Mix old with new Lifepo4?

[Frank in Thailand]

I was thinking about the future.

Now I have 456Ah new Lifepo4.
In 3 to 5 years they have 80% capacity left, if I'm lucky.
Could be less.

While 80% is still enough for our usage, it will decrease to a point where it is not.

For lead acid, you can not mix old and new.
The old ones have absurd bad effectiveness and hardly hold a charge.
Most of all... They never are full.
You keep putting energy to it and it just gets lost (heat)
When you add new battery to that parallel string, the old will discharge the new one in not nice speed.

For Lifepo4, things are different (are they?)
The effectiveness is already at start a lot higher then Lead acid.
(+95 Vs -80%)
If my understanding is correct, the total capacity of older lifepo4 reduce, but that's it.

It still holds this reduced capacity like new ones (not leaking like lead acid who is empty after a week)

Also, the charging efficiency stay the same.
The 456 Ah battery just reduced capacity to 250Ah.
And is like 250Ah cell
It probably will decade faster then new cells, but that's OK.

In a parallel string of 3 x 152= 456Ah, who have reduced capacity to "only 250Ah" could you add 300Ah cells?
Making together 550Ah.
And yes, reduce faster the 250Ah part.
The 300Ah cells are not affected by the reducing capacity of its neighbour cell in the parallel set.

This is theoretically.
And only holds up if the charging efficiency and possibility to hold the charge stays unaffected.

If it is like lead acid, the new 300Ah cell will just try to charge the "now together" 250Ah set.
With doing so, it will discharge it self, and the old set will just lose that energy into heat.

To my understanding, the parallel sets in a string of series don't have to be identical.
Better, yes, for sure.
It can have a few 280Ah cells and a few sets made of 2*120Ah and one 40Ah., Together still 280Ah.
Naturally, there is difference in resistance for 1 cell of 280 and 3 cells combined.

An active balancer could compensate this without any real problems.

Please : shoot!! (Be kind and constructive!!)

I'm thinking about the future.
Could I just add cells and get away with it, or will this be big problem?
(I do have active balancer)

It doesn't have to be in 5 years, maybe in 2 or 3 years we need more energy.
Good to know if I just can add 16 cells to my 3*16 152Ah cells setup.
Those new cells can be anything lifepo4 that I would like to have more.
In 3 years.. who knows what the prices are.....
500ah for 50 USD??

 

[jasonhc73]

Can you please summerize all that to one question?

If you 100% discharge and 100% charge the lifepo4 battery, it will have 3 to 5 thousand cycles, then 80%.

If you 90% discharge and 99% charge, there is no measurable limit to how many cycles they have. Especially true if you never see freezing or below freezing temperatures.

 

[Frank in Thailand]

My cells cycle every day from +/- 85% charge (3.35v is the planning) down to??
I think we use about 50% so down to 35% and then recharged again next day up to 85%

The % are not that relevant.

Fact is, the cells will decrease in maximal current.
Let's say form 100 to 80% and from 80 down to 60 ...
Up to the point where there is not enough power available.

Can you add to the parallel set new fresh batteries?
If it would go down to 55% this will make my 456Ah @ 51.2 volt a 250ah @ 51.2 volt
But I need (for the sake of argument) minimal 300Ah @ 51.2 volt.

Can I add 200Ah to make it 450Ah again?

If not, why?

For lead acid, it's clear.
For Lifepo4...
Not yet.

 

[jasonhc73]

Internal resistance is what is at play here. The capacity of the system is always limited by the worst cell.

You add new freshly cells, great. Your system is still limited to the worst cell.

 

[Knighthawk]

In a parallel setting I would think you will be ok.

 

[Jeremiah]

Internal resistance is what is at play here. The capacity of the system is always limited by the worst cell.

You add new freshly cells, great. Your system is still limited to the worst cell.

Even when they are added in parallel?

 

[jasonhc73]

Even in parallel, if one cell drops below the BMS limit, if the BMS is worth it's salt, it will either shut everything down for Cell Difference, or Low Voltage.

You certainly will have a much more usable battery in the middle of the "legs", sure. But if you remove the wacky cell from the entire pack (and replace with a non-degraded cell), the entire pack will now how more voltage range.

Lithium Iron Phosphave Charge Range

Sheet5 Lithium Iron Phosphate Charge Voltage Range (LiFePO4) What to charge your cells/batteries/packs to. (SOC),Never More than,3.650,Shown is the "FLAT CURVE" of LiFePO4 cells/batteries/packs. What to not discharge past. (DOC),Nominal,3.200,The steap parts on the left and the right are the "Le...

docs.google.com

 

[zorlig]

Yes you are just fine doing this. Increasing the number in parallel over time is a great strategy for battery bank expansion!

 

[Frank in Thailand]

Even in parallel, if one cell drops below the BMS limit, if the BMS is worth it's salt, it will either shut everything down for Cell Difference, or Low Voltage.

You certainly will have a much more usable battery in the middle of the "legs", sure. But if you remove the wacky cell from the entire pack (and replace with a non-degraded cell), the entire pack will now how more voltage range.

Lithium Iron Phosphave Charge Range

Sheet5 Lithium Iron Phosphate Charge Voltage Range (LiFePO4) What to charge your cells/batteries/packs to. (SOC),Never More than,3.650,Shown is the "FLAT CURVE" of LiFePO4 cells/batteries/packs. What to not discharge past. (DOC),Nominal,3.200,The steap parts on the left and the right are the "Le...

docs.google.com

I'm sorry, I think you are missing the word parallel here.

There is not 1 single cell bad.
There are 48 cells, all same good or bad.
48x Original 152Ah.
In set of 3 cells, seen as one big cell of 456Ah.

Now those 3 cells slowly decade due usage, and have lower total capacity.
No longer 456, but in time less then 250Ah.

Together they make 16 parallel strings = 51.2volt (each cell/string 3.2v)
They used to be total +23kwh, now 12kwh
Reason for decaying?
Easy, age/cycles

Now..
Back to topic.
While 12kwh is quite a lot, and for many people enough, our home required more.

Do we need to replace the whole battery array?
Trow away 12kw/h??

I understand that decaying doesn't go linear. It starts slow and go faster bad over time.

It might have took (in theory) 5 years for the cells to go below 60% rest capacity (= losing 40%) it will not lose 40% again in the next 5 years.
The 60% left will be next year about 45, 25, 10%.
No exact figures, you get the idea.
Still 3 good years of reduced capacity left over to use.

When I would add new cells to the parallel string (who is now seen as 1 big cell, but consists of 3*152 cells with reduced capacity, and one new cell let's say 500Ah,) who will be at start 500 (new) + 250 ( 3* 152 in reduced capacity) is 750 Ah.

We already know that cycle a cell damage it.
Cycle a cell a little bit, top charge 75%, deepest discharge 25% will increase the total cycle count. / Life time.

If I would just replace the old 250Ah left over cells / parallel sets with new 500Ah, they will be used in this 50% way.
We already concluded that our home needs little more then 250.
That's this 50%

If I would not replace, but add, it would be total 750Ah (at start)

As we use little over 250Ah a day(night) it will be used at start for only 33%, not even 50%

Naturally, this is good news for the new cells who have a really kind start, and as result have increased lifecycle.
The old cells in this string also are treated more kind, the same 33% discharge.
While they will decade more fast then the new , those old cells will live longer then the original thought 3 years till 10%
making this a win win situation for both old and new cells.

there is not 1 bad cell.
there are 48 "bad" cells
and 16 new

adding new cells to old parallel strings is for lead acid bad, bad, bad.

for Lifepo4, I don't know!!

I understand that the new cell needs to be protected by increasing its internal resistance to the same as the old ones.
if it have lower resistance, it will go empty first before it will use the old cells.
preferably the same or little higher.
as we like to use the old cells to "protect*" the new cell.
* Protect like in less discharge

That is a challenge to face then

The internal resistance increase when the cell gets older.
How high it will be over 5 years?
Problem for then on how to measure.

One thing good it know now already
(And good to know for people who already face this situation), it is good or bad practice to mix old and new cells in the same parallel string!!

For lead acid, it's clearly a no go.
For Lifepo4, I'm not sure yet.

@zorlig , to you have technical information to backup this claim that it is good to mix/add new cells on old strings??

 

[Sojourner1]

Make it easier for everyone.. cells don't make cells, cell make batteries packs.

In order for you to use up the "rated" cycle capacity of the batteries you would need to be full cycling couple times aday and if that's the case it probably would be better to just install all new and give the old batteries an easier end of life job.

Or make you battery packs bigger now for less DOD.

 

[Frank in Thailand]

Make it easier for everyone.. cells don't make cells, cell make batteries packs.

If you like to be exact..
Pouches make cells
More then one Cell make a set
A set can be parallel or series
Better even parallel and series
Together they make battery packs...

Yet...
The BMS treats a parallel set as one cell, sees it as one cell not a set of individual cells.

And this is what the thread is about..
Mixing cells in a set who is seen as one cell...

To keep it confusing

Unless there is a new universal naming I'm unaware of...

Yes it can be confusing for who reads fast..
You might even need to read it a few times..

BMS doesn't manage individual battery packs, in their terminology they manage cells that make up a battery pack.

Yet those cells can be made up of one or more individual cells who need to have the same total capacity, but individual they can be different, and on different location...
A parallel set of 120 100 120 is the same as a parallel set of 100 120 120
Or 120 60 100 60
They are all seen by the BMS as one cell of 340Ah...

One 4S can be 4 x 340 who is
120 120 100
60 120 40 120
100 120 60 60
100 120 120

They are for the BMS 4 x individual cells

And to keep it on topic..
One of the 120Ah in this list is not really 120Ah but old 280Ah who have 120Ah capacity leftover.
The others are new cells


Yes.
This can be easily be confusing.
But that's the question!

Is it OK to mix old cells with new?
Or do you always need to have to make up the pack with new or old cells?

If the capacity is the same, does it really matter?
Do the old cells damage the newer cells?

 

[Sojourner1]

I understand everything you wrote it's only been 4 years living of them everyday and researching before that and seeing that you have all the answers and you prefer to possible confuse folks or maybe just like to post extra words. I'll back out now of the post.

Have a nice day.

 

[Ped]

Its much more helpful to phrase an OP "can anyone point me to tests or data or have direct experience with...."

 

[Jeremiah]

If you like to be exact..
Pouches make cells
More then one Cell make a set
A set can be parallel or series
Better even parallel and series
Together they make battery packs...

Yet...
The BMS treats a parallel set as one cell, sees it as one cell not a set of individual cells.

And this is what the thread is about..
Mixing cells in a set who is seen as one cell...

To keep it confusing

Unless there is a new universal naming I'm unaware of...

Yes it can be confusing for who reads fast..
You might even need to read it a few times..

BMS doesn't manage individual battery packs, in their terminology they manage cells that make up a battery pack.

Yet those cells can be made up of one or more individual cells who need to have the same total capacity, but individual they can be different, and on different location...
A parallel set of 120 100 120 is the same as a parallel set of 100 120 120
Or 120 60 100 60
They are all seen by the BMS as one cell of 340Ah...

One 4S can be 4 x 340 who is
120 120 100
60 120 40 120
100 120 60 60
100 120 120

They are for the BMS 4 x individual cells

And to keep it on topic..
One of the 120Ah in this list is not really 120Ah but old 280Ah who have 120Ah capacity leftover.
The others are new cells


Yes.
This can be easily be confusing.
But that's the question!

Is it OK to mix old cells with new?
Or do you always need to have to make up the pack with new or old cells?

If the capacity is the same, does it really matter?
Do the old cells damage the newer cells?

I'm glad you are asking about this because I'm also interested in planning for the future and I don't have infinite budget to just "get more cells now."
Now I'm just thinking aloud, and I'm inexperienced, so don't flame me guys if this is off.
If we put new cells in parallel with the old, won't the new cells absorb the brunt of the work first since they have lower internal resistance (I think you hinted about this)? So I'm wondering if you'll have kind of a tug of war where the new cells will drop in voltage faster than the old cells, then the old will follow as the voltages equalize? Or will they both discharge together equally with the old cells just producing more heat? If this is true, then maybe it's a good idea to space the newer cells in this scenario a bit away from the older cells? I really like the idea of finding a way to keep the old cells in the game (preferably without having to split them out as say "just house lights" scenario or something like that.
@Sojourner1 this is just @fhorst 's style to think through things discussing the whole process. It'd be good if you keep contributing to the discussion if you change your mind. Especially if you've been using LifePo4 already for some years we value that experience. Have you experimented with adding used cells to your set up? If so can you share what happened? Thanks

 

[Frank in Thailand]

Lol.

You back out now??

You don't reply to the topic about the topic, only choice of words.

I live of them as well.
First Lead acid, now moving to Lifepo4.

My 48 X 152Ah will decade.
Your cells will do to.

It's easy if you have a few cells.
You don't use much energy, so your investment to get new is not high.
- not high today. 4 years ago they where expensive!! Even in low Ah quantity.

My battery pack needs something when it have 50% capacity left.
That is still 12kwh (!!)
Not something just to ignore.
Most lifepo4 users don't even have 12kwh

Yet, our household need more backup power, as we live off grid.

We cycle, as you probably do, once a day, not deep discharge, but as the capacity reduces, this depth of discharge while per kWh stay the same, the dept will increase, up to 100%.

That is when I need to act.
(Or before)

The new cells, probably 16* 500Ah by then, are enough to replace all.

And will start at 50% discharge.
Or..
If I add the new to the old, together they are 750, and will do only 33% discharge.

I'm brainstorming here.
I don't know any true or false.

I ask if it is good or bad to mix, and why this would be good or bad.

Preferably with some technical explanation on why this will be good, or bad.

I can explain why having 750 Ah is better to start with then 500 Ah

I can't tell you if this is partly old cells, if this will damage the new cell!

If you have something to contribute to the topic, please share!

In 2 to 4 years you will have this situation yourself.
You will just dump your old and buy new?
If so.. Why?

 

[Frank in Thailand]

Its much more helpful to phrase an OP "can anyone point me to tests or data or have direct experience with...."

If that would be needed then 80% of all the threads should be renamed...

We ask questions on this forum to learn from people who have had that experience.
Or know answers as they have studied it and can backup their statements.

People who just like to say " hello, I'm reading this" ....
I don't know what I'm talking about but..
I have solar panels also, a company installed them for me and they work perfectly."
Then you can reply if your type of panels work as you expected, but not on how to install them !!!
All you can say is "call company XYZ".

That's not a bad thing. !!

But don't reply to topics that ask about solar installation.
You don't know.

Unless you are solar panel installer who does have in dept knowledge, just did not install on your one home.

Same same for this question.

Who have knowledge or experience??
Those who think have knowledge..**
Please do backup the claims you make with data.
That is links to websites who so really know.

Better??


Peace ☮

** I often think I have knowledge as I read about it on the internet.
That doesn't mean I'm always right.
I can point to the location where I gained the knowledge.
And no, that's not "Fakebook" or other unreliable source.

I'm always glad if I'm wrong, then I'm learning some new.

Please don't just say "your wrong" without having any source of information to back this up.
That puts you in the line of "Fakebook" and other unriliable sources.

 

[Ped]

If that would be needed then 80% of all the threads should be renamed...

We ask questions on this forum to learn from people who have had that experience.
Or know answers as they have studied it and can backup their statements.

People who just like to say " hello, I'm reading this" ....
I don't know what I'm talking about but..
I have solar panels also, a company installed them for me and they work perfectly."
Then you can reply if your type of panels work as you expected, but not on how to install them !!!
All you can say is "call company XYZ".

That's not a bad thing. !!

But don't reply to topics that ask about solar installation.
You don't know.

Unless you are solar panel installer who does have in dept knowledge, just did not install on your one home.

Same same for this question.

Who have knowledge or experience??
Those who think have knowledge..**
Please do backup the claims you make with data.
That is links to websites who so really know.

Better??


Peace ☮

** I often think I have knowledge as I read about it on the internet.
That doesn't mean I'm always right.
I can point to the location where I gained the knowledge.
And no, that's not "Fakebook" or other unreliable source.

I'm always glad if I'm wrong, then I'm learning some new.

Please don't just say "your wrong" without having any source of information to back this up.
That puts you in the line of "Fakebook" and other unriliable sources.

People want to be helpful. I catch myself sometimes.

 

[Frank in Thailand]

I do also...
I often start with "I have no knowledge about this question, but it looks like ..
That I can talk about and might help you.

Ok..
Mostly not that extensive...
I say, 'I don't know" but think that this and this will be a good solution

 

[Sojourner1]

Jason summed it up pretty well in post 4. It's one thing to replace 1 or 2 cells that might go a stray for what ever reason but to replace a 1/3 or 1/2 the pack trying to extend the life of already aged cells, you still have underperforming cells in the packs.

I have read about individual cell replacements on other off grid forums that folks have done after 4-5 years in heavy daily service but not replace many cells like what you're asking.

From my experience after 4 years of daily use with roughly 525+ accumulated full cycles, the cells stay with in 40mv between the 20 cells. Late fall to early spring DOD is 40-45% (200-225ah), rest of the the time 30-35% (150-175ah). Charging to 14.1 or 14.2 (3.52 - 3.55) and floating at 13.6v. Even now inverting loads at 180 -200a the cells seem to perform as they did when first put in to service years ago.

If I had a cell showing issues I wouldn't think twice of replacing with a new one, if the entire pack starts to show aging then either generator time will be needed if solar can't keep up for the possible deeper DOD. Me personally I will replace the entire pack, there is always someone looking for used cells just look at this forum.

 

[zorlig]

Batteries only have so many variables. Capacity. Internal resistance. Self discharge.

As long as self discharge isn't a problem and the set is otherwise still balanced and you don't change that then you can add on to your hearts content.

If you have weak cells you can always supplement them slightly to balance the pack again, but that's hard work!