Drop-in LiFePo4 balance issue

[lpbvi]

Hi all, I'm new here and very grateful for the wealth of experience and different ideas and techniques on this forum. I have an issue with a 120AH drop-in battery:
-One cell always lags behind the others at the top of the charge and the same cell drops below the others at low states of charge (about 20%).
-Anothercell always spikes at high SOC, triggering the cell high voltage protection in the Daly BMS.
-Mid range voltages are generally within a few mv.
-I reduced my charge controller to 13.8v to make sure the highest cells wouldn't go too high. That works, but when the charge current starts to drop because the spiking cell raises the pack voltage, the reduced current causes the lagging cell to fall even more in voltage (up to 0.3v difference) so I feel it will never charge as much as the others.
-The BMS appears to only activate balancing (active balance, according to the SmartBMS app) when charge currents are above 2A, so the time it can spend balancing is limited. I understand the balance current is pretty small, so it might take forever to make difference.

If I want to limit regular discharge to about 80%, I'm worried I'll be cycling that cell deeper than the others if it doesn't balance. I'd like to set the balance cut-in voltage low, such as 3.2vpc so that the pack spends more time balancing, but I read on another thread that would undo the top balance. Is this the case? Surely the balancing works to keep cell voltages close, regardless of the state of charge and the more time spent with it on, the more time the low cell has to catch up.

Why would one cell stay low at both high and low SOC? Could it be a larger Ah capacity than the others (in which case why does it drop more than the others when low)? A less charged state than the others which will blance out eventually? Higher internal resistance? I can live with it, since the battery capacity is roughly within the advertised range, but of course if there's a way to improve things I'm all ears.

 

[zorrobyte]

Not much you can do without opening it up and manually balancing I'm afraid. What brand is it?

The cells are likely "grade B" and not matched from the OEM, pretty common on inexpensive batteries.

 

[Goboatingnow]

You can just live with the Imbalance , all it is effectively reducing the overall battery capacity. . With cheap cells you should plan to seriously over provision, 30% capacity headroom or more.

Plan to limit DoD to 50% or less if charge sources are online always recharge as soon as possible.

 

[lpbvi]

Thanks for the replies The brand is AMPS (sold by Sterling Power in the UK). The company have been very helpful on this and another issue I had, so I'm still experimenting with different charge settings and a few more cycles before really being able to give a verdict on whether it's any good or not. The battery wasn't the most expensive option, but certainly not the cheapest either.
Shame about limiting to 50% DoD - kind of defeats many of the benefits of having lithium batteries rather than lead!

 

[zorrobyte]

I don't agree with needing to limit SoC, cells being out of balance just limits your min/max capacity. Well, as long as the BMS high/low cell voltage isn't too low or high.

I'd try and confirm for sure that their BMS can do passive equalization/balancing. It should slowly balance over time

 

[Goboatingnow]

DoD is one of the major factors in cell life. With cheap cells your already skirting the boundaries anyway

Hence the more headroom in capacity you build in the longer these cheaper cells will last before degradation

So use the cost savings to build more headroom capacity and by definition this gives you lower DoD figures. Don’t delay any recharging restarts if available .

A certain degree of out of balance can be lived with all it really results is lower capacity then the sticker value. What you will find is over time with medium C discharge and Charge rates ( ie 0.5>C<0.25 ) is that actually cells will converge.

Again the headroom capacity means you’re less sensitive to having all cells perfectly balanced

If you need high capacity and high DoD don’t buy cheap LFP , buy Winston or CALB

 

[Substrate]

Whatever you do, don't set your balance trigger at 3.2v. That is too low, and YES, it will unbalance the cells because balance is only done once you are well into the upper-knee of charge. Even 3.4v is skirting the hairy edge.

At a nominal 3.2v, due to LFP's flat charge profile, that voltage could mean anywhere from say 20-90% SOC. Heh, so at 3.2v as a balance trigger, it isn't "top" balancing, but "mid balancing", which is a misnomer and leads to endless bleeder resistor burn off.

Your best bet at this point is to cycle the AMP batt normally to full, since you have no manual access. The first handful of cycles will probably trigger on that one cell's high voltage. Hopefully, as cycles go on, you'll get better. If you never charge to full (or at least attempt to in the beginning), then the bleeder resistors will never get a chance to do their job. As painfully slow as that might seem.

One method suggested by Renogy under similar conditions with a 100ah batt, is to use a very low amperage charger, like 1 to 3A to help give the bleeders time until the balance improves. Obviously charging from a heavy discharge will take insanely long, so charge up to as full as you can get it with your higher powered panels / higher powered normal charger, and then discharge a few amps with a load. Then recharge to full with the dinky 1-3A charger. A Noco 2A lithium charger might be helpful here, and may be faster than just normal high current charge cycling until the balance becomes reasonable.

 

[sunshine_eggo]

+1 to what @Substrate said plus...

-One cell always lags behind the others at the top of the charge and the same cell drops below the others at low states of charge (about 20%).

Probably just at a lower state of charge than the others.

-Anothercell always spikes at high SOC, triggering the cell high voltage protection in the Daly BMS.

The fullest cell.

-Mid range voltages are generally within a few mv.

Always the case even in batteries with notable cell imbalance. LFP voltage curve is FLAT.

the battery capacity is roughly within the advertised range, but of course if there's a way to improve things I'm all ears.

Thanks for the replies The brand is AMPS (sold by Sterling Power in the UK). The company have been very helpful on this and another issue I had, so I'm still experimenting with different charge settings and a few more cycles before really being able to give a verdict on whether it's any good or not. The battery wasn't the most expensive option, but certainly not the cheapest either.
Shame about limiting to 50% DoD - kind of defeats many of the benefits of having lithium batteries rather than lead!

Will the battery charge to the spec charge voltage without triggering OVP? If no and/or if you are unable to extract the full 120Ah due to this issue, the battery does not meet spec. A warranty replacement is in order. You shouldn't have to mess with it continuously.

 

[lpbvi]

Whatever you do, don't set your balance trigger at 3.2v. That is too low, and YES, it will unbalance the cells because balance is only done once you are well into the upper-knee of charge. Even 3.3v is skirting the hairy edge.

At a nominal 3.2v, due to LFP's flat charge profile, that voltage could mean anywhere from say 20-90% SOC. Heh, so at 3.2v as a balance trigger, it isn't "top" balancing, but "mid balancing", which is a misnomer and leads to endless bleeder resistor burn off.

Your best bet at this point is to cycle the AMP batt normally to full, since you have no manual access. The first handful of cycles will probably trigger on that one cell's high voltage. Hopefully, as cycles go on, you'll get better. If you never charge to full (or at least attempt to in the beginning), then the bleeder resistors will never get a chance to do their job. As painfully slow as that might seem.

One method suggested by Renogy under similar conditions with a 100ah batt, is to use a very low amperage charger, like 1 to 3A to help give the bleeders time until the balance improves. Obviously charging from a heavy discharge will take insanely long, so charge up to as full as you can get it with your higher powered panels / higher powered normal charger, and then discharge a few amps with a load. Then recharge to full with the dinky 1-3A charger. A Noco 2A lithium charger might be helpful here, and may be faster than just normal high current charge cycling until the balance becomes reasonable.

Thank you for the advice about balance settings. I will set it higher. I have been experimenting with a charge of a few amps and it does seem to help with bringing all the cells up with less variation. That one cell is still behind, but I'll see how it goes with time.

 

[mikefitz]

Hi, I have been assisting others who have had similar issues and other problems with batteries from Sterling Power and have been in communication with Sterling's technical staff. I no longer recommend batteries from Sterling
It's clear that there are issues with cell balance and I suggest you return the battery.
You may also have issues with the Daly BMS reporting incorrect current values.
Alternative ' low cost' batteries from

Roamer - Premium LiFePO4 Leisure Batteries for off grid

For continuous power for non-stop adventures, say hello to Roamer – the next generation in SMART 12v and 24V LiFePO4 leisure batteries for campervans, boats and off grid life. Equal parts power, intelligence and reliability - managed with a smartphone app that puts the power in your hands. 10...

roamerbatteries.com

And,

Lithium Leisure Battery | Fogstar UK

‘Fogstar Drift’ Lithium Leisure Battery | All models include Grade A EVE Prismatic LiFePO4 cells, High Discharge JBD BMS, Bluetooth, Heating | Lowest Prices in the UK. 10 Year Warranty.

www.fogstar.co.uk

As yet no experience with Fogstar products but the claim to use grade A EVE cells and JBD bms suggests a good product.

I am aware that Roamer have had issues with a small number of their batteries but have replaced faulty ( poor balance) batteries promptly. I have communicated with Steve who runs Roamer batteries and although a small organisation makes an effort to keep customers happy.

These are small companies relying on quality control from their suppliers in China. What does not seem to be happening is testing in the UK to ensure the batteries are useable.
I think that most users of these batteries have no understanding of cell balance and BMS protection, just charge and use the battery.

Mike

 

[lpbvi]

Will the battery charge to the spec charge voltage without triggering OVP? If no and/or if you are unable to extract the full 120Ah due to this issue, the battery does not meet spec. A warranty replacement is in order. You shouldn't have to mess with it continuously.

No, not without lowering the charger voltage. Capacity is ok, I think. I'm a long way from where the battery was bought (a replacement would be 3 months to return the battery, then another 3 months to get a replacement). Oh well. If things improve I'll count myself lucky and make a solemn promise to buy better quality next time. If not, I'll break into it and see what I can do.

 

[Substrate]

Thank you for the advice about balance settings. I will set it higher. I have been experimenting with a charge of a few amps and it does seem to help with bringing all the cells up with less variation. That one cell is still behind, but I'll see how it goes with time.

Cool - what's happening here is that unless a cell reaches 3.4v (3.45 if you want to get more accurate), then the cell will never absorb and accept more current to fully charge. To achieve a full charge, LFP needs to see 3.4v or more per cell. Otherwise it will just "stall" at about 80% soc.

Yeah, try lower current and use a 3.45v balance trigger (or higher if you see improvement like to 3.5).

Or do what Will suggests - use the thing like 15 times and check for improvement.

 

[lpbvi]

Hi, I have been assisting others who have had similar issues and other problems with batteries from Sterling Power and have been in communication with Sterling's technical staff. I no longer recommend batteries from Sterling
It's clear that there are issues with cell balance and I suggest you return the battery.
You may also have issues with the Daly BMS reporting incorrect current values.
Alternative ' low cost' batteries from

Roamer - Premium LiFePO4 Leisure Batteries for off grid

For continuous power for non-stop adventures, say hello to Roamer – the next generation in SMART 12v and 24V LiFePO4 leisure batteries for campervans, boats and off grid life. Equal parts power, intelligence and reliability - managed with a smartphone app that puts the power in your hands. 10...

roamerbatteries.com

And,

Lithium Leisure Battery | Fogstar UK

‘Fogstar Drift’ Lithium Leisure Battery | All models include Grade A EVE Prismatic LiFePO4 cells, High Discharge JBD BMS, Bluetooth, Heating | Lowest Prices in the UK. 10 Year Warranty.

www.fogstar.co.uk

As yet no experience with Fogstar products but the claim to use grade A EVE cells and JBD bms suggests a good product.

I am aware that Roamer have had issues with a small number of their batteries but have replaced faulty ( poor balance) batteries promptly. I have communicated with Steve who runs Roamer batteries and although a small organisation makes an effort to keep customers happy.

I sure wish I'd done a bit more research and shopped around more. I went with Sterling in the end because a lot of their other products (like B2B chargers) are widely sold and recommended by quite a few retailers. As I said in reply to another post, returning the battery takes so long because of where I am that it's a real last resort. Thanks for the links to other some other UK battery suppliers. I know where I'll be buying from now if the need arises.

 

[lpbvi]

Cool - what's happening here is that unless a cell reaches 3.4v (3.45 if you want to get more accurate), then the cell will never absorb and accept more current to fully charge. To achieve a full charge, LFP needs to see 3.4v or more per cell. Otherwise it will just "stall" at about 80% soc.

Yeah, try lower current and use a 3.45v balance trigger (or higher if you see improvement like to 3.5).

Or do what Will suggests - use the thing like 15 times and check for improvement.

That's very useful info. Basic stuff to everyone who's been dealing with it for long enough, I'm sure, but I needed it spelled out clearly. I will keep at it with a low charge rate and see if I can at least get that one cell above 3.4 without the others running away.

 

[Substrate]

Cool - yeah this is a last resort solution if return is impractical.

The other question though is should a bms be user-programmable with known problem points (like too low of a balance trigger voltage) creating problems instead of fixing them?

 

[Goboatingnow]

I’ve had a few Sterling Power products , won’t touch them now.

 

[Goboatingnow]

Whatever you do, don't set your balance trigger at 3.2v. That is too low, and YES, it will unbalance the cells because balance is only done once you are well into the upper-knee of charge. Even 3.3v is skirting the hairy edge.

At a nominal 3.2v, due to LFP's flat charge profile, that voltage could mean anywhere from say 20-90% SOC. Heh, so at 3.2v as a balance trigger, it isn't "top" balancing, but "mid balancing", which is a misnomer and leads to endless bleeder resistor burn off.

Your best bet at this point is to cycle the AMP batt normally to full, since you have no manual access. The first handful of cycles will probably trigger on that one cell's high voltage. Hopefully, as cycles go on, you'll get better. If you never charge to full (or at least attempt to in the beginning), then the bleeder resistors will never get a chance to do their job. As painfully slow as that might seem.

One method suggested by Renogy under similar conditions with a 100ah batt, is to use a very low amperage charger, like 1 to 3A to help give the bleeders time until the balance improves. Obviously charging from a heavy discharge will take insanely long, so charge up to as full as you can get it with your higher powered panels / higher powered normal charger, and then discharge a few amps with a load. Then recharge to full with the dinky 1-3A charger. A Noco 2A lithium charger might be helpful here, and may be faster than just normal high current charge cycling until the balance becomes reasonable.

Charging LFPs using less the 0.1C , really contributes to SEI layer growth and damages battery life. Best avoided if at all possible.

 

[Substrate]

I agree with that with. Surely one isn't going to do a bulk-charge at dinky currents at less than 0.1C. Not only is it very inefficient, it is just purely annoyingly slow.

The recommendation by Renogy as a quick-fix bandaid to use a very small current for charge is meant for near the end of charge, to try and give bleeders a bit more time to operate without driving the runaway cells into an immediate over-volt bms disconnect.

So yes, use a charger / solar setup larger than 0.1C for bulk charge. Then discharge maybe 5-10% with a load, and then recharge with the dinky charger. Not as a permanent solution mind you. Rinse and repeat if needed until the runaways don't immediately trigger the bms disconnect.

While 0.1C is also my personal minimum charge current goal, under some solar conditions, you may get even less than that on a regular basis, but yeah I know what you are saying.

Used to drive me nuts back in the day - guy spends $500 on nice battery, and ruins it with a $2 unregulated wall wart.

 

[lpbvi]

Charging LFPs using less the 0.1C , really contributes to SEI layer growth and damages battery life. Best avoided if at all possible.

Almost every time I check this forum I discover some new quirk or limitation of LFPs! I'm using my batteries (2x120Ah) in a liveaboard sailboat and I'll very often have less than 0.1C charge from wind or solar. I am counting on bulk charging them with the engine as little as possible. I have a feeling these batteries will be my moderately expensive personal experiment, a punishment for not learning absolutely everything there is to know about LFP before buying! I'll set reasonable limits for everything and they will last as long as they last. Thank you all very much for the replies, though, much appreciated.

 

[Goboatingnow]

Almost every time I check this forum I discover some new quirk or limitation of LFPs! I'm using my batteries (2x120Ah) in a liveaboard sailboat and I'll very often have less than 0.1C charge from wind or solar. I am counting on bulk charging them with the engine as little as possible. I have a feeling these batteries will be my moderately expensive personal experiment, a punishment for not learning absolutely everything there is to know about LFP before buying! I'll set reasonable limits for everything and they will last as long as they last. Thank you all very much for the replies, though, much appreciated.

The issue is you need to assign” levels of concern “ to all aspects of LFP

battery life is affected by
DoD. Higher is worse
Low C charging (<= 0.1C)
Float charging , ie voltage impressed though battery is full
Temperature
Just sitting there at high SOC fir too long

But let’s say you have a 10000 cycle battery and you charge it so you get 5000 cycles. This may be way more then you care about. Hence the degradation factors like DoD or C rate may be irelevant to you

The same is true of top balancing. It’s no skin of LFP to sit at 75 , 80 % SOC. other then energy storage efficiency do NOT charging to 100% on all cells may not give you any cause for concern

So your nominal 200AH battery doesn’t top balance perfectly and hence is 170Ah , if your application only needs 130Ah , do you really care

Too much advice on LFP is presented as “ do this or it’s all wrong “ balancing is a classic one.