In California, the solar installation (panel - Inverter) before incentive should be between $2 and $3 per k|W of panel. You have 10kW so between 20k and 30k
2 power wall should add about $22k to that.

Solar panels are pretty much all the same, no need to worry there.

Using micro inverter or optimizer (versus string inverter) is really only needed if you have shading issues.
If you do not have shading issue, The next question is where the panels will be and how accessible they will be. This is because when Micro inverter or optimizer go bad, you will have to get to that panel to replace them. If it is on a high pitch 3rd story roof in the middle of 4 rows of panels, this can turn into a nightmare, and more often than not labor is not covered by the warranty.

The more complicated the electronic module under each panel, the more likely it is to fail. That said, even with a string inverter, you would likely need quick disconnect devices like the Tigos under the panels, although these are not as complicated as a microinverter.

Pay back is fairly easy to estimate based on the current amount of your yearly electricity bill that you will save. What will make a big difference and is a wild guess, is what percentage increase to apply every year to that yearly electricity bill being saved. The more you forecast electricity prices to go up the faster you will amortize the system. This is how installer exaggerate the return on Investment, forecasting exaggerated yearly increase in electricity prices. I suggest being conservative like 1% a year to make a worst-case scenario.

The most important beside finding someone with reasonable prices, is to find a reputable installer that will be there for the initial years of the system, when problems are most likely to arise. that is of course easier said than done, given the high percentage of charlatans out there....

Thank you. When we had our system installed, at that time LG and Panasonic panels were recommended more often alongwith Solaredge inverter and optimizers. Fortunately none has failed so far.
It is a single story ranch with easy access to panels in two rows on not too steep of a pitch. There are no shading issues. Should we exclude the optimizers?
Considering the usage, May-September being the highest usage months and with reduced credits under NEM 3.0, what is the largest this system should be sized?
Should we consider panels with higher wattage than 390W?
The yearly electricity billed for the past 12 months was $6,221. Not sure how much of that would be saved under NEM 3.0 for the proposed configuration.

Month Usage kWh Cost
Aug 1,698.90 $ 594.88
Sep 1,595.97 $ 547.57
Oct 1,100.74 $ 364.34
Nov 1,032.72 $ 338.77
Dec 1,214.15 $ 408.62
Jan 1,056.35 $ 358.55
Feb 896.86 $ 301.40
Mar 1,162.25 $ 416.33
Apr 1,088.91 $ 391.43
May 1,623.45 $ 611.72
Jun 2,110.43 $ 790.25
Jul 2,730.81 $ 1,097.53
Total 17,311.54 $ 6,221.39

If you have no shading issues, and are interested in saving money, I would definitely get some quote with a string Inverter to see how much you could save.

If the panels have easy access, the one thing I do like about optimizers are the individual panel monitoring. This proved invaluable for my installation....

I made a simple ROI excel sheet a while back, you can download it below and play with the variables
Simple Solar ROI Tool.xlsx

I entered your numbers in the green variables

Array yearly prod (kWh) 14,000
Year Needed power (kWh) 17,311
Install Cost after incentives $47,000
Average $/kWh today $0.36 (6221/17311)
Electricity Rate yearly increase 1.0%
Max Panels yearly loss 0.5%
Panel end of Year one minimum 98.0%

That gives an ROI of 10 years

Of course this is a rough estimate that does not take into account effects of NEM3 and how you could use the battery to minimize consumption during peak hours. But it should still give you an idea.

To come back on microinverters or optimizers. If going for a regular string inverter, you would probably need modules like the Tigo rapid Shutdown (TS4-A-F @ $27), and can add the monitoring to it for an extra $10 (TS4-A-S @ $37).
TS4 Flex MLPE | Tigo Energy

These would give you the comfort of individual panel monitoring without the unneeded complexity of optimizer or microinverters.

As for adding more panel power, this is all about whether you want to cover ALL your consumption or not, and the corresponding financial implications.

I addition to what scrambler has already stated, I would also have your friend write down what they are trying to achieve. Yes we know they want to save money, but what other goals do they have if any?

i.e.

I want to minimize my power bill.

Or

I want to minimize my power bill and provide power for 3 days during wildfire season public safety power shutdowns.

Or.

Their stated goal here.

Monthly usage is one thing, but you have to understand all the loads your friends are using. It is an interesting eye opening project to list each and every device hooked up to the electric panel (voltage and start up amps/continuous amps).

In a power outage situation, you might find that two powerwalls cannot start up a 5 ton AC unit.

I don't have a tesla powerwall or gateway. I have read numerous threads on dark start tesla powerwalls and it would be wise to read up on this depending on your friends goals. I have a NG generator and don't have to worry about dark start issues.

Thank you. I noticed you had a sheet comparing panels in 2017. Did you ever update that sheet?

I'd say their goal is: I want to minimize my power bill and provide power for single days during wildfire season public safety power shutdowns and around 7 storm related tree damage caused outages/year which last no more than 10 hours.

Like most consumers, they do not have detailed loads calculation. It is a 3,100 sq ft home with two AC units of sizes unclear.

The installers proposing systems are claiming the two Tesla Powerwall 2 can handle the two AC units.
The idea behind Powerwalls is primarily to shift load to batteries during the 4-9PM high power rate time windows and during planned and unplanned outages.
It's my understanding that with NEM3, their use of the grid for credits generated to offset power used during the night is practically non-existent so obviously they need local storage.
Are you suggesting that having generators is a more cost and operation effective option than batteries?

I generated the attached monthly usage vs production sheet based upon last year's actual usage data and NREL PVWatts calculator. Are we sizing it properly based upon this data?

No did not update panels, but it would be easy enough to plug in new panels specs. That said panels really are not critical these days...

AC should really not be considered critical during an outage unless your are in desert country....

That is what makes most sense for the battery. Easy automated use during peak rate hours to minimize cost, and outage for critical things.


Very probable that generators are more cost effective, but they are a pain in the butt to maintain and operate, and they are noisy and smelly. Given one of the key usage is to minimize grid usage during peak hours, battery are a much easier and comfortable way to do that (although probably more expensive)

Finally "proper sizing" is about how much money you are willing to spend, and how much space you have for panels.
From the data above, the array will not cover 100% of the current usage, so there is justification for more production capacity.

I'm not in desert country, but we had wildfire smoke and ash in 90 to 95 degree weather that had our power shut off for days. The smoke and ash reduced solar generation to practically zero. So for me, a generator is the only solution. Batteries would be nice for the short power outages we have, but at this point they are not necessary for my situation. For me, I want to have AC available in my house during wildfire power shutdowns. I have a smaller than average house with an inverter drive AC condenser so it takes a lot fewer amps to start and run than a traditional single stage AC condenser unit.

Does the solution include running AC during power outages for your friends? If yes, this should be added to their goal statement.

I would suggest pursuing the detailed loads calculation spreadsheet. I don't understand how an installer can say yeah the 2 Tesla Powerwall 2's can handle the two AC units without having a complete load spreadsheet. A single outdoor AC condenser unit can have locked rotor amps of 50 to 80 amps.

We used an LA headquartered installer the last time. Now they make their own panels. Is it not important to choose a panel brand that has a better chance of being in business 20+ years from now? No-name vs major brand name?

What's the best and simplest way to perform a detailed electrical load calculation in absence of a smart panel?

OK you got me
Sure definitely pick a company that has a chance to stay in business...