New LiFePO4 installation charge parameters

[Muskoka]

Unplug the BTS aka Battery Temp Sensor...

Per some documentation here on how to charge Lithium, I was using a stated feature. https://www.midniteftp.com/support/kb/faq.php?id=75

"The Classic now has a few new features that can be of use for Lithium Battery Systems.
Lithium batteries generally do not require temperature compensation like Lead-acid batteries do. Setting the mV/degC/cell to the lowest value without flipping the setting to DISABLE is about all we can do right now. This allows the BTS (Battery Temp Sensor) to function and this allows us to set an emergency HIGH BATTERY SHUTDOWN temperature in the LIMITS MENU under the CHARGE Menu in the Classic. When this temperature is sensed the Classic will stop charging to avoid a catastrophe"

I have it set to the lowest value as instructed. Maybe It isn't a good idea to have it on, but I was using a documented feature, maybe I shouldn't use it, as you suggest?

Maybe even at the lowest setting it's driving up the voltage and causing a problem? My battery sits in my main room, on the floor, sitting in the styrofoam it came in, probably about 18 celsius.

Even if it had an effect, it couldn't have been the issue when I ran the first charge cycle and Absorb was set to 28.2v, lowest setting would have rose the voltage a few 10'ths maybe, not enough to cause the issues I'm having?

[boB]

What version is your Classic code ?

The Lithium version isn't released quite yet.  Well, a beta was released here on the forum but it wasn't  good enough we found out.  Also, it worked for one version of the Classic I think but not all Classic voltage versions.

boB

[Muskoka]

Classic Monitor app.



Sounds like I should remove it to test the next charge cycle, easy enough to do.

[ClassicCrazy]

Here is what I recommend for your battery settings - and I am giving you the 48v versions so dived by 2 for your 24v settings.
55.2 for the Absorb ( you could go up to 56v but cells are almost full below that)
53.6 to 54v for Float
You don't need to use end amps
set the absorb for 15 minutes to half an hour to watch what is going on.
If you see the current drop to near zero when in absorb ( via Whizbang system amps) then you know the battery cells are full . If they are still taking in power then make the absorb time longer.
If in Float and you have loads on when there is plenty of power in from pv and you see a negative current on the Whizbang system amps then you need to raise the float value - start at 53.6 and go up as needed to 54v.
The Classic sometimes will go to absorb below the Absorb voltage setting - at least it has done that with mine . So you need to tweak the Absorb and Float to suit your needs.
Also you need to use the voltage offset tweak for the voltage drop in the cables. Do that when the voltage is highest which will be your absorb voltage. When you do that you will see the Float voltage may not be accurate for where the battery voltage is . That is why you need to tweak all the settings until you get them just right .
With lead acid it didn't make much difference but with Lithium it does. You also may notice that when it goes from Absorb to Float and show resting -  your batteries will start discharging even though there is power available. That is because the voltage on lithium stays up high for awhile. It will probably go to Float and provide power to the loads again once the SOC drops to around 99 percent from 100 . If you don't have much of any loads on the system this could take awhile.
I didn't study all your batteries parameters or try to figure out how your bms is working - I am just telling you that these general settings are right for most LiFePo4 cells. The only consideration is the balancing your bms may do - but if it is a passive balancer it doesn't shunt much power from the high cells and if you don't have any software or bluetooth that shows what each cell is actually at during Absorb you won't really ever know how well balanced they are. But at the 55.2v it won't matter too much if they are way out of whack anyway.
Larry

[Muskoka]

Larry, thank you very much for the detailed post, I will do as you suggest. Unfortunately I have no access to the BMS, it might have made things easier if I could see what was going on inside that battery, other than a  voltage reading.

All set, ready for next charge cycle. I did disconnect the BTS.

I posted these in another post, will do so again for reference. These values came from the battery seller.

High voltage cut-off: 31.2V
Charging limited voltage: 29.2V
Balance charging: 29V
Boost charging: 28.8V
Float charge: 27.6V
Boost charge return voltage: 25.6V
Overcharge return voltage: 23.6V
Low voltage warning: 23.6V
Over-discharge voltage: 22.4V
Discharge limited voltage: 21.6V
Over-discharge delay time: 10s
Balance charging time: /
Boost charging time: 240 mins
Balance charging interval: /

[ClassicCrazy]

Larry, thank you very much for the detailed post, I will do as you suggest. Unfortunately I have no access to the BMS, it might have made things easier if I could see what was going on inside that battery, other than a  voltage reading.

All set, ready for next charge cycle. I did disconnect the BTS.

I posted these in another post, will do so again for reference. These values came from the battery seller.

High voltage cut-off: 31.2V
Charging limited voltage: 29.2V
Balance charging: 29V
Boost charging: 28.8V
Float charge: 27.6V
Boost charge return voltage: 25.6V
Overcharge return voltage: 23.6V
Low voltage warning: 23.6V
Over-discharge voltage: 22.4V
Discharge limited voltage: 21.6V
Over-discharge delay time: 10s
Balance charging time: /
Boost charging time: 240 mins
Balance charging interval: /

Some of those seem way off

High voltage  off should be 3.65v per cell so 29.2 v would be highest they should ever go.
 
Most of that info is for newer  inverters/ chargers that use all that terminology and don't apply to the Classic.
We have no idea what that bms is set up for so stick with something sensible. Read some manuals from other lithium batteries that are similar and see what they suggest for theirs.
for example https://kilovault.com/wp-content/uploads/2023/03/KiloVault-HLX-User-Manual-2023-02-08.pdf

I would suggest the 27.6v for Absorb ( you could go up to 28v if you want to )
Float at 26.8  ( and go higher as I said before if needed )
Absolute lowest cell voltage is 2.5v  (20v) but I would go  higher so their recommended 22.4v is good for inverter shut off to protect battery harm.

I watched lots of Off Grid Garage youtube videos to start to understand all the info about lithium lifepo4 cells. This is for single cell so multiply by 8 for your 24v values.
https://youtu.be/hLf1FdRjlaw?si=r_T4e4JR2LlN2_WU
Larry

[Muskoka]

Thank you again for the feedback, I have been watching Off Grid Garage.👍

[Muskoka]

Got a full charge cycle today, all went well. No tripping the inverter, no crazy voltage spikes.

Have Absorb set to 27.6v, absorb time 30 mins, Float 26.8v, Rebulk 26.4v, end amps set to 0.

Thank you for all the help and feedback, much appreciated.

[Muskoka]

Another sun day, after about 10am. Thought I'd try a test, and bumped up the Absorb set point to see what would happen.

Well, it didn't work out too well. Yesterday I had Absorb set at 28.0v and everything was fine, other than messing with differning voltage readings, 2/10's, as discussed in my other thread. Classic went from Bulk to Absorb, then to Float, when it was supposed to.

So today I thought I'd try 28.4v for Absorb. Everything was going fine, it got to Absorb, soc was getting close to 100%, then everything started going haywaire again. Voltage spikes, tripped my inverter which is fixed to 30.5v, I can't change it. The Classic started the cycling again of the charge stages. So, I disconnected the panels, rebooted the Classic, set Absorb back to 28.0v. turned the panel breakers back on, and it was fine. Happily sitting in Float now.

What does that sound like to you guys, a BMS issue? I had my meter on the battery terminals when it was in the Absorb cycle and it matched the Classic cause I used the Tweak menu to match things up. It was only out 1/10 nearing the end of Absorb, which I corrected using Tweak menu.

Battery seller is telling me the battery needs 28.8v to 29.2v to balance, and I'm assuming get to 100%. Well, I'm nowhere near that when set at 28.0v. If that extends the life of the battery fine, but why is it being so problematic when the voltages are set to where they are telling me they should be.

[boB]

Do you have any loads turning on and off ?

May have already chatted about this one ?

boB

[Muskoka]

Internet router and inverter, that's it, 50-55w.

[Muskoka]

Think I'll try tomorrow without the end amps. The other day it took 18 mins for the current to drop below the 4 amps I had set for end amps while in Absorb, that was when I had Absorb set to 28.0v. I'll set Absorb to 20 mins maybe, no end amps, and see what happens. Still going to use the 28.4v Absorb setting though, need to know whats causing the issue with raising the Absorb above 28.0v.

[boB]

Tell me if you can...  How long are the spikes above your set point voltage up into the 30 V range ?

[Muskoka]

A second or less, it "flashes" on the MNGP display, and is gone. It doesn't stay there.

So fast they dont seem to get picked up by My Midnite either, the voltage spikes don't show on the graph, but I definitely see them on the MNGP.

Again, it's only happened when Absorb was set to 28.2v and 28.4v, everything worked fine when Absorb was set to 28.0v.

And while it's happening the voltage is all over the place, in the low 30's, down around 27, back and forth. The mode is also cycling, Bulk, Absorb, Float, it all happens in a second, the cycling, but would likely go on for minutes if I didn't intervene.

Should add, the current is also jumping around while this is going on, 45a, down to 0, back up again. Seems to repeat itself like everything else.

Like I've mentioned before, it's like the controller doesn't know what it's supposed to be doing, and starts jumping around, if that make sense.

That's why I was thinking maybe getting rid of the end amps would help, maybe. One less requirement that had to be met. Then again, it was all fine at 28.0v Absorb, so who knows.

[boB]

Do the spikes happen when the voltage is near Absorb and the current is near Ending amps ?

Again, we may have chatted about this but maybe once question at a time will help.

boB