First Charging and top balancing Lifepo4

[Flod]

So I hooked up my 8 winston cells in parallel (100ah) with the standard copper connectors and put a 100a fuse on the positive side. Then hooked up my 3.65v 5A lithium charger and plugged in. It has now been charging for 14 hrs. The voltage on all cells was 3.303v when starting and now I have 3.350v. What worries me is that I only measure 0.06A going in. I would suspect 5ish A since the charger is 5A. It is still charging though and the LED’s on the charger indicates that it’s not done. I realise that it will take like 6 days to charge the full 800ah/3.65 V bank but why is the current so low?

Any comments?

[sailormed]

The batteries are deciding what they want, not the charger pushing load down to the batteries. Apparently they are close to fully charged.

[Flod]

Ok but at 3.350v ? I would expect closer to 3.65v based on the charging curves I have seen. I am more leaning towards a faulty amp meter

[Flod]

Ok so you can call me stupid. I had the multimeter set to AC. Tried again with DC and now it shows 5.11A. Problem solved!

[Maine Sail]

It is good that it is still in constant current mode..

Your charger however most likely lacks a dedicated voltage sensing circuit, likely has tiny leads, and the charger end can be seeing what it thinks the voltage is at the batteries but there is voltage drop in the wiring so this reading is incorrect and the charger end enters constant voltage mode while the battery end of the circuit is still well below that. Rx = Be patient when it comes up to target voltage.

Charging a large pack like that, 800Ah, with a 5A charger 0.006C), means your cells will be at a high voltage for a longer period of time. Sometimes best to charge them individually then wire in parallel and do a final top balance push. Also that end cell on the left appears to have a bulge...?

[Flod]

It’s this charger: https://www.ev-power.eu/Chargers-6V-...1-3.html?cur=1

It is suppose to stop at the correct voltage but maybe what you say about voltage drop is valid. I’ll measure the current going forward to see when it drops

[Flod]

This one I mean

https://www.ev-power.eu/Chargers-6V-...l-1.html?cur=1

[Cpt Pat]

Originally Posted by Flod

Ok so you can call me stupid. I had the multimeter set to AC. Tried again with DC and now it shows 5.11A. Problem solved!

Please be careful. I can't stress this enough: you cannot determine state of charge (SOG) by measuring the charging voltage alone. At low charging current (like the current you are using) you can seriously overcharge a LiFePO4 battery by applying any voltage over 3.4 volts per cell for too long. See the charging voltage versus SOC curve below.

By the time that massive battery bank charge current begins to taper below your charger maximum output of 5 amps, the batteries will be seriously overcharged. The charge cutoff threshold is 0.05C = 40 amps for an 800 AH bank. You won't be able to detect that threshold with a 5 amp charger. You will charge way past that 40 amp threshold.

Instead, stop the charging process, let the cells rest for a few hours, and measure the terminal voltage. When the resting open-circuit cell voltage reaches 3.35 volts (measured with a very accurate meter), you are done. Stop there!

With the equipment you have, that is the only method available to you for measuring SOC. Charge, rest, measure, charge, rest, measure, stop.

The voltage you report now while charging could be very near the 100% SOC cutoff (again, you can't know precisely by measuring terminal voltage while charging). Please measure the resting voltage now.

Curve below excerpted from http://nordkyndesign.com/practical-characteristics-of-lithium-iron-phosphate-battery-cells/. I strongly recommend reading that entire page.

[john61ct]

If you continue to charge at such a dangerously low current rate

why are you doing that?

you definitely want to get accurate meters and keep them well calibrated.

You have very likely been overcharging, losing lifetime cycles off the back end. Sure maybe not by a lot, not saying you've "damaged" the bank enough to worry about, but stop doing that if you care about longevity.

It would be much safer to get the charge amps up above .1C as a minimum, either by

upgrading the charger to a higher-current version, or

charging at lower Ah units, even one cell at a time up to a common precisely defined point, and

**then** paralleling them, letting them sit overnight, giving a 30min "balancing finish" the next day, then letting them sit again before assembling into your final bank layout.

Cpt Pat's suggestion of using 3.35V **resting** voltage as a working definition of Full in this scenario is a good one.

Also not just the too-low a current, but a charger that lets you tweak adjust the finish voltage would be a lot safer, that 3.65V makes it much harder to avoid overcharging.

[Cpt Pat]

john61ct is correct. You should be charging with at least an 80 amp charger (0.1C) when charging all 800 AH batteries in parallel. That way you can see when the charge current tapers off to 40 amps, which is the generally accepted charge cut-off point.



Beware of the chargers that many LiFePO4 resellers offer. Most a junk.

[Flod]

Ok Thanks for the advice guys. I now stopped the charging and the terminal voltage is 3.366v I will let the bank rest and measure again. The cells was 3.303v when I started charging and I have been charging for 36 hrs. The current was still
5.11A when I stopped.

[Flod]

And the reason I started this from the beginning was this video from
GWL where I bought the cells. https://youtu.be/wY81iMYB29w

[john61ct]

OK, you're doing a commissioning top-balancing protocol.

Not saying it's OK for normal daily cycling, but as a rarely used maintenance / testing routine, you can in such a case, accept charging closer to the

"theoretical / vendor 100%" level SoC

higher than you'd want to use in normal daily cycling, so long as this overall procedure is performed

**as quickly as possible** not letting the cells sit around at that stressful too-high level any longer than necessary.

For others reading in future, OP should have spent the little extra for the higher-amp version 1S-voltage charger, even though it will be so rarely used. The ideal as Cpt Pat suggested would have been 80A, but even 40A would be OK.

Note it does not need to be a x-stage charger, in fact a lab-style variable power supply would be better, perhaps rented as good ones can be very expensive.

If you're checking in regularly, let's see if we can adapt their spec'd protocol a bit to compensate, still getting the cells effectively top-balanced,

in the context of how **you** plan to handle charge-termination in normal daily cycling, but also shortening the process to reduce unnecessary wear on the bank.

How confident are you in the resolution and accuracy of your best available voltage measurement tool? What is it, link?

Same for your ammeter please.

Do you plan in daily charge cycling, to just follow the common path of charging

#1. to the maximum do-not-exceed mfg spec'd profile, or

#2. more gently, in order to get better longevity, lots more cycles in the long run?

The latter involves sacrificing a bit of theoretical capacity off the top, maybe 3-7%, but still likely greater than the nameplate rating.

What is your assembled bank layout going to be, 2P4S for 12V @200Ah, or 8S for 24V @100Ah?

What shore charger will you be using for bulk charging, i.e. at that assembled pack level? Link?

Do you know its default Absorb voltage? and whether or not that can be tweaked, adjusted in precise increments?

What is its current rating and what kind of AC mains circuit do you currently have available?

Finally, assuming the paralleled group has been resting isolated from any loads and from the charger, what is its voltage reading new?

If still over 3.35V, and assuming your DMM is accurate, I'd say the next step for now, should be to

break up the group, wait an hour, and log the individual cell voltages.

For discussion purposes, let's label your four top SoC definitions

A. nameplate Ah rating of the pack (as opposed to actual tested Ah capacity)

B. 3.60 (or 3.65V), stop-charge when current tapers to 0A absorption, "theoretical / vendor 100%" SoC level (may never be used in practice)

C. 3.xxV taper CV stage to 0.03 - 0.05C and Stop, "actual usage 100% Full, for regular testing or BM calibration" (requires at least 0.1C charge rate)

D. 3.xxV and Stop, no-CV "daily usage cycling Full" (depends on charge rate)

[Flod]

The batteries have been resting for 5-6 hours and now read 3,351v so I guess fully charged?
I’m using a brand new “Malmberg” multimeter of Mosel hk588d. For the 12v charging I plan to use an older version of this: https://www.cristec.fr/english/our-p...e-power-units/
It has 3 dc outputs, one goes to my starter battery (still lead) on goes to my bow thruster battery (agm) and the third I plan to use for the LiFeYPo4. My plan was boost of and set the charging voltage to what the LiFeYPo4 needs (don’t remember now). The charger output have a potentiometer so the voltage can be adjusted to just right. Then my plan was to let the bms123 cut the Charing current via a
Relay at 95% SOC and don’t restart charging before like 80% soc again (or similar). I have a master volt sunpanel regulator also and I don’t know what to do with that but basically the same I guess.

[x2y3z4]

Why don't you integrate their BMS?
https://www.ev-power.eu/Battery-Mana...4-0.html?cur=1