&E is not harassing them to upgrade. Also no turnkey company wants to try to give them a new inverter using the existing strings ?‍♂️
I don’t think NEM 3.0 is ready yet (was told not before early next year).

As long as their billing remains unchanged, they are likely being left on NEM1…

When the billing changes, they’ll know they have been moved to NEM 3.0.
1741SB seems designed to support minor customer side upgrades (voluntary based on incentives or mandatory) to give the utility more control. Since the new features are not used by a complete system yet.
It will be great if the utilities elect to allow grandfathered equipment to continue to bd used until it fails.

Just not assured until we know what the carryover rules will be.
 
I am starting my 5th year under NEM 2.0 so I should have 10 more years before they mess with me. The only rule that really bothers me is if they do end up trying to charge just for having solar panel wattage available. A fixed fee seems wrong. If they did charge when exporting at noon, I get it. And shifting some of the electric bill from "energy use" based to a "maintenance fee" does seem fair to a point. The one proposal I saw where they base this fixed maintenance fee on the household income feels blatantly wrong. Of course it will flip in my favor in 10 years when I retire, but still, it should certainly not cost ever me more than I used to pay for electricity.

The plan my system is working under now is trying to flatten my grid energy usage to essentially no import or export. But as long as I am getting some credit for export, I am just letting that happen after my battery is full. Today, the battery hit full charge in the XW-Pro at 12:50 pm and so the Enphase system has been exporting since. My 2 KW of DC coupled panels are set to charge to 0.9 volts higher so they keep charging the battery for another 2 to 3 hours, pushing in another 2 to 4 KWHs.

If they do change the rules and try to charge me for the energy export from the Enphase system, I will look into the electric water heater route. I use a gas water heater now. I may also be adding either a plug in hybrid or full EV car. The extra power I export now is only enough for about 12 miles a day. My typical commute is about 50 miles.
Does your utility know about your Conext XW-Pro?

I’m assuming it’s kosher / allowed to install a true off-grid system with AHJ approval only but concerned that even an off-grid system using a grid connection for charging / backup may require utility review and approval (at least in their mind).

If you have been powering a large % of your overnight load using battery power and your Conext XW-Pro without issue ;from your utility), that’s encouraging (though the arrival of dual-channel metering could change things next year…).
 
Officially... No, my local AHJ and Electric utility have not inspected/approved my Schneider XW-Pro install. I did all of th wiring to code, and actually a bit above with the wire being 1 or 2 gauge steps heavier than required. The only reason I have not had it inspected is that by battery bank is not UL listed. I am using repurposed Chevy Bolt cells. They are in a pair of steep cabinets with proper BMS units and Class T fusing. It is just a tick more than half of what is in an actual Chevy Bolt that I could have parked in the garage, so what's the problem? And I am also running them in a narrower range and at far less current than in the car.

There is no doubt the electric utility can tell I am using energy time shifting storage.
About 3 years before I installed the Enphase system, they had already changed out my meter for a digital bidirectional unit. My old mechanical meter was constantly being miss read because the pointers were warping from the crazy heat here. One month they would bill me over 1,000 kilowatt hours extra, then the next month I would have a negative bill. But on the tiered billing, that didn't erase the incorrect high usage demand charges. I complained several times and they had to issue refunds. So they finally replaced the meter and me reading were then fairly consistent, so it was obvious meter reading errors.

The new meter has a separate counter for energy imported and energy exported. The energy in read out will never go backwards. If power flows out, it counts up the second counter instead. And since it is "Time of Use" it tracks the total power that goes in, AND the total power that goes out, and that is reported back to the utility, So Cal Edison in my case. The bill only shows the full hour total difference. I do not know how often the meter transmits the data. It could be every 5 minutes, or 15 minutes, I don't know. But it really does not matter. If I pull 10 KWHs and then export 10 KWHs, both counters go up by 10, but they bill me zero for that total hour.

When the Enphase solar originally went in, I was typically exporting close to 10 KWHs every day between 10 am and 3 pm. But then I would import pretty much all of it back again overnight until the sun came back up. Peak export rate could hit 3,000 watts. Peak import rate could hit over 5,000 watts in the summer with A/C running and using a hair dryer or the microwave. My daily totals would sit right about 2 KWHs of consumption. And the bulk of it was in the high rate 4 pm to 9 pm Time of Use block.

So I decided to add the battery system. At first, I had to manually command it to charge when I left for work each morning. I set the charge current fairly low, and had the charge limit and low voltage set close enough, that it took 10 KWHs to go from the low to high cut and then back down again.

So my consumption would go up an extra 1,000 watts or so if I left before the sun was up. Then my export would be 1,000 watts less because it was charging. And the battery would go full by 2 pm, and the export would jump up a bit before the sun went down. Then I had it set to just grid sell 2,000 watts for the 5 hours from 4 pm to 9 pm. It didn't even know how much power I was using. When the A/C was off, it might be exporting over 1,000 watts. And with the A/C and cooking etc., we would still consume up to 1,000 watts from the grid. Total daily usage actually increase a little due to the efficiency losses of the charging and inverter steps of the XW-Pro. But since I time shifted 10 KWHs from the cheap time of day to the high rate time of day, that cut my bill in half again. I was saving over $2.00 a day with that dumb setup. Both the import and export counters in the utility meter were counting about 10 KWHs less per day. My energy use "Duck Curve" was flattened dramatically. There is NO doubt that they can see this. But they have never asked me about it. I am using less power, and exporting less power. How is this a bad thing for them? They offer a cheaper total rate if I had an approved battery system. So I am actually paying them a little more then if I had a "legal" Powerwall installed. So I don't think they want to complain.

But then, about a year later, I finally got my PLC and the power meters working properly. I no longer have to manually start the charging, AND it tracks the extra power and adjusts the charge and export currents every 5 seconds. When the sun is up, I let it export up to just 60 watts until the battery is full. And it exports enough, so that I am powering all the loads, even back in my main panel, and a tiny 20-30 watts goes out to the grid while I am on battery power all night. This is what the energy consumption looked like with just the Enphase solar.
SCE_02-25-20.PNG
My net export was 2.56 KWHs, but it is a huge Duck Curve.
With my system charging properly, it now looks like this.

SCE-03-23-22.PNG
This is 13 months later, after I got the PLC dialed in and working great. Since it is a month later into spring, the solar produced a bit more, so my net export is up a little to 2.90 KWHs. But instead of drawing 1,000 watts all evening and over night, and then exporting 3,000 watts, I only exported 1,300 watts for 1 hour, and less than 600 watts for an hour, and 60 watts the rest of the time the sun was shining as the battery charged. The export spike is after the battery is full. Too bad the graph scales to always fill the height. That big spike is well less than half the export power from before the battery was working.

So yes, they know I am doing something, but it's not hurting them at all.

Now for th last 9 months, I also have DC panels helping to charge the battery bank. The curves don't look much different except the battery becomes full 1 or 2 hours earlier, and even on low production days, I usually make it all the way to sun up now, where without the DC panels, I would go to the 1,000 watts of grid power usage in the morning if the battery fell short.

My NEM 2.0 agreement states I am allowed to export up to 16 amps at 240 volts, but not to exceed 900 KWHs of export in any 30 day billing period. There is no place in the contract where it states when I can import or export energy. But there is also a rule for battery storage systems that you are not allowed to sell energy from a battery that was charged from grid energy. That rule is to prevent people from storing up 100 KWHs of cheap rate power and stuffing it into the grid at the high rate and making money off of the utility using their power. My system never exports more than my solar panels make, so I am not violating that rule either.

If I sell the house, I would either have to install UL rated batteries, or just remove the battery system. I am totally aware of that. And, if my house catches fire from an electrical problem, even if it has nothing to do with my battery system, it is possible my home insurance might not cover it, if they discover the un-permitted battery equipment in the garage. This is my bigger worry. My plan is that the batteries I have will last about 7 to 10 years with how I am running them. At that point, I will likely switch to something like server rack batteries, and then have the system permitted and made fully "legal".

My additional DC charging equipment may also have to go. The panels are CE rated to meet UL specs, so I might be able to get them through an inspection, an they are on the garage, not over the living space. But I currently don't have Rapid Shut Down on the DC panels. And the DC charge controller is a cheapo BougeRV unit from Amazon. It is only rated for "Off Grid" systems. All it does is charge the battery. But... That battery is them feeding into the grid interactive XW-Pro inverter. That is another grey area. I can pull out one Anderson plug, and then the DC system is just charging a spare battery. The other half of the battery bank will stay as the storage battery for the XW-Pro. "Oh, I just plug that in if the grid is down and I am Off Grid." That way the whole DC system is a separate off grid system.
 
The current switch would be located on the home run wire from the roof into the CLP. No direction needed since there can only be solar power passing through that wire from Microinverters into CLP.
Is this intentionally an approximation? It does not take into account the draw from the CLP loads. But in exchange you don’t need power direction.

I assume you are asking about the rectifier (or dump load) failing (hardware failure). This would cause power into the SW to surge past its ability to absorb it so frequency should shift to the level that Microinverters turn off.
Maybe, it depends on what kind of guardrails the SW has on detecting this situation. I think there are going to be a lot of situations where your 1741 inverters will drop off real quick due to voltage or frequency excursions. 1741 is before ride through, it uses the old philosophy of disconnecting quickly if grid is funky. So It probably won’t tolerate much overvoltage.

And in any case, this does not address the central concern you raised with true off-grid (no connection to grid) because of frequent grid qualification by the Microinverters…
The weird thing is that the official anti-islanding algorithm used my Enphase should not be this sensitive. And note because some of these techniques encourage feedback between multiple inverters the behavior may depend on how many you have installed.

1695894934255.png

https://support.enphase.com/s/quest...tation-for-the-inverters-antiislanding-method

https://gridworks.org/wp-content/uploads/2020/03/PGE-generic-response-model-v12-3.23.pdf
 
Officially... No, my local AHJ and Electric utility have not inspected/approved my Schneider XW-Pro install. I did all of th wiring to code, and actually a bit above with the wire being 1 or 2 gauge steps heavier than required. The only reason I have not had it inspected is that by battery bank is not UL listed. I am using repurposed Chevy Bolt cells.
View attachment 169856
My net export was 2.56 KWHs, but it is a huge Duck Curve.
With my system charging properly, it now looks like this.

View attachment 169857
I use much less power than you (~12kWh/day) but your system is a reference for exactly how I’d like my ultimate system to perform.

With an Conext SW with AC input tied to the grid, I can get close. Since the SW will not be able to absorb 100% of peak-solar production, I’ll have a modest export ‘hump’ whether battery is still charging or not (which can be reduced using an external switched battery charger).

You have the option to ‘go official’ with your Conext XW (with a new battery) while the SW is not on the CEC list, so it’s less clear to me whether I’d have that option.

If the SW ever makes it onto the CEC list (or regulations come out making it clear that spurious export below some threshold does not require utility review or permission once the new era of dual-channel metering is upon us), that’s probably the direction I’ll go (or something similar).

That’s all I’m the context of moving my grid-tied solar string from main service panel to SW (either AC output or generator input) which allows excess grid-tied Microinverter to pass through the SW for export to grid.

I was hoping the SW could AC-couple well-enough to also manage / accept that Microinverter string power even when off-grid (no grid connection to AC input), but that Sy be a pipe-dream.

If I converted my 3.5kW array to a DC-coupled string connected to an SCC (or an AIO) that would be easy - DC-coupling is nicely self-limiting once the SCC enters CV mode.

I’d then be able to draw some energy from grid to charge the battery only on overcast days.

But sounds like the current generation of AC-coupled off-grid inverters is not ready for prime-time.

Hoymiles Microinverters have a zero-export solution through PLC communication (from their gateway communicating with an energy meter) so switching to a different brand of Microinverter may be an avenue to stay AC-coupled.

If on-grid, this would allow zero export and if off-grid, the newly-introduced Hoymiles hybrid inverters will hopefully be able to successfully couple with their Hoymiles Microinverters.

I’ll need to wait for things to shake-out before I can finalize my decision but here is a ranking of my concerns about how the ‘rules’ can change by phase:

NEM 1.0 (at least another 7 years for me, hopefully 13):

-about the worst they can do under the rules of NEM 1/2 is to drop midday rates under $0.10 while cranking overnight rates up over $0.40. Whether I can get away with being unofficial like you or I need to get AHJ approval for a small off-grid system to take my ~300W of average load (fridges) off-grid, I should be able to survive any change in rates to that extreme.

-means-based rates - as long as NEM1/2 customers get credited at the same rate they are charged, this amounts to the same thing as above.

-increase in Minimum Monthly Bill including ‘Solar Tax’ (Minimum Monthly Bill computer based on kW of installed solar, possibly also means-based): there would be no escaping this save to decommission, bring as much consumption as possible off-grid, and return to be a ‘regular-ol’ electricity consumer’ (which would only avoid the Solar Tax bit).

Successor Tarrif (NEM 3 or perhaps it will be NEM 4.0 by then):

-Export rates can now be dropped to $0 or even go negative, so this is where having a battery to absorb all AC-coupled energy that otherwise would get exported (or dumping / throttling it) will be a necessity.

-Acceptable Solar Tax: if the cost to stay on-grid is not prohibitive, an AC-coupled off-grid solution such as what the Conext SW can deliver should suffice. At most, you might need to use a gateway-controlled zero-export solution such as what Hoymiles is currently offering. This scenario means the utility knows you have grid-tied solar and are controlling / throttling it to avoid export during times it is disadvantageous to export. Highly likely that utility review / approval
will be required in this case and unclear what new regulations / requirements may be involved. It’s possible that only the panels, racking and home-run wires can be reused and the rest of the system (Microinverters, hybrid-inverter, battery) will all need to be updated / upgraded.

-Prohibitive Solar Tax: this is actually the scenario that caused me to start this thread. This means decommissioning and becoming a non-solar customer from the utilities perspective (meaning no AC connection from grid to AC input / True Off Grid). DC-coupling would make this straightforward, as I stated, but using frequency-shift-based AC-coupling with True Off Grid sounds problematic, certainly today and possibly forever. A closed-system PLC-based AC-coupling solution such as that offered by Hoymiles could be an alternative and hopefully others will emerge over the coming years.

I appreciate all the contributions to this thread - it’s helped me to understand that counting in frequency-shift as the foundation of a future True Off Grid AC-coupled system is probably a mistake and I’m better off to wait for full walled-garden solutions such as that offered by Hoymiles to emerge and be proven (as well as for the future rules to emerge).
 
Is this intentionally an approximation?
Intention. As long as incoming AC-coupled solar power is well below the amount the Conext SW can manage, no need to divert 2kW into the rectifier…
It does not take into account the draw from the CLP loads. But in exchange you don’t need power direction.
CLP loads jus make the situation easier to manage for the SW. any yes, directionality is thus irrelevant (even though my current switch is directional). If the current switch was on the AC output connection between CLP and SW, there would be no hysteria and it would thrash…
Maybe, it depends on what kind of guardrails the SW has on detecting this situation. I think there are going to be a lot of situations where your 1741 inverters will drop off real quick due to voltage or frequency excursions. 1741 is before ride through, it uses the old philosophy of disconnecting quickly if grid is funky. So It probably won’t tolerate much overvoltage.
I can check with my Microinverter manufacturer, but I am coming to the conclusion that True Off Grid AC-coupling to Microinverters is far from simple in the best of cases and is hopeless using old / recycled Microinverters…
The weird thing is that the official anti-islanding algorithm used my Enphase should not be this sensitive. And note because some of these techniques encourage feedback between multiple inverters the behavior may depend on how many you have installed.

View attachment 169859

https://support.enphase.com/s/quest...tation-for-the-inverters-antiislanding-method

https://gridworks.org/wp-content/uploads/2020/03/PGE-generic-response-model-v12-3.23.pdf
I can ask my Microinverter manufacturer (NEP) about deactivating anti-islanding, but I’m not optimistic.

It seems more realistic to hope for a walled-garden full Hybrid + Microinverter solution that supports effective AC-coupling when Off Grid by using communication to the Microinverters from a gateway (PLC, WiFi or whatever).

I’m just a bit ahead of my time (again :)) and need to schedule this change for when the solar is all being pulled-down anyway to replace my aging asphalt-shingle roof…

If I could, I’d wait until my NEM1 period is ending, disassemble and decommission my grid -tied array, then use the panels, racking and home-run wiring to build an AC-coupled True Off Grid system with new Microinverters and matching Hybrid / Off Grid inverter…

But I doubt I can push things out that far.
 
It seems more realistic to hope for a walled-garden full Hybrid + Microinverter solution that supports effective AC-coupling when Off Grid by using communication to the Microinverters from a gateway (PLC, WiFi or whatever).

I've been wondering whether the problem is mainly just Enphase. It's possible a microinverter that depends on PLC to support AC coupling within its ecosystem has less need to play nice interfacing with another brand via the "analog" style interface of UL1741SA.

Hedges seems to not have much trouble with Sunny Island / Sunny Boy combinations (though that's in the same ecosystem, and likely a more complex grid tie inverter than can fit in a micro).

Too bad there isn't someone on the forum with all the hybrid inverters people might be interested in, and a ground mount on which they're willing to wire on a representative deployment of microinverters. Not sure what's a fair donation $$ to compensate for the testing time. It's probably 8 hours of work to swap on 10 panels worth of microinverters (10 is probably a pretty middle of the road size for home systems, and I think a mostly populated branch is enough to shake out a lot of issues) + new trunk.

I’m just a bit ahead of my time (again :)) and need to schedule this change for when the solar is all being pulled-down anyway to replace my aging asphalt-shingle roof…
By that time you can probably just get a hybrid for a decent price and turn on zero-export when on-grid to absorb all the AC coupled power. Also basically every hybrid will have a bigger AC charger than on the SW.
 
I've been wondering whether the problem is mainly just Enphase. It's possible a microinverter that depends on PLC to support AC coupling within its ecosystem has less need to play nice interfacing with another brand via the "analog" style interface of UL1741SA.
That’s what I’m assuming, but it needs to be proven.

I believe at this stage that the Enphase AC-coupled solution must play nice off-grid when it’s Enphase Microinverters and Enphase AC-coupled batteries/inverters. Can’t remember whether I read somewhere that Enphase uses PLC communication within that solution, but it wouldn’t surprise me.

Hoymiles will be the first that I’m aware of that has in-house Microinverters as well as an in-house hybrid that all communicate with the same gateway, so it’ll be interesting to learn from Todd how that solution is working. Todd has already proven Hoymiles’ zero-export solution which relies on protocol communication between gateway and Microinverters (WiFi or some other wireless standard).

And it’s hard to say with Solark. Unclear to me at this stage whether the ‘Solark’ Microinverters are also manufactured by Deye or not. And the current Solark solution for AC-coupling sounds pretty brain-dead when off-grid (ON / OFF only).

If Schneider purchased NEP, now that could make for interesting competition.
Hedges seems to not have much trouble with Sunny Island / Sunny Boy combinations (though that's in the same ecosystem, and likely a more complex grid tie inverter than can fit in a micro).
Has he tested extensively off-grid?
Too bad there isn't someone on the forum with all the hybrid inverters people might be interested in, and a ground mount on which they're willing to wire on a representative deployment of microinverters. Not sure what's a fair donation $$ to compensate for the testing time. It's probably 8 hours of work to swap on 10 panels worth of microinverters (10 is probably a pretty middle of the road size for home systems, and I think a mostly populated branch is enough to shake out a lot of issues) + new trunk.
I believe Todd is planning to do exactly that with his new Hoymiles hybrid, certainly with his Hoymiles Microinverters and hopefully also with his Enphase Microinverters.
By that time you can probably just get a hybrid for a decent price and turn on zero-export when on-grid to absorb all the AC coupled power. Also basically every hybrid will have a bigger AC charger than on the SW.
If a zero-export hybrid will be allowed to have a connection to grid and to solar without triggering an unacceptable solar tax, I’d be all set.

This entire thread was predicated on the eventuality that going off-grid (meaning no connection of grid to AC input) may be the only way to ‘decommission’ on-grid solar to get out of an unacceptably-high solar tax (which may or may not materialize in the future).

I suspect the utilities are not going to let customers with solar panels and a hybrid inverter connected to the grid for any help at all (other than consuming power i.e. powering loads including a battery charger) be treated as “regular-ol’ customers” to get out of paying the solar tax…
 
Hoymiles will be the first that I’m aware of that has in-house Microinverters as well as an in-house hybrid that all communicate with the same gateway, so it’ll be interesting to learn from Todd how that solution is working. Todd has already proven Hoymiles’ zero-export solution which relies on protocol communication between gateway and Microinverters (WiFi or some other wireless standard).
Yes HoyMiles uses Sub-1G or 2.4G for zero export. It's Nordic protocol (for at least one), not WiFi, the radio details are available on some github project.
And it’s hard to say with Solark. Unclear to me at this stage whether the ‘Solark’ Microinverters are also manufactured by Deye or not. And the current Solark solution for AC-coupling sounds pretty brain-dead when off-grid (ON / OFF only).
There's a couple of suspicious things going on with SolArk AC coupling, including the extra recommendation of a minimum DC solar size.

Has he tested extensively off-grid?
I do not know. But note Sunny Island / Sunny Boy are same vendor.
If a zero-export hybrid will be allowed to have a connection to grid and to solar without triggering an unacceptable solar tax, I’d be all set.
This was one of the solar industry's backup plans if the NEM2->NEM3 transition had happened in early 2022 with the 2021 draft decision, according to some people I talked to on reddit back then. Zero export allows you to apply for interconnect under one of the non-NEM interconnects, which presumably means you don't pay the monthly solar tax of that decision.

Zero grid access fee is probably not sustainable long term, so I would have kind of expected NEM4 to show up some time to impose a retroactive fee if that netted out as $0/month grid charge for a lot of customers while also reducing the amount those customers paid in usage. We'll never know. I'd say something proportional to the AC capacity you need for backup seems more fair than DC-side solar tax. I did not read through the proposals / decision to see why DC side solar tax was the chosen method.
 
Yes HoyMiles uses Sub-1G or 2.4G for zero export. It's Nordic protocol (for at least one), not WiFi, the radio details are available on some github project.
They have used 2 different standards, one of which is WiFi, I believe, but the Nordic Protocol may be the only option going forward…
There's a couple of suspicious things going on with SolArk AC coupling, including the extra recommendation of a minimum DC solar size.
Yeah, I don’t expect that to change unless Deye is manufacturing the new ‘Solark’ microinverters as well… (and not terribly relevant to me in any case, since Solark is out of my price league).
I do not know. But note Sunny Island / Sunny Boy are same vendor.
Wonder if they are using a communication protocol or just frequency-shift…
This was one of the solar industry's backup plans if the NEM2->NEM3 transition had happened in early 2022 with the 2021 draft decision, according to some people I talked to on reddit back then.

Zero export allows you to apply for interconnect under one of the non-NEM interconnects, which presumably means you don't pay the monthly solar tax of that decision.
Interested to learn anything more about zero-export solar connection process / requirements. Does this mean that you must inform your utility before connecting a solar zero-export system to the grid (but you are still treated as a regular electricity consuming customer - just your regular monthly bill?
Zero grid access fee is probably not sustainable long term, so I would have kind of expected NEM4 to show up some time to impose a retroactive fee if that netted out as $0/month grid charge for a lot of customers while also reducing the amount those customers paid in usage.
The monthly minimum bill is an absolutely fair way to treat all customers with grid connections whether they consume much electricity or not.

I’d have no problem with the minimum monthly bill increasing to $25, $20, or even $50 as long as it is applied equally to all customers with grid connections whether they consume any electricity or not.

It’s where there is any kind of ‘access fee’ or ‘tax’ charged only to customers with solar panels on their roof that rubs me the wrong way.

If those fees are only charged because you are using the grid as a buffer to get your AC-coupling to play nice, I can accept that. As long as going to a true off-grid setup with no grid support avoids those fees / taxes.
We'll never know. I'd say something proportional to the AC capacity you need for backup seems more fair than DC-side solar tax.
If I can set up a small off-grid system to power my fridge 24/7 with one or two DC-coupled solar panels only relying on the grid to power an AC battery charger during extended periods of overcast weather, why should I have to pay any solar tax at all. I’m already paying the Minimum Monthly Bill as well as any more beyond that due to the electricity I’ve consumed, so why should powering my own fridge off-grid be treated any differently than throwing my fridge away???
I did not read through the proposals / decision to see why DC side solar tax was the chosen method.
The precedent of CPUC based on panel PTC x inverter efficiency goes back to the earliest days of NEM…

I just have a big problem with the idea of any solar tax at all if you have a true off-grid system with no support of any kind from the grid other than occasionally powering a battery charger…

Increase my Minimum Monthly Charges and my TOU electrical charges all you want, as long as those increases are being imposed on all other (non-solar) customers as well.
 
Does this mean that you must inform your utility before connecting a solar zero-export system to the grid (but you are still treated as a regular electricity consuming customer - just your regular monthly bill?
I don’t know what tariff you would be on or if there’s a grid charge. But zero export parallel system you have to tell them about.
They have used 2 different standards, one of which is WiFi, I believe, but the Nordic Protocol may be the only option going forward…
I don’t think the protocol really matters, maybe the radio robustness would. Both are open radios in the sense that you can buy the modules. 900 mhz is probably cleaner these days and penetrates better ?‍♂️
Yeah, I don’t expect that to change unless Deye is manufacturing the new ‘Solark’ microinverters
I think they (or other vendors) can still shop for or customize a microinverter that is electrically compatible wrt anti islanding.
As long as going to a true off-grid setup with no grid support avoids those fees / taxes.
Some places require you to pay for an expensive disconnect and reconnect fee to disincentivize going full off grid. Arguably a disconnect fee proportional to your share of the recent capital investments on infra seems reasonable.

If I can set up a small off-grid system to power my fridge 24/7 with one or two DC-coupled solar panels only relying on the grid to power an AC battery charger during extended periods of overcast weather, why should I have to pay any solar tax at all
If you go fully off grid but keep grid backup you should pay for the T&D and standby generation infrastructure.

If it’s a pretty small amount of load they should probably just exempt it from regulation.

I wouldn’t be surprised if winter electricity rates will be much higher in the future than today.
 
I don’t know what tariff you would be on or if there’s a grid charge. But zero export parallel system you have to tell them about.

I don’t think the protocol really matters, maybe the radio robustness would. Both are open radios in the sense that you can buy the modules. 900 mhz is probably cleaner these days and penetrates better ?‍♂️

I think they (or other vendors) can still shop for or customize a microinverter that is electrically compatible wrt anti islanding.

Some places require you to pay for an expensive disconnect and reconnect fee to disincentivize going full off grid. Arguably a disconnect fee proportional to your share of the recent capital investments on infra seems reasonable.


If you go fully off grid but keep grid backup you should pay for the T&D and standby generation infrastructure.

If it’s a pretty small amount of load they should probably just exempt it from regulation.

I wouldn’t be surprised if winter electricity rates will be much higher in the future than today.
When I talk about ‘True Off Grid’ I’m not meaning Completely Off Grid (I.e. no grid service).

I wold leave many infrequent high-power 240VAC loads such as the electric oven on grid as well as a 240VAC battery charger, so I’d be happy to pay all the charges any other light grid consumer would pay (including whatever monthly minimums all customers at as well as whatever TOU rates).

There is a certain amount of ‘capital investments’ etc… is built into the bundled rates (NBCs) and whatever amount should be paid regardless of how little electricity is consumed should be built into minimum monthly bills.

There should be a way that.customers offsetting a % of their consumption with solar power without relying on the grid (True Off Grid) are treated no differently than customers who have reduced consumption by purchasing more efficient appliances or eliminating some appliances.
 
I’ve been thinking about this some more and have one more crazy idea to throw out there.

Seems to me when we look out to whatever the NEM 3.0 successor may be (or at least how the rules of NEM may be changed going forward) we’re likely to end in one of three buckets;

1/ Export will still be allowed (meaning export won’t be unreasonably-priced in terms of any eventual ‘solar tax’)

2/ Export will not be allowed (meaning too expensive to be worthwhile) but zero-export with grid-connection will be allowed (meaning frequency-based AC-coupling will be a viable option).

3/ The only way to avail a prohibitive solar tax will be to go True Off Grid meaning no grid connection (and no possibility of spurious export).

The first scenario is easy-peazy and the 3rd scenario may never be possible based of frequency-shift (as GXMNow pointed out) so the second scenario is the most interesting alternative to focus on (as well as the most likely to eventually materialize IMHO).

My crazy idea is to always drive a battery charger consuming more 240VAC power than the incoming solar power (similar to what I suggested earlier).

For example, if using a Conext SW limited to 2.4kW / 10A of AC power, once incoming AC solar current in excess of 8A or 9A arrives, a 240VAC charger consuming 2kW (or at least absolute max AC-coupled solar current - 8A or 9A) is turned on so that the incoming AC-current the SW must manage is always ‘within range.’

Both AC-coupled solar and battery charger must get shut down for the day whenever battery reaches a preset nearly-full SOC / voltage but this should allow AC-coupled solar power to contribute to charging the battery even with a hybrid inverter like the SW that has to little charging capacity to manage maximum incoming AC-coupled solar power.

This should work with grid connected to AC-input by avoid the complexity of frequency-shift-based throttling when off grid (as experienced by GMMNow).

Whether it will also work for the 3rd scenario without that connection is questionable (at least today) but the same architecture with both Microinverters as well as off-grid hybrid inverter that throttle / control Microinverters through the use of communication protocols when off-grid (and possibly also disable anti-islanding protections when off-grid) should work as well (ie: either Enphase or hopefully also Hoymiles today).

At just over 8 or 9A of incoming solar power, there would be 2kW of AC power consumed to generate ~1.9kW of DC charge power contributing to ~2.22kW of DC charge power needed to offset that 2kW of AC consumption, so as much as ~322W of net DC power may be getting consumed / wasted, but given the likely over-solar-generation available ~90% of the days, that seems like a mesninglessly small price to pay in the grand scheme of things.

Does anyone see an issue having an off-grid-inverter power a battery charger supplying 85.5% of the DC power being consumed by the inverter to generate that offsetting power (the balance being provided by the hybrid reducing it’s battery charging power)?
 
Seems to me when we look out to whatever the NEM 3.0 successor may be (or at least how the rules of NEM may be changed going forward) we’re likely to end in one of three buckets;

1/ Export will still be allowed (meaning export won’t be unreasonably-priced in terms of any eventual ‘solar tax’)

2/ Export will not be allowed (meaning too expensive to be worthwhile) but zero-export with grid-connection will be allowed (meaning frequency-based AC-coupling will be a viable option).

3/ The only way to avail a prohibitive solar tax will be to go True Off Grid meaning no grid connection (and no possibility of spurious export).
One other scenario I forgot to mention, let’s call it scenario:

1.5/ Export is allowed by expensive (as in -$s per each kWh of export, at least during the peak solar production hours of the day when that export is not wanted.

The solution I outlined in the previous post for scenarios 2 and 3 can also support this scenario 1.5 with the addition of a transfer switch controlled by a timer.

Hours when export is compensated the AC-coupled array is connected to the grid (AC input); hours where it becomes unattractive to export, the timer switches the AC-coupled array off the grid and onto the off-grid system (AC output).
 
There are many ways to try and get around the issues, but the rules are going to keep changing. And it's not just the electric utilities.

In my community, a few years back, the water company was all worried we were going to have a major shortage and they told everyone we need to reduce water use. They even put restrictions on use lawn sprinklers, and most of the lawns all turned brown. Many people installed low flow faucets etc.

Then 6 months later, the water agency applied for a big rate increase because their revenue tanked. Of course, the state gave them the rate increase, so now while using 30% less water, we have to pay more than before the whole crisis hit. Hmmm. So now the water agency makes more money for supplying less water.

Back to electricity...
Even looking ahead at the utility commission reports and filings, it is tough to know what rules are going to end up being put in place, but the one thing for sure, they will find a way to make more money. For today, we can look at our electric bills and see what they are doing. I used my electric bill to decide that I really needed to do self consumption with the battery bank.

I feel the best we can do is to try and produce and use as much of our own electricity as possible. Any power we need to get from the grid is going to cost more and more, and any credit we get for exporting to the grid is going to keep going down. With the current time of use billing, it only makes sense for me to not buy any power during the higher rate times. But under my current rules, I do still get export credit. So once my battery is full, letting the extra power export to grid is still my best action.

Now if they change the rules to where export is anywhere from where I am now, down to no credit at all, it becomes less beneficial, but it still won't hurt, so it is still easiest to just let the export happen. Again, this is for my situation, and I do have an NEM agreement.

But if I didn't have the NEM agreement, or if they change the rules in the future and either start charging a penalty for export or ban export all together, then I will need to reprogram my system a bit. Stopping all export from the XW-Pro coming from battery is easy. My current program has offsets written in already so I can shift it to block it from exporting most of the time. But I will still have short term export spikes when a large load shuts off. Getting rid of that type of export would be extremely difficult with the inverter connected to the grid and loads that are not in the backup panel. The only realistic way I can see to completely stop those export spikes would be to have every load powered by the battery inverter. Then have the inverter running off grid. To make this work, I would need at least double the inverter capacity to be sure every load in my home could be covered from it. My existing 36 KWH battery bank might be enough, but it would be on the edge. I have shown over this past year that my PV solar panels can make enough energy to run my home about 90% of the time. Maybe even 95%. To cover that little bit more I might need, I could have a charger that would run off of grid power that turns on when the storage batter falls below a threshold level. Something like a Chargeverter that can push up to 100 amps into the battery would probably cover my needs for those days when I need some extra energy. The absolute worst I fell behind all last year was only by about 10 KWHs on the third day into horrible cloudy weather. So the Chargeverter would need to be setup to run for a bit over 2 ours at the cheapest grid time to get that energy back into the battery bank.

After messing with my system for 4 years, if grid export became banned, I would maybe try adding the Enphase Energy Consumption CTs and loading the zero export grid profile. That uses the data from the grid side power meter to send commands to the inverters to ramp down their output to ensure they don't export to grid. But how fast is that? If my Central A/C is running at 2 pm while my battery bank is full, the Enphase inverters will be supplying their full power to try and cover the A/C load. But then when the compressor shuts off, the grid consumption CTs will show current exporting, then the Envoy will send the command to the micros to reduce their output. My code takes 10 to 15 seconds to balance the inverter output to the load. Is this delay an issue? Can the Enphase setup do it faster?

If the grid operators are going to call you out for imbalanced current and/or spurious export for less than a minute, then I feel the only real solution will be only DC coupled PV solar. Run the system completely off grid with the grid input as just a generator when the battery bank runs too low. I do not see any other way to completely eliminate any chance of export pulses when the amount of load changes. There is a work around for the A/C compressor, but it is not ideal. Have the thermostat command a small PLC or Pi etc. that you want A/C. Then the controller can send the compressor start signal to the contactor. There will be a surge of power drawn from the grid. Then the controller ramps up the inverter power to cover the load and grid consumption drops back to zero. But then when the thermostat commands the A/C to stop, it actually tells the system controller. This can tell the inverter to ramp down the power being pushed into the house. For a few seconds, the A/C compressor will be running on just grid power. Then the controller can shut off the compressor contactor and the grid draw will again drop to zero power.

But how many devices can you operate like this? You might be able to do it with a microwave oven. But it might get tricky. When the cook timer get's down to 5 seconds remaining, have the inverter begin ramping the power down so that the grid is running it for the last half a second before the over stops. But if you use a lower power level, do you do this for every cook pulse? What about turning off a hair dryer or a toaster? It's just not practical to do this for everything.

Without going fully off grid, I think we need to accept that there will be some export from time to time. Even while my system is at "zero export", I am actually exporting 1.9 amps on L1 and consuming 1.8 amps on L2 while the XW inverter is actually pushing 5.8 amps back to the main panel to run the loads. This offset is cause by my L2 loads drawing 3.7 amps more than my L1 loads. My utility meter does not appear to be able to tell there is an imbalance like that. As far as I can tell, it is only reporting the total watts from both legs. So this is reported as just a 0.1 amps of export at 240 volts, or about 24 watts going out to the grid. I am intentionally adding the 0.1 amp offset to ensure I am exporting and not importing. I could easily set this to have the total be closer to zero, or even a little grid consumption, but unless I go up to over 1.9 amps of import on L2, L1 would be exporting. To get around that, I would need separate 120 volt inverters that could be programmed independently on each leg.

For now, I know my setup is working well with the rules I need to deal with. I am certainly willing to help work out ideas for changing rules as I may need to deal with them in the future. But I think w will all agree that the solutions won't be easily solved with a cheap inverter. I almost went with a Schneider SW. But I was turned off because I knew I would end up having to push energy towards the grid for my situation. The SW can't do that. The AC coupled solar limit also became an issue. I knew I needed at least a 4,000 watt inverter to be able to accept the back feed from my Enphase system. I looked into putting together a system of chargers and inverters with a controller to adjust things. Buying devices off Ali Express could have saved a fair bit of money. But the thought of pushing around 4,000 watts to and from a battery bank and into my house wiring, I decided, even if it was not inspected, I wanted a UL listed system to handle all that power.
 
Back to electricity...
Even looking ahead at the utility commission reports and filings, it is tough to know what rules are going to end up being put in place, but the one thing for sure, they will find a way to make more money. For today, we can look at our electric bills and see what they are doing. I used my electric bill to decide that I really needed to do self consumption with the battery bank.
Yes, and with NEM 3.0, it’s clear rate changes will slowly and inexorably move more and more in that direction.

Under NEM1/2 export credit must match consumption rates, but there is nothing to stop them from moving some or all of overnight rates to peak or reintroducing a ‘partial peak’ rate that increases overnight rates above (off-peak) rates during peak solar production hours.
I feel the best we can do is to try and produce and use as much of our own electricity as possible. Any power we need to get from the grid is going to cost more and more, and any credit we get for exporting to the grid is going to keep going down. With the current time of use billing, it only makes sense for me to not buy any power during the higher rate times. But under my current rules, I do still get export credit. So once my battery is full, letting the extra power export to grid is still my best action.
Absolutely. In my case, I just want to be sure my net consumption during peak hours is negative (meaning net export exceeds net import during peak period) all through summer period and as close to the minimum solar production months of December/January as possible.
Now if they change the rules to where export is anywhere from where I am now, down to no credit at all, it becomes less beneficial, but it still won't hurt, so it is still easiest to just let the export happen. Again, this is for my situation, and I do have an NEM agreement.
Which means that as long as annual kWh of export exceeds annual kWh of import, you will get compensated at Net Surplus rates (one aspect I doubt will ever change).

So the only scenario where I see export not making sense anymore is where a new solar tax gets introduced that is so prohibitive it’s more attractive to go off-grid for as much covdunotion as possible and go back to being a regular consumer of electricity subject to the same rules and rates as all non-solar consumers of electrical power.
But if I didn't have the NEM agreement, or if they change the rules in the future and either start charging a penalty for export or ban export all together, then I will need to reprogram my system a bit. Stopping all export from the XW-Pro coming from battery is easy. My current program has offsets written in already so I can shift it to block it from exporting most of the time.

But I will still have short term export spikes when a large load shuts off. Getting rid of that type of export would be extremely difficult with the inverter connected to the grid and loads that are not in the backup panel.
While we are on-grid and under NEM agreements, offset of grid-side loads is easy and undetectable during daylight hours.

If we are forced to go off-grid and can no longer export even spurious amounts when loads switch off without getting into trouble, offsetting grid-side loads is problematic.

In my case, it’ll only be infrequently-used high-power loads that I leave on-grid so their overall contribution to my annual consumption will be small and I’ll just accept to have to pay when I use them.

We’ll be subject to minimum monthly bills in any case, so we might as well do something with that amount of power we’re forced to purchase :).
The only realistic way I can see to completely stop those export spikes would be to have every load powered by the battery inverter. Then have the inverter running off grid.
Or just the loads that dominate your daily / annual covdunotion pie…

Not worth it to me to try to move everything off-grid.
To make this work, I would need at least double the inverter capacity to be sure every load in my home could be covered from it.
Over 50% of my annual consumption is my 5 fridges / freezers averaging 300W 24/7 and never exceeding 650W. So I’m going after that first. Add in lights and the most heavily-used kitchen circuits / appliances and I’d need to move up to an inverter in the 5kW-per-leg class.

The 3kW electric oven would be next but I don’t think it’d be worth it…
My existing 36 KWH battery bank might be enough, but it would be on the edge.
My daily consumption is only ~12kWh (out of which over 4kWh is during solar priduction hours) so my 14kWh battery is more than enough to get through the night.

I’d need a bigger battery to get through a day of overcast weather without having to pay far an energy boost, but again, I don’t think it’s worth it (plus, my eventual V2L-capable EV will give me that extra battery capacity ‘for free.’ :)
I have shown over this past year that my PV solar panels can make enough energy to run my home about 90% of the time. Maybe even 95%. To cover that little bit more I might need, I could have a charger that would run off of grid power that turns on when the storage batter falls below a threshold level. Something like a Chargeverter that can push up to 100 amps into the battery would probably cover my needs for those days when I need some extra energy. The absolute worst I fell behind all last year was only by about 10 KWHs on the third day into horrible cloudy weather. So the Chargeverter would need to be setup to run for a bit over 2 ours at the cheapest grid time to get that energy back into the battery bank.
Precisely - that is the ‘energy boost’ I was referring to (which can also be supplied by an EV).
After messing with my system for 4 years, if grid export became banned, I would maybe try adding the Enphase Energy Consumption CTs and loading the zero export grid profile. That uses the data from the grid side power meter to send commands to the inverters to ramp down their output to ensure they don't export to grid. But how fast is that? If my Central A/C is running at 2 pm while my battery bank is full, the Enphase inverters will be supplying their full power to try and cover the A/C load. But then when the compressor shuts off, the grid consumption CTs will show current exporting, then the Envoy will send the command to the micros to reduce their output. My code takes 10 to 15 seconds to balance the inverter output to the load. Is this delay an issue? Can the Enphase setup do it faster?
Export being ‘banned’ would only apply to the case you have decommissioned your solar. If you are still a solar customer export would not be banned, merely discouraged.

So your above solution should be fine if you are still a sola customer and connected to grid while generating.

If you are not a solar customer and have no grid connection to your off-grid power (other than your battery charger / booster), your off-grid system needs to be working off-grid, so either it works fine or you need a new system…
If the grid operators are going to call you out for imbalanced current and/or spurious export for less than a minute, then I feel the only real solution will be only DC coupled PV solar. Run the system completely off grid with the grid input as just a generator when the battery bank runs too low. I do not see any other way to completely eliminate any chance of export pulses when the amount of load changes.
It’s an interesting question. When your XT is working during a grid outage (backup power), would it be able to cycle a 3000W electric oven element?

I’ll wrap it up here by saying this is a relatively easy problem to solve if you are a known generator with a grid connection (regardless of however severe the disincentives to export become).

It’s a significant challenge if you have decommissioned and are a standard energy customer with no grid connection to your off-grid power system other than a battery charger (at least with AC-coupled solar power, you’ve helped me to understand).

And the less-clear ‘grey area’ (from whee we sit today) is whether there will be options to decommission with off/grid solar power allowed to maintain a connection to grid subject to clear limits in allowable spurious export.

I started this thread primarily from an interest to explore what types of hybrid systems would also support this capability in case it ever becomes necessary / sensible.

As long as any solar taxes charged can be largely offset by credit for net export (at NSC rates or whatever), staying on-grid will make the most sense.

But once you can supply most of your own needs and are willing to pay for whatever boost you need in winter, it becomes cheaper to just decommission, add mode solar panels and forget about the power that is getting wasted in summer months…
 
Hedges seems to not have much trouble with Sunny Island / Sunny Boy combinations (though that's in the same ecosystem, and likely a more complex grid tie inverter than can fit in a micro).

By that time you can probably just get a hybrid for a decent price and turn on zero-export when on-grid to absorb all the AC coupled power. Also basically every hybrid will have a bigger AC charger than on the SW.

The proportions of Sunny Boy Smart Energy look interesting, if they do support pairing it with ABU (200A transfer switch and auto-transformer). It is a high frequency hybrid like the other brands popular on the forum, but is packaged as a PV + battery inverter that AC couples to the grid, so an extra box will be required for backup or stand-alone.

The largest model SBSE 11.5 will accept 23kW of PV, deliver 11.5kW of AC and simultaneously charge HV battery with 10kW. Or draw from battery as needed.
Unfortunately it has no surge capability, so probably about 2.5kW motor is the largest it would start.

With PV DC coupled internally, it doesn't have to ramp down AC coupled GT PV gradually, so better able to avoid export when blending its power with the grid.

The 32kWh BYD battery it can work with is about $18k. Still haven't figured out what BMS for DIY will talk with SBS or SBSE.
 
Back
Top