Residential Payback for Off-Grid Solar Q1 2023 (grid connected)

[SignatureSolarJames]

Guys, I figured I would start a thread here, I only ask that all comments contain math in them as this is a math problem, too many words only comments on other threads IMO

here are the numbers I am seeing in Texas:

1600kwh/kw/year (grid tied) -battery system inefficiency (~10%) =1440kwh/kw/year
Let's do a 2000kwh per month target

-16.7kW of solar needed @ $9k
-30kwh of LFP batteries at 46% nighttime vs daytime usage @$9k
-12kw ac output inverter/ inverter stack with mppts for DC coupling @ $3k for 6.5x2 or $8300 for solark 15k
-roof racking (ground mount will add) at $.10/watt @ $1600
-Wiring at $700
-Rapid shutdown at $.05/watt @ $800

$24100 for "basic"

$29400 for SLRK//Deye

if average life of parts is 20 years then 480000 kwh in the lifespan

24100 + rebuild fee x2 ($1000 for stack) $26100 Divide by 480,000 kwh = .054/kwh - 30% = 3.8 cents per kWh
if your cost of capital is a 20 year 6% loan then $18270 after credit = $31414 total payments = 6.5 cents per kWh
If you get an installer to be fair and do the work for $15k in 2 days tops then you say $24100+ $15000 - 30% = $27,370 = 5.7 cents per kWh
if your cost of capital is a 20 year 6% loan then $27370 after credit = $47060 total payments = 9.8 cents per kWh
All of this is under the 12-cent mark and 12 cents WILL go up at a rate exceeding inflation over the next 20 years

there are other pitfalls/ mods here but the 20 year life is unfair to the panels, they are 33% + of the costs and they will go for 30 years, ground mount bifacial may have a higher ROI as well, SLRK/Deye style inverters add 20% here but that is still in the ballpark, and we know those are coming down drastically in the next 2 years (which will make installing easier and ownership trouble less) The SLRK/Deye style most likely will have an overhaul cost at 10-12 years average, Schneider may outlast them but will not see a reduction in costs coming up

Post anything that can modify this equation please

 

[sunshine_eggo]

No math, but comments on assumptions:

I don't see a generator or fuel cost. In that case, you need to size your system for the worst case, i.e., typically Nov-Jan when you have shitty sun or have sufficiently large battery to weather those days.

$1600 racking for 16kW of panels feels REALLY optimistic.

Is roof angled for optimal solar? If not, did you factor this in?

 

[sunshine_eggo]

Lots of days where your battery doesn't get fully charged:



Fair amount of uncaptured solar:

 

[SignatureSolarJames]

No math, but comments on assumptions:

I don't see a generator or fuel cost. In that case, you need to size your system for the worst case, i.e., typically Nov-Jan when you have shitty sun or have sufficiently large battery to weather those days.

$1600 racking for 16kW of panels feels REALLY optimistic.

Is roof angled for optimal solar? If not, did you factor this in?

roof mount rack, mini rails are that cost, working on rail-based in that range


south facing unshaded roof
rapid shutdown is added, the alternative is rapid shutdown plus racking money placing a groundmount where needed

 

[SignatureSolarJames]

generator and fuel are not in this argument, if this was non-grid connected then they would be, but the competition is $0.45/kwh generator power, a much different equation

 

[Checkthisout]

You should do a thread analyzing the benefit of your hybrid a/c units.

Just panels and a changeover to minisplits with the complications of tying into house power or batteries.

I bet those numbers look good in hot sunny climates.

 

[SignatureSolarJames]

I

You should do a thread analyzing the benefit of your hybrid a/c units.

Just panels and a changeover to minisplits with the complications of tying into house power or batteries.

I bet those numbers look good in hot sunny climates.

maybe will add categories

 

[GregTR]

-30kwh of LFP batteries at 46% nighttime vs daytime usage @$9k

What UL Listed batteries are these?

Also, 46% nighttime usage is unrealistic. You'll be on batteries from 5PM to 8AM this time of year for sure and probably also in the summer once you consider A/C usage. Your peak usage will be 5PM to 9PM, all have to be supplied via battery.

I also find your budget estimate extremely conservative, I would be amazed if you can build and install this system for that much. As a comparison, here is my build and price breakdown: https://diysolarforum.com/threads/s...megrid-stackd-38-4kwh-battery-pack-ama.51948/

 

[robby]

What UL Listed batteries are these?

Also, 46% nighttime usage is unrealistic. You'll be on batteries from 5PM to 8AM this time of year for sure and probably also in the summer once you consider A/C usage. Your peak usage will be 5PM to 9PM, all have to be supplied via battery.

I also find your budget estimate extremely conservative, I would be amazed if you can build and install this system for that much. As a comparison, here is my build and price breakdown: https://diysolarforum.com/threads/s...megrid-stackd-38-4kwh-battery-pack-ama.51948/

Yep, it's the reverse of the power equation that dictates that power companies need to produce a lot more power in the day because manufactures and businesses are using it.
At night it's the home owners that are using a lot of power. Especially between the hours of 6pm to 11pm.
My 1600KWH usage per month breaks down on weekdays to about 70% at night and 30% in the day and weekends about 50/50 due to everyone being home plus the laundry, house cleaning etc.
I need 35KWh of batteries to cover that completely if I use 80% DOD daily. That is also based on having good to average PV production.

 

[SignatureSolarJames]

What UL Listed batteries are these?

Also, 46% nighttime usage is unrealistic. You'll be on batteries from 5PM to 8AM this time of year for sure and probably also in the summer once you consider A/C usage. Your peak usage will be 5PM to 9PM, all have to be supplied via battery.

I also find your budget estimate extremely conservative, I would be amazed if you can build and install this system for that much. As a comparison, here is my build and price breakdown: https://diysolarforum.com/threads/s...megrid-stackd-38-4kwh-battery-pack-ama.51948/

Lifepower4 is pending the announcement of UL9540A,

the 50/50 split or less just comes from data we have collected from hundreds of sites; this is over a year average with significant cooling loads, it is easy to look at one day and get a different average.

 

[Quattrohead]

Not enough battery.
I have 66kwh of battery and the last 7 days have been shit sun, slowly loosing capacity until last night. Went to bed with 24% remaining.
Next morning, her indoors has a shower and I see 1 of the batteries hit 0% so I ran out to the garage and turned the HWT off LOL.
I finally had to turn the utility breaker on and took 32kwh at a cost of $4.53.

 

[GregTR]

1600kwh/kw/year (grid tied) -battery system inefficiency (~10%) =1440kwh/kw/year

I'm going to say that these numbers are also rather optimistic based on my actual data.

First of all, very few homes would have south facing unobstructed roof areas to support that size of solar installation AND also consume so little power. My home is a 4,000 sq ft house and I could only put 16/44 my panels on the south roof and I use over 25,000 kWh/year. Expecting that size system to fit on the south roof of a house is unrealistic.

Second, the system efficiency is far worse than 10% when it's all combined, a Sol-Ark will use 2.4 kW/day as will any battery system just sitting idle, not counting any other losses.

Again, if you find an installer that will put a 17kW PV on my roof with a Sol-Ark 15K for $2/W without battery and before tax incentives you're selling snake oil. I dare you to find ANYONE do this work for this much today.
The $9K for 30kWh battery is also awful generous, it is if you don't pay sales tax and you don't mind 6 LG4 units just laying around without a rackmount.

Bottom line, your numbers are the absolute bare minimum in a theoretical dream house that I highly doubt anyone can get to on their own and anyone trying to get into solar based on these numbers will be very disappointed to learn that they're not attainable.

the 50/50 split or less just comes from data we have collected from hundreds of sites;

I'm going to repeat what I said, the number is not reasonable. My energy provider who offers free electricity from 9PM to 7AM and charges $0.29/kWh from 7AM to 9PM estimates that the average user will end up paying $0.17/kWh on average. This would indicate they estimate that the average user uses ~58% of their electricity between 7AM and 9PM. if we assume even distribution within this time period (which we shouldn't as we all know about the peak use being after 6PM) it would mean that a mere 37% of electricity usage would occur during daylight hours (8AM to 5PM). The real percentage is probably even worse given the real use models indicating higher use in the evenings.

 

[wattmatters]

the 20 year life is unfair to the panels

But is overly optimistic for the electronics, i.e. inverter.
20 years for batteries is also a pretty heroic assumption.
Don't forget decommissioning and disposal costs for end of life items.
Costs for any additional infrastructure required to safely house such a system.

I only ask that all comments contain math in them as this is a math problem, too many words only comments on other threads IMO

Any maths problem needs a discussion to define what it is trying to solve, which includes understanding all relevant inputs and assumptions used.

By restricting the discussion to only one type of solution to begin with, you are putting the maths cart before the horse. It's making the bold assumption that grid supported off-grid solar + battery is the right solution to begin with, when there are a raft of options for how one might allocate their limited resources.

In essence the problem with threads like this is the maths is completely different for every individual so you will inevitably end up with a discussion in order to define the optimisation problem to begin with (and what people are looking to optimise varies). Only until that has been done can the maths follow.

It depends on each individual home's priorities/objectives, where they are located, the sort of home they occupy, their energy consumption and consumption patterns, what other energy sources their home uses, and the local grid regulatory and tariff regime they are operating within, how long they expect to be in that home, the space available suitable for solar PV, what additional infrastructure might be required to install such systems etc etc.

As a result the type of home energy solution which is financially rational/optimised can be highly variable.

It should also be considering the demand side of the equation, not just supply. Part of the answer is finding ways to consume less energy to begin with, or to work on shifting consumption to when it is least expensive.

And anything which makes an assumption about the future (e.g. grid tariffs) needs to include some sensitivity analysis to understand whether it is overly reliant on assumptions being correct when predicting the future might be notoriously difficult.

e.g. where I am (Australia) it's almost impossible to make a grid-supported off-grid PV + battery system financially viable no matter how you slice and dice the numbers. Grid-tied solar PV however is almost a complete no-brainer. As for grid-tied batteries, you would be better off just burning a $20 bill every week.

 

[GregTR]

It's making the bold assumption that grid supported off-grid solar + battery is the right solution to begin with

OP is concentrating on Texas and in Texas I would agree with the assessment that self-consumption is the name of the game, having no reasonable NEM agreement available for anyone.

You can skin the self-consumption cat two ways: 1) only install enough PV that you can use during the day or 2) add battery to your system.

Given the 80-20 rule applies here just as it applies to most things in life, I would say that the best system is one that produces the largest amount of self consumption with the least amount of wasted energy and/or storage space.

The proposed 17 kW system with 30 kWh battery is probably pretty close to what achieves this goal for an average Texas suburban home of ~3,000 sq ft and your average A/C using, gas furnace winter use household. In the U.S. especially in the south with high A/C needs this sounds about right.

Trying to shoot for 100% consumption coverage with 100% self-use is a fool's errand and not worth the investment, especially in Texas where grid access is relatively cheap but without any kind of NEM availability. Obviously other areas with other costs and benefits may vary greatly from what is a solid choice in Texas. A setup that is proposed is most definitely a decent startup plan for your average Texan suburbanite from a capacity/generation standpoint which may or may not work for someone in Europe or California.

I already laid out my contentions with the proposal, namely that it is way too optimistic on pricing, especially for people who are looking for a turn-key UL listed solution that actually doesn't look like a heap of garbage when installed. It is also way too optimistic on mounting/shading and therefore possible energy generation on an annualized basis as well as the life expectancy of the components.

These factors skew the cost per kWh significantly and therefore present an irrational ROI expectation in hopes of presenting such system being a viable choice for more people than it actually is.

Currently solar install costs are relatively high and most would probably do better if they postponed their install a year or two when pricing would regress back to the natural declining cost curve of recent years. But hey, if you can have a Sol-Ark 15K system with 17kW of PV installed at $2/W I'd say that's a heck of a deal and you should jump on it!

 

[wattmatters]

OP is concentrating on Texas

An example was given for Texas but there was nothing implying a Texas only thread.

If the thread title had said it was for Texas only then I'd not have bothered commenting.

 

[GregTR]

That is fair and I think I misread it, and you are correct.

TL;DR: there is no once size fits all optimal solution in Q1 2023.

 

[GregTR]

BTW what is off-grid (grid connected)? You're either grid connected or you're off-grid.

I am absolutely positively sure that 100% consumption offset with 100% self consumption is not an ideal strategy for a grid-tied solution due to the rate of diminishing returns. You will either oversize the system to cover the worst case cloudy periods or you'll undersize the PV and won't be 100% offsetting or undersize the battery and won't be 100% self consuming.

 

[wattmatters]

I think of those systems as being halfway between fully off-grid and fully grid-tied.

e.g. a system which can pass through grid power as required but has no means to synchronise with nor feed power back to the grid.

For instance I have a grid-tied PV system (no battery). Fully synchronised with the grid, always generates the maximal capacity of the PV arrays and any power in excess of household demand is exported to the grid. It cannot operate without the grid. If grid is down, so too is it.

I also have an off-grid inverter with its own PV array and battery which supplies power to the house but it can pass through grid power if so desired. It has an AC input (grid or generator) and of course an AC output.

Because I have a grid tied PV system (11 kW) and the off-grid PV array is small (2.22 kW) then as a general mode of operation I set up the off-grid system to pass through "grid" power during the day (which in reality is being supplied by my grid-tied PV), while the small off-grid PV array is dedicated to recharging the off-grid battery. Then as we approach the evening the system switches over to powering loads from the battery and any solar PV from the off-grid array.

So that I am not reliant on the off-grid system always working, it supplies power to the house via a transfer switch, so I still have the ability to flip the house back to grid only supply if needed.

In terms of financial impact, our grid-tied PV resulted in an annual bill reduction of ~A$3k/year. It cost A$12.8k (professionally installed).


That indicates a reduction of A$3.4k/year but some bill reduction was via some consumption reduction strategies.

My DIY off-grid system was initially built as a grid outage backup system (with lead acid batteries). It was not intended to run the home nor reduce bills. But I soon realised it could actually do that if I added some LiFePO4 capacity. It's too early to say what the full annual impact on our bills will be but the impact on 4th quarter 2022 is shown below.



That's a reduction compared with same quarter in previous years of ~$2.70/day.

I do not expect to achieve the same level of saving across a full year of operation. I think ~$1.50/day is more likely, about A$550/year.

The 10 kWh of LiFePO4 cost A$4k.

In this instance I don't count the cost of the existing off-grid system since that was a separate investment in energy security which I was making either way. Being able to use it for reducing our energy bills is just a nice bonus and as it turns out, reducing bills by $550 for a $4k outlay seems a reasonable outcome.

 

[Shimmy]

Trying to shoot for 100% consumption coverage with 100% self-use is a fool's errand and not worth the investment, especially in Texas where grid access is relatively cheap but without any kind of NEM availability.

That's the bottom line for the analysis. Specific to Texas, things like the "free nights" plans might also skew the optimal solution.

I've gone through a few 8760-hour analysis to compare options in the past, really it is the only way to get to your best solution... based on today's rate structures.

 

[SignatureSolarJames]

But is overly optimistic for the electronics, i.e. inverter.
20 years for batteries is also a pretty heroic assumption.
Don't forget decommissioning and disposal costs for end of life items.
Costs for any additional infrastructure required to safely house such a system.


Any maths problem needs a discussion to define what it is trying to solve, which includes understanding all relevant inputs and assumptions used.

By restricting the discussion to only one type of solution to begin with, you are putting the maths cart before the horse. It's making the bold assumption that grid supported off-grid solar + battery is the right solution to begin with, when there are a raft of options for how one might allocate their limited resources.

In essence the problem with threads like this is the maths is completely different for every individual so you will inevitably end up with a discussion in order to define the optimisation problem to begin with (and what people are looking to optimise varies). Only until that has been done can the maths follow.

It depends on each individual home's priorities/objectives, where they are located, the sort of home they occupy, their energy consumption and consumption patterns, what other energy sources their home uses, and the local grid regulatory and tariff regime they are operating within, how long they expect to be in that home, the space available suitable for solar PV, what additional infrastructure might be required to install such systems etc etc.

As a result the type of home energy solution which is financially rational/optimised can be highly variable.

It should also be considering the demand side of the equation, not just supply. Part of the answer is finding ways to consume less energy to begin with, or to work on shifting consumption to when it is least expensive.

And anything which makes an assumption about the future (e.g. grid tariffs) needs to include some sensitivity analysis to understand whether it is overly reliant on assumptions being correct when predicting the future might be notoriously difficult.

e.g. where I am (Australia) it's almost impossible to make a grid-supported off-grid PV + battery system financially viable no matter how you slice and dice the numbers. Grid-tied solar PV however is almost a complete no-brainer. As for grid-tied batteries, you would be better off just burning a $20 bill every week.

Sorry bud, it is one case study and it is a math convo. Plato is for another thread