diy solar

diy solar

LiFePO4 voltages - new battery

Don B. Cilly

Energetic energy padawan
Joined
Aug 24, 2021
Messages
1,263
Location
Mallorca ES
I need some pointers here. Just installed a new battery. 110Ah, 24V. It's a Pylontech UP 2500.
Not much sun today - well, not constant, charging it at 15-20A - with plunges at around 5-6, when clouds obscure. discharging almost nothing.
SOC, it has 4 LEDs. 3 are lit. According to the manual, it means it's between 51 and 75%. Voltage is 26-6 to 26.9, depending on clouds.
I measured it with three different voltmeters, right at the battery leads. They all agree.

Now, shouldn't it be higher - at that SOC - and above all, constant?
I have no other batteries connected.
 
No - it's LiFePO4. Voltage is not an indicator of state of charge, and a full cell will settle to 3.4 something, at 20 - 80% state of charge somewhere in the 3.3V range or thereabout. Notice the very sharp voltage drop at 100%, and the flat curve over the majority of the capacity.

xLiFePO4-discharging.gif.pagespeed.ic.EQVEXWFoto.png
 
Yes, that's exactly what I was worried about.
But when charging, it would vary with clouds/amps, right?
It now goes to 27.1 (13.55 on 12) at 20+ amps, drops to ~26.5 with clouds.
LEDs are still 3 out of 4...
'saul good, man, right? ;·)
 
Yeah, I wouldnt worry. If you want a good idea of its actual voltage (notice I didnt say state of charge) you can remove all loads and charging, then let the batteries rest. 1 hour is spec, but 15 minutes should give you a decent idea. If you want more accurate SoC you can always add a shunt/coloumb counter.

I would let it charge to full and also full cycle it a few times, as well.
 
I was somewhat worried about my LiFePO4 batteries and their SOC when I first started using them, as well. Then I wrote in these forums about it and many users recommended shunt/coloumb counter/monitor as @DerpsyDoodler suggested. I have been running that device for about a month and I've really learned a lot about how my batteries behave, eg, discharge, charge, SOC, etc. I have it mounted so when I walk by the system I can easily see what's going on.

Sometimes the voltage is seemingly low but the SOC is very high. My only real struggle was calibrating it correctly. But with the instructions and good ol internet searching, I got it sorted... These monitors are great.
 
OK, so. Charge is now almost zero (1.5A). Discharge is more - like 4A... fridge, computer...
LEDs on batteries blinking (means discharge). Still 3 out of 4. Maybe tomorrow -we should get some sun - I can get a full charge.
Voltage is stable at 26.5.
If I see it approach 25.6 (the 12.8 in the graph above) I panic, switch everything off and turn the generator on, right? :)
 
OK, so. Charge is now almost zero (1.5A). Discharge is more - like 4A... fridge, computer...
LEDs on batteries blinking (means discharge). Still 3 out of 4. Maybe tomorrow -we should get some sun - I can get a full charge.
Voltage is stable at 26.5.
If I see it approach 25.6 (the 12.8 in the graph above) I panic, switch everything off and turn the generator on, right? :)
you're good till around 22.5-23v and that's still conservative. yes, a victron smart shunt is a coloumb counter.
 
I would guess the internal BMS protects the cell from over discharge, so I wouldn't worry too much.

Well, I would :·)
For one thing that little baby just set me back €1200 - and as soon as it tests out I'll add another one to it - so you can guess I want to be as careful as I can.

But mainly, so the BMS kicks in, what happens? It disconnects the battery from the system, right?
Everything goes dark... OK, no big deal, but... my all-in-one inverter says, Warning! Do not disconnect the battery without first disconnecting PV input! This might cause Vogon spaceships to... er well.
So what if it disconnects early morning, when the charge is lowest, the panels are beginning to charge but don't quite make up for load yet, and my inverter panics and calls the Vogons? ;·)

Also, "Lithium can be safely discharged almost completely (within BMS parameters) and won't be too grumpy about it".
Yeah well. My new baby. I'll try and avoid that myself if I can :)
 
For one thing that little baby just set me back €1200 - and as soon as it tests out I'll add another one to it - so you can guess I want to be as careful as I can.

Good time to make sure the system works as intended; you'd want to know this early on so if it doesn't, warranty can kick in.

Everything goes dark... OK, no big deal, but... my all-in-one inverter says, Warning! Do not disconnect the battery without first disconnecting PV input!

This is a major issue if this is the case. There will be a time when there is some kind of issue and the BMS kicks in, disabling the output (maybe some dark weather, drained the battery a little too far, ...). If this kind of incident can destroy the inverter, I would get a different inverter and return this one (if you're still in warranty, return policy, ...).

My new baby. I'll try and avoid that myself if I can

In general, this is true. But for a critical system, you want to make sure it works as designed if something happens (which, trust me, it will over time). Better get to the bottom of what happens in these cases early on and within the warranty/return period.
 
Good time to make sure the system works as intended; you'd want to know this early on so if it doesn't, warranty can kick in.

Indeed. I'm trying to do just that. I connected the second battery this morning, they've been charging for some hours, got some sun, they're at 26.9. What slightly worries me (in my ignorance of lithium) is this:

pyloleds.jpg

The LEDs on the bottom one are 3, the top, 2. They've been like that for a while too.
Imbalance? They're paralleled with rather thick cables. Positive is from top battery, negative from bottom.
BMS(s) not calibrated? I haven't managed to get a full charge yet.

[EDIT] Sun came out full, close to ideal angle, charging at 25A, 27.1V, LEDs are the same...
 
Last edited:
Probably normal. If one is at 50% state of charge, and the other at 70%, they can be at the same voltage. Just paralleling doesn't change anything about that, since there is no balance current between them. Now that you are charging them, they should get closer to each other eventually and those LEDs will be more in sync.
 
As upnorth said, voltages might be same but soc different.

As they climb in charge, one will come up in voltage slightly faster, creating resistance and less current flow. the excess flow will channel into the lower voltage paralleled battery.
 
Back
Top