diy solar

diy solar

Parallel LiFePO4 12V 90Ah pack with 12V 280Ah pack, any issue?

So, I'm pretty new here (there, that's out of the way ;) ).

I've been reading the board for a quite a while, from way before I ever posted. I'm amazed at the knowledge and sharing of information. It's awesome!

My background: I'm a retired Electrical Engineer, and I designed my own solar power system, but I used Lead-Acid AGM batteries in my system, so I'm mostly new to the idea of LiFePO4 batteries. I admit I don't really understand how charging and discharging of LiFePO4 (or battery in general) really work (there, that's out of the way too ?).

A few folks in this thread (and others) have said that when they have packs/batteries of mismatched Ah tied together in parallel, it all works great and they don't see any current / power moving from the bigger pack (say, 280Ah) to the smaller pack (say, 90Ah) during discharge. Further, they report that the current delivered from the two disparate packs is proportional to their overall contribution to the total. I am confused by this, but I have no doubt of the authenticity that it is true. I just don't understand how this works.

Let's walk through it...

A 280Ah pack and a 90Ah pack are wired in parallel. They are both nearly fully charged, and so are quite happy at say 3.5V. Then someone turns on the loads. Both packs seek to fill whatever they can of the current demand of the load. It appears to me that the cells in a 90Ah pack and those in a 280Ah pack are not significantly different in internal resistance (from what I have been able to find), or perhaps it is somewhat unpredictable. Nevertheless, it seems that the 90Ah pack will deliver pretty much (90/(280+90) of the drawn current. Why is that?

As the two packs are drawn down, the voltages obviously are held together. However, it would seem like the smaller pack would be depleted faster. Eventually the smaller pack would not be delivering any current towards the load, and to maintain the same voltage, it would have to be receiving what is effectively a charge from the larger pack, while the larger pack is also serving the demand of the loads. However, it has been reported that there is not current flowing INTO the smaller pack in this instance, so.... Why is that?

Eventually, the loads quit and a source (solar, generator, shore, etc.) start recharging the two packs. Again, the two packs are maintaining the same voltage. Since the larger pack will rise more slowly, this part maybe makes more sense to me: The small gets what it needs to be at 3.3V, but the larger pack needs more current to get it up past that. So, the larger pack takes more current. Do I have that right?

This is probably a little thing, and not worth much discussion. However it is driving me crazy! I just don't get it.

Caveat: I'm coming from the Lead-Acid / AGM world. In that world, there are lots of rules you hear when you are initiated. One is that you don't put batteries that are not of the same Ah and same age - and maybe not even if they are not the same brand and date - together. To do so is asking for unspecified catastrophes! I didn't really understand that much either, but I could some sense of it, but it was mostly just the magic of batteries. Now I'm more prone to question things, I guess...

Any thoughts? Again, I'm just trying to learn here, and you guys know what you are talking about.
 
This is what are shown with 20A load, it has about 7A drawn from 90Ah pack, 13A from 280H pack.


90Ah paralleled with 280Ah_20A load.jpgdrawn, the 280Ah is about 13A drawn.
 
Last edited:
So, I'm pretty new here (there, that's out of the way ;) ).

I've been reading the board for a quite a while, from way before I ever posted. I'm amazed at the knowledge and sharing of information. It's awesome!

My background: I'm a retired Electrical Engineer, and I designed my own solar power system, but I used Lead-Acid AGM batteries in my system, so I'm mostly new to the idea of LiFePO4 batteries. I admit I don't really understand how charging and discharging of LiFePO4 (or battery in general) really work (there, that's out of the way too ?).

A few folks in this thread (and others) have said that when they have packs/batteries of mismatched Ah tied together in parallel, it all works great and they don't see any current / power moving from the bigger pack (say, 280Ah) to the smaller pack (say, 90Ah) during discharge. Further, they report that the current delivered from the two disparate packs is proportional to their overall contribution to the total. I am confused by this, but I have no doubt of the authenticity that it is true. I just don't understand how this works.

Let's walk through it...

A 280Ah pack and a 90Ah pack are wired in parallel. They are both nearly fully charged, and so are quite happy at say 3.5V. Then someone turns on the loads. Both packs seek to fill whatever they can of the current demand of the load. It appears to me that the cells in a 90Ah pack and those in a 280Ah pack are not significantly different in internal resistance (from what I have been able to find), or perhaps it is somewhat unpredictable. Nevertheless, it seems that the 90Ah pack will deliver pretty much (90/(280+90) of the drawn current. Why is that?

As the two packs are drawn down, the voltages obviously are held together. However, it would seem like the smaller pack would be depleted faster. Eventually the smaller pack would not be delivering any current towards the load, and to maintain the same voltage, it would have to be receiving what is effectively a charge from the larger pack, while the larger pack is also serving the demand of the loads. However, it has been reported that there is not current flowing INTO the smaller pack in this instance, so.... Why is that?

Eventually, the loads quit and a source (solar, generator, shore, etc.) start recharging the two packs. Again, the two packs are maintaining the same voltage. Since the larger pack will rise more slowly, this part maybe makes more sense to me: The small gets what it needs to be at 3.3V, but the larger pack needs more current to get it up past that. So, the larger pack takes more current. Do I have that right?

This is probably a little thing, and not worth much discussion. However it is driving me crazy! I just don't get it.

Caveat: I'm coming from the Lead-Acid / AGM world. In that world, there are lots of rules you hear when you are initiated. One is that you don't put batteries that are not of the same Ah and same age - and maybe not even if they are not the same brand and date - together. To do so is asking for unspecified catastrophes! I didn't really understand that much either, but I could some sense of it, but it was mostly just the magic of batteries. Now I'm more prone to question things, I guess...

Any thoughts? Again, I'm just trying to learn here, and you guys know what you are talking about.

DUUUUUUUUUUUUUUUUDE

Like a brother from another mother. Makes no electrical since to me either. Largely mismatched capacity should cause issues in my mind. Yet a few folks have posted some measured results that show quite the opposite :unsure:
 
Welcome to a world of confusion.
I went from Big Heavy Lead to LFP and had to retrain my brain a bit... Basically & simply put, forget all Lead Acid information ! They are batteries but that's it for commonality on many points. It's as bad as thinking in VAC terms and them flipping modes into thinking VDC mode... Certainly has headache potential.

1) The aging between battery packs is not an issue like it is with FLA, where the lowest denominator rules.
2) LiFePo / Lithium battery assemblies have a BMS while Lead does not...
The BMS controls and manages the battery pack and the cells within.
It cannot distribute voltages in/out. It can only connect/disconnect the pack when conditions require it. Low/High Volt disconnect (at cell level) and pack level, temp disconnect (if so equipped) and so on.
Some BMS' can passively Balance (discharge high volts from hi cells) or actively balance (transfer volts from hi to lo cells) but this is low end stuff.

When properly paralleled battery Packs are working correctly within a battery Bank the packs will share Charge/Discharge proportionaly. It has to do with the collective resistance / impedance of the assembled packs, which is directly correlated to the capacity of the collective cells. For all intents & purposes, from low SOC state like 10% (2.950 per cell +/-) to 80/85% (3.350 +/- per cell) is where the power lives. When all packs are within their functional range they stay very level and will distribute power as shown above by Bud Martin's images.

The tricky part where the confusion starts and I likely will not help .... LOL (okay, it's 04:13 atm and I'm edging really tired)
The BMS part...
All BMS' will cutoff for Lo or Hi Volt cell states. Default is 2.50V & 3.65V respectively at cell level.
Hi/Lo volt cutoff is NOT dictated by the pack assembly voltage.

No BMS "distributes power". Meaning that if you are pushing 28V/30A (to a 24V pack) the cells will independently absorb what they can, based on their internal chemistry, resistance & impedance. Generally, within the main power curve, the cells will absorb pretty evenly within the pack itself and not usually deviate too much (< 30mv) but as they start to reach 80/85% and up above 3.400 cells will start to deviate more and can be as much as 1mv per AH capacity of the cells in question. So 100mv for 100AH cells. At this point, all a BMS can do is to run its balancing process if so equipped and it is slow... This is happening within Every Pack inside a battery bank when charging. Once a pack reaches Full Cutoff the remaining charge potential goes to the battery packs which are still taking charge until they all reach full. As all Battery Packs are now "full" and you will be able to observe, that they will "pack balance" as they settle post charging for roughly an hour (if not loaded). The various packs may fluctuate in voltage a bit as the cells settle and these packs will share their charge state across the "bank bus" and level up, usually within 0-20mv which will float a small amount up & down (more notable if there is BMS Balancing going on).

Discharging is again an odd situation... NB there is BMS differences which affect how this works and I am not getting into a long discussion on BMS tech. 04:30 now... Simply put, a BMS will cutoff for Low Volt Disconnect or LVD setting as programmed by user. With independent packs, each with a BMS they are "free agents". Now take the 100AH pack sitting next to a 200AH pack in a bank. Of course the 100AH will not have the same capacity / watt hours as the 200. It will reach FULL before the 200 when charging as will hit Empty before the 200 when discharging.

The confusion and some of this is BMS related....
Lead, AGM have none, they go until they die. Take a 12V Lead battery and you can discharge it to ZERO VOLTS. You can overcharge it to 16.0V and it will have no idea that it is in danger but we know that's not good. With Lithium, the BMS stops "stupid" from wrecking the battery pack and cells within. The BMS will cutoff at damage level points, when it cuts off, it is cutoff, nothing in or out. Some BMS' require a manual reset, others use "test pulses" and open momentarily to see if charge / discharge state exists and act accordingly.

LFP while discharging has no clue about where the juice is going, it does not care. When charging, it can take a low amp charge to full 1C or better (depending on your cells type). When in an LVD state and pulse testing, the BMS won't reopen to charge unless it's significant enough, so it should NOT parasite the other full packs. *** BMS's are different and work differently, I am describing a typical good design. *** The 200 will continue to discharge and service the load until it reaches LVD. Much of the confusion results from the BMS' interactions and the control they have because Lead & AGM do not have this as they are "stupid".

WHAT MAKES IT WORSE ~ Confusion wise ~
We talk about Parallel Battery Packs within a Bank.
Pack (aka battery) = Complete battery assembly with a BMS which can act as an independent entity.
Bank = An assemblage of Battery Packs into one common DC Bank.
~~~ but then we have paralleled cells which does not exist in Lead Land the same way. Take a typical 6V Lead battery, it has 3x 2V cells internally linked and no smarts between the cells. The lead cells are in series to make the 6V and that's it.
With Lithium Based batteries, we can Parallel Cells within the battery pack which increases the packs overall Amp Hour capacity, which is a very different process. This is often a "Mental Trip" especially for people transitioning off Lead Variants.

Something old time Lead Heads will recognize...
When you've tested the electrolyte levels of the cells within your lead battery, ever notice how some cells may present a different gravity compared to their sibling ? When we see that, it's Equalize Time after a watering (if required) to bring the cells within the batteries up to snuff again. With LFP, the BMS (if it has balancing) will level the cells out). With lead, all the batteries in series will collectively go up or down and the parallel set will mimic it (if properly wired) and stay fairly level. They'll also go down to 100% DOD if allowed to should the inverter fail to cutoff for low voltage. They will also go into Overcharge if the chargers fails to cutoff at a specific voltage or End Amps. Lithium BMS prevents that.

OKAY... I gotta call it, 05:00 nap time....
I hope this helps rather than confuses... It is not just a simple sentence to explain and there is a heap more to it... just tried to keep it simple but somehow I figure I may have missed the mark. Ohhh well I tried.
 
Wow @Steve_S - sorry for having ruined your sleep.

I've read it twice, but I think I need to read it again. I'm guessing I will have more questions based on what you have written. Thanks for spending the time. I really appreciate it!

Steve
 
Wow @Steve_S - sorry for having ruined your sleep.

I've read it twice, but I think I need to read it again. I'm guessing I will have more questions based on what you have written. Thanks for spending the time. I really appreciate it!

Steve
I sleep very strange and weird hours and often have to take at least two naps during a day. It's a direct side effect of the cancer and my continued deterioration. Some days it's worse than others and the worse days are becoming ever more frequent. Plugging forward anyhow and doing the good things to do.. like helping folks improve our world by using Solar even a little bit gives my 3 surviving daughters a chance at some sort of future. Sometimes I get a tad OFF when I am really tired or in a lot of pain and what I wrote above actually is not bad considering how I was feeling at the time.
 
Here is a discussion of some of the issues that are encountered when paralleling strings or packs.
It does not answer all the questions that have been raised but some may find it useful.
That actually is very helpful, and makes me feel better. The paper actually confirms my suspicion that there is current flowing between packs (the paper calls them "Eddy currents") and those currents reduce the total Ah. The example he gives is two 10Ah strings in parallel, will always have less than 20Ah of total capacity.

I still have more to read in the document, but skimming through the rest he never addresses the concept of paralleling disparate capacity packs. Judging from what I've read so far, I think the author would not be recommend it.

Thanks @Ampster !
 
I sleep very strange and weird hours and often have to take at least two naps during a day. It's a direct side effect of the cancer and my continued deterioration. Some days it's worse than others and the worse days are becoming ever more frequent. Plugging forward anyhow and doing the good things to do.. like helping folks improve our world by using Solar even a little bit gives my 3 surviving daughters a chance at some sort of future. Sometimes I get a tad OFF when I am really tired or in a lot of pain and what I wrote above actually is not bad considering how I was feeling at the time.
I'm very sorry Steve. I wasn't aware.
 
I'm very sorry Steve. I wasn't aware.
No worries No problems, it is what it is. Stupid thing is, had I been able to get a Prostate PSA test when I was 40 they may have caught it... But Ohh No, not till after 50 they say... The PSA test was the Big Flag at 2.19 which resulted in more test and well, so much for that, way too late and it had spread and others flourished in the meantime which is why the PSA numbers were so high.

MEN, don't WAIT till your 50 to get a PSA Blood Test ! It is quick, painless and easy, if it comes back with a high number follow up quickly. If your doctor says you do not need it till your 50 and over, Cold Cock the bugger on the spot !
 
I sleep very strange and weird hours and often have to take at least two naps during a day. It's a direct side effect of the cancer and my continued deterioration. Some days it's worse than others and the worse days are becoming ever more frequent. Plugging forward anyhow and doing the good things to do.. like helping folks improve our world by using Solar even a little bit gives my 3 surviving daughters a chance at some sort of future. Sometimes I get a tad OFF when I am really tired or in a lot of pain and what I wrote above actually is not bad considering how I was feeling at the time.
Thank you Steve
I've watched other post and always appreciated the comments you've offered up. You've helped us all grow in this forum.
Will put you in my prayers.
I sleep very strange and weird hours and often have to take at least two naps during a day. It's a direct side effect of the cancer and my continued deterioration. Some days it's worse than others and the worse days are becoming ever more frequent. Plugging forward anyhow and doing the good things to do.. like helping folks improve our world by using Solar even a little bit gives my 3 surviving daughters a chance at some sort of future. Sometimes I get a tad OFF when I am really tired or in a lot of pain and what I wrote above actually is not bad considering how I was feeling at the time.
 
I would expect, all other things being equal, same or very similar battery chemistry that the higher capacity cell would have a proportinately lower internal resistance, so higher capacity cells will tend to source more of the current load. As long as the battery + and - terminals are tied together with a good conductor the voltage potential should be the same at both pack terminals. If the chemistry is the same the two packs should have very similar discharge curves, so one would expect that the percentage capacity of the two batteries would remain close to each other. Therefore they will both drain at rates roughly proportional to their capacities. The problem comes in at the top of the discharge curve and at the bottom. One BMS will cut off before the other, so you have to be sure that each pack BMS can handle the current draw or the charge current that is provided to the whole assembly. Also, I do have concerns about what happens near the end of life of the cells. But for new or mid-life LiFePO4 packs I don't see where there is a problem in tying them together in parallel.

I have a an 8S 202AH pack which I built a few months ago. If I can snag 8 280AH cells at a good price I might buy those to increase the capacity of my system. My power and current requirements will not change. I'll still be using the combined pack to power my 2000W inverter, so I know that either pack by itself will be able to handle my solar charge current and the current draw. I'll even think about paralleling each 202AH cell with a 280AH cell if they are close enough in physical dimensions. That way I wouldn't even have to necessarily buy another BMS. Only problem is I don't know whether the Overkill Solar BMS I have has enough cell balancing current to keep such a large battery well balanced.
 
Intermixing different capacity cells within a single pack ?
THAT is begging for calamity & an open invitation for Murphy to drop his laws all over your backside.
Putting a 280AH pack next to a paralleled in 202AH pack will work fine but will present a few minor issues when charging and packs reach full (202 will get there 1st and when discharging the 202 will reach 0% SOC 1st as well).

K.I.S.S. Applied is always best.
 
Intermixing different capacity cells within a single pack ?
THAT is begging for calamity & an open invitation for Murphy to drop his laws all over your backside.
Putting a 280AH pack next to a paralleled in 202AH pack will work fine but will present a few minor issues when charging and packs reach full (202 will get there 1st and when discharging the 202 will reach 0% SOC 1st as well).

K.I.S.S. Applied is always best.
No. Since the batteries are at the same chemistry and are in parallel (and therefore at the exact same voltage) they will have the same S.O.C. in theory. Now, I know there are sometimes deviations between theory and reality, but I wouldn't be afraid to give it a shot. The smaller battery would reach empty first if they were in series, but that doesn't happen when they are in parallel.

To be clear, I'm talking about paralleling a 200AH 8S1P to a 280AH 8S1P. I would also not be averse to paralleling each 200AH cell to a 280AH cell, then assembling them all together to make an 8S2P configuration with 480AH. Each parallel group of 2 cells, even though mismatched will be at the same S.O.C as long as the bus bars are very good. I am not 100% sure on this last setup, but I wouldn't be afraid to give it a try. I'm pretty sure there wouldn't be any issues as long as none of the cells are near end of life.
 
Last edited:
No. Since the batteries are at the same chemistry and are in parallel they will have the same S.O.C. in theory. Now, I know there are sometimes deviations between theory and reality, but I wouldn't be afraid to give it a shot. The smaller battery would reach empty first if they were in series, but that doesn't happen when they are in parallel.
No. Since the batteries are at the same chemistry and are in parallel (and therefore at the exact same voltage) they will have the same S.O.C. in theory. Now, I know there are sometimes deviations between theory and reality, but I wouldn't be afraid to give it a shot. The smaller battery would reach empty first if they were in series, but that doesn't happen when they are in parallel.

To be clear, I'm talking about paralleling a 200AH 8S1P to a 280AH 8S1P. I would also not be averse to paralleling each 200AH cell to a 280AH cell, then assembling them all together to make an 8S2P configuration with 480AH. Each parallel group of 2 cells, even though mismatched will be at the same S.O.C as long as the bus bars are very good. I am not 100% sure on this last setup, but I wouldn't be afraid to give it a try. I'm pretty sure there wouldn't be any issues as long as none of the cells are near end of life.
Now, of course it is important to either top or bottom balance all cells before assembling and double checking cell voltages before paralleling them to make sure they are very close.
 
There is no problem in paralleling 2 LFPs with different capacity (or more). This is done all the time with capacitors of different sizes in electronics!

If you take for instance a 100AH + 200AH LFPs in parallel and start discharging with a 300A load you will see that the 200AH is supplying 200Amps of current and the 100AH is supplying 100A of current while discharging. There are no large currents going in between the parallel connection and they share the same potentiale (voltage). Remember to connect the load diagonally to the two batteries in parallel

You could look at it as two containers of same heights but different in volume being connected together with a pipe in the bottom and filled with water. You will see that the height of the water in the two containers equalize and becomes the same heights but different amount of liters. If you drain water from in between the two containers by connecting a hose to the pipe in the middle you will see that the height of the water in the two containers will drop at the same rate. The same analogy can be used for capacitors and batteries
 
I posted the screen shot of 2 different batteries ( one is a 90 amp hour and the other is a 180 amp hour) and they have running together paralleled now for 4 months in home solar setup. The loads and charges vary. When they are first connect they should be close in charge but little off is fine. I notice when charging or discharging the larger battery takes more of the currant and if out of sync then the smaller ones takes the load until they are the same state of charge. They both top out the same time being top balanced.
 
I saw some of the data shots of mis-matched capacities from a few folks earlier in the post. That fascinates me. I would have never believed it, and still wary to try it.

But I would love to see some more folks with this type of data to make me brave enough to try it :cool:
 
As @jpimlott indicates, I have 2x175AH & 2x280AH in Parallel it's very similar observations. I use Chargery BMS' which will drop the 175's into "storage mode" after they cutoff for Low Volt while the 280's keep pumping out energy. Different BMS' behave differently.

As soon as charge voltage/amperage is sensed they switch to charge mode and start charging up. Even when all four packs have hit LVD and start taking charge, If pushing 75A, the 280's will take 25A each and the 175's will take 12A each, the amperage floats around a little as they take charge which is expected.

When the 175's hit Full and start the HVD cutoff "cycle" as the 280's obviously continue taking more charge. The HVD cycle happens because when this cells hit HVD cutoff they settle and then reconnect to take more charge and then disconnect again when a cell hits HVD and the cycle starts over again. This is when things get tricky because of the difference in pack capacities.

When charging from an Inverter/Charger, it only sees the Battery Bank as one collective and only reads the Battery Bank Voltage and the collective resistance. Trying to gauge "End Amps" does not work by the way with multiple packs. The Charger is for all intents & purposes "dumb" and it will keep pushing out juice while "a" battery pack is taking it, without regard. Now the two 175's are Full and doing their HVD Cycle, then a 280 hit's full and starts the same dance, BUT when the 2nd 280 hit's HVD point this is when Nasty behaviour starts. When all 4 packs hit onto HVD simultaneously and the entire bank cut's off the charger will HATE IT ! A Good Inverter/Charger will do a Safety Trip and stop charging and flip cycle to Invert Mode BUT if Bank is still in HVD Cutoff you get a big surge dump and shutdown.
* Cheaper "Value Inverter/Chargers" Can just trip and shutdown "hard" and the really cheap HF models can FRY in such instances. Seen it Happen, more than once (luckily not on my own gear, phew....)
- IF the Bank happens to come back online fast enough, the Inverter/Charger may attempt to continue charging until it gets tripped again.
- IF the Bank is Not back up by the time the Inverter cycles (this is FAST), it will surge dump it's capacitors and Safety Trip and shutdown in "error mode". (No Batt sensed).
- IF there is a Generator on the backend providing AC to the Inverter/Charger and the Charger all of a sudden drops cold, the Generator will NOT like that fast drop and depending on the load demand a sudden drop CAN trip the Genset circuit breaker. If the load is "hi" for the Genset (at the top of its specified range), this is really hard on its electronic hardware. *This can kill smaller Inverter Generators if they are at the top of their output capacity... That was a COSTLY Lesson Learned ! Anyone want a nicely fried 4Kw Inverter Generator ? All out of Magic Smoke.
*NB* Almost all Inverter Generators are "High Frequency" Inverters which are intolerant of spikes & surges and do not handle hard drops very well at all.

The 105AH difference between the packs is a dance and not that pleasant because of the above. END Emps are more or less useless because it based on the concept that you are taking a battery to 100% SOC and that everything is perfectly matched & uniform. With LFP we are not getting all the cells to 3.65V "100% full" which is where you would try to calculate that end amp resistance. At 3.500V the cells can still take a lot of amperage. Unless ALL of the cells are perfectly Matched/Batched & Binned (NONE of the Lower Cost ESS Grade cells we use here are, "NONE" ! ) you will never get all of the cells to a uniform end voltage & amperage. There will always be a "Runner" that reaches HVD in a pack before the other cells. You will also have a "Lazy" that will dip to LVD (Lo Volt Disconnect) and trip cutoff before the others (most noticeable when cells are under load and usually below the 2.900V threshold).

I use a heavy 7200/9000 Genset as a backup power source. When the Samlex EVO-4024 Inverter/Charger is pushing 80A to the battery bank, it is pulling 120V/22A (+/-3A) and 2500W (+/-200W) from the L5:30 Line. This is also providing "pass-through" power as well at the same time. Any solar power coming in just get's added into the charge voltage/amperage.

Hope this explanation helps enlighten on the issue of using different capacity battery packs in a paralleled battery bank.
If there are batteries in Series, then that is another can of worms which is really best to avoid.
.
 
Back
Top