PV System for My Own House

[curt swartz]

Hey PV experts, I would like to get your input on a PV system for my own house.

I purchased the house I grew up in from the family estate and in the process of doing a major remodel with additions. Hopefully I will be able to afford move in when completed. I had to leave 50% of the exterior wall framing for the city to consider it a remodel. The remainder of the exterior walls, all interior walls and the entire roof structure were replaced.

Anyway, I plan to install a PV system as part of the work. I work around PV systems everyday but never do any installs or pay much attention to what is being installed. Usually we do all the electrical work on a project and a separate PV contractor comes in at the end to install the system.


I’m planning to install roughly a 10KW system. In doing online research I was thinking of using LG panels with SolarEdge inverter/optimizers.


My longtime salesman at my primary wholesale supply house connected me with a sales person at their solar division branch. He said that their main line was Trina but they could get LG if that is what I preferred. He ended up connecting me with his manager since the system was for my house and his manager could discount the material more. After speaking to the manager, he said that they are not authorized for LG sales. He said he could get the panels through another wholesaler but if I ever had a warrantee claim it could be an issue. He asked if I would be willing use Panasonic panels instead of the LG.

I don’t see as much info online on the Panasonic panels compared to the LG but what I have read they seam to be pretty good panels. The LG’s I was considering were either the 350 or 360 watt panels and the Panasonic’s are only 330 watts. From what I’m seeing the price per watt of the Panasonic’s is going to be less than the LG’s so the system would be less expensive with only a minor reduction in output.

So my main question is what are your thoughts of Panasonic panels compared to LG?

My other question is making sure I understand 705.12.

My plan was to install a 200 amp meter main then feed a 400 amp bussed panel next to it. The reason for the 400 amp panel is to allow the 10Kw PV system plus connecting Tesla or some future similar batteries to the system. The 400 amp buss will provide plenty of cushion for the PV system and batteries to stay under the 120%.

The Square D meter mains used in my area all have a main breaker with an 8/16 circuit buss bar then lugs on the bottom. Normally when sizing service equipment we allow for the future PV. We frequently have 30+ branch and feeder breakers in the service panels so the back feed PV and main breaker is limited to 120% . In my case if I don’t install any breakers in the service panel portion I don’t believe I need to worry about the combined utility plus PV supply since there is no way it could overload the buss bars. I would just need to keep the combined feeds of the 400 amp panel under 120%.

I would like to put my A/C and a couple other larger loads in the service panel instead of the 400 amp panel incase I do add batteries in the future. I would not want a large automatic load on the batteries so placing it before the gateway is a better choice.

If I understand 705.12(12)(B)(2)(3)(c) correctly as long as the combine rating of the branch/feeder breakers in the meter main are less than 200 amp I should be ok. Am I understanding this correctly?

Thanks!

 

[jaggedben]

Panasonic bought Sanyo's high efficiency cell technology a few years back. Meaning, it has a long (for the industry) track record. I think any panel from a company that is likely to be around to honor the warranty is fine to go with. In that respect LG and Panasonic are top tier, Trina maybe a notch below but you could do worse. .

Planning for batteries, so many details to think about...

The 400a panel does make sense because it will pretty much allow you any of the 705.12(B)(2) options down the road. You're on the right general track with your thinking about leaving the high power loads in the service panel. However, what I would recommend if possible is to bring all the branch circuits to a large adjacent gutter, so that it is easy to relocate circuits as necessary when the time comes, and also easy to insert a Powerwall Gateway or other micro-grid disconnect device. Unless you have lottery winnings type money to spend on Powerwall, a backup system is not going to power more than 60A, and 30A might be more realistic, but my point is that it's hard to predict what your backup system will support, so give yourself a way to easily change things later. Who knows, maybe you'll even decide to put the AC on backup after all.

Also, avoid MWBCs, unless you are positive that you'll want both branches either backed-up or not down the road. Both branches have to terminate wherever the neutral does. Dealing with this issue has been one od the most annoying wrinkles I've encountered with installing backup systems.

Finally, I'd avoid using feed through lugs. That complicates 705 rules and depending on how the 2020 NEC comes out (I'm drafting a comment on the very subject) there's no telling that won't throw a monkey-wrench into the works.

 

[PWDickerson]

You might consider installing a 200A panel with a 225A bus. This increases your PV system backfeed breaker to 70A under the 120% rule, which is plenty of capacity for a 10 kW inverter. Manufacturers have begun marketing these as "Solar Ready". I would second the advice to avoid feed-through lugs.

 

[ggunn]

On batteries...

What is your need for batteries?

Do you want them for a backup when the grid goes down? That begs the question of how often and for how long that happens. In my case, I'd estimate that the grid has been down for a total of a couple of hours in the last ten years; it doesn't make any sense for me to spend a lot of money on batteries that I will virtually never need. Batteries are very expensive.

Or do you want them for temporal load shifting to get around Time Of Use (TOU) tariffs? In my case there are no TOU tariffs in place, so I have no need to do that.

Depending on which strategy is your motivation, the design of your system and the type of batteries you need may be very different. In the first case your batteries will just sit there for long periods of time, while in the second they will cycle every day. In the first case your protected loads panel will contain the loads you want to keep active to bridge a grid outage and your batteries will be sized for how many days of autonomy you'll need, while in the second the point is to shift as many essential loads as you can to the batteries during the times of day when kWh are expensive. In the first case lead acid batteries might be your best option while in the second case lithium ion might be better.

How important is it to you that the PV continues to run and service your loads when the grid is down? That will significantly affect the design of your system; you'll need to run the numbers to see how much that would be worth to you and how much it would cost.

Grid tied PV without batteries is pretty straightforward and relatively inexpensive. Once batteries come into the picture, though, things rapidly get more complex and expensive. Many people who think they need batteries really don't. I, for example, don't.

 

[GoldDigger]

Bringing the PV backfeed into the 400A panel does not mean that you do not have to do the 120% calculation on the meter main panel, especially if you will also have loads in that panel.
The backfeed calculations, in one form or another, apply to the panel in which the PV is connected *and* to all upstream panels.

Sent from my XT1585 using Tapatalk

 

[jaggedben]

Bringing the PV backfeed into the 400A panel does not mean that you do not have to do the 120% calculation on the meter main panel, especially if you will also have loads in that panel.

Actually, one option to avoid the 120% rule on the main is using 705.12(12)(B)(2)(3)(c). He mentioned a strategy for that, and his thinking on it is conceptually correct. Whether he can actually pull it off given all the loads he will have is a more open question.

 

[electrofelon]

My thoughts on panels: I dont worry too much about what panels I use. I do try to avoid "xyz" panels that I have never heard of from China, but other than that its usually what I kind find good deals on. Check out sun electronics, sunelec.com, they often have good deals. They buy out older stock and bankrupt lots, but have current big name panels too. Obviously you will have to weigh in freight costs with that route, but I find it is usually worth it. Could also check craigslist. Often you kind find pretty good deals there as someone has bought a container for a price break and is selling the extras or something.

 

[jaggedben]

On batteries...

What is your need for batteries?

Do you want them for a backup when the grid goes down? That begs the question of how often and for how long that happens. In my case, I'd estimate that the grid has been down for a total of a couple of hours in the last ten years; it doesn't make any sense for me to spend a lot of money on batteries that I will virtually never need. Batteries are very expensive.

Or do you want them for temporal load shifting to get around Time Of Use (TOU) tariffs? In my case there are no TOU tariffs in place, so I have no need to do that.

Depending on which strategy is your motivation, the design of your system and the type of batteries you need may be very different. In the first case your batteries will just sit there for long periods of time, while in the second they will cycle every day. In the first case your protected loads panel will contain the loads you want to keep active to bridge a grid outage and your batteries will be sized for how many days of autonomy you'll need, while in the second the point is to shift as many essential loads as you can to the batteries during the times of day when kWh are expensive. In the first case lead acid batteries might be your best option while in the second case lithium ion might be better.

How important is it to you that the PV continues to run and service your loads when the grid is down? That will significantly affect the design of your system; you'll need to run the numbers to see how much that would be worth to you and how much it would cost.

Grid tied PV without batteries is pretty straightforward and relatively inexpensive. Once batteries come into the picture, though, things rapidly get more complex and expensive. Many people who think they need batteries really don't. I, for example, don't.

In California, the situation is basically that people get it for an emergency backup solution, but at a discount from the cost for that. The discount comes from a state rebate that covers approximately 30% of the cost, plus rolling in federal tax credit on the remainder, plus using time-of-use management to recoup maybe 10-20% of the cost over the life of the battery. So it is not exactly either/or.

 

[baserunner4]

I would not recommend SolarEdge but it also depends what NEC version your state is using. Also is the house a regular pitched roof? What racking are you planning on using?

 

[electrofelon]

Not sure which the OP was planning, or if he has room for a ground mount, but I would go ground mount if possible. Probably will be able to get more ideal angles and better production than the roof, unless the roof gets you above obstructions. Also one can "homebrew" ground mount mounting structures (assuming you want to invest some time to save some money) much easier than roof mount systems. Lastly, if you need another structure anyway, like a carpart or wood shed, it can make it really cost effective.

 

[ggunn]

Not sure which the OP was planning, or if he has room for a ground mount, but I would go ground mount if possible. Probably will be able to get more ideal angles and better production than the roof, unless the roof gets you above obstructions. Also one can "homebrew" ground mount mounting structures (assuming you want to invest some time to save some money) much easier than roof mount systems. Lastly, if you need another structure anyway, like a carpart or wood shed, it can make it really cost effective.

Ground mounting is generally more expensive than roof mounting. You might be able to "homebrew" a ground mounting system but you might not, depending on where you are. Some AHJs require a PE stamped structural drawing or letter for a ground mounted PV system.

One advantage, however, to a ground mounted system is that in general rapid shutdown rules do not apply.

 

[ron]

I am not a PV expert, but I'll tell you what I did at my house.

I did a 200A main with a 50A branch breaker in position 1,3 with a mechanical slide interlock (Reliance). The only thing I would do different would be to add extra space for watt meters.

Then a line side tap service fused disconnect for PV supply. I only had enough South facing roof for 8kW of panels and a 6kW inverter (actually it has (2) 3kW circuits for better max power point tracking). I live in the Northeast, so I only make 6kW on few and far between periods.

That way I have the ability to use a gen and PV at my choice (not at the same time).

Because of my utility pays me the same $$$ as it costs me to buy kWh from them, the battery is not cost effective for me, so I don't worry about it

 

[curt swartz]

Hey…….Sorry for the delay in responding. I have been on the forum some for only enough time to quickly read a few topics.

First, I want to thank all that replied.

It sounds like I’m fine going with the Panasonic panels if they are the most economic for me to get over the LG’s.

As far as installing a Tesla or other battery. This is just a possible future option. It’s definitely not mandatory and probably will not happen in the near future if ever. I just want the service equipment to support adding it.

I’m mounting my 200 amp meter main and 400 amp panelboard about 3’ apart. I will nun a raceway from the meter main down, over that up to the panelboard. Since the 400 amp panel board has quite a bit of space for only 200 amp conductors my though is if I ever add a gateway I could mount it next to the panel then extend the line/load conductors into the panel. My other option is I could easily add a gutter under the panel later.

One poster suggested using a 200 amp service panel that has 225 amp bussing. This is actually how most of our combination service entrance equipment is built now. This would be fine if just adding the 10K of PV but not if adding a battery or batteries.

This system is a rooftop not a ground mount. I have 1 good large area of roof so we filled it up as much as possible. That worked out to 30 panels. Its 3 rows of panels between the ridge and eve. Should make installation easy.

 

[jaggedben]

...

One poster suggested using a 200 amp service panel that has 225 amp bussing. This is actually how most of our combination service entrance equipment is built now. This would be fine if just adding the 10K of PV but not if adding a battery or batteries.

...

It would be fine for one Powerwall unit (30A breaker) and a 40A solar breaker (7600W inverter, not atypical for a 10kW DC system). However it would leave you no room for additional Powerwall units, or any other AC-coupled storage solution requiring more than a 30A circuit.

 

[ggunn]

... a 40A solar breaker (7600W inverter, not atypical for a 10kW DC system)...

That would be a 1.32 DC:AC ratio, which is more than I use for any systems I design. Some AHJ's I deal with dictate a 1.20 max ratio to qualify for incentives.

 

[jaggedben]

That would be a 1.32 DC:AC ratio, which is more than I use for any systems I design. Some AHJ's I deal with dictate a 1.20 max ratio to qualify for incentives.

I have pushed ratios that high when the cost of a service upgrade isn't justified by the extra energy gained with a higher-power inverter. Also some systems with panels facing different directions never get above the inverter nameplate, even at that ratio. When both factors are present it's kindof a no-brainer. I have learned to explain it clearly to the customer though.

 

[ggunn]

I have pushed ratios that high when the cost of a service upgrade isn't justified by the extra energy gained with a higher-power inverter. Also some systems with panels facing different directions never get above the inverter nameplate, even at that ratio. When both factors are present it's kindof a no-brainer. I have learned to explain it clearly to the customer though.

Still, I wouldn't call 10kW DC STC on a 7.6kW inverter "typical", although the optimal inverter "overloading" can be an elusive number. What software are you using to determine the maximum output of a module at a specific location and orientation?

 

[jaggedben]

Still, I wouldn't call 10kW DC STC on a 7.6kW inverter "typical", although the optimal inverter "overloading" can be an elusive number. What software are you using to determine the maximum output of a module at a specific location and orientation?

I said 'not atypical'. Maybe 'not unheard of' would have been a better choice of words.

I use SolarEdge's software for determining how much is lost to clipping on such systems. I haven't done that kind of overpowering in other cases (except microinverters where a higher power inverter wasn't available).

 

[curt swartz]

I have another question. My supplier sent me a material quote for all material. It includes the standard SolarEdge inverter not the newer Wave unit. I don't know if this is what they stock or if it has to do with pricing. Is the Wave better or worse than the older model?

They also included a Zigbee module but I don't believe I need that if I have an Ethernet connection correct?

 

[jaggedben]

The HD Wave is smaller and lighter, which is great for us guys installing them everyday but maybe not as important to the owner if space is not an issue. They are in my experience a little bit less reliable. They also limit power output to actual nameplate, wheresas the older ones will go a few percent over if voltage is above nominal. However they are also 99% efficient instead of 98.

If you're getting a 10kW inverter, the HD wave version just came out. I wouldn't be surprised if your supplier still has some older stock they'd like to ship. The lower power versions came out last year.

You don't need the Zigbee if you can bring a physical ethernet hookup to tue inverter.