Energy Storage Replace Gas Plants

[Jay McKinsey + 1,489]

7 minutes ago, KeyboardWarrior said:

Jay, solar has 24% capacity factor because of how much it generates annually, not how much is sold. 

Solar is not capable of generating more than 24% of its capacity on annual average. Storage allows you to SELL all of it by releasing it when there’s demand.

You are correct, but capacity factor is based on reported nameplate power of your plant. Which I think developers soon begin listing as the 100% capacity of the battery. The solar will just be treated as an internal part of the process just like the fuel efficiency of a gas plant doesn't affect its nameplate power.

Edited May 15, 2020 by Jay McKinsey

[Jay McKinsey + 1,489]

@KeyboardWarrior

So a 100MW/2400MWh battery would be the same as a 100MW NGCC. Asking how much solar the battery requires is like asking how much gas the NGCC requires.

Edited May 15, 2020 by Jay McKinsey

[KeyboardWarrior + 527]

15 minutes ago, Jay McKinsey said:

You are correct, but capacity factor is based on reported nameplate power of your plant. Which I think developers soon begin listing as the 100% capacity of the battery. The solar will just be treated as an internal part of the process just like the fuel efficiency of a gas plant doesn't affect its nameplate power.

Oh I understand. 

[KeyboardWarrior + 527]

@Jay McKinsey

Understanding how the definition changes, I refer back to my original statement: that gas should be able to sell more power, and that net earnings make gas a very attractive investment. 

edit: spelling 

Edited May 15, 2020 by KeyboardWarrior

[Jay McKinsey + 1,489]

16 minutes ago, KeyboardWarrior said:

@Jay McKinsey

Understanding how the definition changes, I refer back to my original statement: that gas should be able to sell more power, and that net earnings make gas a very attractive investment. 

edit: spelling 

Not if I scale up appropriately. I've been working on what that scaling factor is and I see the mistake I made in the NPV. Will post new calculation in a bit.

[Jay McKinsey + 1,489]

@KeyboardWarrior

The mistake was to compare nameplate values. The right thing to compare is output. So I scaled my solar only plant up to the same output as your 1000MW NGCC. The kWH/yr is just about the same. 

 

 

Edited May 15, 2020 by Jay McKinsey

[Jay McKinsey + 1,489]

@KeyboardWarrior

Fixed errors

 

Edited May 15, 2020 by Jay McKinsey

[KeyboardWarrior + 527]

59 minutes ago, Jay McKinsey said:

@KeyboardWarrior

The mistake was to compare nameplate values. The right thing to compare is output. So I scaled my solar only plant up to the same output as your 1000MW NGCC. The kWH/yr is just about the same. 

 

 

No, that was just fine because we're comparing yield on investment capital. If you make power output the same you're just investing more capital into the solar farm. It doesn't really mean anything in terms of yield per dollar spent. In fact, it may get slightly worse as you increase size UNLESS you have batteries. 

Edited May 16, 2020 by KeyboardWarrior

[KeyboardWarrior + 527]

@Jay McKinsey To demonstrate the point more clearly, if I would have simply taken that 2.2 billion dollars and bought a big gas plant where the power was needed, I would have a very rich cash flow and a great return. 

What this means to the value investor is that solar currently has an opportunity cost in certain locations. However, gas has an opportunity cost in regions where the grid is saturated (like where ERCOT operates). 

Edited May 16, 2020 by KeyboardWarrior

[Jay McKinsey + 1,489]

12 minutes ago, KeyboardWarrior said:

No, that was just fine because we're comparing yield on investment capital. If you make power output the same you're just investing more capital into the solar farm. It doesn't really mean anything in terms of yield per dollar spent. In fact, it may get slightly worse as you increase size UNLESS you have batteries. 

Yes, in the real world storage is required. But our hypothetical shows the power of scaling when you have zero marginal cost, free fuel. At the same Wh output my solar plant has greater yield than your NGCC at today's numbers. The same with batteries if cost gets low enough. I''ll be working on that number.

[Jay McKinsey + 1,489]

19 minutes ago, KeyboardWarrior said:

@Jay McKinsey To demonstrate the point more clearly, if I would have simply taken that 2.2 billion dollars and bought a big gas plant where the power was needed, I would have a very rich cash flow and a great return. 

What this means to the value investor is that solar currently has an opportunity cost in certain locations. However, gas has an opportunity cost in regions where the grid is saturated (like where ERCOT operates). 

I agree, solar + storage is not competitive in all markets. But wind + storage often is in those regions and importing renewables with underwater/ground HVDC is going to make solar competitive in dark regions where people live. People mostly live on the coast or on a river.

[KeyboardWarrior + 527]

1 hour ago, Jay McKinsey said:

At the same Wh output my solar plant has greater yield than your NGCC at today's numbers.

I don't believe so, according to division:

Example 1:

$200 M / $1 B = 20% yield

$150 M / $1 B = 15% yield

Example 2:

$200 M / $1 B = 20% yield

$345 M / $2.18 B = 15% yield

Scaling doesn't seem to improve yield per dollar. Am I missing anything?

Edited May 16, 2020 by KeyboardWarrior

[footeab@yahoo.com + 2,135]

On 5/13/2020 at 12:57 PM, Jay McKinsey said:

Yeah, very awesome! It has really had me wondering what is the optimal ratio of battery to solar capacity? I realize it won't be the same in all cases but I suspect the industry will settle on a rough standard. For max revenue the batteries should be fully filled and discharged every day plus some electricity will be sold during the day while production is occurring. Australia is going to be covered by projects like this in a few years.

Simple, get an average of about 4--6 hours of sun a day with non tracking and 8 with tracking over a large portion of earth with a few rare exceptions.  We want power 24 hours a day, so need (24hr-8hrs) name plate capacity = 16X nameplate capacity is a good place to start. Now if  you want batteries to last a LOT longer, need to only use roughly +50%, so you actually need roughly 32X nameplate capacity to cover battery capacity factor and degradation over time.   

[Dan Clemmensen + 1,011]

18 hours ago, KeyboardWarrior said:

Since these plants have access to high temperatures, a more efficient storage medium might be syngas. 

I have to believe this is a losing proposition. CCGTs are already connected to a storage system with multi-month capacity, namely the NG system, so converting their excess capacity back to gas fails. the whole reason to use batteries in this scenario is peak shaving so you can meet peak power demand with less total CCGT peak capacity.

[KeyboardWarrior + 527]

3 hours ago, Dan Clemmensen said:

I have to believe this is a losing proposition. CCGTs are already connected to a storage system with multi-month capacity, namely the NG system, so converting their excess capacity back to gas fails. the whole reason to use batteries in this scenario is peak shaving so you can meet peak power demand with less total CCGT peak capacity.

You aren’t converting it back to “gas”. It’s a mixture of carbon monoxide and hydrogen, and can be made at 80% efficiency. I’m not making claims about expenditure, but I’m wondering if a syngas reactor setup would cost less than batteries. It’s all about finance. 

The main drawback is that you lose 60% of the stored energy because you’re running through the Brayton cycle again.