Understanding MPP Solar charge settings for LifePO4 batteries

[KenDan]

Because LifePO4 have such a flat discharge curve, you can't really balance based upon cell voltage in the 10% to 90% SOC range. If you have cells with only a moderately small difference in their state of charge, you probably don't see any noticeable difference in their voltage. That basically means the balancer probably can't do much.
It's only at an very low or very high state of charge where voltage differences will show up. That's why you have to "top" balance (or "bottom" balance). Trying to balance in the middle of the charge range is basically futile with LifePO4.
This was the thing that was tripping me up on balancing LifePO4 cells. All my previous experience with lithium batteries was Lithium Polymer. LiPo has a lot more slope on the discharge curve and can actually be balanced to a certain extent even in the middle of their operating voltage range.

 

[Capt Bill]

Re my MPP Solar Powered Charging Cycles on my LV2424s. I got in more lessons via observation of a charge cycle after draining my battery down lower than I wanted to ... With my charge setting at User Defined: Bulk 28.8v / Float 27.3v / & Cut Off 22.6v ... and charging my battery up from its' bottom voltage at about 90 Amps via Solar In alone, ... I did see the bulk charging mode via the mppt Solar Charging, via seeing my battery voltage (as indicated on the LV2424/ might be 0.5v higher at battery itself) go up beyond my float v setting to my Absorption volatage (Bulk charge) setting of 28.8v, before throttling back to Float v. PLUS I noticed this: When I got above 28v at full charge amps, one cell w the most resistance peaked up past my BMS's 3.65v cut off before the other cells. Seems my cell stay well balanced in big range below 3.5v per cell. / plus with Charge Amp switched off/ my battery settled down to 27.6v within 10 or 15 minutes. (I wonder if that is normal/ had recent overnight drain down of my battery / from a failed protection protocol that briefly got me down to 12v on my 24v LiFePO4s (Yes/Wow/ from a huge mistake/ from experiment and/or bad roll of dice. Wrote about that; an experiment that turned haywire / in the "LV2424 Dry Contacts" thread. I did get my battery up to 24v within minutes via my 10amp charge from my 30v 10A Bench power supply. ... I am still recovering from that one, as in still figuring out the details of why, how & exactly what happened/ on that unfortunate roll of dice.

I also have noticed, since putting my LV2424 online; from before my mishap; I have had a few low battery voltage fault alarms on my LV2424. including turning itself OFF (though solar in might still work? ... that I have not fully understood/ as in; once or twice my LV2424 would not reboot on a low battery (when it was above a 22.5v cut off volt setting/ and above 22v I read is min. need for LV2424s? , more likely had 23v going by memory). I recovered from those incidents via easy disconnect/ off to standby, then using my 30v/10a bench power supply for five minutes to get it all rebooted and functioning like new again. Most recently, got low volt fault beeping even though my battery was show 23 + volts on the LV2424 interface (wondered if form memory/ was recently after big mishap/ but at 23 + Battery volts showing at my LV2424 lcd?) ... I mainly want to share: from my most recent experience I do see MPP's LV2424 solar power bulk charging works. got reason to believe the grid charge has simialr charge profile; but w both do not have a bulk charge "timer setting" like the bigger MPP 48v 5000 watt version has (so think the LV2424 has some kind of auto protocol).

... For now, given how I have seen my individual cells getting out of balance in the very low volt range, and the higher than 3.5v per cell volt range, and staying in very good balance in a large section of mid volt range where most of the Ahs are ... I have now settled on these User Defined Voltage setting for my three LV2424s that I have configured for 240AC split phase / wired up as my Island style option for powering my main home sub panel, with and easy Grid Input option to either my LV2424, or House Sub panel (only solar or grid one at a time) to Grid to both without LV2424 powering up m house loads (like for a possible grid base battery charge cycle) .

I have now set my user defined setting to: Bulk Charge: 28.0v (to dent or stop my BMS triggers of high individual cell voltage alarms + hopefully extend my life cycles)
Set Float: 27.2v & Cut Off: 22.8v

I have a second set of 8 280 Ah LiFePO4s in transit/ on way to up my 280 Ahs to a 560 Ah (possibly might be less from mishap?) Battery Bank. I will hopefully get my third 2000 watt set PV panels (obtain locally in No. California @ $.50 per watt total cost including the Ca sales tax) ... wired up and dialing in to my 3rd LV2424 in next week or two for a full 6000 watts of pv panel power connect to 3 LV2424s. I will have to figure out more ways to use my Solar Generated Electricity when the Sun is Shining, and believe I will eventually be charging up a used Electric Vehicle (maybe a 2016 Nissan Leaf w that 30 kWh battery?) on top of a few water heating tasks. ... Very Open to Feedback ... :+) Happy Days, Bill

 

[KenDan]

I recently discovered that my cell balance problem is really caused by one of my cells having lower capacity than the others. Cells with mismatched capacity can not be properly balanced. It reaches full charge before the others and the voltage goes dangerously high. If I try to top balance, the voltage on the lower capacity cell drops faster than the others when I start to discharge which leads to large imbalance on the low end.
These are 150Ah aluminum cased cells - 15 are good and one now appears to be going bad. The worst part of this is that I can't seem to find anyone who will sell a single 150Ah cell anymore.
I'm tempted to just order 16 of the new 280Ah cells and then try to figure out what to do with the 150Ah cells which are only a few months old.

 

[Capt Bill]

Testing my 3 x LV2424s configured for 240ac split phase output, on my 280Ah LiFePO4 set (with update to 560 Ah coming soon), ... I just figured out how to make use of my potential solar panel watts to get a higher Amp Battery Charge cycle initiated; by simply Flipping my all my PV Solar IN current lines to all my LV2424s to OFF, and then back ON (easy by my switched circuit breakers): Before and After RESULTS: BEFORE: as morning sun is powering up my PV panels, .... battery charge definely looks like it is in float charge mode with 230Ah (for 280Ah set as full) SOC showing on my Chargery BMS8T (not convince that reading is totally accurate) ... and my Battery Voltage showing as 27.0v at my LV2424s (w Absorption config at 28.3v / Float 27.2v) ; I had just 21 Amps on Solar powered Battery Charge going into my Battery Bank (shwn on BMS); with only 900 Watts of my PV power being taken in by my LV2424s (shows on LV2424s PV page); ... AFTER: (via flipping Solar IN to off, then back on): My Solar Powered Battery Charge Cycle goes up to 65 Amps into Battery; & I immediately am using 2000 watts from my Solar PVs (vs just 900 watts previously being used) ... & half hour later, with more sun, I have a 75 amps charge going into my battery with my LV2424 showing 27.7v battery, and BMS now showing SOC @ 277Ah (again, not convince of accuracy /configured for 280ah full). ... I got about an Hour of Absorption Charge at 65 - 75 Amps before my Charging settled back to Float mode. My battery is currently showing 27.2v at both my LV2424 / and BMS is showing 279Ah SOC, and also 27.2v. :+) ... It would have taken a longer time to get my battery topped up at previous lower 21 Amp float kind of charge if I did not flip my solar IN off and back on !!!

WOW for that kind of discovery for me; who likes using solar watts to charge my battery asp when my PV are producing potential watts. ... I think I am gona automate a daily OFF / then ON cycles to make up for what I currently view as one of MPP's LV2424 shortcomings. I will be interested in your testing of this procedure to see what you think.

Fuller disclosure of my 3 x LV2424 / 240vac split phase set up. I only have two of my 2 LV2424 currently wired to Solar arrays IN; and will retest this discovery when I get my third 2000 watt array set up and wired to my 3rd LV2424 (coming up soon. :+) :+)

 

[Dauntless]

I have three of the lv5048's..The time at absorbtion can be set in the menu, the rub is that the absorbtion clock starts counting as soon as float voltage is reached. You would think you would want the clock to start once the bulk voltage is reached, but sadly no.. The automatic setting basically goes straight to float once float voltage is reached.

This does not work well for lifepo4 batteries as they need different absorbtion times for different depth of discharge..

My work around is to set a high float voltage and change the absorbtion time depending on the time of year.

End amps would be the way to go but unfortunately the lv5048 are not capable of this..

This, resetting the Float is exactly what I must do, if I want the batteries at 100% Normally i don't though, so I am happy to set Float to 13.3v, which is where my MPP 1012 stays when connected to shore power, which is most of the time.
So the only "fault" I see with the MPP is that once it goes to Bulk, it gets out of Bulk too soon and goes to Float once Float voltage is reached, thus the batteries don't get to 100%.

 

[Csimmer]

Testing my 3 x LV2424s configured for 240ac split phase output, on my 280Ah LiFePO4 set (with update to 560 Ah coming soon), ... I just figured out how to make use of my potential solar panel watts to get a higher Amp Battery Charge cycle initiated; by simply Flipping my all my PV Solar IN current lines to all my LV2424s to OFF, and then back ON (easy by my switched circuit breakers): Before and After RESULTS: BEFORE: as morning sun is powering up my PV panels, .... battery charge definely looks like it is in float charge mode with 230Ah (for 280Ah set as full) SOC showing on my Chargery BMS8T (not convince that reading is totally accurate) ... and my Battery Voltage showing as 27.0v at my LV2424s (w Absorption config at 28.3v / Float 27.2v) ; I had just 21 Amps on Solar powered Battery Charge going into my Battery Bank (shwn on BMS); with only 900 Watts of my PV power being taken in by my LV2424s (shows on LV2424s PV page); ... AFTER: (via flipping Solar IN to off, then back on): My Solar Powered Battery Charge Cycle goes up to 65 Amps into Battery; & I immediately am using 2000 watts from my Solar PVs (vs just 900 watts previously being used) ... & half hour later, with more sun, I have a 75 amps charge going into my battery with my LV2424 showing 27.7v battery, and BMS now showing SOC @ 277Ah (again, not convince of accuracy /configured for 280ah full). ... I got about an Hour of Absorption Charge at 65 - 75 Amps before my Charging settled back to Float mode. My battery is currently showing 27.2v at both my LV2424 / and BMS is showing 279Ah SOC, and also 27.2v. :+) ... It would have taken a longer time to get my battery topped up at previous lower 21 Amp float kind of charge if I did not flip my solar IN off and back on !!!

WOW for that kind of discovery for me; who likes using solar watts to charge my battery asp when my PV are producing potential watts. ... I think I am gona automate a daily OFF / then ON cycles to make up for what I currently view as one of MPP's LV2424 shortcomings. I will be interested in your testing of this procedure to see what you think.

Fuller disclosure of my 3 x LV2424 / 240vac split phase set up. I only have two of my 2 LV2424 currently wired to Solar arrays IN; and will retest this discovery when I get my third 2000 watt array set up and wired to my 3rd LV2424 (coming up soon. :+) :+)

I have noticed the same on my system (ms5048). Float is initiated way too early and the amps are throttled. That is why I'm thinking of just charging on bulk but have the BMS control the system. The problem with this is the power spike that occurs when the charger is disconnected when it is pumping can trip the inverter high DC volt disconnect. I have ordered some ultracapacitors and diodes to reduce the impact of the BMS cutting off the charging. Circuit simulation

 

[Capt Bill]

just charging on bulk but have the BMS control the system.

You might have different options on the all in one you mentioned; not sure how I could config for just bulk charging alone. ...Will share from my studies on this forum, and via Will P and other lesson on YouTube, I have bought into idea that my BMS is my #1 back up protection, and also my individual cell v balance helper and indicator; while my Solar Charge Controller and/or All In One is the main battery charge controller that reacts to total battery voltage in way that also protects the battery with proper configuration. ... that being said, I currently have my bulk charging voltage to 28.4, and Float V to 27.4 ... & on some charging cycle, I see the charging cycle go up to full bulk voltage before relaxing back to float voltage. I sometimes have flipped my Solar IN current lines off, then back on mid morning, before fuller PV watts are being directed IN to my Controller/Inverter system, and see it reset from Float charge mode to bulk charge mode; and some other times, like if the battery was low enough last night, the bulk charging happens on its' own.

I have most recently reduced my Bulk Voltage to 28.4v (after testing configs up to 29.2v); partly because of info on how to get more LiFePO4 battery cycle life from a bulk charge setting of 28.2v vs 29.0v ; AND from noticing, for my set up with LV2424s, @ higher than a 28.4v Bulk Charge setting with 60amps or more charging up my 280Ah battery bank, I get one individual cell hitting my 3.65v High v limit, triggering my high v BMS disconnect. ... I do not like the idea of that high v BMS trigger cycling on and off and on etc. when I am not there (seen the possibility in my trails and testing) ... so I picked 28.4v for my bulk charge setting ... thinking that is low enough to prevent those high cell v BMS triggers.

 

[Csimmer]

You might have different options on the all in one you mentioned; not sure how I could config for just bulk charging alone. ...Will share from my studies on this forum, and via Will P and other lesson on YouTube, I have bought into idea that my BMS is my #1 back up protection, and also my individual cell v balance helper and indicator; while my Solar Charge Controller and/or All In One is the main battery charge controller that reacts to total battery voltage in way that also protects the battery with proper configuration. ... that being said, I currently have my bulk charging voltage to 28.4, and Float V to 27.4 ... & on some charging cycle, I see the charging cycle go up to full bulk voltage before relaxing back to float voltage. I sometimes have flipped my Solar IN current lines off, then back on mid morning, before fuller PV watts are being directed IN to my Controller/Inverter system, and see it reset from Float charge mode to bulk charge mode; and some other times, like if the battery was low enough last night, the bulk charging happens on its' own.

I have most recently reduced my Bulk Voltage to 28.4v (after testing configs up to 29.2v); partly because of info on how to get more LiFePO4 battery cycle life from a bulk charge setting of 28.2v vs 29.0v ; AND from noticing, for my set up with LV2424s, @ higher than a 28.4v Bulk Charge setting with 60amps or more charging up my 280Ah battery bank, I get one individual cell hitting my 3.65v High v limit, triggering my high v BMS disconnect. ... I do not like the idea of that high v BMS trigger cycling on and off and on etc. when I am not there (seen the possibility in my trails and testing) ... so I picked 28.4v for my bulk charge setting ... thinking that is low enough to prevent those high cell v BMS triggers.

I think part of the problem is the charging algorithm is designed for AGM so with the flat voltage curve for lithium the float triggers earlier than we want. I wish I could find out the actual algorithm they use but I don't think that's available.

I turn float off by setting bulk time to 900 minutes.

I am also wary of using my BMS to control the charging end point but if the BMS is oversized and if I reduce the shock on the system with my capacitors I hope it will be all good.

 

[jimjones26]

I have read through this thread several times. I have an lv5048 with (4) commercial 24v lifepo4 batteries with a 92.8ah capacity each, wired in series and then parallel.

I have noticed what seems to be a pretty severe imbalance between the 4 packs when charging them. Today I came home and noticed the voltage meters on each battery were pretty far off. One was at 29.4, one was at 28.5, and the other two were at 27.3 and 27.4 respectively. I am assuming the best way to resolve this is wire them all in series and put a charger on them?

I am also noticing that even in full sun, the batteries are getting up to a certain point, they stop charging and voltage starts falling off, and it seems as thought the inverter is ignoring the PV energy available and falling back to using the battery to power the loads.

 

[jasonhc73]

Take a picture.

I have read through this thread several times. I have an lv5048 with (4) commercial 24v lifepo4 batteries with a 92.8ah capacity each, wired in series and then parallel.

I have noticed what seems to be a pretty severe imbalance between the 4 packs when charging them. Today I came home and noticed the voltage meters on each battery were pretty far off. One was at 29.4, one was at 28.5, and the other two were at 27.3 and 27.4 respectively. I am assuming the best way to resolve this is wire them all in series and put a charger on them?

I am also noticing that even in full sun, the batteries are getting up to a certain point, they stop charging and voltage starts falling off, and it seems as thought the inverter is ignoring the PV energy available and falling back to using the battery to power the loads.

Take a picture.

 

[jimjones26]

Take a picture.

Take a picture.

Take a picture of the voltage readings of each battery? They have since fallen back to 26.6v.

 

[Capt Bill]

I have read through this thread several times. I have an lv5048 with (4) commercial 24v lifepo4 batteries with a 92.8ah capacity each, wired in series and then parallel.

I have noticed what seems to be a pretty severe imbalance between the 4 packs when charging them. Today I came home and noticed the voltage meters on each battery were pretty far off. One was at 29.4, one was at 28.5, and the other two were at 27.3 and 27.4 respectively. I am assuming the best way to resolve this is wire them all in series and put a charger on them?

I am also noticing that even in full sun, the batteries are getting up to a certain point, they stop charging and voltage starts falling off, and it seems as thought the inverter is ignoring the PV energy available and falling back to using the battery to power the loads.

Did you Top Balance them before hooking up to lv5048?

 

[jimjones26]

Did you Top Balance them before hooking up to lv5048?

No I did not. They all showed 26.6 v, and back then I thought that meant they were all at the same level. After some research tonight I now realize that is not a valid assumption with lifep04. So to top balance these, my thought was to let the solar charge them up tomorrow until it switches off, then turn the lv5048 off, put the batteries in series, fire up the generator and let my AIMS power top them off.

The question I have though, is should I let the AIMS charger run at 100% charge current?

 

[Capt Bill]

No I did not. They all showed 26.6 v, and back then I thought that meant they were all at the same level. After some research tonight I now realize that is not a valid assumption with lifep04. So to top balance these, my thought was to let the solar charge them up tomorrow until it switches off, then turn the lv5048 off, put the batteries in series, fire up the generator and let my AIMS power top them off.

The question I have though, is should I let the AIMS charger run at 100% charge current?

I would charge em up near full, with a BMS configured to keep any one cell from going over 3.65v (each cell which tend to peak out of balance near full charge); ... then top balance them. .... That might be the fix right there, and if not; IMO: Needed to properly move forward w a diagnosis. ... It will be interesting to hear your story after that.

 

[jimjones26]

These are commercial packs that came from bigbattery. Each 24v battery does have its own bms, the single cell overcharge protection params are:

• Min: 3.72V
• Standard: 3.75V
• Max: 3.78V

I m not sure I have access to change the settings of the BMS, as it is installed in the battery case.

I am guessing that when big battery built these they balanced the individual cells. So as I understand it, each battery is at a different state of charge right now, and I need to get them all fully charged, which is difficult as we are completely off grid. So the idea is to let the sun charge them up until the inverter stops charging them, and then shut the whole system down, rewire the batteries in series, fire up the generator and charge the batteries with my AIMS Power charger. Is this the correct action to take?

 

[Mac6792]

I have read through this thread several times. I have an lv5048 with (4) commercial 24v lifepo4 batteries with a 92.8ah capacity each, wired in series and then parallel.

I have noticed what seems to be a pretty severe imbalance between the 4 packs when charging them. Today I came home and noticed the voltage meters on each battery were pretty far off. One was at 29.4, one was at 28.5, and the other two were at 27.3 and 27.4 respectively. I am assuming the best way to resolve this is wire them all in series and put a charger on them?

I am also noticing that even in full sun, the batteries are getting up to a certain point, they stop charging and voltage starts falling off, and it seems as thought the inverter is ignoring the PV energy available and falling back to using the battery to power the loads.

How are the two 48 volt packs wired to the inverter? Are they connected to a bus bar or directly wired? If not via a busbar then be sure to wire them diagonally across the packs. Positive from one pack and negative from the other. Given that one pack had both batteries at higher voltage I'm guessing you had positive and negative wired to the same 48 volt pack?

Regarding charging.. post your inverter settings. There are lots of nuances to these poorly documented settings and how they interact

 

[jimjones26]

How are the two 48 volt packs wired to the inverter? Are they connected to a bus bar or directly wired? If not via a busbar then be sure to wire them diagonally across the packs. Positive from one pack and negative from the other. Given that one pack had both batteries at higher voltage I'm guessing you had positive and negative wired to the same 48 volt pack?

Regarding charging.. post your inverter settings. There are lots of nuances to these poorly documented settings and how they interact

I have a bus bar. My guess is the batteries were at different rates of discharge when I hooked them up. I did a little but of hillbilly balancing the other day and they are all pretty close now, however I think they would benefit to ire them up in series and let them slow charge for a few days to equalize out, if that will even work with lifep04.

 

[Darren Orange]

I am seeing this same issue on my LV5048. The MPPT seems to do things completely different from the AC charger. I will change the bulk time today and see if that will fix the issue. Changing the bulk/float does briefly get me max output from my system and then it quickly drops off quickly. to stay around 10 amps charging. My batteries can handle 30-40 amps all the way to their max voltage that I have set which is 57V without problem and taper there. It seems like the bulk thing is the problem here mine is currently on auto so I will try changing that to 900 or max out that time.

Does that seem correct to you guys?

More details on what I experienced is here. https://diysolarforum.com/threads/lv5048-solar-charging-confusion.18882/#post-218610

 

[Csimmer]

I am seeing this same issue on my LV5048. The MPPT seems to do things completely different from the AC charger. I will change the bulk time today and see if that will fix the issue. Changing the bulk/float does briefly get me max output from my system and then it quickly drops off quickly. to stay around 10 amps charging. My batteries can handle 30-40 amps all the way to their max voltage that I have set which is 57V without problem and taper there. It seems like the bulk thing is the problem here mine is currently on auto so I will try changing that to 900 or max out that time.

Does that seem correct to you guys?

More details on what I experienced is here. https://diysolarforum.com/threads/lv5048-solar-charging-confusion.18882/#post-218610

I have a few MS5048's and I have the same issue. What I have done is I have put a contactor on the solar input side which is controlled by the SOC relay on a Victron BMV700. It works a charm but I have since discovered there is another solution. Some folks downunder in the AEVA forum have produced new firmware for some MPPs to make them more lithium friendly. The firmware, amongst other things, stops the nuisance going to float to early problem and it also stops the annoying under volt warning for the narrower lithium voltage spread.

 

[Bob Steel]

Cells are not full at 3.4V. Cells are full at 3.65V.

The purpose of top balancing is to be able to fully charge all cells to their peak voltage. Top balancing explicitly means they are only balanced at ONE point in the curve - fully charged.

Don't tell people to do this . Its wrong and will damage their LFP cells. I'm not going to argue it with you . If you don't know why its wrong then research SEI layers. You are acting as a moderator . You need to have it right because most will be scared to argue with you.

No higher than 3.5 volts fellas and in practice 3.4v is a safer level and they are full at that if you look on a discharge graph.
You only go to 3.65v on the very first charge on cells you are sure are brand new.

Here is a graph produced by EVPower that reflects what I believe to be correct . It matches graphs I have produced during testing . Also its available on their website so see for yourself. I think the sellers draw theirs themselves to enhance sales.

You will notice there is nothing above 3.4volts.
This also points out the latent error in trying to estimate the SOC of a LFP cell . It can't accurately be done on a voltage basis . 3.2 volts could be anywhere on a 80% spread.