lifePO

[scott p]

Optimal charge/discharge long life cycle for LifePo4. I have been advised 90% cutoff, (53.6) Volt charge, 10% discharge cutoff  (48) volt. Or per cell 3.35V charge , down to 3 volt discharge.

[AdeV]

According to this website: https://www.evlithium.com/Blog/lifepo4-battery-care-guide.html

You should charge at a constant current until the battery reaches its 100% voltage level - then charge at constant voltage until the current is minimal. I suspect, without an actual lifepo4 charger, that would be quite challenging to do...

Not sure about discharge (the link above is vague to the point of uselessness), 10% sounds like a sensible margin. You may be able to pick up a BMS (Battery Management System) which will help in that respect - either by sending a signal to tell you to disconnect the load (or start charging), or by actually disconnecting it.

[BruceM]

Current limiting with a fixed voltage charge is well supported by most commercial PV charge regulators and is not usually not a problem even for older charge regulators.

LFP has to be regularly full charged if you have any hope of having an reliable state of charge indication and to do non-manual balancing.  The cell charge-voltage curve is flat with some variations until you reach around 90% of full charge.  Likewise, LFP cells should only be balanced by your BMS when approaching full voltage (voltage is rising appreciably), very unlike the lithium ion cells.  You can decide how full is full  if you like, but you must pick a value in the nearly full range where the voltage is rising substantially.  Most have so far found that even keeping the cells at full charge is not causing significant degradation over time.  See Will Proust's diysolarform.com to get some decent info. He does his technical homework well before presenting info.

I am not using LFP yet but just 5 months ago was working on an all linear, high voltage BMS design for 38 LFP cells in series with another EE friend.  In my research I found that the differences between LFP and Lithium Ion cells was causing trouble for many early LFP adopters with BMS hardware that was not well suited for them.  Balancing LFP by any means (active or passive, high or low current) at other than near full was always a mess.  The voltage-charge curve is just too flat, and instead of balancing, it was actually unbalancing.

Apparently large balancing current is not needed despite the more limited balancing time window, and 100ma to 1 amp seems to be working fine. The cells are just that good at least when young (under a decade old). 


Best Wishes,
BruceM

[scott p]

Well, the reason I decided to poke around here is that I bought 100 Lifpo 22 AH  cylindrical cells. The business ends have a flat surface with a raised circular ,10mm dia.  column that is internally threaded.  I use aluminun with 10mm holes as buses.

I already had 16 of these cells. I put them together as a 4P4S 12 volt  battery. I charged them with a multi use charger set to lifepo, balancing not included. It, as adev indicated,  charged  to  a constant voltage (13.5)  and stayed there for a long time and then shut down.  In some circles 13.5 volts is not  considered to be fully charged, talking over 14 volts by some.

I did a simple resister /amp test  of eleven amps, which did last for four  hours  but just barely. At 12 volts the voltage began to drop quickly, talking only a couple minutes ,if that, to drop to eleven volts.

Based on that  a 10% cutoff is right on the edge.

There is of course two methods of arraigning these  cells.  High current or high voltage. High voltage requires  a high order of balancing. A possible would be 38 cells in series times three. Whereas according to some , cells in parallel are self balancing, which cuts down  the  wiring by a large margin.

I lean to high voltage but can these balancers be trusted ?

I found this formula to calculate balancing current divide the total Ah capacity by two and then multiplying it by 0.1

[AdeV]

A BMS would be essential even in a mixed serial/parallel setup, I'd suggest. It will do the job of keeping your cells in balance, it should also cut off the power before any cell becomes dangerously undercharged.

Fogstar Batteries is a good place to look around - they might even be worth a call, they seem to know their stuff when it comes to LiFePo4 batteries...

https://www.fogstar.co.uk/collections/bms-battery-management-system

[scott p]

Understood,  if you have a large number of cells in a box that you can't open.  Balancing is a very good idea.
     
However, is it fair to say that sometimes the need or degree of balancing can lead to a band aid approach?
     
My situation is a DIY project.
   
I have watched  some DIY battery building where thin, nickel buses are happily spot welded by hand,  to the cells and there was a lot of cells, a lot.  They could not use aluminum or copper on account the welder was not  hot enough but given a powerful enough welder I think I saw some success with copper. Cost wise there is a wide range of small spot welders in the market but it's not wise to skimp.
     
Given the number of  welds I can probably say that overall the quality of the welds will not be consistent. If the welds are not exactly the same distance apart there will be lack of continuity. Miner and perhaps not so minor factors over time. There  was also talk about cell temperature and cell swelling, which perhaps is more of a design problem. Like so many things in today's world  maybe there is a certain amount of hype concerning BMS that needs to be considered.

My situation will be a bit different, I won't  be welding. I will be screwing aluminum busses to the cells and I am using the mill to drill the 10 mm holes center to center in the 1/8th inch  busses. Also the cells have a 22AH capacity so there aren't so many and each cell will be open to check . 
   
I downloaded  an open source 3D cad program, which is very helpful to see how a certain run of cells will look and how they can be bussed and connect to other runs.
   
Over time I will most likely use balancing but when I do I want to know what and why but within  reason. Perhaps some times it is simply a need to clean up a connection to bring the cell back into sync.
   
Recently I have been  looking at a old computer power supply to aid in balancing these cells. Two of the various voltages available are  3.3 and 5 volts The states on them indicate a fairly high amp but not so much, I found if you let them output full amps they will blow. Perhaps Five volts with variable  resistance can be a way to balance?
   
If I use the formula above, (22 / 2 * .1) I get a BMS charge current of  1.1 amp. That seems high to me and like Bruce said shouldn't it be related to the time and charge rate of your LifePo charger?  This little so called LifePo charger I have puts out a steady four amps for the entire duration (several hours) of it's charge cycle and powers down at 13.8 or so voltage. The cells seem to like it, there is still 12.8 or so volts left after a test. You can't compare lead acid to LifePo.
   
I have been spending a lot of time looking at inverters. There is some pretty fancy stuff out there. The surge in lithium cells has apparently caught some of the older inverters off guard, but most are now making room for lithium including BMS capability but the advised conditions for lithium varies from inverter to inverter.  Usually there is (some times vague) user settings available to set the inverter to your Lithium not necessarily LifePo battery and not necessarily over/under cutoffs appropriate to the batteries . There will be a lot of note taking before a proper inverter pops up.
   
Off  shore companies are turning out inverters, it's a buyer beware situation.

As I recall both Bruce and Ade are off grid I appreciate your feedback on inverters and other.

[AdeV]

I'm not off-grid, although I would like to be one day (or, at least, generating more than I use all year round).

I'd still recommend a BMS, even if you're making a DIY setup. The BMS will spot a cell going bad, for example, before it causes a fire. It will also make sure no cell gets overcharged, or over-discharged. Even with a DIY system, it's a vital safety component - and a good one will give you data as well, so you can see how your battery bank is performing over time.

[keith71]

You will not get 38 cells in ser/par to balance very good if at all. You will not get a 4p4s to even balance itself properly. Have you been to diysolar forum?  Great source of info. A bms with 500ma to 2amp balance is pretty much a must.. Even a $130 12 volt lifepo4 100 amp/Hr from amazon all come with a bms.
 
I am no expert, but the people over there while very helpful will explain it properly, but they will not put up with much nonsense.. Such as debating whether a balancer is needed for your bank and most banks.

Extremely smart people over in that forum. You will get solid advice.

Not that you havent here, we have a few smart people as well.

[BruceM]

I agree with Keith71 that the diysolarforum.com is a good place to learn. 

All the LFP capable balancers will work, and obviously, even high voltage series/parallel configurations work fine with a suitable passive balancer, and that's what most EV's use. 

Many people have managed to do manual top balancing and monitoring and  re-balancing of new matched cells, but I feel that is asking for trouble. I like idiot proofing for myself.  Even high voltage balancers are available on Aliexpress for around $300 and up.  Daly makes one as well, though you can't find it in the US. 

The commercial digital based balancers which support LFP will only balance when charging, and at the specified (higher) voltage.  For higher voltages they typically use opto isolated SPI bus wuth a master controller board and multiple 16 cell BMS IC controllers. The master then can read and compare all the cells, and decide when and which cells to load, while monitoring temperature of each BMS IC.  It's the simplest, lowest cost method.

I prefer higher current (3A) passive cell balancers which only start when the "full" voltage is hit, and the peak balance current (analog opto isolated) is used to throttle back the charge current. This can be done entirely analog, which is my preference.  I use this method for my existing battery bank and I have spec'd out the design and parts for a 38 cell LFP version. 

The big question is whether LFP cells will be worth all that custom build effort for me, since they may be buggy whips in 10 years.

[dax021]

Another really good forum with very knowledgeable members here in SA.

https://powerforum.co.za/

Merry Xmas everyone.

[scott p]

 My hat is off to you ade  and bruce for your patient replies.

Bruce has put a name to my  initial thoughts concerning balancing.

Many people have managed to do manual top balancing and monitoring and  re-balancing of new matched cells, but I feel that is asking for trouble.

I can attest to the asking for trouble part of his quote but that is another story.

I have been spending a lot of time looking for information concerning inverters and LifePo cells. Predominately on the diysolarforum and the power forum za cite.

On diy solar I plugged into a thread titled  (How to Parallel Balancing. (YEP 99% of us is doing it wrong)(PART#1) posted 2020

Lots of  pushing back and forth concerning LifePo cells.  Finally the general conscience was that no one fully understood  these new cells.

On the first page of the thread Solar rat introduced a link (marine how to lifePo batteries on boats.) This fellow, based on a lot of hands on testing of LifePo cells, claims to have a good grasp of said cells. It's a long article whereas a few statements on his part tweaked my bias.  Last edit;3/26/23

I quote ::

These batteries can take immense current, and charge extremely fast, but really tend to do extremely well with .3C to .5C in charge current.
   
LFP batteries do not need to get back to 100% SoC frequently, like lead acid does. In fact, keeping LiFePO4 cells at 100% SoC can actually negatively impact cycle life.
   
Unfortunately most commercially available chargers, solar controllers and alternator regulators are of extremely limited design and are just not well suited to charging LFP banks. (Scott p adds standard  inverter/chargers to that comment)
   
 In a well designed fractional “C” system where the charging voltages used are not pushing into the knees regularly, the need for cell balancing, with well matched cells, can be extremely rare and you should rarely have a need to push the cells to cell balancing levels. For our lack of a need to re-balance I credit good initial cell matching, top balancing and sufficient, but not extreme, charging voltages. I purposely keeping my own bank out of the knee ranges on charge and discharge cycles.

He is not the only person holding that point of view, but he goes on to discuss BMS  and what to expect.  That might explain why some some choose to balance manually.

quote ::

 I am also a believer that high charging voltages, above 14.2V, per 12V nominal pack, can result in more of a need for balancing. Pushing the charge voltages too high tends to result in more need for balancing, and it becomes a vicious cycle. A real catch 22. Higher charging voltages actually tend to serve to create a need for a balancing BMS system. A well built and well cell matched DIY bank will deliver all the capacity you’ll need, when charged to just 13.8V – 14.0V. Why go any higher if it is not necessary.
   
When looking at the attachments below I can appreciate why staying out of the knees looks like a good plan.  Bad things can happen quickly and at a charge/discharge ratio around 80/20% the cells are virtually 100% charged anyway.

Also, when looking at a 3D picture of say a 6P?S block up it seemed logical to separate the 6 cells into groups of two and a buss between each two as they went series so that they shared a bus equally current wise rather than a cell pushing it's way over other cells to get to a buss. I call it balanced electron flow since electrons are considered to be particles and not waves of pure energy.

Concerning BMS, I saw a Utube fellow applying two little wires to each cell X a lot of cells. I also saw a 16s BMS with one black com  wire connected to the negative battery post and one positive red wire for each 2P12S bank of 24 cells X 16 banks. 

Which one is right ?

[scott p]

I have no inclination or need to step on anyone's toes. I am just playing the part of a messenger.

Do I even need a BMS ?? Absolutely, for the very same reasons put out by Ade and bruce to keep an eye on things.
I have spent more time than I should looking at inverters and BMS units, mostly inverters . I found a workaround to charge the batteries based on lithium rather than lead cells. CHARGEVERTER-GC by EG4 electronics.
(Quote)

The EG4 Chargeverter-GC by EG4 electronics  is a robust AC-to-DC converter designed specifically for 48 VDC Lithium battery charging in on-grid and off-grid Energy Storage Systems. The charger can operate on 120, 208,
or 240VAC and is capable of up to 100 Amps of DC charging current.

Fully programmable for lithium (lifePo) with automatic generator start / stop and a RS485 (RJ45 Jack) for communication purposes to monitor a batteries SOC , but also a battery voltage over\under charge if battery SOC is unavailable. This will be an option if a chosen inverter charger is not fully capable with lithium (lifePo). Either a standalone off grid or I have noticed some inverter chargers can shut off the charger (double check on that, (human error) the two chargers would not get along.) Perhaps, if possible and to be safe no inverter solar or generator wire ups.
It is based on a generator to keep the battery charged but that is fine for now. I would like to have more detail on how they handle the auto start  process. That could be tricky.

And a  sweeping, large, expensive BMS.   quote

(The Orion BMS 2 by Ewert Energy Systems) is the second generation of the Orion BMS. The Orion BMS 2 is designed to manage and protect Lithium ion battery packs and is suitable for use in electric, plug-in hybrid and hybrid electric vehicles as well as stationary applications.)

They also say. 

Lithium ion batteries, unlike lead-acid batteries, tend to stay in balance very well once they are initially balanced.
Differences in self discharge rates, cell temperature and internal resistance are the primary causes of an unbalanced
battery pack in a properly designed system, and these differences in self discharge rates are typically measured in micro amps (uA). The BMS must be able to correct these discrepancies in order to keep the cells balanced. The purpose of balancing a battery pack is to maximize the usable capacity. Even in an ideal battery pack, all cells will have slightly different capacities and will be at slightly different balance points. The total usable capacity of the battery pack is limited to the lowest capacity cell, less the difference in balance from the strongest to weakest cell.
Last Updated: 1/9/2018

Based on that scenario, passive balancing, which they do, would be totally viable, contrary to some opinions.  They mention a power draw for 108 cells to be 40 W. It appears, if I read right, the BMS unit is connected to a isolated Power supply of 12 or 24 volts. I have noticed some inverters have a out put to charge a generator battery.

These bits of information on their part is just rubbing the surface. They have a lot to say.