I wanted to get started in building a powerwall, thinking I could do it relatively cheaply. I love the DIY/reusing junk/getting something for super cheap with a little work aspect of this hobby, which is how I tend to live my life.
So I bought some ebay cells to get started and as an experiment to see what it would really take to make this happen. I paid more than I wanted, but I was OK with the price if I could recover 66% of the cells. They were all Lenovo/IBM packs, with cells of either 2200 or 2400 mah, based on the pack's label. I also had just about 6 packs from other brands. I had no other leads on cheaper batteries at the time other than quantities of 1 or 2 at time.
Here are my disappointing results:
So, Lets say I want to make a 7Kw powerwall, like a tesla. Here is what I would need.
150 cells @ 2 Ah = 300 Ah
300 Ah * 3.7v = 1110 Wh = 1.11 KwH
150 cells * 4 (in series) = 600 cells to make one 14.8 volt- 1.11 KwH battery
600 cells * 7 = 4200 cells to make 7.7 Kwh -14.8v Powerwall
Buying cells Vs. Harvesting:
Cheapest 2 Ah cells are $1 each, so $4200
or $0.50 per Ah
Using higher capacity cells would have fewer cells, but would still need to be less than $0.50/Ah
So lets sayI re-test my 1900 cells and they all come out over 2000.
That gives me 20% harvest success rate. I would need to process 21,000 cells to get 4200 good cells.
My local battery recycler wants $2.65 per pound.
There are 9 cells per pound.
21,000/9= 2333 pounds.
2333*2.65=$6183
A 7 KwH Tesla wall costs $3000.
To get the cost below a Tesla, I would have to get cells significantly cheaper, or have higher harvesting success rate.
To break even with the cost of a Tesla wall at my 20% success rate, I would need to get cells for $1.30/pound, or double my success rate.
These would still be used cells with unknown history/life span, and there would be months of labor involved.
I've seen others post higher success rates, which is why I bought those ebay batteries. Although 370 cells is a pretty good sample size.
I've also seen others post lower cost per pound than what I can currently get ($2.65), but $1.30/pound is pretty cheap from what I've seen posted here.
A Tesla wall has a 10 year warranty, built in cooling, and BMS.
And although I really want to continue this project and I have the ambition to do the work, Im not sure the economics are really there.
Convince me otherwise please!
EDIT:
As requested by another member...I made a math error on the number of cells required to create a pack similar to a tesla powerwall.
It turns out I only need about 1200 cells for 8.8 Kwh, not 4200.
So with a 20% success rate, I would need to process 6000 cells.
6000 cells @ 9 cells per pound = 667 pounds
667 pounds * $2.65/pound = $1766 total cost of cells
with a 50% success rate, the cost drops to $706
The cost is not as bad as I thought when I first wrote this post, but also not as good as I had hoped getting into this project (was thinking about 66-75% success rate, and lower battery pack pricing based on other people's accounts)
So I bought some ebay cells to get started and as an experiment to see what it would really take to make this happen. I paid more than I wanted, but I was OK with the price if I could recover 66% of the cells. They were all Lenovo/IBM packs, with cells of either 2200 or 2400 mah, based on the pack's label. I also had just about 6 packs from other brands. I had no other leads on cheaper batteries at the time other than quantities of 1 or 2 at time.
Here are my disappointing results:
So, Lets say I want to make a 7Kw powerwall, like a tesla. Here is what I would need.
150 cells @ 2 Ah = 300 Ah
300 Ah * 3.7v = 1110 Wh = 1.11 KwH
150 cells * 4 (in series) = 600 cells to make one 14.8 volt- 1.11 KwH battery
600 cells * 7 = 4200 cells to make 7.7 Kwh -14.8v Powerwall
Buying cells Vs. Harvesting:
Cheapest 2 Ah cells are $1 each, so $4200
or $0.50 per Ah
Using higher capacity cells would have fewer cells, but would still need to be less than $0.50/Ah
So lets sayI re-test my 1900 cells and they all come out over 2000.
That gives me 20% harvest success rate. I would need to process 21,000 cells to get 4200 good cells.
My local battery recycler wants $2.65 per pound.
There are 9 cells per pound.
21,000/9= 2333 pounds.
2333*2.65=$6183
A 7 KwH Tesla wall costs $3000.
To get the cost below a Tesla, I would have to get cells significantly cheaper, or have higher harvesting success rate.
To break even with the cost of a Tesla wall at my 20% success rate, I would need to get cells for $1.30/pound, or double my success rate.
These would still be used cells with unknown history/life span, and there would be months of labor involved.
I've seen others post higher success rates, which is why I bought those ebay batteries. Although 370 cells is a pretty good sample size.
I've also seen others post lower cost per pound than what I can currently get ($2.65), but $1.30/pound is pretty cheap from what I've seen posted here.
A Tesla wall has a 10 year warranty, built in cooling, and BMS.
And although I really want to continue this project and I have the ambition to do the work, Im not sure the economics are really there.
Convince me otherwise please!
EDIT:
As requested by another member...I made a math error on the number of cells required to create a pack similar to a tesla powerwall.
It turns out I only need about 1200 cells for 8.8 Kwh, not 4200.
So with a 20% success rate, I would need to process 6000 cells.
6000 cells @ 9 cells per pound = 667 pounds
667 pounds * $2.65/pound = $1766 total cost of cells
with a 50% success rate, the cost drops to $706
The cost is not as bad as I thought when I first wrote this post, but also not as good as I had hoped getting into this project (was thinking about 66-75% success rate, and lower battery pack pricing based on other people's accounts)