Float charge voltage for LiIon cells. (Solved)

[Jim0000]

I am planning to wire up a LM1084 regulator and adjust it to float charge a single LiIon cell from a small 12 volt solar panel.
This is intended to replace the original 3 AA cells that power an automatic door opener/closer on a chicken pen.

Question:

Is 4.1 volts a suitable potential to keep this single LiIon cell charged?

I have yet to swap over the batteries to test the 4.1 - 4.2 volt fully charged LiIon cell to see if it would provide sufficient power for the tiny motor driving this auto door., but I am confident that it will.
I have done a similar swap of battery power to electric scissors that I use for fabric cutting.


Jim.

[thm_w]

LM1084 seems a bit overkill, what will be limiting the current into the cell? Or is the solar panel small enough its not an issue?
4.1V is ok, 4V will increase lifespan a bit longer.

I would rather use a buck regulator and standard TP4056 linear charger, or a buck lithium charger.

https://www.eevblog.com/forum/projects/float-charge-voltage-for-liion-cells/
https://www.eevblog.com/forum/chat/lipo-battery-on-holdingfloat-charge/
https://www.aliexpress.us/item/2255800978401541.html

[NiHaoMike]

4.1V/cell should be fine given it's not applied 24/7. But most likely you'll be able to get by with less.

[Jim0000]

Thanks for the replies.
The buck converter option is a good tip. However, I have yet to find a 4.1 volt, or 4 volt one. I will keep looking.
And yes, the lower voltage 4 volts, will likely do the job and provide longer cell life.

Jim.

[IanB]

Anecdotal data point number 1:

I left an unused phone plugged permanently into the charger for several weeks so I wouldn't have to remember to keep recharging it. After a while I found the battery had failed.

Anecdotal data point number 2:

I left a laptop in the docking station and powered continuously for several months when working from home. After a while I found the battery had failed.

Tentative conclusion:

Do not float charge lithium ion batteries.

[SiliconWizard]

Couple points regarding IanB's notes:

1. Leaving Li-ion-powered devices plugged into their chargers doesn't mean the chargers will "float" charge the batteries. Actually, they usually won't, as most Li-ion charging ICs out there are designed to stop charging automatically and restart only if the battery voltage dips below a certain threshold. So a charger usually won't float charge whatsoever.

2. In your case, it's more likely that leaving a Li-ion battery fully charged for extended periods of time (which is considered "storage") WILL degrade it and eventually kill it. It's not that it's float-charged, it's just that it's stored fully charged with no opportunity to discharge significantly (except for the small threshold of the charging IC.) It's not the same thing as float charging, though. But I can confirm this is a good way of killing the battery of many devices - it's happened to me numerous times as well.

3. Float charging is definitely not recommended for Li-ion batteries, mostly because of the above - meaning it's akin to leaving a battery fully charged for extended periods of time - but also depending on how you float charge it. Barely floating it with a voltage source for a long time is no good. Your voltage source should at least have a current limiter and monitor current at all times. While the current should drop to a minimal amount when the battery is charged to the float voltage, chemistry is a bitch and it can go up in some conditions (like temperature).

Point 3. can be alleviated a bit by using a lower float voltage as some suggest, but even 4.1V is too high then. Even 4V IME. Floating to the nominal voltage (3.7V - 3.8V) is a much safer bet, although obviously you don't get a full charge then. But even doing this, I don't like much.

I do recommend playing safe with Li-ion batteries.
But you guys are adults (or so it seems), so this is your call!

[thm_w]

Thanks for the replies.
The buck converter option is a good tip. However, I have yet to find a 4.1 volt, or 4 volt one. I will keep looking.
And yes, the lower voltage 4 volts, will likely do the job and provide longer cell life.

Jim.

Depending on the IC, you might be able to get away with a single schottky diode on the output, or two anti-parallel. You could also add a TL431 or maybe zener to bleed power and prevent the voltage rising too high.
A few days ago I opened a lithium pack with a built in schottky, which drops the charge voltage, first time I've seen that.

Do not float charge lithium ion batteries.

Float charge is fine when you control the parameters, read the thread I linked.

[ledtester]

Float charge is fine when you control the parameters, read the thread I linked.

Do you mean the link in Reply #1:

LM1084 seems a bit overkill, what will be limiting the current into the cell? Or is the solar panel small enough its not an issue?
4.1V is ok, 4V will increase lifespan a bit longer.

I would rather use a buck regulator and standard TP4056 linear charger, or a buck lithium charger.

https://www.eevblog.com/forum/projects/float-charge-voltage-for-liion-cells/
...

because that link refers to this thread.

[thm_w]

because that link refers to this thread.

oops thanks for that.

Here are some threads:
https://www.eevblog.com/forum/chat/lipo-battery-on-holdingfloat-charge/
https://www.eevblog.com/forum/chat/is-big-clive-right-about-lithium-battery-charging/
https://www.eevblog.com/forum/projects/float-charging-lithium-ion-batteries-at-lower-voltages/

[thm_w]

Thanks for the replies.
The buck converter option is a good tip. However, I have yet to find a 4.1 volt, or 4 volt one.

Did you misread thm_w's post? If you use a TP4056 (like this $1 module) you can power it with anything from 4.5-6V.

Powering it was not OPs concern, but what voltage comes out of the IC.
TP4056 is preset to 4.2V, I think there are other ICs you can get with customized output voltage, but they are harder to source.
So its easier if you can just drop that 4.2V a few hundred mV or so.

[IanB]

Float charge is fine when you control the parameters, read the thread I linked.

I have indeed read and taken note of many threads on this subject.

I would have to say I don't know for sure what values to select for the parameters for the best outcome. Like for example, how much might it depend on the specific battery? There are so many with different specs from different vendors.

Even in this thread there might be debate about the safest float voltage. 4.1 V? 4.0 V? Less?

[Psi]

I don't see any problem float charging lithium batteries (24/7), but you have to lower the voltage considerably to keep them happy and long-lasting.

I would go with 3.95V personally, but you can do whatever you want it will just slowly damage them the higher you charge to.
And capacity will obviously be less the lower you charge too.
3.95V gives you ~70% capacity which is not too bad and is also a good voltage to avoid stressing the cell.

I would make sure they are kept in a container/location that will be somewhat safe if they decide to catch of fire for any reason.
That's just a good plan for any DIY lithium battery charging but probably more important if you keep them on 24/7.

I would also MAKE SURE you put some thermal fuse/fuses in the center of the battery pack, so any dangerous temps immediately cut off the charge and disconnect the battery.  It's safe to charge up to 45C/113F so you might want 2 levels of protection. A system that just measures temp and stops charging or reduces charging at 40C and a 2nd 'failsafe' thermal fuse which blows at 45C or 50C, or something like that. Make sure you use the center of the pack as the location you detect heat as that is where the hotspot will be.

A system to measure the temp and stop charging could be something simple like an NTC resistor as part of the voltage divider on the charging system feedback/vdetect. So as the thing gets hotter and hotter the max voltage its charging to gets lower and lower. Thus charging will slow and stop if it gets too hot. Or you could add a mosfet so as soon as the NTC temp gets to a set point it starts to clamp the feedback line high and stops charging altogether.
 
The failsafe thermal fuse can just be in series with the cell output, assuming it can handle the current.

Make sure you have a low voltage cut-off for the cells as well. That is important as you don't want your load to overly discharge the cells causing damage and then the charger to try charge them back up in a damaged state .

[thm_w]

Even in this thread there might be debate about the safest float voltage. 4.1 V? 4.0 V? Less?

Its a trade off of capacity and lifespan. Anything under 4.2V is "safe", but the cell will quickly degrade at that voltage.
Psi's recommendation seems reasonable, 70% state of charge.

edit: https://www.researchgate.net/figure/Calendar-aging-measured-on-6-Li-ion-commercial-cells-according_fig1_261297316

[Jim0000]

Couple points regarding IanB's notes:

1. Leaving Li-ion-powered devices plugged into their chargers doesn't mean the chargers will "float" charge the batteries. Actually, they usually won't, as most Li-ion charging ICs out there are designed to stop charging automatically and restart only if the battery voltage dips below a certain threshold. So a charger usually won't float charge whatsoever.

2. In your case, it's more likely that leaving a Li-ion battery fully charged for extended periods of time (which is considered "storage") WILL degrade it and eventually kill it. It's not that it's float-charged, it's just that it's stored fully charged with no opportunity to discharge significantly (except for the small threshold of the charging IC.) It's not the same thing as float charging, though. But I can confirm this is a good way of killing the battery of many devices - it's happened to me numerous times as well.

3. Float charging is definitely not recommended for Li-ion batteries, mostly because of the above - meaning it's akin to leaving a battery fully charged for extended periods of time - but also depending on how you float charge it. Barely floating it with a voltage source for a long time is no good. Your voltage source should at least have a current limiter and monitor current at all times. While the current should drop to a minimal amount when the battery is charged to the float voltage, chemistry is a bitch and it can go up in some conditions (like temperature).

Point 3. can be alleviated a bit by using a lower float voltage as some suggest, but even 4.1V is too high then. Even 4V IME. Floating to the nominal voltage (3.7V - 3.8V) is a much safer bet, although obviously you don't get a full charge then. But even doing this, I don't like much.

I do recommend playing safe with Li-ion batteries.
But you guys are adults (or so it seems), so this is your call!

Thanks for the detailed reply.
I recognise your caveat about leaving LiIon cells at a constant charge, and have taken it seriously.
So much so, that I have decided against doing this and staying with the original 3 x AA dry cells power supply as original.

Jim.

Did you misread thm_w's post? If you use a TP4056 (like this $1 module) you can power it with anything from 4.5-6V.

No. I was deterred by the 4.2 volt output of those devices.

I am abandoning this concept, for reasons mentioned above.
So, I will mark the thread as solved.
Thanks for all replies.

Jim.

[shapirus]

I have a LiPo battery lying on my desk right now. It was charged with to 4.1 V per cell (lower than the normal 4.2 V charge -- to extend its life span) and the charger was disconnected about two months ago.

Its cells now measure: 4.09 V, 4.07 V, 4.08 V.

Now, for the sake of simplicity, let's take one cell, the 4.07 V one, and say we are going to float charge it.
Take a CC/CV power supply and set its voltage limit to 4.07 V. Connect it to the cell. What's going to happen immediately? Nothing: no current will flow in either direction, because the potentials are equal. What's going to happen in a longer period of time? An extremely tiny current, equal to the self-discharge current, will be flowing into the battery.

What is the difference, from the cell's point of view, between it being not connected to anything and being connected to a voltage source with a potential equal to that across the cell's terminals (the very definition of float charging)? I say there is no difference, because there is no current flow, except for the self-discharge compensation.