Rooftop solar - what inverter

[jyl]

Question for the solar gurus - for residential rooftop solar, do you prefer a string inverter (like SolarEdge) or microinverters per panel (like Enphase)? What are the pros and cons? Are there better brands?

Starting to cost out a system.

[red-beard]

It depends on your roof. If you have a flat open roof, no trees nearby and all the panels are pointed the same direction, string inverters are the best value.

If you have an irregular roof, have potential shading or some of the panels will point a different direction, individual panel micro-inverters are the way to go.

Systems will cost what the market will bear. In Houston 5KW is about $15K. My sister just installed 7KW east of Oakland for about $22K. Prices are actual, before tax credits.

[jyl]

Thanks. Any thoughts on brands?

[hcoles]

$22k buys a good amount of electricity. The breakeven point might match the end of life replacement point.

[hcoles]

You might want to play around with this calculator. I trust it but I'm biased.
https://pvwatts.nrel.gov/

[red-beard]

Not a fan of Sunny Boy (SMA). I've replaced several of their string inverters at less than 10 years old. One client has a Fronius (11 kW) that is going strong after 12 years.

Micro-inverters I like Enphase.

One client had Solar Edge optimizers and they never worked right. Optimizers are a DC "micro-inverter".

For battery systems, we used Outback. But I probably would go with Sol-Ark these days. They are an "all in one" system optimized for batteries and on-grid/off-grid setup. With proper LIFEPO4 batteries, you can net-meter yourself (store excess and use later in the day). More expensive, but might be worth in in some cases. And you can connect a backup generator to work with the system.

[flatbutt]

Is it possible to install a system that is isolated from the grid? I've heard that when the gris goes down so does your system.

[Icemaster]

Originally Posted by flatbutt

Is it possible to install a system that is isolated from the grid? I've heard that when the gris goes down so does your system.

Curious what you've heard...mine is connected, I can isolate if I want and have zero failover time when I do have an outage. Being fully isolated means you can't really do anything with excess power once you max storage, or have a backup if something.

Not sure there's a benefit to that...?

[flatbutt]

Originally Posted by Icemaster

Curious what you've heard...mine is connected, I can isolate if I want and have zero failover time when I do have an outage. Being fully isolated means you can't really do anything with excess power once you max storage, or have a backup if something.

Not sure there's a benefit to that...?

I don't have room for batteries so I was looking at a system that would feed the house during daylight and on the grid at night. I was told that if I'm on the grid and it goes down I'd lose my daytime solar power.

[red-beard]

Unless you have batteries, grid down then solar down.

[hcoles]

Originally Posted by flatbutt

Is it possible to install a system that is isolated from the grid? I've heard that when the gris goes down so does your system.

If your system is isolated from the grid then there are no connections to the grid so it is not affected by the grid. The utility doesn't want your batteries "back feeding" them when they go down.

[red-beard]

Let me expand on what I was writing. If you have a "Grid Tied" system, it consists of solar panels and a DC to AC inverter. UL1741 sets the safety regulations for Grid Tied inverters. One part is that to prevent backfeeding the utility, the Grid Tied inverters will only work if Grid Power is available. This way linemen do not get killed by power backeeding into the utility system when the utility is down.

Another issue with grid tied systems, you can't use them tied to a generator. You do not want to backfeed a generator with excess power. Grid tied systems, in general, produce what they produce. They are not modulated to produce what is used. There are some exceptions (Sunny Island). They could be, but that would be a feature which would add significant cost. Plus you would need something like a transfer switch to separate yourself from the utlility. Again, this adds cost.

Outback and Sol-Ark "fix" this by having internal large relays which work as transfer switches. They can be used purely off-grid or used in grid tied applications. But they are battery based. When the solar produced exceeds the power used in a home, the excess is stored in the batteries. If the batteries become full, and the system is grid connected, it can send power out to the utility. If the system in not grid connected, or the utility is offline, and excess power is cannot be stored (batteries full), then the charge controllers will reduce the solar output of the panels to match demand.

Now I want to add, there are purely off-grid inverters/systems. And there are some that are "one-way" when grid tied (SAMLEX) where they can use Grid power but cannot send power to the grid.

It is a complicated field. And you have to know what you are doing to make NEC and UL compliants systems. LOTS of the "cheap Chinese" inverters and charge controllers are not NEC and UL compliant. If your house or RV burns down, your insurance may not cover your loss. I saw some of these at an RV business and asked someone why they would sell stuff that was not UL compliant. They told me that "UL" was a "Marketing Thing". Yep, they were not going to buy my products.

I am "Out of the Business". 1) The systems were expensive and the market was small. 2) I did not want to sell Grid Tied systems and flat out lie about the savings.

Solar "CAN" save you money, IF your power is expensive AND you are paid for your excess solar.

[ADDvanced]

Highly suggest you start an account here, and start asking questions. Community is awesome.

https://diysolarforum.com/

[jyl]

OK, I took my bills for the past 12 months including kwH and all charges-fees-taxes, reverse-engineered the bill calculation, used the NREL site to estimate kwH generated per each month, subtracted the generated kwH from the actual kwH to estimate net-new kwH, and recalculated the monthly bills using net-new kwH. This assumes the local utility will simply charge me on the net-new kwH, without extra fees for having a solar PV system etc - which may not be true, I dunno.

Lessons learned:
- I use what seems like an awful lot of electricity. 25,909 kwH/year.
- The effective $/kwH, including the dozens of extra charges and fees, is $0.16/kwH.
- I pay what seems like an awful lot for electricity, $4,142/year.
- The largest system that will fit on my roof is 10.1 DC kw. I used the tool to estimate usable roof area and eliminated the roof pitches that face north and east (house has hipped roof, fairly shallow pitch), specified "premium" "roof-mounted" panels, and accepted all the other defaults in the NREL online tool.
- That will (may) generate 11,584 kwH/yr, leaving me to buy 14,325 kwH/year from the utility (that's the net-new kwH).
- My annual bill, calculated on net-new kwH, will be about $2,268/yr. By these numbers, I never have excess kwH to sell, not even in the height of summer.
- So I'd save about $1,874/yr.
- Assuming system lasts 20 years (?), that would be saving about $38,000 over 20 years. That ignores both inflation and time value of money, which may more or less wash each other out.

So if I can install such a system for less than $38,000 net of all credits/subsidies, it is arguably worth considering? And probably not really worth considering unless it is under $30,000? Is that how you'd think about it?

I'm looking at this because I figure on replacing my roof in the next couple of years. I've never been interested in putting a new solar system on a near-end-of-life roof. Seems the ideal time to install solar PV is when the roof is replaced.

I suppose I could also, or alternatively, reduce my electricity consumption. It is extremely seasonal (summer and winter kwH is 2X the shoulder months) so the biggest factor must be running the central AC/heat. I have gas forced-air furnace w/ central AC, gas water heater, gas range, electric clothes dryer, all lights are LED, zero insulation in the walls, single-pane windows. The interior lights are on a lot (in the Portland winter, I need all the daylight spectrum light I can get). 3-5 people living here (three generations). I plan to add mini-split ACs to the second floor, which might reduce running the central heat/AC.

Feels to me like a big house insulation project, or a major change to lifestyle/living temperature, could reduce my kwH/yr, but not sure by how much. I am not into being a hair-shirt.

Portland seems like not a great place for solar PV? Solar radiation drops by 2/3rd in winter, the system generates 1600 kwH in July but only about 400 kwH in Nov Dec Jan. It will probably get sunnier (and hotter) here over that 20 years.

Maybe a solar system adds some value to the house?

[rwest]

Originally Posted by jyl

OK, I took my bills for the past 12 months including kwH and all charges-fees-taxes, reverse-engineered the bill calculation, used the NREL site to estimate kwH generated per each month, subtracted the generated kwH from the actual kwH to estimate net-new kwH, and recalculated the monthly bills using net-new kwH. This assumes the local utility will simply charge me on the net-new kwH, without extra fees for having a solar PV system etc - which may not be true, I dunno.

Lessons learned:
- I use what seems like an awful lot of electricity. 25,909 kwH/year.
- The effective $/kwH, including the dozens of extra charges and fees, is $0.16/kwH.
- I pay what seems like an awful lot for electricity, $4,142/year.
- The largest system that will fit on my roof is 10.1 DC kw. I used the tool to estimate usable roof area and eliminated the roof pitches that face north and east (house has hipped roof, fairly shallow pitch), specified "premium" "roof-mounted" panels, and accepted all the other defaults in the NREL online tool.
- That will (may) generate 11,584 kwH/yr, leaving me to buy 14,325 kwH/year from the utility (that's the net-new kwH).
- My annual bill, calculated on net-new kwH, will be about $2,268/yr. By these numbers, I never have excess kwH to sell, not even in the height of summer.
- So I'd save about $1,874/yr.
- Assuming system lasts 20 years (?), that would be saving about $38,000 over 20 years. That ignores both inflation and time value of money, which may more or less wash each other out.

So if I can install such a system for less than $38,000 net of all credits/subsidies, it is arguably worth considering? And probably not really worth considering unless it is under $30,000? Is that how you'd think about it?

I'm looking at this because I figure on replacing my roof in the next couple of years. I've never been interested in putting a new solar system on a near-end-of-life roof. Seems the ideal time to install solar PV is when the roof is replaced.

I suppose I could also, or alternatively, reduce my electricity consumption. It is extremely seasonal (summer and winter kwH is 2X the shoulder months) so the biggest factor must be running the central AC/heat. I have gas forced-air furnace w/ central AC, gas water heater, gas range, electric clothes dryer, all lights are LED, zero insulation in the walls, single-pane windows. The interior lights are on a lot (in the Portland winter, I need all the daylight spectrum light I can get). 3-5 people living here (three generations). I plan to add mini-split ACs to the second floor, which might reduce running the central heat/AC.

Feels to me like a big house insulation project, or a major change to lifestyle/living temperature, could reduce my kwH/yr, but not sure by how much. I am not into being a hair-shirt.

Portland seems like not a great place for solar PV? Solar radiation drops by 2/3rd in winter, the system generates 1600 kwH in July but only about 400 kwH in Nov Dec Jan. It will probably get sunnier (and hotter) here over that 20 years.

Maybe a solar system adds some value to the house?

Most of the information I have seen on buying things that are supposed to have a payback is that the payback should be 8 years or less to be something that you should for sure do. I think you would need to get some sort of rebates or credits to have it make financial sense.

[jyl]

Looking at what credits, rebates, etc are available here.

- Oregon Dept of Energy incentive, $0.20/DC kw up to a limit. For my hypothetical system, that’d be about $2,000.
- Energy Trust of Oregon incentive, $1,500 for a 10 kw system.
- Federal clean energy tax credit, 30% of total cost (possibly less the Energy Trust credit).

So if the gross cost is $30,000 (not even a guess, just a what-if), the net cost after incentives might be $17,950 = $30,000 - ($30,000 - $1,500) * 0.30 - $1,500 - $2,000. I think.

If so - payback might be around 10 years.

Maybe I should also look into heat pumps. It looks like there is a federal tax credit of 30%, up to $2,000/yr, for qualifying heat pumps which can be ductless (minisplit) types. If I could put heated and cooled air directly into the bedrooms, I might run the central HVAC less and that seems to be a big electricity use for me.

In fact, looks like almost any energy conservation work you do on a house is potentially eligible for a 30% tax credit, although there is an annual limit (per improvement and per year) so you may have to spread the work out over multiple years. Windows, insulation, efficient furnaces, efficient water heaters, etc.

[Zeke]

Without overthinking this it would seem that the more money you throw at the house for energy savings extends your payback w/o even considering the added expense. Just using less power due to increased insulation and better overall energy envelope. Therefore, it would seem that starting with the solar upgrade is first in line.

[Icemaster]

We went through the same cost/benefit analysis and netted out neutral on the costs, meaning our electric costs would match the costs of the package and install BEFORE the rebates etc., our payoff period was a hair over 8.5 years.

What sealed it for us was a couple things:

1. A whole house generator was significantly less, but it still was a net cost for us to run, had zero option to spin the meter backwards, and wasn't going to save us anything except for loss of food in one of the many outages we deal with.

2. Those calculations were based on the costs at present value which as of today was 18 months ago. Meaning, if the cost of that electricity stayed the same we'd net out in under a decade, if the cost went up the payoff horizon would decrease. Guess what.

Effectively, we bought and installed a system that gives us whole house backup, generates and sells excess back to PG&E (granted, for a pittance but it's something), and one could argue locks in our cost for electricity for the payoff period of the purchase.

They can get pretty creative with battery locations, have the folks you've talked with given you a definitive 'no' on storage?

[red-beard]

Jyl, the other thing you must check, to get maximum savings, you have to have net-metering with your electricity provider.

[jyl]

If a system can only generate half the kwH consumed, would batteries make any sense?