What SOH% do use in you're packs, how and why

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
1,355
Good day all,
Maybe a very stupid question, but my mother taught me:
The only stupid question, is the question that is never asked!
So here it goes:

I am segregating my cells in there different stadia of health that will make it to my powerwall:
Between 80 -90% and between 90 -100%
(freaking IR meter did NOTarrived yet, BAST*** going to open a dispute)

For the mah is clear(special thanks to Daromer!!!!), but do you use everything above 80% in youre packs, or:
You use 80%-90% in one string and 90%-100% in another string.
Or you just mix it up?
What is wisdom.
I am able to make my first 2 strings (48v x 2) from 90-100%
And then 1string(48v x 1) from 80-90% soh cells.

Or just........
Everything above 80%soh is fine and the mah are more important?

So quiastion a would be:
Does it matter that you divide at 10% soh in your powerwall?
So question b would be:
Would it make a significant difference, if I put 80-100% together?

I think, the strings with only 90-100soh%will last longer
and 80-90% soh strings will need maintenance earlier?

My first setup, needs only 600-800 watt an hour 24/7 @230v for the pound(got ~3500 cells ready @~2750mah)

Thanks in advance, best
 
I love that you have brought this up. My early thinking originally was that 85% - 100% cells would last longer and behave better. True in short run. As my project reached operational status it became clear my my ROI is currently 22.17yrs. SO... I become more focused on long run. As part of this my battery bank is built to allow 100% consumption of PV array with average 40-45% DOD of cells specs which according to 'some' info such as battery university might result several thousand cycles before the entire pack degrades into 50% of original specs range.

Now - to your point. It occurs to me that the real question is that in the long run, no matter how the packs begin, the packs will age and *require attention*. Next thought is what kind of attention? 2 major categories come to mind:
- Pack no longer delivers required ah compared to others in series OR entire series is so low in ah that it no longer meets your needs.
- Individual cells at random lose more % than neighbors and so overall pack requires attention to stay 'in balance' with other packs.

Some long term questions:
* If a pack has matched 85% cells in beginning, will they degrade together/gracefully? or will some cells go low before others. If some cells go before others, then there's no point in obsessing on 85% vs 90% etc... as some original 85% cells might continue while some original 90% cells go more quickly.

* If a pack has a group of 50% cells while rest of pack is still at 70% - is it practical to replace the 50% cells with 85% cells to extend the life of the 70% cells ... OR is this to much work / folly and entire pack should be dropped.

Many question in my mind and no clear info :)
 
My head hurts right now thinking about all this.

As far as building my14s200p pack I personally will use 80% to100% SOH cells strewn randomly throughout the whole build.
The only thing that will be an important factor is that all cells will have an IR of ?75m?. If the cell count permits I may go with an IR as low as ?70m? or even lower It all depends on the cell count. I think the closer the IR of all cells in the pack are, the better the longevity of the pack will be.
My goal is to get ~45A out of my pack which translates to 225mA per cell. That is the max my 2 GTILwill draw to supply my house with 1800W at any given moment. As long as the sun shines my Grid consumption will be next to zero with only the occasional spike pulling from the grid.
Also I should be able to keep my cycle voltages between 4v to 3.3V

I may also go up to 85% SOH as my starting point asOffGridInTheCity has done and that may be a good possibility as with my latest harvest many cells are over 85%.

I think the powerwall building will be a continuesproject only to slow down a little in the battery harvesting area as I will always want to have spare packs in reserve to replace the potential "bad" pack.
We will see.

Wolf
 
@ offgrid in the city,
We discussed 50% dod in some earlier threat, and i still agree with youre thought.
BU has proven the same, iow if you go low on your dod, your cells will last longer, because there is no cycle anymore.

To give you my thought on you're questions, when the packs need to have attention:
If a pack can not give my the ah anymore i want to be able to toss the whole pack, and not go thru the whole testing again.
So after every year (or two) every pack will be taking out and will be fully tested with the big discharge tester 200-300w.

If there is an individual cell that is going to lose more than i want, i think i must be able to see that with the balancer?
The batrium will tell me witch pack is out of balance and that will require desoldering one side of the pack, and check after 30 days witch cell lost the most volt?
Measured with the ir meter so you see immediately also the remaining ir.
But before the pack is being tested there is a "new" ready to assemble. One out...one in.
So the older pack, after being tested and approved/rebuild will get his spot back in his string, with a label on it.
What if the whole string get failures, you replace pack for pack, or after a certain amount of replacement you replace the whole string and test 1400(in my case) cells again on individual level?
But like wolf is stating and also my thoughts: its a ongoing process of harvesting, so when you're bank is growing, is it getting older?

For the long term questions, with ir you can see how hard the cell life was, if you take out every cell that has to much ohms(above 70 mohm or below 25 mohm, before wolf is going to bark:))
I really think the combination of those three: SOH%, IR and you're DOD% will give you packs that will last for a very long time

offgrid wrote: * If a pack has a group of 50% cells while rest of pack is still at 70% - is it practical to replace the 50% cells with 85% cells to extend the life of the 70% cells ... OR is this to much work / folly and entire pack should be dropped.
My thought: before coffee: drop, after coffee on a rainy day, maybe?
 
@ wolf,
You're head hurts after thinking about this? The king of ir, with the logic of spock?

I can understand, however(captain:)) i think it matters to have you're SOH% as close as possible in a pack, with the same IR.
For me its quite simple, i guess, i am used to segregate certain things at a more difficult level.

My demands for a cell that will make it into my pw are 80-100% SOH and with ir no higher than 70 mohm at a minimum of 2000mah
The range of 50-80 soh is going to be used for testing and small packs, or even use/test them in my pw.

I think i will start building my packs if i have more than 1400 cells at 90-100 soh and with a lower ir than 70 mohm and above 2700mah each
Then create a second string with the cells at my minimum specs.
For the next upcoming strings they will be only added if i can make a string:
90-100 soh with an ir 35-70 above 2000mah each
80-90 soh with an ir 35-70 above 2000mah each

And a string for "left overs" (incl 1600-2000 mah)

Because its going to be a continuous process with harvesting and testing ect, it can be easy to place you're tested cell in a predefined bucket or immediately in there future "prison"
With those criterias for that particular string or pack.

Or create packs with those demands and make a string asap.
With a newer pack finished that is better in the demands i will take out the weakest pack?
This is something time will tell me, it depends on what was brought home after the hunt.


Wolf wrote:
I may also go up to 85% SOH as my starting point as OffGridInTheCity has done and that may be a good possibility as with my latest harvest many cells are over 85%.

When my pw is big enough to power the pound for a minimum of 5 days, i will add for testing purpose 70-80% cells, lets see what happens, and if it will add some value into the project.

But still, the question remain: will this segregating pay off its effort in long term use for in the powerwall? or is this just bullocks.
And just slap every cell within my minimum specs in a tray.
Of course every pack in a string will have the same amount of amp.

This brings also a other question in my mind.
Cells that have over 80 mohm had a very hard life, but with a good soh, aren't we able to give them a easy and soft life?
The "hard lifer's" had to give and take 10 to 20 amp hour, we give and take from them 200mah - 500mah.

Enough with the rambling for now, any info, feedback, comment or food for thought is very welcome and very appreciated.

Thanks for your replay's, May you're cells "live long and prosper".
 
@100kwh-hunter

I appreciate your comments. During thehighest PV array power months, the system generates45kwh/day. Thebattery bank is'cycling'each day between 52-55v hi (3.7-3.93v/cell) and 48v low/inverter off(3.43v/cell). Batrium shows 30mv to 80mv min/max difference between the 42packsthru-out the cycle which I take to be 'excellent'. Overall, for the year, I'm projecting an average of38% DOD. In simple terms,38% DOD (of 40kwh battery) is15kwh. So I need 15kwh from my battery to enable 100% consumption of all my PV array generates.

Over time, the battery bank will degrade. But even at 50% degradation (to 20kwh) there will still be enough capacity for15kwh per day - in theory. I think it all depends on how far do cells / packs degrade gracefully (e.g. in similar amounts) so that reasonable balance can be maintained between packs. Do things go wacky at 80% SOH (probably not)or 70% SOH (maybe not) or even 60% SOH ??

I think, with my current focus, I'm going to try to document this and see what happens year over year and contribute it back to this forum somehow.
 
100kwh-hunter said:
This brings also a other question in my mind.
Cells that have over 80 mohm had a very hard life, but with a good soh, aren't we able to give them a easy and soft life?
The "hard lifer's" had to give and take 10 to 20 amp hour, we give and take from them 200mah - 500mah.

@100kWh-hunter

Captain it is not logical to assume that if a cell has an IR >80m? to have had a hard life.
There are many cells that by manufactures standards claim an acceptable m? IR of 80+
I also believe that certain chemistries in certain cell manufactures and part numbershave by nature a "higher" IR.
As I also believe most "high drain" cells have a low IR <30m?. Mostly due to themanganese /manganese nickel chemistries
giving them high drain low IR properties.

image_thvcfu.jpg


Now that being said I find that mostly the LGs that have high IRare good performers with many of them giving between 80% to 100%SOC.
Example: LGABB41865 High IR

image_xmfuzi.jpg

Would I use them in my powerwall? Not a chance. The IR is too high and the cell in the future will probably only cause megrief.

My indicators show for that cell (LGABB41865) anything <60m? is really good.
Example:LGABB41865 Low IR

image_covhiy.jpg


Logic prevails as always.


image_hmtaga.jpg



Wolf AKA Spock :D
 
Back
Top