Off-grid residential sizing: size for min production months or min excess?

[wayne530]

Just curious what your sizing strategy would be for an off-grid (i.e., can't export excess production), residential system. I have sufficient roof-top space to install enough panels such that all of our usage would be covered by solar production (on average - we have a large 30kWh battery bank), but that would mean significant wasted excess production in higher production months. Alternatively, I could size to minimize excess production in higher production months but that would mean buying power from the utility in lower production months, and of course all options in between.

Our average daily usage is about 16-18 kWh and I'm comparing a ~4 kW solar array vs. an ~8 kW array. January (lowest in my area) daily production would be 10kWh vs. 20kWh and July (highest) daily production would be 23kWh vs. 46kWh. Even with two EVs, we could not ever hope to use nearly 28kWh in excess per day in the summer months. I'd love not to give the utility a single dime, but it just feels terribly wasteful to have 28kWh of unused excess production.

For completeness, I currently pay 27 cents/kWh during off-peak hours (midnight-3pm), 47 cents/kWh from 3-4pm and 9pm-midnight, and 58 cents/kWh between 4pm-9pm, so it's pretty expensive to buy at the moment.

Anything else I should take into consideration? What would your strategy be?

 

[ElectricIslander]

You need to consider the daily statistics behind your estimates of "average monthly" January production. In other words, in your climate (like ours here in the cloudy PNW) are there stretches of 5 days without any meaningful solar gain? If so, even if the monthly average meets your needs, unless you have over 5 times the storage than your daily loads, it will be lights out.

When off grid, design is all about minimum production periods vs days autonomy from storage. PVWatts and other calculators can help you develop these statistics for a variety of return periods and probabilities of outage for your PV capacity, loads and storage.

What I'm saying is that you have to consider the current and past variability of weather in addition to just average monthly data developed from long term averages.

 

[Danke]

Just curious what your sizing strategy would be for an off-grid (i.e., can't export excess production), residential system. I have sufficient roof-top space to install enough panels such that all of our usage would be covered by solar production (on average - we have a large 30kWh battery bank), but that would mean significant wasted excess production in higher production months. Alternatively, I could size to minimize excess production in higher production months but that would mean buying power from the utility in lower production months, and of course all options in between.

Doesn’t sound like off grid. More like grid tied.

 

[Mattb4]

I understand your point however there is no wasted production. There is only non utilized potential production. One of the advantages of a grid assist setup is that you can exist more easily somewhere between having too little and excess by relying on the grid to cover any shortfalls. You can't take advantage of the grid to bank all PV production like the grid tie folks so you must make use of on site storage batteries.

The grid tie people used to be constrained though to losing PV when they lost the grid. Newer setups that allow hybrid energy storage systems (ESS) have changed that.

As to sizing a grid assist system? It comes down to money and our physical layouts for most of us. If you got the ability to setup a system that covers 99% average of all your needs through out the year you will still have feast and famine periods.

 

[DIYrich]

You don't have off grid, you have grid tied with no export. Big difference. With grid tied, the grid can make up shortfall.

Size for min unused production. Then calculate production of the next panel for times the production is used. What is the value of the electricity used (at grid rate). How long to pay for that panel. Rinse and repeat until the payback period is too long for you.

You can also factor in grid reliability. How valuable is the additional production worth it to you when the grid is down.

Do you have a dump load that you could use? Preheat water for gas water heater. Mine bitcoin.

 

[Mattb4]

You don't have off grid, you have grid tied with no export. Big difference. With grid tied, the grid can make up shortfall.

...

Definitions. Grid tie means being in grid parallel so that you could export even if you are setup not to. Grid assist or grid as backup is off grid (no parallel capability) using the grid as an alternative source. Completely off grid is there is not grid available.Some people refer to grid tie with ESS as hybrid. Others equally call off grid with grid assist, or as backup, hybrids.

So the OP may not be grid tied.

 

[OffGridForGood]

Even with two EVs, we could not ever hope to use nearly 28kWh in excess per day in the summer months.

14 kWh each per (avg) day = about 50-70 miles commuting/daily driving, would be most of that exess for an average household.
A/C use typically parallels sunny summer days, sounds like you are 'right-sized' to me.
Didn't see location listed, the utility rates you quote sound like California prices (yes?)
Options for excess: Ask your neighbours if they want to buy electricity at 25% off utility rates?

 

[Doggydogworld]

Our average daily usage is about 16-18 kWh and I'm comparing a ~4 kW solar array vs. an ~8 kW array. January (lowest in my area) daily production would be 10kWh vs. 20kWh and July (highest) daily production would be 23kWh vs. 46kWh....

For completeness, I currently pay 27 cents/kWh during off-peak hours (midnight-3pm), 47 cents/kWh from 3-4pm and 9pm-midnight, and 58 cents/kWh between 4pm-9pm

Oversimplified version -- you use 17 kWh every single day and a 4 kW system produces 10 kWh every January day, 12 kWh every December/February day, 14 kWh every November/March day, etc. Your 30 kWh pack can sustain you an entire day plus two overnights, so all purchased electricity costs 27 cents.

You purchase 7 kWh * 31 days in January + 5 kWh * 59 days in Dec/Feb + 3 kWh * 61 days in Nov/Mar + 1 kWh * 61 days in Oct/Apr = 756 kWh
756 * 0.27 = $204/year

In our simple model a 8 kW system would save you all $204/year. If the 8 kW system costs $4k extra, that's a 20 year payback. If it costs $12k extra, well.....

In reality you'll probably buy more than 756 kWh with the 4 kW system, because daily production fluctuates. Your battery smooths out a lot of that, but a week-long cloudy stretch in September could cost you 20 bucks or so vs. 0 in our simple model.

Consumption also varies seasonally for most households, due to A/C and electric heat and hot water. Even if you don't have any of those, you can get daily variations, e.g. laundry day.

If you install a 4 kW system designed for expansion you can log actual daily production, consumption and grid purchases for a full year and get a much better idea of how much the extra 4 kW would save you.

 

[DIYrich]

So the OP may not be grid tied.

OP talks about their TOU rates, so grid tied.