D
Deleted member 1888
Guest
Can you mix LiFePO4 and Li-ion NMC in the same bank? What would happen if you mixed them in the same bank?
FilterGuy
Solar Engineering Consultant - EG4 and Consumers
Yes, it is a horrible idea. Never connect different chemistries of battery.
I actually experimented with this about a year ago and my results were horrible. Ran into a lot of problems and pulling full capacity was problematic. One battery would constantly feed the other battery depending on voltage. The charge/discharge SOC curve is different for every chemistry, so they were never able to equalize. And performance was really bad.
I actually experimented with this about a year ago and my results were horrible. Ran into a lot of problems and pulling full capacity was problematic. One battery would constantly feed the other battery depending on voltage. The charge/discharge SOC curve is different for every chemistry, so they were never able to equalize. And performance was really bad.
D
Deleted member 1888
Guest
Thank you
Dave.l
New Member
Can you add a Lifpo (Drop-in equivalent) battery to an existing AGM bank?
No
Considering your earlier concerns (well more accurately your families concerns) about not wanting 48v due to safety concerns, I would think NMC would be off the table.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,703
We have discussed this before on other threads. Due to the different voltages involved, it is not possible to make an LFP and NMC pack work at exactly the same voltages. The most common arrangement for a "48 volt" system is 14S for NMC and 16S for LFP.
Her is the story during discharge..
LFP at 10% SOC is about 3.35 volts per cell or 53.6 volts total with 16 cells in series.
NMC cells at 53.6 volts / 14 cell is still 3.83 volts per cell. This is above 70% charge. You are hardly using the NMC pack.
If you go up to a 15S for the NMC cells, you are then at 53.6 / 15 = 3.67 volts per cell. This is better, but still over 30% SOC.
So, on the discharge side, you will not be able to use all of the NMC and the LFP pack will do virtually all of the work.
Here is what happens when charging.
LFP at 90% SOC is about 3.6 volts per cell or 57.6 volts total on the 16 cells.
If we use the 14S NMC again, this is 4.11 volts per cell. That is also 90% SOC so this looks pretty good. But then you only get to use 20%
If you go to the 15S, then the 57.6 / 15 = 3.84 volts per cell. This is only bringing the pack up to 70% SOC. Oops. With this setup, you can use about 40% of the NMC pack while using about 80% of the NMC pack.
It is actually even worse than this. As Will has said, due to the flat discharge curve of the LFP and the linear slope of NMC, the NMC cells will give out most of their 40% first as the voltage comes down to the flat part of the LFP curve. Then the LFP will take most of the current while the NMC battery just floats. When the LFP hits the lower knee, the current will again go back onto the NMC cells.
This is not a good situation. Both packs would need to be able to handle virtually all of the current, depending on where you are in the discharge curve. And when the load is light, the cells will be trying to "equalize" the voltage between them.
Bottom line, if you need more capacity, get more of the same chemistry. You can easily parallel different size packs, as long as you are not trying to get more current, and just want more time. My main pack can easily feed my inverter to full output. But for more time, I may end up adding another cabinet of batteries next to it. Fuse it at a lower current and use slightly thinner cables and it will lag behind the main pack and just sort of feed charge to the main pack until they equalize. I have even thought about doing something like that to get more capacity in my hybrid car.
Her is the story during discharge..
LFP at 10% SOC is about 3.35 volts per cell or 53.6 volts total with 16 cells in series.
NMC cells at 53.6 volts / 14 cell is still 3.83 volts per cell. This is above 70% charge. You are hardly using the NMC pack.
If you go up to a 15S for the NMC cells, you are then at 53.6 / 15 = 3.67 volts per cell. This is better, but still over 30% SOC.
So, on the discharge side, you will not be able to use all of the NMC and the LFP pack will do virtually all of the work.
Here is what happens when charging.
LFP at 90% SOC is about 3.6 volts per cell or 57.6 volts total on the 16 cells.
If we use the 14S NMC again, this is 4.11 volts per cell. That is also 90% SOC so this looks pretty good. But then you only get to use 20%
If you go to the 15S, then the 57.6 / 15 = 3.84 volts per cell. This is only bringing the pack up to 70% SOC. Oops. With this setup, you can use about 40% of the NMC pack while using about 80% of the NMC pack.
It is actually even worse than this. As Will has said, due to the flat discharge curve of the LFP and the linear slope of NMC, the NMC cells will give out most of their 40% first as the voltage comes down to the flat part of the LFP curve. Then the LFP will take most of the current while the NMC battery just floats. When the LFP hits the lower knee, the current will again go back onto the NMC cells.
This is not a good situation. Both packs would need to be able to handle virtually all of the current, depending on where you are in the discharge curve. And when the load is light, the cells will be trying to "equalize" the voltage between them.
Bottom line, if you need more capacity, get more of the same chemistry. You can easily parallel different size packs, as long as you are not trying to get more current, and just want more time. My main pack can easily feed my inverter to full output. But for more time, I may end up adding another cabinet of batteries next to it. Fuse it at a lower current and use slightly thinner cables and it will lag behind the main pack and just sort of feed charge to the main pack until they equalize. I have even thought about doing something like that to get more capacity in my hybrid car.
I can't help but feel that you are making your life substantially more difficult by this piecemeal accumulation of dissimilar batteries and incrementally designed system.
With a system as large as you want to build, I think it would be easier and cheaper to take some time to clearly define your needs and goals, and then develop a plan to implement them, rather than piecing together a few components at a time and trying to make them play well together.
Building your bank out of LFP raw cells might be the most logical and cost effective option choice. At first I thought you were avoiding raw cells to avoid complexity, but at this point, I think the complexity involved in the piecemeal approach and some of the workarounds being considered has surpassed the complexity of just buying buying and building with all one type of raw cell. I know you already have sunk costs in drop-ins, but I think its at least worth considering. In a lot of ways it would avoid complexity in design and in purchasing.
With a system as large as you want to build, I think it would be easier and cheaper to take some time to clearly define your needs and goals, and then develop a plan to implement them, rather than piecing together a few components at a time and trying to make them play well together.
Building your bank out of LFP raw cells might be the most logical and cost effective option choice. At first I thought you were avoiding raw cells to avoid complexity, but at this point, I think the complexity involved in the piecemeal approach and some of the workarounds being considered has surpassed the complexity of just buying buying and building with all one type of raw cell. I know you already have sunk costs in drop-ins, but I think its at least worth considering. In a lot of ways it would avoid complexity in design and in purchasing.
.... Going to post in the General Discussion Category ....
Reposting so the answer makes sens for anybody else .....
Hi all, I am a newbie to solar power.
My systems is:
6 Solar Panels Rizen of 450watts each
2 x Growatt SPF3000TL of 24volts - Split Phase
2 x LiFePo4 12v 100ah in series
Settings:
Max. Charge Amps 20 at 28 volt with float at 27.7
My big mistake is that I wanted to start with a small system and later upgrade by steps. I thought that in the first step only to use solar power in the daytime. It seems it is not very stable because the BMS of one of the LiFePo4 Battery keeps shutting down at less than 50 amps.
The other problem is if one of the batteries shuts down the whole system collapses even with grid power as the growatt needs 24v at the busbar.
Now I know that I should have at least 400ah of batteries for this system.
To avoid a sistema shutdown I added some Lead Acid batteries (4s 6v/200ah) in parallel with I previously charged to the same volt levels as I have seen in this video.
I have amp+volt meters on both batteries and there is little current flowing between them in idle (no charging).
Amp Loads are depending on voltages.
With a 1000watt load everything works, but if I go a little higher the BMS of one LiFePO4 shuts down.
Some Questions:
Can I leave both types of batteries in the system for some time (a month) without damaging the LiFePo4 batteries until I get a bigger LiFePo4 battery?
It is normal that the BMS of one of the battery shuts down?
I have checked the batteries disconnected and both of them have the same voltage (I have tried to balance them putting them 1 day parallel and charged parallel), but in series I get up to 1 Volt difference (14.2 +13.2 volts) in charging and discharging state. Is this normal?
How I can add a 24v permanent source to the DC input of the Growatts to avoid a complete shut down (house without power) even when grid power is available (My Idea is to add a Lead Acid, AGM or Gel battery)?
I hope somebody with more experience can help me and that I am in the correct post category.
Best regards from the Caribbean (Dominican Republic).
Reposting so the answer makes sens for anybody else .....
Hi all, I am a newbie to solar power.
My systems is:
6 Solar Panels Rizen of 450watts each
2 x Growatt SPF3000TL of 24volts - Split Phase
2 x LiFePo4 12v 100ah in series
Settings:
Max. Charge Amps 20 at 28 volt with float at 27.7
My big mistake is that I wanted to start with a small system and later upgrade by steps. I thought that in the first step only to use solar power in the daytime. It seems it is not very stable because the BMS of one of the LiFePo4 Battery keeps shutting down at less than 50 amps.
The other problem is if one of the batteries shuts down the whole system collapses even with grid power as the growatt needs 24v at the busbar.
Now I know that I should have at least 400ah of batteries for this system.
To avoid a sistema shutdown I added some Lead Acid batteries (4s 6v/200ah) in parallel with I previously charged to the same volt levels as I have seen in this video.
I have amp+volt meters on both batteries and there is little current flowing between them in idle (no charging).
Amp Loads are depending on voltages.
With a 1000watt load everything works, but if I go a little higher the BMS of one LiFePO4 shuts down.
Some Questions:
Can I leave both types of batteries in the system for some time (a month) without damaging the LiFePo4 batteries until I get a bigger LiFePo4 battery?
It is normal that the BMS of one of the battery shuts down?
I have checked the batteries disconnected and both of them have the same voltage (I have tried to balance them putting them 1 day parallel and charged parallel), but in series I get up to 1 Volt difference (14.2 +13.2 volts) in charging and discharging state. Is this normal?
How I can add a 24v permanent source to the DC input of the Growatts to avoid a complete shut down (house without power) even when grid power is available (My Idea is to add a Lead Acid, AGM or Gel battery)?
I hope somebody with more experience can help me and that I am in the correct post category.
Best regards from the Caribbean (Dominican Republic).
Last edited:
grizzzman
Photon Vampire
- Joined
- Sep 21, 2019
- Messages
- 2,749
"Some" 12 volt Lifepo4 do not deal with series connection very well. With that said, I would charge the LFP batteries using a 12 volt charger tell the BMS disconnects then inspect voltages the next day. If they match try again. (You may be exceeding the amps that the battery can output causing the BMS to disconnect.) I run a LFP-FLA bank (for the last two years) A couple of months would be no problem.( If you pull many amps from the FLA bank, you could disconnect the FLA and equalize them) Please let us know what you did and how it worked for you.
That's what I thought. I wanted a 24v 100ah battery, but they were out of stock and they sold me 2 x 12v 110ah. I will try to change them.
I will do that or let the store do that and test them.
The label says: Cont. Discharge: 100A and Peak 200A and I am far away. I am pulling only 50 amps in day time with clouds.
Thanx, I was worried that something blows up or I damage the batteries. Now it's working for 3 days.
I will post my results.
