LiFePo4 monitoring

Started by Offgridiot, February 20, 2024, 01:54:30 PM

Previous topic - Next topic

ClassicCrazy

Watch Off Grid Garage video where he did some battery capacity discharge testing at various voltages . We are talking about LiFePo4 cells.
The conclusion is that you can fully charge but it takes longer at lower voltages. So not very practical for solar where we have a limited charge time .
The most practical charge point for absorb is 55.2v which gets them 95% full at the bottom of the steep curve. That is 3.45v per cell. You will also notice the current they will accept around this point, and since it is quite a bit lower than 3.65v max - you don't have to worry much about over charging a cell.
The bms settings are best if you can pick the point they start balancing. From off grid garage testing point of view it doesn't make much sense to start balancing below 3.45v per cell. If you have a bms without programmable parameters you can't adjust this.
My bms I have set to start balancing 3.45v and 20mv differential. It has to have both of these and you will only see it at the top charge. I absorb time them for about 15 to 30 minutes which I have determined by watching my battery cells and their needs. My monitoring software watches all 16 cells in each battery. I don't use ending amps.
The Float point really is just finding the point where the batteries are not charging but the controller will kick in to supply all the loads. This is around between 53.6 to 54 v. Since the Classic doesn't have voltage sense you just have to watch how it sees the voltage at the point you want it - because it won't be accurate at the float range if you used the tweaks to offset the voltage . The offset is done for the top of the absorb charge point.
Passive bms usually only balance by shunting some top voltage into resistor and more like 70 to 150 mv from what I have read. My bms is in that range . So what I do is I can adjust the point where the cell goes into over volt protection so I set it a bit above the 3.45v . That way it will turn off charging to cells so the small amount of current shunted off has a better chance at balancing. It seems to be working well for my situation.
One battery I installed an active balancer and that is a different story - it will balance cells fast with 4 amps . Unlike the passive balance that does any and most unbalanced cells at once, the active balancer does one cell at a time , moving current from the highest cell to the lowest cell. It gets them in balance much faster.
I think in reality it doesn't matter too much at the lower 3.45v range if the cells are more out of balance because you are staying away from the top critical points and are in the range where you will be getting most of the power out of your cells anyway.
I have notice the balance getting better on my batteries - it takes a long time to do it with passive balancer of small shunt current. Also it isn't worth in my opinion being obsessed with some perfect balance differential - under 80 to 100mv differential at the 3.45v point is probably good enough .
 You can of course do more controlled charging to a higher voltage point near maximum 3.65v with controlled current of small amount with passive balancer , or using active balancer , or doing it manually on each cell.
Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

ClassicCrazy

In this you can see the balance . Battery 2 has active balancer and battery 1 is passive.
Both of the batteries drop their current to near zero as they fill up at 55.2v which is 3.45v.
I think I had 35 minute absorb time and then you see it drop to float voltage.
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

mahendra

Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMWatch Off Grid Garage video where he did some battery capacity discharge testing at various voltages . We are talking about LiFePo4 cells.
The conclusion is that you can fully charge but it takes longer at lower voltages. So not very practical for solar where we have a limited charge time .
The most practical charge point for absorb is 55.2v which gets them 95% full at the bottom of the steep curve. That is 3.45v per cell. You will also notice the current they will accept around this point, and since it is quite a bit lower than 3.65v max - you don't have to worry much about over charging a cell.
The bms settings are best if you can pick the point they start balancing. From off grid garage testing point of view it doesn't make much sense to start balancing below 3.45v per cell. If you have a bms without programmable parameters you can't adjust this.
My bms I have set to start balancing 3.45v and 20mv differential. It has to have both of these and you will only see it at the top charge. I absorb time them for about 15 to 30 minutes which I have determined by watching my battery cells and their needs. My monitoring software watches all 16 cells in each battery. I don't use ending amps.
The Float point really is just finding the point where the batteries are not charging but the controller will kick in to supply all the loads. This is around between 53.6 to 54 v. Since the Classic doesn't have voltage sense you just have to watch how it sees the voltage at the point you want it - because it won't be accurate at the float range if you used the tweaks to offset the voltage . The offset is done for the top of the absorb charge point.
Passive bms usually only balance by shunting some top voltage into resistor and more like 70 to 150 mv from what I have read. My bms is in that range . So what I do is I can adjust the point where the cell goes into over volt protection so I set it a bit above the 3.45v . That way it will turn off charging to cells so the small amount of current shunted off has a better chance at balancing. It seems to be working well for my situation.
One battery I installed an active balancer and that is a different story - it will balance cells fast with 4 amps . Unlike the passive balance that does any and most unbalanced cells at once, the active balancer does one cell at a time , moving current from the highest cell to the lowest cell. It gets them in balance much faster.
I think in reality it doesn't matter too much at the lower 3.45v range if the cells are more out of balance because you are staying away from the top critical points and are in the range where you will be getting most of the power out of your cells anyway.
I have notice the balance getting better on my batteries - it takes a long time to do it with passive balancer of small shunt current. Also it isn't worth in my opinion being obsessed with some perfect balance differential - under 80 to 100mv differential at the 3.45v point is probably good enough .
 You can of course do more controlled charging to a higher voltage point near maximum 3.65v with controlled current of small amount with passive balancer , or using active balancer , or doing it manually on each cell.
Larry

Larry ! I could not have said it better !
1.5kw on Midnite classic 150(whizbang jr.) networked 0.660kw on classic lite 200 wbjr ,180ah CALB Lifepo4 48v battery bank,jkbms 150amps bms(top balanced) Outback vfx3648.
Soon to be added Sol Ark 8k with 5Dyness Bx51000 ,7.8kw pv

boB

Quote from: ClassicCrazy on February 24, 2024, 01:06:19 PMIn this you can see the balance . Battery 2 has active balancer and battery 1 is passive.
Both of the batteries drop their current to near zero as they fill up at 55.2v which is 3.45v.
I think I had 35 minute absorb time and then you see it drop to float voltage.

WoW !

K7IQ 🌛  He/She/Me

Offgridiot


Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMWatch Off Grid Garage video where he did some battery capacity discharge testing at various voltages .
He has a LOT of videos. I watched one where he was discharge testing some cells but without a focus on different voltages. I'll keep looking. In fact, I went to his oldest video and started watching in order (but skipped a bunch I thought were completely irrelevant). Should keep me busy for a minute....

Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMWe are talking about LiFePo4 cells.
Sorry, I'm still not sure of some of the vernacular. Do you mean raw (?) cells as opposed to ready made units like Battle Born (or Power Queen, like mine) or server rack? Or LiFePo4 cells as opposed to lead acid batteries?
I made the (questionable/regrettable?) decision to buy 8 Power Queen 12.8V 300Ah units for my 24 volt system (mostly because I didn't (don't) have the confidence/knowledge/time to learn to build a bank out of raw cells and choose BMSs and all that), and the BMSs in my units don't have anything in the way of individual cell monitoring accessible to me. So balancing for me is going to be limited to placing a balancer (I think the Vicrton unit) between the two halves of my series-connected bank. I'm interested in eventually learning about the ins and outs of managing an open rack of cells. I figure that when my 5 year warranty runs out, I can open up my units and test individual cell health, and maybe add a Bluetooth BMS setup. Hopefully by then I'll have gained a better grasp of what's going on with such things.
The process for me of typing out more questions has, by itself led to little incremental gains of understanding. Type a followup question to you, answer part of it myself, Google more ideas, erase followup question to you. 
I guess my brain is in the bulk stage and a little unbalanced. I feel like I'm at or near the point where I'm about to transition to absorbing the info you've passed along. I found a few more Off Grid Garage videos that talk about this stuff and it's helped quite a bit (I guess having the visuals along with the explanations bumped up the experience?). Thank you for thinking of pointing me toward them.


The conclusion is that you can fully charge but it takes longer at lower voltages. So not very practical for solar where we have a limited charge time .
Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMThe most practical charge point for absorb is 55.2v which gets them 95% full at the bottom of the steep curve. That is 3.45v per cell. You will also notice the current they will accept around this point, and since it is quite a bit lower than 3.65v max - you don't have to worry much about over charging a cell.


So with my 12.8V units in a 24V system, I'm taking my time extrapolating your help for my situation.
The math is easy enough. Your charge point of 55.2v means 27.6v for me. The concerning point for me is that that also means charging my units at well below the manufacturer recommendation 14.4v-14.6v. I guess I need to reach out to Power Queen to find out about how their BMS handles balancing the cells within their units. Would 3.45v (or lower) even trigger their (internal) balancer? My paranoid tendencies lead me to suspect that maybe their motivations are that they would love me to overcharge them and burn them out shortly after the 5 year warranty runs out, instead of treating them in a manner that would help them last 20 years or more.
As far as worrying about overcharging any cells, I now can see where that could happen if I follow the battery manufacturer's guidelines with the settings in my Classics and/or my inverter/charger. My thoughts are that I should generally follow the advice you here (and Off Grid Garage) are presenting, and every now and then (every 6 months?), either crank up the charging parameters to closer to the manufacturers recommendations for a day or a week, or take the batteries offline (when they are close to fully charged) and finish them off running my generator with the two 40 amp chargers I bought from the company (Power Queen) along with the batteries. These plug-in chargers shut off completely when the batteries are full. Seems like a bit of a PITA but shouldn't take excessively long.

Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMThe bms settings are best if you can pick the point they start balancing. From off grid garage testing point of view it doesn't make much sense to start balancing below 3.45v per cell. If you have a bms without programmable parameters you can't adjust this. My bms I have set to start balancing 3.45v and 20mv differential. It has to have both of these and you will only see it at the top charge. I absorb time them for about 15 to 30 minutes which I have determined by watching my battery cells and their needs. My monitoring software watches all 16 cells in each battery. I don't use ending amps.

Yeah obviously I can't adjust this.  :(


Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMThe Float point really is just finding the point where the batteries are not charging but the controller will kick in to supply all the loads. This is around between 53.6 to 54 v.

The small amount of info that came with my batteries doesn't refer to a float stage at all. Just bulk and absorb (both @ 14.4-14.6V). It does state a parameter for what it calls "tail current" (6Amps (0.02C)) which is what I'm assuming everyone else refers to as "ending amps". I guess I'm not really sure what the difference even is between float charging and ending amps. Now I'm thinking that ending amps isn't the amps that are applied near the end of the charging cycle but rather, the (adjustable) point at which the absorb charging basically ends. For some reason, I had the former stuck in my head.
Do you choose not to use ending amps simply because you charge at such a low voltage?


Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMSince the Classic doesn't have voltage sense

As in, it can't distinguish individual cell SOC?

Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMyou just have to watch how it sees the voltage at the point you want it - because it won't be accurate at the float range if you used the tweaks to offset the voltage . The offset is done for the top of the absorb charge point.

Is this relevant to my (non-open-cell bank) situation? Is "offset" a parameter that can be manipulated with the balance portion of your 'controllable BMS'?


Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMOne battery I installed an active balancer and that is a different story - it will balance cells fast with 4 amps . Unlike the passive balance that does any and most unbalanced cells at once, the active balancer does one cell at a time , moving current from the highest cell to the lowest cell. It gets them in balance much faster.

So, on another forum, I had been discussing an alternative to the Victron balancer. Again, the balancer I'm talking about is not for balancing individual cells but rather, balancing two halves of a series-connected 12.8V battery bank for a 24V system. The Victron balancer burns (as heat in a resistor?) around 1 amp from the higher side, only after the battery system increases to more than 27V. Victron doesn't say specifically that it's a passive balancer, but can you confirm whether or not (from my description) that it is? The alternative (ANGUI (brand) HC01 active balancer, found on Amazon.com) transfers up to 5 amps from the high side to the low side if the difference is more than 20mV. Apart from the reputations of Victron (solid) and ANGUI (who?), where do you lean on these choices for my situation? The 1 amp of the Victron unit seems like it would take forever to make an impact on four 300AH 12.8V batteries if they wandered a bit higher than the other four.


Quote from: ClassicCrazy on February 24, 2024, 10:45:38 AMI think in reality it doesn't matter too much at the lower 3.45v range if the cells are more out of balance because you are staying away from the top critical points and are in the range where you will be getting most of the power out of your cells anyway.
I have notice the balance getting better on my batteries - it takes a long time to do it with passive balancer of small shunt current. Also it isn't worth in my opinion being obsessed with some perfect balance differential - under 80 to 100mv differential at the 3.45v point is probably good enough .
You can of course do more controlled charging to a higher voltage point near maximum 3.65v with controlled current of small amount with passive balancer , or using active balancer , or doing it manually on each cell.

This alleviates so much of my stress! I can't thank you enough. Trying to sort through the opinions, and the actual importance of many of the details is a thing. I obviously care foremost about the safety factors (....overcharging leading to a fire?), and then system health (....but there's varying degrees of that concern. Overcharging leading to the battery bank failing within 1 year? OOF! Overcharging leading to it failing within 18 instead of 22 years? Meh.). You and the other helpful guys are to be commended for your input here. You're why the internet was invented.
I feel less like I'm sinking, and more like it's sinking in. Slowly, but....
P.S. I hope this (very long) post to your excellent response doesn't scare you off.  :-X
Now, back to the cat videos! LOL

mahendra

#20
You are worrying too much .Power Queen already have BMS .
The seem to be doing a decent job as well so not to worry.
I have been fondling with redodo,chins ,eco worthy ,some other brands,I can't remember and Ampere time they all seem to do ok to good not excellent like the top tier brands that you mentioned but the work ,Some are better than others though.

Your focus should be following the manufacturer's charge and discharge parameters .

Some word of comfort; I saw calb cells gasses on more than one occasions because a guy was too stubborn to put a bms on his pack and misinterpreted the manufacturers charge profiles and recommendation ; He charged his pack to 3.65v per cell every time with no bms ,no balancer and no top or bottom balancing .

It took two years for the overcharge cells to gas and got swollen.He didn't have a fire as yet but I trust that he may eventually.He just replaced the damaged cells and continue business as normal .

Off grid garage seems to be the lifepo4 Guru,that guys does a lot .You have been to the right place .
1.5kw on Midnite classic 150(whizbang jr.) networked 0.660kw on classic lite 200 wbjr ,180ah CALB Lifepo4 48v battery bank,jkbms 150amps bms(top balanced) Outback vfx3648.
Soon to be added Sol Ark 8k with 5Dyness Bx51000 ,7.8kw pv

ClassicCrazy

#21
yes you are going to have to work with the bms that you have.
LifePo4 I meant that particular lithium chemistry as other lithium chemistries have different charge parameters.
You don't need to worry so much about balancing if you are using the power between maybe 35 and 90 percent because the cells won't show much imbalance at those levels. It is only when you try to squeeze out all the power that you can from the battery that balanced cells makes a difference. Since you don't have a bms that shows cell voltages you won't really know what is going on inside.
I don't remember  your original concerns. As I may have said before - you charge to your setpoint where you want - then cells are full. After that , you go to float voltage where it lets the controller supply loads but won't keep charging the battery. That is between 53.6 and 54v for Classic.
They recommend higher charge voltage because it charges the batteries cells faster. If you don't need to do that fast charge - it will charge lower voltage for longer time to get them full.
You are right about Off Grid Garage having too many videos - I just looked and can't even find right now the one I was recommending.
But I found this one that might help
https://youtu.be/xBu7ScAdKrg?si=2vB_ykI2WzYInZcM
So in a nutshell
You have the batteries and bms you have. You don't have to worry about the balance so much unless you are trying to squeeze every amp possible out the battery. If you stay lower than the suggested top charge point you will still be getting up to 90% out of them.
It took me a long time to understand how  lithium LiFePo4 behave.
I spent a lot of time watching the actual cells voltages in my batteries. You can't do that so have to just be happy with if you get the power you need out of them.
I worried a lot about the cells being balanced when I got my batteries and didn't understand it all and made videos and posts on forums about it. Now I understand how it works and wouldn't have worried so much.
Here is another video with load and charge test graphs - I started it in middle of video but watch the whole thing if you want
https://www.youtube.com/watch?v=FjZEjmJiqTo&t=1298s
Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable