How do you calculate your $/kWh for a off-grid PV system that's competing with grid-tied?

[45North]

Let's get the ball rolling with an over-simplified example:

The goal is to make a PV system that can produce electrical power cheaper that your local utility. (Let's assume rates in NA are around $0.10 to $0.30 per kWh, depending on cost of production at your location, time of day metering, etc.) For a PV system you will be buying equipment once for an up-front capital cost. After that, your operating costs can be assumed to be 0 for the service life of your equipment. So your cost per kWh will be reduced for every hour you operate.

• Let's say you can build a system for $1/w (to keep the math simple) So that's $1000 per KW.
• Let's also leave out the time value of money for now (interest rates are low) and the opportunity cost of not investing elsewhere.
• So, let's say you build a 1Kw system and operate it for one service life of all components - if it's 10,000 hours that would bring your cost down to $0.10 per kWh; and if it's 20,000 hours that would bring your cost down to $0.05 per kWh. Which is more realistic?

What do you think? How do you figure your costs? Are you able to compete with your local utility? When do you replace your equipment? When do you upgrade?

 

[svetz]

I believe mine was expensive.... around $3/w, but if I recollect, based on electricity cost the payback was under 10 years... so I suspect most PV systems will be cheaper than the utility. But that's primarily because I'm grid-tied. If I had to add batteries into the costs, I'm sure I wouldn't have a positive return on the investment.

...When do you replace your equipment?

25 year warranty on panels and microinverters.

... How do you figure your costs? ...

SAM. It takes into account the inflation costs of electricity, degradation of the panels, incentives, electric rates, insurance, shading, salvage value, grid limiting (e.g., 1741), operating costs, etc. Doesn't consider $ if you had invested the same amount into other investments, but with an online calculator for investments it isn't hard to compare the two numbers.

 

[Hedges]

Check a website for "insolation", equivalent hours of full sun (1000 W/m^2) for your location.
Some include historical weather (cloudy days), some do not.

For my location (San Jose) I'm given 5.5 effective hours sun, so 1000W peak PV system is expected to generate 5.5 kWh/day average around the year.
365 x 5.5 = 2000 kWh/year.
Expect the system to last 10 years.
PV panels ought to be 25 to 40 years (with some degradation in output as time goes by), but some have quality issues and fail sooner.
Expect inverters to last 10 years. Out of 5 owned for 17 years, one failed and was repaired in-warranty around 2 years, one out of warranty around 10 years. Other 3 and repaired one worked until swapped for other models at end of 17 years.

10 x 2000 = 20,000 kWh for $1000 investment.
$1000/20,000 kWh = $0.05/kWh

So I see that as break even in 3 years (or less for power during higher rate times of day, but those are now late in the afternoon so my production has declined)

At that point I expect you to spend $100 for 1000W of replacement inverter (I've paid $500 for 5000W), and operate another 10 years.
Second decade, $100/20,000 kWh = $0.005/kWh
(Or, over 2 decades $1100/40,000 kWh = $0.0275/kWh)

Compare those costs to $0.15 or $0.45/kWh, typical utility rates in my area now.
IF (big if) you can have net metering with no overhead cost and credit for power same as price to purchase power, it is a big win.
Reality is, you are probably forced on a higher priced time of use schedule designed to let you enjoy less benefit.
For the time being, I think we are best off if we over-build solar to ensure we are a net exporter of energy, not a net consumer.

These figures are different from 17 years ago when I paid $5/W for PV panels, $0.80/W for inverters, total $8/W for the complete system (installed my own labor) and got a $4/W rebate. Utility rates were 1/2 to 2/3 of what they are today.

 

[FilterGuy]

Let's get the ball rolling with an over-simplified example:

The goal is to make a PV system that can produce electrical power cheaper that your local utility. (Let's assume rates in NA are around $0.10 to $0.30 per kWh, depending on cost of production at your location, time of day metering, etc.) For a PV system you will be buying equipment once for an up-front capital cost. After that, your operating costs can be assumed to be 0 for the service life of your equipment. So your cost per kWh will be reduced for every hour you operate.

• Let's say you can build a system for $1/w (to keep the math simple) So that's $1000 per KW.
• Let's also leave out the time value of money for now (interest rates are low) and the opportunity cost of not investing elsewhere.
• So, let's say you build a 1Kw system and operate it for one service life of all components - if it's 10,000 hours that would bring your cost down to $0.10 per kWh; and if it's 20,000 hours that would bring your cost down to $0.05 per kWh. Which is more realistic?

What do you think? How do you figure your costs? Are you able to compete with your local utility? When do you replace your equipment? When do you upgrade?

There is 8760 hours in a year. If you buy decent equipment (not even top end) and do a decent design that does not stress the equipment, you should be able to get several years (5+?) without a failure. Furthermore, if/when you do get a failure, you don't have to replace everything.

Lets assume 5 years. That is 43,800 hours or $.023/KWh.

Note: I put my system in 14 years ago and it is still running strong without a single failure. 14 years ago the $/KWh was a *lot* higher than it is today but it has paid for itself many times over.

 

[Hedges]

There is 8760 hours in a year. If you buy decent equipment (not even top end) and do a decent design that does not stress the equipment, you should be able to get several years (5+?) without a failure.

In the life of an off-grid battery inverter. Fewer hours per year for grid-tie.
What sort of system have you run for 14 years? (I suspect it isn't the micro-hydro and LiFePO4 in your profile.)

 

[FilterGuy]

In the life of an off-grid battery inverter. Fewer hours per year for grid-tie.
What sort of system have you run for 14 years? (I suspect it isn't the micro-hydro and LiFePO4 in your profile.)

This is the system currently on my house.



I should have pointed out it is a grid tie system..... Sorry.
The reason I used 5 years for the calculation is that battery based systems are more complex and therefor are more likely to fail.

As an aside: Just this past week I got notified by PG&E that my very favorable E-6 billing schedule is going away. Consequently, I am probably going to add storage to the system to do peak shaving. When I do that, I will add critical circuit battery back-up to the system but that is not the primary design objective. I am debating all of my options. (DC Coupled, AC coupled, High or Low Voltage Battery, Use the solar during black-outs or not, etc) The options run from fairly straight forward to somewhat complex.

The easiest would be to add an inverter and some batteries running the critical subset of the house circuits. The only complexity would be picking an inverter that can easily switch to battery during peak hours and then switch to grid and charge the batteries during off-peak hours. This would go in parallel to the existing system, but I would not be able to use the solar during black-outs.

Since this thread is about return on investment: There is probably no financial justification for me adding batteries. The ROI will be horrible.....I am looking at it as a combination of Hobby costs, Sticking it to PG&E and the limited value of battery back up. (I have been spared the blackouts so far... I am not in a fire prone area so it is limited risk for Public Safety Power Shutdown events. Furthermore, I was once told I am on the same main circuit as a nearby hospital and they try not to let it go down if there are grid problems.)

 

[Dave (Boog)]

Check a website for "insolation", equivalent hours of full sun (1000 W/m^2) for your location.
Some include historical weather (cloudy days), some do not.

For my location (San Jose) I'm given 5.5 effective hours sun, so 1000W peak PV system is expected to generate 5.5 kWh/day average around the year.
365 x 5.5 = 2000 kWh/year.
Expect the system to last 10 years.
PV panels ought to be 25 to 40 years (with some degradation in output as time goes by), but some have quality issues and fail sooner.
Expect inverters to last 10 years. Out of 5 owned for 17 years, one failed and was repaired in-warranty around 2 years, one out of warranty around 10 years. Other 3 and repaired one worked until swapped for other models at end of 17 years.

10 x 2000 = 20,000 kWh for $1000 investment.
$1000/20,000 kWh = $0.05/kWh

So I see that as break even in 3 years (or less for power during higher rate times of day, but those are now late in the afternoon so my production has declined)

At that point I expect you to spend $100 for 1000W of replacement inverter (I've paid $500 for 5000W), and operate another 10 years.
Second decade, $100/20,000 kWh = $0.005/kWh
(Or, over 2 decades $1100/40,000 kWh = $0.0275/kWh)

Compare those costs to $0.15 or $0.45/kWh, typical utility rates in my area now.
IF (big if) you can have net metering with no overhead cost and credit for power same as price to purchase power, it is a big win.
Reality is, you are probably forced on a higher priced time of use schedule designed to let you enjoy less benefit.
For the time being, I think we are best off if we over-build solar to ensure we are a net exporter of energy, not a net consumer.

These figures are different from 17 years ago when I paid $5/W for PV panels, $0.80/W for inverters, total $8/W for the complete system (installed my own labor) and got a $4/W rebate. Utility rates were 1/2 to 2/3 of what they are today.

"10 x 2000 = 20,000 kWh for $1000 investment.
$1000/20,000 kWh = $0.05/kWh"

I'd be interested in what kind of 1000 watt Solar system you were able to buy or build for a total $1000 investment? Is it a grid tie or off grid system?

 

[Hedges]

"10 x 2000 = 20,000 kWh for $1000 investment.
$1000/20,000 kWh = $0.05/kWh"

I'd be interested in what kind of 1000 watt Solar system you were able to buy or build for a total $1000 investment? Is it a grid tie or off grid system?

Just notional size. OP wrote,
"Let's say you can build a system for $1/w (to keep the math simple) So that's $1000 per KW."

What I can do is build a 5000W grid-tie PV system for $5000.

New old stock SMA 5000US inverter $500 (I've bought before on eBay)
Or, current model for $1500 to comply with latest codes

SMA Sunny Boy SB5.0-1SP-US-41 Grid Tie Inverter | eBay

Find many great new & used options and get the best deals for SMA Sunny Boy SB5.0-1SP-US-41 Grid Tie Inverter at the best online prices at eBay! Free shipping for many products!

www.ebay.com

21, 285W PV panels for 6000W (STC), $1685 (probably 7s3p array, but I haven't checked the voltages)

SunTech 285W | SanTan Solar

store.santansolar.com

That's $3185 for the core materials. Easily under $5000 for everything need.
If you have to pay $50/panel for RSD modules, that'll add $1000 but should still be within budget if you skip on mounting hardware.

 

[Dave (Boog)]

Just notional size. OP wrote,
"Let's say you can build a system for $1/w (to keep the math simple) So that's $1000 per KW."

What I can do is build a 5000W grid-tie PV system for $5000.

New old stock SMA 5000US inverter $500 (I've bought before on eBay)
Or, current model for $1500 to comply with latest codes

SMA Sunny Boy SB5.0-1SP-US-41 Grid Tie Inverter | eBay

Find many great new & used options and get the best deals for SMA Sunny Boy SB5.0-1SP-US-41 Grid Tie Inverter at the best online prices at eBay! Free shipping for many products!

www.ebay.com

21, 285W PV panels for 6000W (STC), $1685 (probably 7s3p array, but I haven't checked the voltages)

SunTech 285W | SanTan Solar

store.santansolar.com

That's $3185 for the core materials. Easily under $5000 for everything need.
If you have to pay $50/panel for RSD modules, that'll add $1000 but should still be within budget if you skip on mounting hardware.

Ok I was just wondering where the $1000 figure came from. I couldn't do much with a 1000 watt system. Like you said, it would produce between 5-6 kwh per day as long as the sun was shining.
However, in my experience, when you add up all of the parts you need, including Solar panels, mounting, cables, Grid tied inverter, distribution panel(s), breakers, paying someone to install, and electrician etc. etc. I don't think I could cobble together a 5k system for $5000.

 

[Hedges]

Yes, hired labor costs more.

Using the $286 zero-export 1000W inverter from following post, I think you can build a 1000W grid-tie zero-export system for $1000.
Very little other cost to make it work as a DIY project. Four panels could be mounted as awnings over your windows (or ground mount) and the included power cord hooks it up.

Planning to reduce electric bill by solar push, no feed back to grid. Daylight hours only.

Here's my thoughts, I have a swamp cooler to cool my home (mobile home), it uses 222 watts when running on low, 120 volt, 440 watts on high. Fridge uses 125 watts when it kicks on, 4.24 KWH in 48 hours time, no big deal. I want to purchase 2x 100 watt panels, and a 600 watt, grid tie inverter...

diysolarforum.com

 

[toms]

It’s difficult to accurately cost off-grid power, as ironically the more kwh you use the lower the cost.

Every off grid system i set up uses more power than the client would use if they were grid connected.

Each case is different, but as a rule of thumb i use a 10 year lifespan on quality components.

 

[Steve_S]

Off-Grid here.
Cost for Power Co to bring power to property & house estimate was $45-50,000. That would have been a 240V/200A service. Rural & Remote on a Granite Ridge with the closest powerline 1km away through an old forest.
Cost for Initial Solar System $11,000 with 856AH(gross) FLA Bank. 2015 pricing.
Upgrades & Add-ons 6,000 which includes 2K for Inverter & 1190AH of LFP and all the goodies to with that.
So, that is $17,000 CAD ($14K USD +/-)

The Midnite SCC tracks AH & kWh Generated over it's operational life, unfortunately I lost 50% of the data when the SCC was sent to repair & got a Factory Reset which zeroed it out... My numbers therefore don't make sense to calculate cost of equipment against kWh generated (major piss off for me).

Grid Hookup costs can be crazy or not depending on location, access to grid power & Power Co fees which are all over the place pending where you are. Then of course if you intend to Push power to the grid, tack on another 20-35% in fees & costs above and beyond for that privilege. The IRONY IS, with the falling cost of Batteries, that extra 20-35% loped on top for fees, permits etc for grid tie, would actually pay for a good percentage of Batteries that would be required. Then when you work out that Grid Power pays you 0.10 per kWh but charges you 0.15+ pr kWh to buy it back, take that "cream" off, add it up over 365 days and it's significant AND there is no guarantee that rate will stay the same for any period of time either. Many Big Power Co's are trying to reduce what they pay out and increase the charge rates at the same time, this is happening in many places right now with some even being in court to do so.

At least when you are off grid, it does not matter WHAT the Big Power Co charges or changes it's rates to, nor when they drop power or have brownouts. That extra cash that stay's in your pocket every month is certainly used for MANY better things than to pay for Power Hostage Fees. If you're paying out $100 a month / $1200 a year for power that adds up over time... Plopp that on your mortgage instead and it will pay you back even faster... Many people spend more than that per month on Grid Power.