Adding storage to my Enphase system

[Hedges]

Of course doesn't make sense to charge battery from AC and inverter battery to AC at the same time. Power is flowing one way or the other at any given instant. Does XW have SCC built in? If so, can it do what you want?

External shunt, maybe plus passive filter to reduce ripple, should make it easy to read with PLC.

 

[svetz]

On the dark start, could swapping in one IQ8 microinverter resolve that issue? You already have the Envoy, but you might not be able to convince the Envoy to put the IQ8 into islanding mode without the Enpower (you don't need the Enpower as you're AC coupled, so there might be some profile for it). Might be worth a call to see what Enphase says. Might need to call Schneider too, they might be expecting to set the phase of the islanding rather than synchronizing to it on that port.

 

[400bird]

Then I still would need 9 of the Tigo RSD units at $30 per panel as well. The cost is a wash. Q cable vs MC4 cables inline fuses, combiner and breakers.

They have the dual disconnects for $45. That saves a little. Also, I'd have to double check the regs, but I think you just need to get under 80 or 50 volts? I don't believe you need to shut down every panel, just the voltage leaving the array.
Maybe, I might be remembering incorrectly.

In the end, I am looking at an all up cost of about $3,600 to add the 2,880 watts of panels. Shipping the cheaper 400 watt panels from San Tan would have been over $400, and those only had a 1 year San Tan warranty. AltE had some nice 425 wat panels, but again, shipping was $600. I am only at $1.25 per watt. That is still very low. By the time I buy the little things locally, it will likely hit $1.50 a watt. One big part of the savings for me is the fact I already have an Enphase iQ system. So I have the combiner and Envoy and outdoor legal disconnect. All things that basically would need to be duplicated if I went DC coupled.

Have you checked CED Greentech? They don't post prices online, which doesn't give me the greatest hope, but they have locations all over the state.

But, overall I'm in about the same spot. I'm looking to add some more panels, but I don't have a great location. I'm short on roof space. I might end up with some ground mounts in the back yard, not ideal. Plus, $3k+ I'm not sure if I want to spend that this year.

 

[GXMnow]

Of course doesn't make sense to charge battery from AC and inverter battery to AC at the same time. Power is flowing one way or the other at any given instant. Does XW have SCC built in? If so, can it do what you want?

External shunt, maybe plus passive filter to reduce ripple, should make it easy to read with PLC.

No, the XW-Pro is just the Inverter/Charger. They currently have 3 different solar charge controllers. The 150 volt 60 amp which I was looking at, then the 600 volt 80 or 100 amp units.

They make a DC version of the PZEM power meter I am using to measure the house current. The -017 uses a shunt and measures voltage and current, calculates the watts, and kilowatt hour just like the AC version, just no Hz or power factor. I could add one onto my DC buss to the battery banks and then the PLC can read it on the ModbusRTU serial network.

 

[GXMnow]

On the dark start, could swapping in one IQ8 microinverter resolve that issue? You already have the Envoy, but you might not be able to convince the Envoy to put the IQ8 into islanding mode without the Enpower (you don't need the Enpower as you're AC coupled, so there might be some profile for it). Might be worth a call to see what Enphase says. Might need to call Schneider too, they might be expecting to set the phase of the islanding rather than synchronizing to it on that port.

As far as I can tell from all of their online data, the grid forming function in the iQ8 microinverters is all controlled by the Enpower switch unit. It needs to signal the grid status to them. I had thought about the iQ8s, but they are also quite a bit more expensive right now. The efficiency is no better than the 7's, so there is no other real gain. The cost of the Enpower would really add to the budget. And I think I would also need to swap out my iQ Combiner.

The dark start issue has not been a real problem yet. The one time I was nervous about it was when we had the power fail in the middle of the night, after it had already discharged all I was planning to run out of the battery, when I only had the first bank, so the remaining 40% was just 7 kwhs to go. And then the next morning when the sun did peek through the clouds, only 5 of my iQ7's would produce. And we had smoke and clouds in the air. AND to top it all off, my generator would not hold rpm (frequency) stable enough to run into the XW to charge the batteries. I ended up plugging the refrigerator into the generator, and then plugging my 600 watt charger into the generator and having that push some charge into the battery. An hour on the phone with Enphase, and we did get the rest of the iQ7's up and working, but the little sun we were getting only had an hour left. I shut down the generator at 10 pm, that thing is noisy. Got the battery state back up to nearly 60% remaining. With basically the refrigerator and an alarm clock on the XW, I figured it would make sunrise. And the power came back on at 1 am.

With twice as much battery now, and all of the microinverters working during a grid outage, I don't think I will get into that kind of a situation again. But I want to at least build in an easy restart if it does happen. They keep talking about "Public Safety Power cuts" if there is a threat of wild fire near us. So far they only did it once, and it was only for 5 hours. But with what PG& did up north, I want to be ready.

I have not coded it yet, but I have a flow chart for the PLC for when the power is out. I already have a relay in the box with the power meters that will tell the PLC if the grid is down. That way, it does not depend on being able to read data from the XW or the power meters. I don't want a comm error to be mistaken for a grid outage. The first thing it will do when the grid goes down, is it will set the charge rate above what my solar array can produce, so it will get as much charge as it can. The last power failure we had happened when I had the charge rate t just 18% and it kept frequency shifting to curtail the iQ7's. It should have been charging at over 50 amps instead of turning them off. Thanks to some manufacturing tolerances, only 5 of them would shut off, the other 11 did stay up. It was always the same 5, so my guess is their internal frequency measurement is reading a tick higher than the 11 that stayed running. The next thing to do is set it to charge up a little higher, I can push it to 95% and not be too worried. It would take a real good sunny day to get it all the way up, but if it does, why have it stop charging? Then I will set the low voltage cut off down lower as well. I can easily take the batteries down to 30% and still have reserve for a cold start if needed.

If it is a LONG outage, and we have really crappy sun, and I am stupid and don't shut off enough stuff over night, and it does get down to where the XW hits low batt shut down, I will have the DC to DC converters keep the PLC and Gateway alive until the BMS shuts off the battery bank. That should last literally weeks. So it would then just be a matter of lowering the low batt cut off when the sun is able to charge the system. If nothing else, I could connect a button to one of the inputs on the PLC that will send the commands to the XW to lower the cutoff voltage and restart the inverter. Or have a photocell and time of day window do it automatically.

When the grid comes back up, the PLC will just set all of the things it changed, back to my normal on grid settings. The only odd part would be if the battery voltage is still too low. I am already reading that, so I will just have it wait to change that until the battery has charged enough so it doesn't shut itself off.

 

[Hedges]

They have the dual disconnects for $45. That saves a little. Also, I'd have to double check the regs, but I think you just need to get under 80 or 50 volts? I don't believe you need to shut down every panel, just the voltage leaving the array.
Maybe, I might be remembering incorrectly.

Earlier RSD requirement was < 30V outside array.
Later RSD requirement is < 80V between any two conductors, inside or outside array.

Another guy with a ground mount planned appeared to think he could put RSD just on the end panels. I asked a question, saw no reply and can't find it now.

I don't think RSD can block voltage of other panels from reaching wires to inverter. To do so, it would have to hold off 600V.
The SolarEdge optimizers, at least, drop voltage to 1V per panel. That may be a buck/boost circuit, 1V to 60V someone mentioned. If RSD only on the end panels, it would reduce their contribution to 1V but all panels in the middle would still deliver their full voltage. I think the string might drop from 400Voc to 302Voc or something like that.

 

[GXMnow]

They have the dual disconnects for $45. That saves a little. Also, I'd have to double check the regs, but I think you just need to get under 80 or 50 volts? I don't believe you need to shut down every panel, just the voltage leaving the array.
Maybe, I might be remembering incorrectly.

Have you checked CED Greentech? They don't post prices online, which doesn't give me the greatest hope, but they have locations all over the state.

But, overall I'm in about the same spot. I'm looking to add some more panels, but I don't have a great location. I'm short on roof space. I might end up with some ground mounts in the back yard, not ideal. Plus, $3k+ I'm not sure if I want to spend that this year.

I saw the Tigo dual disconnect units, and if I did the larger 6 x 72 cell panels I would be able to just use 3 of those for the array 2S3P. The dual boxes connect the 2 panels in series in the on position. With the 9 x 60 cell panel 3S3P array, I would need 6 boxes, 3 duals and 3 singles. So the $50 per panel is still a decent estimate. With how the pricing works out, I think the addition will be either all Enphase, or all DC Schneider. And I am really leaning towards the all Enphase setup. The cost is so close, and it keeps all of my solar monitoring on a single app.

The main reason I was looking at the DC charging was mostly to do with optimizing the battery charging to get them to use as much of the solar as possible beyond what the house was consuming. I really did not expect my PLC setup to work as good as it is. That is a real game changer. I really want to send a fairly nasty e-mail to the Schneider tech guys, with a video of it controlling the system. "See, this is what the Gateway should be able to do" With no extra hardware at all, I had the PLC making it charge near perfectly. And they already have it able to use the WattNode to control the export power, they could also have that add in during charging like I am doing with the external power meters. As I told them, it is a few lines of easy code. The XW-Pro should natively have this functionality for AC coupled systems. The WattNode box is a 3 phase unit, the third CT could even measure the power coming from the solar inverter, so the Enphase App could show it as an input on the house icon in Insight.

Are they really just worried about losing sales of the MPPT units? Or is it something else? They have "Smart Charge" in the SW series.

 

[Hedges]

The dark start issue has not been a real problem yet. The one time I was nervous about it was when we had the power fail in the middle of the night, after it had already discharged all I was planning to run out of the battery, when I only had the first bank, so the remaining 40% was just 7 kwhs to go. And then the next morning when the sun did peek through the clouds, only 5 of my iQ7's would produce.

A load-shed relay should be a robust solution. So long as you can reliably determine SoC with enough accuracy to know enough power remains to run the inverter no-load until sunup. You just open a relay to disconnect all loads, leave GT PV connected. (to get there on less power, go to a sleep or inverter off state while waiting.)

 

[GXMnow]

Earlier RSD requirement was < 30V outside array.
Later RSD requirement is < 80V between any two conductors, inside or outside array.

Another guy with a ground mount planned appeared to think he could put RSD just on the end panels. I asked a question, saw no reply and can't find it now.

I don't think RSD can block voltage of other panels from reaching wires to inverter. To do so, it would have to hold off 600V.
The SolarEdge optimizers, at least, drop voltage to 1V per panel. That may be a buck/boost circuit, 1V to 60V someone mentioned. If RSD only on the end panels, it would reduce their contribution to 1V but all panels in the middle would still deliver their full voltage. I think the string might drop from 400Voc to 302Voc or something like that.

The way I read the RSD requirements, it really does need to shut of every panel. Even 60 cell panels will creep up to just over 40 volts when unloaded. Just 2 in series is then over 80 volts. I just checked the data sheet. The Qcells 320 watt panels I am looking at have a VOC of 40.56 volts at 25C. And that can go up 0.27% for every degree C colder. Wow, the warranty on these panels is to still produce 85% at 25 years, 90% is way out to 15 years. These are clearly better panels in many ways over my existing SilFab 300M's. Having them flat on my garage roof will reduce their Fall to Spring output a bit, but in Summer, when I need the juice for the air conditioner, they will get almost an extra sun hour each day.

The VOC of my old panels is only rated at 39.85 volts. Maybe I should change out the 8 on my lower roof for the new Q-cells, and use the SilFabs n the garage? Then the RSD could be on every 2 panels.... Nah, I think the Enphase iQ7+ on the new panels is the winning move.

 

[Hedges]

A switch (transistor or mechanical) simply shorting string's PV+ to PV- would drop all wires to zero volts (+/- depending on exact Isc of each panel). (better use a blocking diode so it isn't shorting out capacitors.) That might work, right until a wire gets cut or burned open. Then full Voc would reappear.

There used to be a quad RSD. I was thinking of getting that and plugging 3s of my AP120 into each input. Just two needed for a 24 panel 600V string. Haven't seen those in a while.

 

[Ozark Tinkering]

I'm following this from 30 pages back. This thread is packed full of important information and experience. The prospect of doing what @GXMnow is what brought me here. I knew there had to be a way. Looks like maybe more than 1 way too.

 

[GXMnow]

So we had two bad days of clouds. It's funny how the battery delays the effect of the poor production.

Back on Friday, the production was lousy, but because the battery bank was fully charged the day before, th system still ran off battery to almost 10 am. Solar production was off and on, and I also forced it to charge a little from grid power. So the "Super Off Peak" time still shows a fair bit of consumption. I did manage to store up 4.5 KWH into the battery before 4 pm, so it did run the highest peak rate off battery yet again. But that was all the battery had.

So then yesterday, the house was running on just grid power up to nearly 10 am. Solar production was actually a little better, but still bad. The system managed to push over 6 kwh into the battery, and the clouds cleared a bit, so at 4 pm, the solar was running the house and exporting a little. I had tried to change the grid sell block time in the XW inverter as it was going into sell at 1 pm when the solar would drop out from the clouds. Something in the XW or the Gateway glitched. It didn't set any error, but it was not confirming my settings change. Normally, the field you change will turn green after it sets. This time it went orange?? In the evening, I notice a small amount of current was being imported. It did not set the sell block time properly, and the setting reverted to noon. But when I went in at that time, it works fine, and I fixed the setting. Because of that glitch, 2 things happened. The first one is that I actually did end up buying some of the peak rate grid power, but it still only totaled 0.46 kwh. And since it was not exporting any power to the main panel, there was more left in the battery at 9 pm, so it stayed running on battery to after midnight.

Instead of a net export, Friday ended up a net import of 3.06 kwh and Saturday imported 6.79 kwh. Not too bad considering my solar production was half of normal. Today it is cool and back to being sunny. The production is mirroring what it did last year, hitting 3,800 watts in mid day again. By 1:45 pm, it has stored 12.9 kwh into the battery, and it's still going strong.

I am a little concerned about the setting that didn't change properly from the Insight Local web UI. Did it have anything to do with the PLC talking to it at the same time? Once I was able to change the setting again in the app, it all seems to be working fine again. When the setting went orange, it was not letting me select it again. I ended up closing the tab with Insight, reconnecting on a new tab and it recovered. But I had waited over an hour to do it because I didn't want to just make another open connection. It does look like the port did close. It probably has a time out on it.

As I was looking at the dashboard, I saw the charge power drop from 40 amps down to 14 amps. The battery voltage just exceeded 56.65 volts, so it went to the 10% max charge rate. The new(er) washing machine is really efficient. It just went into spin cycle, and it is still drawing under 260 watts. The old washer was over 500 watts. The real pig now is the dishwasher. I see that hit over 1,200 watts when it's running. And that was "Energy Star" rated back 9 years ago when I bought that.

I have been watching a few videos today abut installing additional Enphase panels. The one part I want to see is the array builder. Just connecting them, they will work and make power, but to have the new panels show up in the graphic with their approximate location, it looks like I will nee the paid version of Enlighten Manager. The Enphase Toolkit will just ad the serial numbers o the system.

With the current price and the free shipping, I am debating just ordering the panels right away. I can get free or cheap chipping on everything else, it is just the panels that are such a pain because of their size. Then I can just get a few iQ micros, hook up 4 panels for now, whatever. Maybe mess around with a charge controller, or even try welding with the panels.

 

[Shimmy]

I really want to send a fairly nasty e-mail to the Schneider tech guys, with a video of it controlling the system. "See, this is what the Gateway should be able to do"

The way they operate, too many options makes it hard to support. Completely agree that it should be a standard profile though.

Are you using a separate transfer switch, or just the internal switching? The 60A limitation on-grid is a little limiting for my place with 7kW of microinverters, so I am contemplating a 200A ATS with the inverter's output hanging on one of the ATS source inputs.

 

[Hedges]

7000W / 240V = 29A

 

[400bird]

7000W / 240V = 29A

I was thinking the same exact thing.

I'd be more concerned about the 7kw on a 6.8kw inverter when the grid is down. I don't remember Schneider's specs for max AC coupled PV.

 

[GXMnow]

The way they operate, too many options makes it hard to support. Completely agree that it should be a standard profile though.

Are you using a separate transfer switch, or just the internal switching? The 60A limitation on-grid is a little limiting for my place with 7kW of microinverters, so I am contemplating a 200A ATS with the inverter's output hanging on one of the ATS source inputs.

Just adding "Smart Charge" to the XW-Pro would be a big start.

I am only using the internal transfer relay. With my current setup, I am well below the power limit. My grid input to the XW-Pro is only coming from a 20 amp breaker, so that is all the transfer relay would ever have to carry.

Are you planning to backup everything in the building? If so, you may need to stack two XW-Pros and split the micros across the two of them. They do say they can handle 6,800 watts of AC coupled solar, but I don't like cutting it that close. Is your system a true 7,000 watts of solar inverter output, or is that DC panel rating? One XW-Pro is able to supply up to 60 amps in pass through from the main panel, the solar could add another 6,800 watts (28 amps) at Solar Noon, and the inverter can also add it's 6,800 watts (28 amps) to it when in "Grid Support" mode.

When it goes off grid, it can supply 6,800 watts from battery, and any solar can add to that as well. If this is not enough power, then it is time to stack two. With 2 stacked, the internal contactor is still fine, unless you really need over 60 amps of pass through power. It has built in control to trigger an external transfer relay.

For my use, the internal transfer is not a limitation at all. Even if I do add the full 9 x 320 watt panels, it brings my DC panel total to 7,680 watts, but the inverters only total 6,450 watts, so it is within spec. That is the old 16 x iQ7 and the new 9 x iQ7+ inverters. My only real concern with this setup is when I have good full sun and the batteries top out. I may need to turn off some microinverters to keep from exceeding my 16 amp back feed limit into the 20 amp breaker. In theory, full solar noon output could hit 27 amps. I often have less than 11 amps of load on.

I could have a little fun with the PLC. Have it monitor the output from the existing 16 panels. It can calculate about how much each panel is able to produce. I could split the 9 new panels into 3 sections. A single panel, 3 panels, and 5 panels. By choosing the combination, it could add 1, 3, 4, 5, 6, 8, or 9 panels to hit as close as it can to 16 amps of back feed, without going over.

Other than my Central A/C, my house is quite efficient. I have not topped out my XW-Pro yet. The max I have seen so far has been about 4,500 watts. I do expect that to change when I have it back feeding the A/C a little later this year. Right now, I have it limited to only back feed 10 amps (has not exceeded 5 amps yet) to the grid input side where my A/C compressor is connected. I will monitor the currents and see how it looks on load and battery current. I want to be sure the load will share across my 4 strings of batteries at that level as well. If all goes well, I can dial it up to back feed as much as 16 amps back through that 20 amp breaker. And it is smart enough to back off the power from the batteries if the solar is already pushing current back. If it is set to 16 amps, and my solar array is producing 12 amps, the XW-Pro will only add another 4 amps. It does come down to a little balancing act. My 100 amp main panel certainly adds a little limitation for me. I had looked into changing out the panel, but long story short, not gonna happen. The integrated meter socket makes it a serious pain.

 

[GXMnow]

Just a fun update this morning. Here is my SCE usage from yesterday.

The power used from 5 am to 7 am is on the cheapest rate, but it still should not have happened. My son let 250 watts of LED and florescent lighting on since 11 pm the night before. That works out to 1,375 watt hours burned up for no reason while we were sleeping. So the battery shut down and we used grid power for about 2 hours in the morning. But then I had my best ever solar production for early March. The baery was fully charged before 2 PM. So the solar then exported the rest of the extra all the way out to almost 5 pm. Then ZERO the gri, and still was a volt above cut off this morning when the sun started charging it again. Here is a snap of the XW page as I began to type this, with my PLC variable window in the corner.

The house at the top it the load side of the XW-Pro. That is showing 1 KW coming in, 0.3 KW going out to the grid input, and 0.7 KW going into the battery. In a snap shot you can't see the direction the dots are moving.

In my PLC window, the variables are a bit cryptic. Here is a simple explanation. B is the battery power as read from the XW. Showing 701 that is 701 watts, and positive numbers are charging. G is 266 watts, with positive meaning exported power. O on the right side is the Output power, the house. showing negative 1032 watts of power coming in the output.

What is missing from the XW-Pro display is that there are loads on the grid side, and solar power as well as loads on the load side. But the XW does not know about that as it only sees the total power on each connection.

That is where my external watt meters come in. N and S are the watts (x 10) being used by loads connected to the grid side of the XW. So in this snap shot, the L1 line is supplying 16.8 watts and the L2 line is supplying just 4 watts. As you can see, my non backed up loads are pretty small at this time, but it can go up a lot. The washer and dryer are about 300 watts each. The toaster is 800 watts. And the A/C compressor is 3,400 watts. Those loads will all show up on the N and S variables. So I add that power to the requested export to zero the grid side power during both charging and when running on the battery.

I don't currently have separate metering for the loads and solar on the output of the XW, it really is not that important. But if I look at my Enphase toolkit I can see the Solar right now is outputting 2.95 KW and the XW is only seeing 1.85 KW coming in, so the house loads on the backup panel at that moment is the difference of 1.1 KW.

I have such mixed feeling about turning on the central A/C. I really want to see my system kick into action and try to support that huge load. I know I can't zero the grid for the whole day at that power level with my current solar array, but I at least want to keep it zero for the 4 pm to 9 pm block. Yesterday, that time frame only pulled about 3,125 watt hours from the battery. If the A/C stayed running that whole time (I really hope that does not happen) it would be adding 3,360 x 5 = 16,800 watt hours of consumption. OUCH! That is more than I am cycling the battery yet, just for the compressor alone. On the bright side, the solar will be making more, and with daylight savings time, once we "Spring Forward", the solar will be producing a full hour later in the day, so hopefully, that will be running the A/C for an hour into the 4-9 block.

This is a big reason why I am looking into a small dual unit minisplit heat pump. Have one indoor unit in my upstairs room, and one in the main living space on the main floor. If I can keep those two areas comfortable on efficient minisplits, then the beast does not need to start.

Mr. Cool has a 2 zone DIY package, with a pair of 12,000 BTU indoor units, the outdoor unit has a maximum draw of up to 21 amps at 230 volts. So it is not really a savings over my Central air, but I think that is with 12,000 BTU indoor units connected. It is an inverter drive compressor that should reduce current based on the cooling load demand. And the 14 amps on my current system is just the compressor, when on the lower cooling mode. I really don't want to measure the current when it kicks to high output mode. That happens if it is running more than 15 minutes, and the thermostat is still calling for cooling. Most of the time, the low setting will meet demand and it will cycle off before it goes to high mode. Even if it is off just 5 minutes, it will come back on in low.

 

[400bird]

I've been looking at mini splits too, I'd probably go with a 4 zone system to deal with the 4 bedrooms overnight.
To me the benefit of a mini split it the multi zone, lessening the energy used when rooms aren't in use.
Not sure I'm ready to pull that trigger yet ($$) But, a mini split and heat pump water heater would zero out my gas bill!

When you add more panels, what's your plan for your 20 amp backfeed?

Are you going to remove and hide your XW and go through the permitting process to up your backfeed and backfeed breaker?
Or add a "reverse load shed" relay to disconnect some of the solar once the pack is full? So you don't exceed your agreement with the power co?

 

[GXMnow]

I have mixed feelings about how to handle this.
If I downgrade the main breaker to a 90, a 30 would be safe and legal, but I really o want to keep the back feed under 16 amps. Today, was a good day. The battery topped out at 2:30 pm, and my existing solar was still making 3,000 watts. 3,000 / 240 = 12.5 amps. So if I do the math from there I could add 3.5 more amps, if there are no loads in the house. But I am ALWAYS pulling 500 watts. So add another 2 amps. Let's call it 5 amps more at 2:30 pm. That works out to 40% more solar panel. I have 4,800 watts now. 40% of that would mean I could add 1,920 more watts of panel and not hit the 16 amp back feed.

If I go with the 260 watt used panels, that would be 7 panels. Or 6 good 300 watt panels. And I could still have the PLC turn off a few if the export goes too high. Or I could turn on a dump load like a hot water heater element. Looking at yesterday's SCE trace, my best day yet, I hit almost 2,200 watts of back feed in the 2 pm hour. That is still under 10 amps, so I think my calculations are pretty good.

But we do have the daylight savings time spring forward happening soon. That will push the solar output an hour later in the day, but it will also start charging an hour later, so it should really stay very similar. I am now up to "driving" my Bolt cells about 80 miles a day. I am using 1/3 capacity, and the car is rated at 240 miles, so 80 miles of use. That is a bit more than my brother typically drives his Bolt. He does about 60 miles a day going to work and back and a little here in town. I may push them a little harder, depending on what the traces look like once I start using the A/C every day. My main goal is to get the battery charged on just extra solar before 4 pm, even when the A/C is running. That might not happen with my current array. This is why I think I can use the extra panels without exporting too much. The A/C and battery will be sucking up all the power while the sun is shining. There won't be any left to export.

But on that odd day when the sun is rocking, and the air is cool so I don't need all the A/C, I guess the extra panels may need to be shut off.

I found a place on Ebay selling iq7 inverters for just $60 each. That makes using the used San Tan panels make sense. 9 older 240 watt panels at $50 each. Under $1,000 for for 2,160 watts of panels AND the microinverters.

 

[GXMnow]

I just peaked at my SCE usage. I knew yesterday had good sun.

I ended up exporting power for the entire day, though several hours are less than 20 watt hours. So Cal Edison must really love this ;-)