What are the issues with mixing LiFeP04 batteries of different capacities?

I have a single brand-name 100Ah LiFePO4 drop-in replacement battery. It's actually serving me well but I am careful with my usage, both in how much capacity I use in a day (say 40 or 50 Ah) before charging as well as with max current draw (less than 100A).

I'm looking to increase both capacity and max current draw and am thinking that a 2nd 100Ah LiFePO4 battery in parallel will achieve this.

Further, I'm going to try to build my own battery. I see, though, that there are options to make a battery that has, for example, 100Ah, 135Ah, 150Ah, etc. I've even seen talk of an EVE 280Ah battery on the forum.

Given that I have only a 100Ah battery now, what are the issues with mixing LiFeP04 batteries of different capacities? e.g. say I can afford to build a 150Ah battery. I would thus end up putting a 100Ah and a 150Ah battery in parallel. What are the issues?

Am I constrained to making a 100Ah battery only?

I don’t think there would be an issue. If you have the batteries connected in parallel, they would be at the same voltage. Because they would have different. BMS, one would cut off before the other but that should be fine. The other would then continue charging (at a higher current) until it also cuts out.

I built a 160Ah battery out of four LIFEPO4 cells. Let me know if you want a link to how I did that, I wrote a post about it.

Good luck!

You can safely have different "Packs" within a Battery Bank.
A pack being an independent battery pack of cells with t's own BMS.
A Bank being the collection of packs assembled into a large power storage bank of batteries.
Packs in Series increase voltage, Packs in Parallel increase Amp-hours.

I am just completing my new 24V battery bank, it consists of two 24V/280AH packs and two 24V/175AH each with a Chargery BMS8T. The four packs are connected to a common DC Bus (Common Port Configuration) and as they are all LFP the charge / discharge settings all work fine. The BMS is the FailSafe and is responsible for its own pack of cells and operates independently of the others. Each pack, of course, needs to have its own fuse (safety first) and when wiring the packs in parallel it is highly recommended to use wires of same length & AWG (ie, all 2/0) to eliminate / reduce any difference in voltage, resistance etc.

An excellent reference for doing this was written by Victron, it is an "essential read" in my opinion and answers most of the questions.

NOTE, that when building your battery bank one must pay close attention to the voltage readings so that you are calibrating your gear properly. See this link for more details. https://diysolarforum.com/threads/c...s-scc-inverter-charger-voltage-matching.5030/

The issue with connecting different packs of different capacity is eddy currents on the buss bars connecting the packs. If there are current measurement devices, their placement may influence the accuracy of their readings. In other words the flow of current between the batteries should not be counted in the Amphours going into or out of the consolidated pack.

Each "Pack" can be monitored for individual performance & statics for AH in/out etc. Additionally, the complete bank as a whole can be monitored as well. This has to be well thought out & planned to work effectively, it is also somewhat dependent on the equipment (BMS' etc) being used and how you wish to monitor the packs & bank.

matthewnz said:

I don’t think there would be an issue. If you have the batteries connected in parallel, they would be at the same voltage. Because they would have different. BMS, one would cut off before the other but that should be fine. The other would then continue charging (at a higher current) until it also cuts out.

I built a 160Ah battery out of four LIFEPO4 cells. Let me know if you want a link to how I did that, I wrote a post about it.

Good luck!

Click to expand...

Thank you. Yes, please send me a link, either here or via PM.

Steve_S said:

You can safely have different "Packs" within a Battery Bank.
A pack being an independent battery pack of cells with t's own BMS.
A Bank being the collection of packs assembled into a large power storage bank of batteries.
Packs in Series increase voltage, Packs in Parallel increase Amp-hours.

I am just completing my new 24V battery bank, it consists of two 24V/280AH packs and two 24V/175AH each with a Chargery BMS8T. The four packs are connected to a common DC Bus (Common Port Configuration) and as they are all LFP the charge / discharge settings all work fine. The BMS is the FailSafe and is responsible for its own pack of cells and operates independently of the others. Each pack, of course, needs to have its own fuse (safety first) and when wiring the packs in parallel it is highly recommended to use wires of same length & AWG (ie, all 2/0) to eliminate / reduce any difference in voltage, resistance etc.

An excellent reference for doing this was written by Victron, it is an "essential read" in my opinion and answers most of the questions.

https://www.victronenergy.com/upload/documents/Wiring-Unlimited-EN.pdf

NOTE, that when building your battery bank one must pay close attention to the voltage readings so that you are calibrating your gear properly. See this link for more details. https://diysolarforum.com/threads/c...s-scc-inverter-charger-voltage-matching.5030/

Click to expand...

I appreciate all the info. Lots to read. And I love learning. Thank you.

Steve_S said:

Each "Pack" can be monitored for individual performance & statics for AH in/out etc. Additionally, the complete bank as a whole can be monitored as well. This has to be well thought out & planned to work effectively, it is also somewhat dependent on the equipment (BMS' etc) being used and how you wish to monitor the packs & bank.

Click to expand...

Yes, I plan to install a number of Coulomb counters at a few places (each battery, the consolidated battery bank). Partly because I'm geeky. Partly because I want to know. (I have a thread about Coulomb meters today.)

One worry I have is that I'll draw greater than the capacity of the smaller battery and it'll cut out. The bigger battery will then shoulder all of the load. Capacity wise, that'll be ok. But the max current capability is no longer 100A + (say) 150A = 250A, but "only" 150A. But if I don't do silly things such as run my induction burner at max while simultaneously microwaving something, I should be fine.

Another worry is that as the smaller battery is nearing its capacity limit, its voltage will fall. And here is where I get fuzzy. Will the bigger battery, presumably pumping out a higher voltage due to being much less depleted, start "charging" the smaller battery? And if so, the max current of the larger battery has to offer to my legitimate load be less than what it should be?

Yes, the load/discharge & charge sharing which occurs within the bank itself. You are quite correct in the theory of it and you do have to account for the Lowest AH Pack in the bank and the potential of it (or any) other pack shutting down and keeping the system running in the meantime. This is where the planning of the whole system kicks in. To put it in the simplest of terms, it is generally best to be able to run what you need from ONE Pack within the bank as the worst-case scenario. Some call it "Limp Mode" which I suppose is a fair way to put it.

The BMS capacity handling for discharge / charge should be relative to the Maximum Load potential of one pack so that if any pack shuts down it will not affect the general operations and keep you going. As such, the system design & components used should be selected with the "worst Case Scenario" in mind, so that you're not caught with your knickers around your ankles, should something go awry.

The amount of amperage you can pull from a battery bank also depends on the voltage being used because there are logistical limits. For example, 12V @ 100A = 1200 Watts (my 1200W Microwave pulls 80A from 24V, on 12V that would be 160A) With 12V the general cutoff is usually 2000W (excluding surge capacity) because after that you get into ridiculous wiring and BOS gear to handle it. With 24V it's normal to 4000W and 48V for 4000W and above.

hope it helps.
Steve

Ample said:

One worry I have is that I'll draw greater than the capacity of the smaller battery and it'll cut out.

Click to expand...

There are two different capacity issues.
One, the Amphour capacity is resolved by the fact that in a parallel configuration each pack will stay at the same voltage by moving current from the larger pack to the smaller pack as they discharge. This will be observed during discharge as one pack will appear to be charging while the other is discharging.
That brings up the other capacity issue which is current flow. Other posters referred to this as the stronger(larger capacity) pack supporting the weaker (smaller capacity) pack. I don't know how to predict or calculate how much that will be. If it is a large current then the larger capacity pack will exceed its current rating as it tries to support the load and the weaker battery. If that happens the larger capacity pack will trip and will go off line. I suspect this will cascade to the smaller capacity pack. As @Steve_S pointed out, planning is important to consider the possible scenarios.
Fortunately, you will have current measurements throughout these packs and can observe this phenomena. It would help if these are logging meters so that one could view the data historically in a format that is easy to interpret. Certainly an alarm or alert that one pack shut down would be helpful to understand what the maximum current that can be pulled from such a pack.

matthewnz said:

Because they would have different. BMS, one would cut off before the other but that should be fine.

Click to expand...

Yes, you are describing the charging scenario. In that case the only defense against over charging is the BMS. There is no second line of defense like a HVD except for the last pack remaining as each pack drops off line. If that were to occur during the Constant Current stage (Bulk) then @Ample needs to make sure the current does not exceed the correct charging current for the remaining pack.

Ample said:

Thank you. Yes, please send me a link, either here or via PM.

Click to expand...