let's talk about battery banks- I reckon the "they aren't worth it" thing is wrong

[PVfarmer]

Let's look at this from the other side and eliminate the situations where we can all agree that batteries as an adjunct to PV are definitely a bad idea, OK? I'll start:

1) For a facility (home or business) where the facility immediately uses all the power the PV system produces, i.e., the meter never "runs backwards".
2) For areas where there is net metering, i.e., the meter runs backwards and forwards and the utility is only concerned with the bottom line at the end of the month - grid storage is free.

The first: why not? When the meter only runs forwards, you can stop it from doing that at night, or for a business in early evening and morning...with a BESS of course.
The second: why would you want the meter to run forward if you could lower that cost with a BESS?

Check this out- the words I've been looking for-
Avoided Cost.
Both a real thing (according to governmental accountants anyway) and a GOOD thing!

measure of what it would cost the grid to generate the electricity = change it to "what it would cost you (the consumer) to pay the POCO to generate the electricity that is otherwise displaced by a new generation project"
New generation project being = to PV system with and without BESS, for accurate comparison.

Since projected utilization rates, the existing resource mix, and capacity values can all vary dramatically across regions where new generation capacity may be needed, the direct comparison of LCOE across technologies is often problematic and can be misleading as a method to assess the economic competitiveness of various generation alternatives. Conceptually, a better assessment of economic competitiveness can be gained through consideration of avoided cost, a measure of what it would cost the grid to generate the electricity that is otherwise displaced by a new generation project, as well as its levelized cost. Avoided cost, which provides a proxy measure for the annual economic value of a candidate project, may be summed over its financial life and converted to a stream of equal annual payments. The avoided cost is divided by average annual output of the project to develop the "levelized" avoided cost of electricity (LACE) for the project.⁴ The LACE value may then be compared with the LCOE value for the candidate project to provide an indication of whether or not the project's value exceeds its cost. If multiple technologies are available to meet load, comparisons of each project's LACE to its LCOE may be used to determine which project provides the best net economic value. Estimating avoided costs is more complex than estimating levelized costs because it requires information about how the system would have operated without the option under evaluation.
http://www.eia.gov/forecasts/aeo/electricity_generation.cfm#4

When avoided costs = the bill that you didn't pay, it isn't too much more complex...

 

[ggunn]

The first: why not?

Simple. If none of the energy you produce ever leaves your home, what would be the point of storing any of it? Even if your utility has onerous TOU charges, the peak production time for your PV system is about the same time as TOU peak; it would be counterproductive for you to store energy and buy what you need from the utility during the day when power is dear and use the stored energy at night when it is cheap. If you could produce energy at night and use it during the day, you'd have something.

 

[TommyO]

Um...what? PV is only worth it in Maine if the cost is zero?
That makes zero sense!

That's right - that makes zero sense!
And it's not what I said.
However Batteries are only worth it in Maine if the cost is zero!
And that DOES make sense.

Let's try again. Person X lives in Maine and wants a PV system. Should they get PV or PV/BESS?
If, in Maine-
* Sales price of power to the POCO = 15 cents
* Cost to store power (BESS) = 7 cents
* Purchase price for power from the POCO =15 cents...

Where's the rest of the cost? Oh right, the panels and BOS.
So panels + BOS = 7 cents.
Total cost = 14 cents.
14 < 15. Good deal.
Right?

No - it's an "OK" deal.

Instead of $.07/kwh that it costs for PV by itself you're paying $.14/kwh for PV + batteries.
$.07/kwh for just PV would be the "good" deal.

STILL a good deal because that 17 cents isn't the whole story- like I keep saying, there is no electric bill!
The fact that there is no bill is what allows you to pay back for the PV, BESS or not.
So yes, believe it or not, you can spend MORE on PV or PV/BESS than your electricity costs.
The amount you originally spent on panels, BOS, BESS goes down, and eventually becomes profit (or just savings when net metering only). That's kinda the whole idea!

No - the whole point of doing an LCOE analysis is that it gives you a number to compare to POCO cost and see if you'll be better off or not.
At $.17/kwh for PV vs. $.15/kwh for POCO means you don't get to the point where you've saved more in utility bills than you spent to purchase the system.
At least not unless you add an assumption that the POCO rates are increasing during that timeframe.

If you pay 15 cents to POCO and spend 17 on PV etc:
Year 1: You are paying 2 cents more than you were. However, you saved X by not paying the electric bill. That X is your "payback money"- it really makes the "C" in LCOE go down,

Can you provide any link that shows that the "C" in LCOE goes down in year 2, 3, 4, etc and therefore the overall LCOE decreases as well?

Year 2: You are paying 1.5 more.
Year 3: 1 cent more
Year 4: .5 cents
Year 5: 0 cents
Year 6: You broke even the year before, now you are making .5 cents.

It really seems pretty straightforward.

So you're saying that LCOE shouldn't be considered a straight line amount over the life of the system, but rather normally will decrease by 2-3% each year (.005/.17 = ~3%)
To put it mildly that seems non-standard.

 

[PVfarmer]

I freely admit that I don't slog through all the laborious numerical analyses (sorry) but a factor I haven't seen in the ones I have read through is just how much storage we are talking about. For any battery storage hardware there will be some fixed costs that do not depend on the magnitude of the storage needed, while the monetary advantage (if any) will be tied inextricably to that quantity. You can't just say that storage in batteries costs some flat $/kWh down to 1 kWh; if you are only generating a kWh above your usage during daylight hours, batteries are not economically viable no matter what else is in play, even if the POCO just takes it without compensating you for it.

You don't have to apologize! We're all using different numbers, and therefore not speaking the same language in a way.
I did just learn a new word, however!

au·tar·ky
ˈôˌtärkē/
noun
noun: autarky; noun: autarchy

eeconomic independence or self-sufficiency.


a country, state, or society that is economically independent.


(Or a house, apparently...)

[h=3]How much is too much?[/h] It is essential to ask what the end-customer wants, whether a large rate of self-consumption from his own power or a high autarky degree. This is linked to the question of whether the consumer wants to achieve high returns, or when it is enough that he does not have to pay additional fees, that a certain level of autarky is appreciated. An accurate return calculation is not really possible as it varies highly with electricity price development prognoses, battery life and price decline of replacement batteries.
A rough estimation shows the sum for a period of 20 years for a system in a household that utilizes 4,700 kWh per year and manages 50% autarky.



The proportion of own consumption from self-generated power and the autarky degree will increase with a bigger storage capacity. The effect is lower with increasing size (see Graph “Influence of battery and solar capacity on the autarky degree”, p. 98).
In other words, for every extra kWh of battery capacity there is a little less extra cost savings. With the 4,700 kWh usage scenario and a 5 kW installation, the possible sensible limit will be a 4 to 6 kWh battery capacity.
The tendency is such that smaller solar installations can be financed easier with smaller batteries than bigger ones due to cost savings. They do not promise exceptionally high autarky.

http://www.pv-magazine.com/archive/articles/beitrag/storage-has-landed-_100009059/501/#axzz3gechEvSw
​

 

[jim dungar]

We're all using different numbers, and therefore not speaking the same language in a way.

Actually, it appears to only be you that is having an issue with not understanding the numbers being used.

Your behavior is seems much closer to 'trolling' than it is to open discussion.

 

[ggunn]

Actually, it appears to only be you that is having an issue with not understanding the numbers being used.

Your behavior is seems much closer to 'trolling' than it is to open discussion.

It is a pleasant fantasy to be the neophyte who comes in with a naive viewpoint not corrupted by conventional wisdom and sees the thing that no one else can see that turns things on their head, but it rarely happens in the real world.

 

[PVfarmer]

That's right - that makes zero sense!
And it's not what I said.
However Batteries are only worth it in Maine if the cost is zero!
And that DOES make sense.

Instead of $.07/kwh that it costs for PV by itself you're paying $.14/kwh for PV + batteries.
$.07/kwh for just PV would be the "good" deal.

No - the whole point of doing an LCOE analysis is that it gives you a number to compare to POCO cost and see if you'll be better off or not.
At $.17/kwh for PV vs. $.15/kwh for POCO means you don't get to the point where you've saved more in utility bills than you spent to purchase the system.
At least not unless you add an assumption that the POCO rates are increasing during that timeframe.

Can you provide any link that shows that the "C" in LCOE goes down in year 2, 3, 4, etc and therefore the overall LCOE decreases as well?

So you're saying that LCOE shouldn't be considered a straight line amount over the life of the system, but rather normally will decrease by 2-3% each year (.005/.17 = ~3%)
To put it mildly that seems non-standard.

There's a Talking Heads song called Stop Making Sense- could you please do the opposite?
Just trying to be friendly there.

If you paid (in Maine) 7 cents for PV only, and that covered ~ ½ the bill, if you instead paid 14 cents for PV + BESS, and had zero bill instead of ½ a bill...
14 cents worth of PV + BESS doesn't put out the same amount of kWh as 7 cents of PV- it might, but someone would have done something wrong in that case.

At $.17/kwh for PV vs. $.15/kwh for POCO means you don't get to the point where you've saved more in utility bills than you spent to purchase the system.

Yes, you do- you only paid the 17 cents once, up front. The 15 cents that you are NOT paying because you bought the PV is what gives you your payback period.
The PV cost = $10,000.
PV output over 20 years = 58,823.5 kWh.
That makes the LCOE (10,000 * 100) / 58,823.5 = 17 cents.

kWh usage a year = 2941kWh (that is = to PV output over 20 years)
Savings per year = $441.25
Payback = 22.6 years. (The 17 cent LCOE used is of course way too high)

If the system cost $7647, for an LCOE of 13 cents, payback would be 17.3 years.

So yes, in fact- let me rephrase it- the LCOE stays the same, and the problem there is that the C you started with is going DOWN in reality- you do in fact get to the point where you've saved more, with or without PV cost > POCO rate.

Can you provide any link that shows that the "C" in LCOE

I did.

So you're saying that LCOE shouldn't be...

I'm saying it is one part of the puzzle that requires other pieces to complete the puzzle.

 

[PVfarmer]

Can you provide any link that shows that the "C" in LCOE goes down in year 2, 3, 4, etc and therefore the overall LCOE decreases as well?

It has to be combined with another "metric" to give the real picture of what is happening.

Avoided cost, which provides a proxy measure for the annual economic value of a candidate project, may be summed over its financial life and converted to a stream of equal annual payments. The avoided cost is divided by average annual output of the project to develop the "levelized" avoided cost of electricity (LACE) for the project.⁴

http://www.eia.gov/forecasts/aeo/electricity_generation.cfm#4

 

[jim dungar]

It has to be combined with another "metric" to give the real picture of what is happening.

You are still avoiding any real answer.

Using any of the numbers given by any one of the other posters, please show how their simple example is either wrong or how you agree with it.
Do not provide any other external link or reference.

 

[TommyO]

If you paid (in Maine) 7 cents for PV only, and that covered ~ ½ the bill,

Why wouldn't I just install PV for 7 cents per kwh and cover the entire bill?

The answer is that is exactly what most people would do.
They install PV such that it eliminates their *entire* bill.

 

[wwhitney]

Gotta run, just a few quick error corections:

If, in Maine-
* Sales price of power to the POCO = 15 cents
* Cost to store power (BESS) = 7 cents
* Purchase price for power from the POCO =15 cents...

Where's the rest of the cost? Oh right, the panels and BOS.
So panels + BOS = 7 cents.
Total cost = 14 cents.
14 < 15. Good deal.
Right?

Slightly better than the POCO, but not nearly as good as PV only, because Maine has net metering. 7 << 14 < 15. 7 cents/kWh (PV only) wins.

If panels + BOS = 10 cents...BESS is the same...total cost = 17 cents.
STILL a good deal because that 17 cents isn't the whole story- like I keep saying, there is no electric bill!

Yes, 17 cents is the whole story, and no it's not a good deal, you are paying more for PV plus batteries than you would for either PV only or straight POCO. LCOE for the PV is the whole story, and LCOS for battery is the whole story.

The fact that there is no bill is what allows you to pay back for the PV, BESS or not.
So yes, believe it or not, you can spend MORE on PV or PV/BESS than your electricity costs.
The amount you originally spent on panels, BOS, BESS goes down, and eventually becomes profit (or just savings when net metering only). That's kinda the whole idea!

No, you can't buy a PV system with an LCOE higher than the POCO prices and expect to save money. While you will save money on your POCO bill each month after the initial outlay for the system, your system will die before it pays back for itself. That is what LCOE > POCO means.

If you pay 15 cents to POCO and spend 17 on PV etc:
Year 1: You are paying 2 cents more than you were. However, you saved X by not paying the electric bill. That X is your "payback money"- it really makes the "C" in LCOE go down, which is why LCOE is not really the whole story.
Year 2: You are paying 1.5 more.
Year 3: 1 cent more
Year 4: .5 cents
Year 5: 0 cents
Year 6: You broke even the year before, now you are making .5 cents.

Too wrong to comment on. Please reread my "race" post.

Cheers, Wayne

 

[wwhitney]

If you paid (in Maine) 7 cents for PV only, and that covered ~ ½ the bill, if you instead paid 14 cents for PV + BESS, and had zero bill instead of ½ a bill...
14 cents worth of PV + BESS doesn't put out the same amount of kWh as 7 cents of PV- it might, but someone would have done something wrong in that case.

What are you talking about? These are all prices per kWh. They all put out 1 kWh, of course they put out the same amount.

Yes, you do- you only paid the 17 cents once, up front. The 15 cents that you are NOT paying because you bought the PV is what gives you your payback period.
The PV cost = $10,000.
PV output over 20 years = 58,823.5 kWh.
That makes the LCOE (10,000 * 100) / 58,823.5 = 17 cents.

kWh usage a year = 2941kWh (that is = to PV output over 20 years)
Savings per year = $441.25
Payback = 22.6 years. (The 17 cent LCOE used is of course way too high)

Right. 22.6 years is greater than 20 years, your system dies before it can pay itself back.

Cheers, Wayne

 

[PVfarmer]

You are still avoiding any real answer.

Using any of the numbers given by any one of the other posters, please show how their simple example is either wrong or how you agree with it.
Do not provide any other external link or reference.

Ok. I said-
The LCOE does in fact matter- it has to be below 5 cents or you are wasting your time.
Right?

The response was not polite or correct.
If the LCOE was a penny as per this example, you would have the least expensive PV system that had ever been installed, and would be very happy.

And then the Scenario C example provided proves that LCOE does matter, as $1.10 is less than $1.50.

To me, "you're still wasting your time if it's $.01/kwh or $.14/kwh" is a very wrong statement. What do you think?


Cost no PV: 10 * .15 = $1.50 per night or
$1.10 if LCOE/PV=$.01/kwh

No - you're still wasting your time if it's $.01/kwh or $.14/kwh


Scenario A: I buy from the POCO (no solar, no batteries)
Cost: 10 * .15 = $1.50 per night

Scenario C: I generate the power, I put it into batteries and then use it.
Cost: 10*L + 10*$.10
$1.10 if L=$.01/kwh
$2.40 if L=$1.40/kwh

 

[TommyO]

Ok. I said-
The LCOE does in fact matter- it has to be below 5 cents or you are wasting your time.
Right?

The response was not polite or correct.
If the LCOE was a penny as per this example, you would have the least expensive PV system that had ever been installed, and would be very happy.

And then the Scenario C example provided proves that LCOE does matter, as $1.10 is less than $1.50.

Sure, $1.10 is less than $1.50.
But $.10 (Scenario B with PV only with that same LCOE) is less than both and is the clear winner.

To me, "you're still wasting your time if it's $.01/kwh or $.14/kwh" is a very wrong statement. What do you think?

Lets look at $0.01/kwh
A: $1.50
B: $.10
C: $1.10
Answer - B is the best

Ok - so let's look at $.14/kwh
A: $1.50
B: $1.40
C: $2.40
Answer - B is the best

SO: I think it's accurate to say that for the entire range, batteries are not a good choice ("wasting your time" as you put it)

 

[winnie]

Hmm, I think I see something that PVfarmer is getting at.

Remember that LCOE and 'LCOS' are both nearly fictional numbers. They are an attempt to make an apples to apples comparison between different sources of electrical supply with totally different cost structures.

In the ideal PV system you pay a large chunk of money up front, but then get electricity for 'free' forever, but only when the sun is shining. Real PV systems will degrade over time, and will require upkeep, and so will have a limited life span and on-going costs.

You know the up-front installation cost. You make guesses about how much energy the system will produce over time, guesses about the 'cost of money' (how much interest you have to pay for the loan, or how much investment income you are foregoing), guesses about the lifetime of the system. You _pick_ a payback period, the time over which you want to spread your initial costs. With all of these parameters you _calculate_ an LCOE. That number is IMHO a nearly fictional number which you use to make comparisons with (say) buying power from the POCO.

The payback period that you select can move this number all over the map. Make the payback period one year, and your LCOE is huge. Make the payback period longer than the expected life of the system and you are deluding yourself. A reasonable question to ask is: what payback period makes the LCOE equal to your current power rates.

After your payback period has elapsed, you are getting 'free' electricity, excepting of course the ongoing maintenance expenses. Just like after you've paid off your mortgage you are living in a free house, excepting of course ongoing maintenance, taxes, etc.

PVfarmer, IMHO the concept of LCOE already factors in a number of factors that you are trying to count _again_. Yes, once you have sunk the costs for the PV system the electricity _is_ essentially free. LCOE helps you figure out if it makes sense to sink the cost in the first place.

You calculate an LCOE, hope your assumptions on things like maintenance costs are correct, and then compare that to what you expect for POCO costs. If your LCOE, over a reasonable payback period compares well to POCO costs, then that is a strong argument for PV. Your calculated LCOE doesn't need to be less than current POCO charges for PV to be a good deal; you have to consider rising POCO rates, cost stability, etc.

For some people it might make sense to sink that cost in the first place regardless of any reasonable LCOE calculation. Some people have a religious reasons for not connecting to the grid; it doesn't matter to them what LCOE versus POCO costs are; they care about LCOE of PV versus LCOE of some other private production system.

-Jon

 

[PVfarmer]

Originally Posted by TommyO No - you're still wasting your time if it's $.01/kwh or $.14/kwh

Sure, $1.10 is less than $1.50.

Dude, the LCOE tells you how many cents it (will) cost to produce a certain amount of kWh.
It isn't an entirely solid number, because while you know how much you will be spending, as you wrote the check, you don't just discard the PV after 20 years; you keep using it or recycle it, and the output is predicted by software, so the LCOE is an estimate, ½ real and ½ educated guess.

Please advise- what meaning does it have beyond that?
You do the equation, you get the number, there is a reason you did that.
What is your own personal reason?

The way you are using the LCOE is foreign to me.
I am honestly trying to help you here.

This is one way LCOE is typically used:
Customer X wants PV.
X pays POCO 15 cents, and will be net metering and wants to pay zero to POCO at the end of the day/month/year.
System 1 output = system 2 output, both = usage exactly.

System 1 uses- panels A - inverter B - BOS C - labor D.
LCOE of system 1 is 17 cents.
System 2 uses- panels E - inverter F - BOS G - labor H. In system 2, either the cost of E or F or G or H, or some combination of any of those, is lower.
So the LCOE of system 2 is 13 cents.

Which is better for customer X, system 1 or 2?

 

[PVfarmer]

Hmm, I think I see something that PVfarmer is getting at.

Before I even read beyond that....
:jawdrop:
:thumbsup:
A THOUSAND THANKS, sir!!
I was starting to wonder if I might be losing my mind...

Now I'll read the rest of you comment!

 

[wwhitney]

You _pick_ a payback period, the time over which you want to spread your initial costs.

For the analysis you describe, I wouldn't call it a payback period. Call it an amortization period.

Payback time is (upfront costs/annual savings). The uncertainty is in maintenance costs, future POCO prices (for the savings), usage pattern, and PV lifetime/degradation rate. Maybe others I forgot.

LCOE should be calculated over the PV lifetime, the only uncertainty is maintenance costs and PV lifetime. If you calculate it over an arbitrary amortization period, don't call it LCOE. You haven't levelized the costs--you are assigning them over a specific period, then they drop to zero.

Of course, I don't know what the industry standard lingo is, my background is in financial math. So take the above with the appropriate grain of salt.

Cheers, Wayne

 

[TommyO]

Dude, the LCOE tells you how many cents it (will) cost to produce a certain amount of kWh.

I believe you have the concept.

Please advise- what meaning does it have beyond that?
You do the equation, you get the number, there is a reason you did that.
What is your own personal reason?

The way you are using the LCOE is foreign to me.

What are you finding that's "foreign" to you?

Why don't you calculate the 3 different scenarios (A: POCO, B: PV-only, C: PV+Batteries) the way you'd like to using the previous values ($.15/kwh for buying/selling power with POCO, $.10/kwh for storing power, $.01/kwh for PV LCOE vs. $.14/kwh for PV LCOE). Most likely you'll find that PV-only is the clear winner in both those scenarios. If somehow you come to the conclusion that option C is the best choice please share your calculations and we can explain where you've gone wrong.

I am honestly trying to help you here.

And with comments like that I think more and more that you're just having fun trolling.

This is one way LCOE is typically used:
Customer X wants PV.
X pays POCO 15 cents, and will be net metering and wants to pay zero to POCO at the end of the day/month/year.
System 1 output = system 2 output, both = usage exactly.

System 1 uses- panels A - inverter B - BOS C - labor D.
LCOE of system 1 is 17 cents.
System 2 uses- panels E - inverter F - BOS G - labor H. In system 2, either the cost of E or F or G or H, or some combination of any of those, is lower.
So the LCOE of system 2 is 13 cents.

Which is better for customer X, system 1 or 2?

System 2 of course. What's your point?
BTW I noticed that you didn't point out the other major factor that affects LCOE - system life. So it could be lower cost for system 2, OR it could be that system 2 is expected to last 5 years longer than system 1 and that is why it's LCOE is lower.

 

[PVfarmer]

For the analysis you describe, I wouldn't call it a payback period. Call it an amortization period.
Payback time is (upfront costs/annual savings). The uncertainty is in maintenance costs, future POCO prices (for the savings), usage pattern, and PV lifetime/degradation rate. Maybe others I forgot.
...arbitrary amortization period, don't call it LCOE. You haven't levelized the costs--you are assigning them over a specific period, then they drop to zero.
Of course, I don't know what the industry standard lingo is, my background is in financial math. So take the above with the appropriate grain of salt.

Hmm, I think I see something that PVfarmer is getting at.

Remember that LCOE and 'LCOS' are both nearly fictional numbers. They are an attempt to make an apples to apples comparison between different sources of electrical supply with totally different cost structures.

You _pick_ a payback period...

PVfarmer, IMHO the concept of LCOE already factors in a number of factors that you are trying to count _again_. Yes, once you have sunk the costs for the PV system the electricity _is_ essentially free. LCOE helps you figure out if it makes sense to sink the cost in the first place.

You calculate an LCOE, hope your assumptions on things like maintenance costs are correct, and then compare that to what you expect for POCO costs. If your LCOE, over a reasonable payback period compares well to POCO costs, then that is a strong argument for PV. Your calculated LCOE doesn't need to be less than current POCO charges for PV to be a good deal; you have to consider rising POCO rates, cost stability, etc.

I'll get back to your previous comments that I skipped, Wayne- I think we're onto something here.
The "lifetime" of a project.
If the PV "lasts" 20 years, lifetime is 20. That's the number you use to "get" (levelize?) the LCOE- the PV puts out ~X a year for 20 years.
You could pick 15 or 17 or 23, nut you have to pick something as a constant.

The lifetime *starts* at zero, so goes 0-20.
Aha! Check this out!

You have another variable, call it P.
P= years elapsed!! Meaning the point you are at between 0 and 20 years...
If you DIVIDE P by 20....!
1/20=.05
2/20=.1
3/20=.15

You get a "factor" called (by me).... CY, or Current Year.

And THAT is how you get to the bolded things in the QUOTE section!!

Feel me?


The CY goes from 0-1, just as the lifetime goes from 0-20.
Not sure how to put it in accounting or mathematical terms.

As P, years elapsed, goes up- and if you factor P into both the LCOE and LACE via the CY factor...they meet at "payback"!
I have this scribbled on a notepad- gotta type it up before I lose it!