LiFe Batteries voltage calibration

[A J Coop]

Hello Forum,
Over the past two years I have had two 12V 175ah (12V 350Ah) LiFe batteries, not connected to the KID, using them & then charging via AC.
I have been building a battery box, allowing time for my existing AGMs to show expiry (still kicking) ...
I have now competed the box and have the KID neatly installed in/on the box.
Doing my due diligence I have been doing a `lots of reading` bit of a headache, taking in the advice and from,
ClassicCrazy and boB.
https://midniteftp.com/forum/index.php?topic=5875.0
Thank you both for your input very helpful.
Over the past few days, I have been putting all suggestions, inputs, advice from MidNite support page and the forum itself!
I have come across a dilemma when it comes to the voltage calibration between the KID and the LiFe batts!!

Rest calibration.
I noted that there was a -0.1*ish difference in the kID and DMM (CALIBRATED) at REST (Solar off in the menu).
Enabling the solar function I started to observe the changes from BULK, (stayed about the same until) ABSORB when the voltage deviated -0.4*ish, allowing the KID to finish, just went to float way too early. *Other factors may have put the KID in float, Absorb time and/or END AMPS.
Absorb Calibration.
A few days of use I tried again this time adjusting the absorb time and END AMPS.
Worked like a charm the KID started BULKING and when ABSORB came about I calibrated as suggested, my -0.4 at the time, after some nail biting, The KID went to float with a 3+ Amp trickle, a short time after that the BMS stopped the current and just gave 0.1A every now and then.
When the sun went behind the hills the KID was float and 0.4ish Volts too high!
Fearing FET burn out or other issues I turned off the Solar in the menu!
I have retuned the KID to the -0.1*ish calibration at the rest voltage.

? Calibrate the KID at the absorb and up the float voltage so the kid can rest!
I fear damaging the FETs! Or something else!!
? Or Manuely switch off the PV!!
I am wonky on how to treat this voltage deviation especially when it comes to REST.
Thank you 

A J
Oh, there was a load, RPi4 and a fan for the KID keeping it KOOL (0.5A total).

[ClassicCrazy]

There are two issues - sometimes the Kid has been reported to not work quite right with Lithium batteries. I experienced this once but that was rare occasion when I was using it to charge someone else's new lithium battery . I don't remember what it was doing - not going to absorb right time ? But I remember someone else had that issue too with a Kid . I didn't spend time messing around with it so don't know if I could have tweaked something. There was a discussion that Mario was going to work on a firmware update for the Kid to fix that lithium issue but pretty sure nothing came of that.
The other issue is what the bms in your battery wants to see for a full voltage for it to start to balance the cells. That might all be unknown . You can charge the batteries until you see the current drop on them, and sounds like you have done this to find an ending amps value that you say works.
But yes, you need to do the voltage offset to compensate for when the batteries are charging at their highest point. But then the voltage will be off when it is done with Absorb , but all you really need to do is set the Float point somewhere that is effective for your batteries. When the batteries are in Absorb and the current drops , they are full, then you drop to float which is a point where the cells don't charge or discharge - but still high enough voltage so the Kid can supply the loads. This is usually 53.6 to 54 volt range for 48v ( divide by 4 to get the 12v value) I only charge to 55.2v ( 3.45v per cell)  but some people charge all the way up to 58.4 volts which is the very top ( 3.65v per cell) .
I guess if you had success with your other charger , just set it for whatever that was doing.
I am glad I have a bms that shows the values in it and also lets me change them.
I go by what Andy from Off Grid Garage recommends because he has done a lot of testing and explains his conclusions showing the graphs of charge and discharge. https://off-grid-garage.com/my-settings/
In a way as long as you charge your batteries 99% full , and give them enough time to balance if needed, you are okay . The only solution to if you don't have any bms monitoring is to either replace what you have with a new bms that will show you all the details via bluetooth app and has active balancing ( like the JKbms) or just don't worry about all that stuff and use them , and replace in 5 or 10 years or more if they ever fail. Price on lithium batteries has gone way down recently . You also have the option to replace your Kid with something else if needed - there are some smaller mppt controllers out there that work well with lithiums and have really good bluetooth apps. Really depends on what kind of pv array you have if that makes sense or not for what you need .
Larry 

[A J Coop]

Thank you, Larry, AGAIN 
When I got these batts there was Bluetooth option the cost pushed me away, I can only KICK myself for that, the RPi is all I have, ALL IS UNKNOWN. I will look at Andys Off Grid Garage later for more info thank you again.
The AC charger did play a major role,  .
I did read a post by Chris, LiFe Charging issues, took it on board.
I just realised 13.2 add 0.4 calibration deviation is 13.6, not to mention I did recalibrate
To REST the KID failed anyway.
The Stack data `Reason for rest` went from 17 (I think) to 5, I have NO idea worth a mention!
I have made some boo boos, WELL I can only learn from them and move on
Finding the sweet spot in float,
As of now Float, 13.4*(adding 0.4 Absorb DEV 13.8 ) was 13.2*(13.6 with deviation) not to confuse myself between Rest* and Absorb calibration results.
It has been over a day and its 13.4 on the DMM 13.4 on the KID (Rest Calibrated -0.1V) with little taken out, the RPi booted 2 or 3 times to log some results.
During the week I have the chance to use the batts and drag the SoC down to 95 ~ 90% recalibrate and leave the calibration ALONE.
The limited info I got from the online shop,
Bulk  14.4 - 14.7 Volts
Float 13.6 - 13.8 Volts
prismatic cell config and other irrelevant info.
I have attached a DoD chart if any hep to anyone... Or to pick it to pieces whatever...



I will update pass or fail. Until then
Thank you

A J

[ClassicCrazy]

Use the Whizbang on your Kid and that would be a better SOC indicator since you can't use that voltage chart unless the batteries were at absolute rest and then maybe not such a good indicator. You take the voltage at the batteries and compare to what the Kid says on display when you calibrate the offset.

[A J Coop]

Oh no, no, Larry that is not for me, I slowly learned from reading up!!
I do have another one for you though   , Lithium SoC under load  .
I post for `others`, `if any help to anyone... Or to pick it to pieces whatever...`
Now the UPDATE:
I have calibrated the KID and ALL seems good.
I am on the next cycle and recheck the KIDs behavior.
The DMM battery is now flat, years later, have a new batt for the recheck...
I still have a bit of a problem with going to rest, lots of V and no A (WhizBang showing -A)! sun behind the hills, I may have to put up with that.
Last but not least,
I had a good look at Andys Off Grid Garage. Well Eye POPPER, that was a lot to take in .
I also forgot to answer you about my PV array!
I have a single LG315N1C-G4 PV (315 Watts), I did pay close attention to the selection, and it does the job on sunny days with a nice, respectable load, I adjust my loads accordingly to make sure the Batts charge. Now I have changed batt-Chem I will be using the same respect to charge the batts, although Lithium is not bothered like AGMs.
Thank you, Larry, I am grateful for your help.

[ClassicCrazy]

The lithium will usually drop a bit , like to 99% SOC on the whizbang , before you see the Kid picking up the loads.
So full , it will go from Absorb 100% ( I would use 13.8 absorb at battery )  to Float very quickly , and then take some power from the battery until it gets to 99% and then the Kid should pick up the load. If not raise the float voltage up some. What I use on mine ( I am dividing my 48v settings by 4 to get 12v for you ) is 13.4 but raise it up to 13.5 or whatever you need to to get the Kid to supply the loads once the SOC is 99%. If you did your voltage compensation right you set that for the absorb point , and then the float point will be off because of the calibration so you just need to put in whatever voltage for float that works .
What you are trying to do is just have the lithium not charge or discharge in the float stage if extra pv is available for the loads.
Hope that makes sense.
Larry

[A J Coop]

I scrapped what the shop suggested I charge them at.
The Batt terminals are very well protected, extremely difficult to probe, I will attempt next cycle fingers X.
I test at the Shunt and Pos Busbar, about ~1/2 way between The KID and batts, very short to short (? how one looks at it!) run of cables.

Knowing that this change would affect the Absorb time I upped it.
  Very happy, the KIDs performance, well I do have to tweak the Absorb time again, floated early, I seen a pattern for each SoC %, so that's not going to need tissue.

REST, oh yes   that is it, I was on the phone when the relay switched off then on for a bit, off/on a few more times then, RESTING, my Cuz on the other end had to tell me to settle down I was that excited.
*No double tapping/latching relay, it has been referred to as woodpecker! in the forum.

At this point the loads while Charging/Float is negligible just the RPi & cooling fan.
When this horrific weather clears up (if), I will be going on camping/road trips around the state. Then the load's vs charging/float will be managed closely.
Still the loads will not be that demanding, I do have a 12V Travel Oven, bit of a Guzzler 10.8, 11A Max (well within the PV), not like a coffee machine, Induction cooker, Hair dryer...
Although Coffee and Induction sounds good.
If I add up all `at once` I`m looking around 250W, hence the 315W PV.
The oven will not be on all day! Neither will be lighting. The load to charging/float ratio is important to me!

Larry,
THANK YOU, for guiding me INVALUEBLE.
My Grandmother would say your blood`s worth bottling .
The RPi will monitor for what you have described. Not forgetting the deviation.
The goal is with or without loads, charging/charged the PV is the sole provider, managed closely, until the obvious! If Weather, clouds... prevent float then tuff.
With these LiFe`s I am not that concerned, there is plenty of Ah there. The most the LG has given me to date is 175Ah (the old AGMs) in a day, that`s 50% of the LiFe`s capacity, 1 Batt, I have 2, thus far not even close to that DoD.
I hope I make sense. I will advise after my next cycle. It`s not over yet...
All is not lost when it comes to The KID and LiFe chemistry.