3 ms·

The key figure of merit for comparing energy storage is not $/KWh, but rather $/KWh*Number of cycles. Li-ion only has about a 1000 cycles. Assuming one cycle pe

The key figure of merit for comparing energy storage is not $/KWh, but rather $/KWh*Number of cycles. Li-ion only has about a 1000 cycles. Assuming one cycle per day (solar charge during day+discharge during night) in 50 years there are 18000 cycles. The figure of merit for Li-ion is 250x1000=2.5e5. My estimate (and those of others) is that this costs up to about $2000/KWh. So the figure of merit for this is 1000x18000=0.9e7. Two orders of magnitude better than Li-ion. Edit: I am ignoring the cost of capital, interest rates etc. Somebody should do this analysis.

mikeyouse 12y ago
Sandia National Labs has done the math but it's quite the PDF at 340 pages describing just about every variety of energy storage: http://www.sandia.gov/ess/publications/SAND2013-5131.pdf http://www.sandia.gov/ess/publications/SAND2013-5131.pdf
jpau 12y ago
I did some quick analysis comparing the proposed project against Li-ion, adjusting for storage and lifespan, and the result is very hopeful. I used values for $/KWh and lifespan from abdullahkhalids' comment. The analysis shows that if you can get more than one cycle per day out, the proposal offers fantastic value. Model visible here - http://imgur.com/KmQPb8P http://imgur.com/KmQPb8P Finance uses something called 'equivalent annual cash flows' (EACFs) to compare projects of different lifespans. Using EACFs makes this analysis very simple. Using abdullahkhalids' figures and assuming a single cycle per day, the equivalent annual cash flows per KWh are equal when the cost of capital is about 4.35%. Something magical happens when you assume can get more than one cycle per day. At two cycles per day, your funding costs could be 10% and the project still feasible! At four cycles per day and with a 5% cost of capital, Li-ion costs almost four times as much as the proposed project. Could someone shine some light on how many cycles per day is realistic for both Li-ion & the proposal? PS if you find this sort of analysis interesting and want to hit me up to talk about such things, feel free to send me an email (available in my profile).

mikeyouse 12y ago

Sandia National Labs has done the math but it's quite the PDF at 340 pages describing just about every variety of energy storage: http://www.sandia.gov/ess/publications/SAND2013-5131.pdf http://www.sandia.gov/ess/publications/SAND2013-5131.pdf

jpau 12y ago

I did some quick analysis comparing the proposed project against Li-ion, adjusting for storage and lifespan, and the result is very hopeful. I used values for $/KWh and lifespan from abdullahkhalids' comment. The analysis shows that if you can get more than one cycle per day out, the proposal offers fantastic value. Model visible here - http://imgur.com/KmQPb8P http://imgur.com/KmQPb8P Finance uses something called 'equivalent annual cash flows' (EACFs) to compare projects of different lifespans. Using EACFs makes this analysis very simple. Using abdullahkhalids' figures and assuming a single cycle per day, the equivalent annual cash flows per KWh are equal when the cost of capital is about 4.35%. Something magical happens when you assume can get more than one cycle per day. At two cycles per day, your funding costs could be 10% and the project still feasible! At four cycles per day and with a 5% cost of capital, Li-ion costs almost four times as much as the proposed project. Could someone shine some light on how many cycles per day is realistic for both Li-ion & the proposal? PS if you find this sort of analysis interesting and want to hit me up to talk about such things, feel free to send me an email (available in my profile).