Author Topic: Trickle charging of 18650 li-ion cells  (Read 1788 times)

0 Members and 1 Guest are viewing this topic.

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Trickle charging of 18650 li-ion cells
« on: March 18, 2023, 11:16:14 am »
It is not recommended, anyhow, would a careful trickle charging of a 3S battery made of the 18650 older li-ion cells work?
I mean for example when AC powered: 12V (or higher) voltage regulator -> schottky diode -> 10 ohm resistor -> battery.
The battery is intended for a voltage reference with aprox 7mA current consumption.
« Last Edit: March 18, 2023, 11:19:27 am by imo »
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #1 on: March 18, 2023, 12:38:35 pm »
You can charge them any way you like, as long as two conditions are satisfied:

1) charge current must not exceed the allowed value (a rule of thumb being 0.5C normal charge, 1C fast charge);
2) voltages on the terminals of each cell must not exceed 4.20 V, or, for some cells, 4.10 V -- use this value to be safe.
 
The following users thanked this post: NiHaoMike, iMo

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #2 on: March 18, 2023, 02:55:47 pm »
This may work..
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #3 on: March 18, 2023, 03:08:34 pm »
What about overdischarge protection? Cell balancing*? You'll also need to make sure that U1 and R1 don't overheat when you begin to charge a fully discharged (that is, 3.0V*3 = 9V) battery, since there's no current limiting other than with R1.

*to prevent cell overvoltage so that no cell goes over 4.20V, you have to have either a balancer circuit, or a per-cell voltage monitor that disconnects the charger when any cell goes over 4.20V, or both.
« Last Edit: March 18, 2023, 03:12:37 pm by shapirus »
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #4 on: March 18, 2023, 03:15:47 pm »
Also, why not use a bridge and a cap on the input? It will likely work as pictured (but check your linear regulator datasheet for dynamic characteristics), but I'd say it looks weird.
 

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #5 on: March 18, 2023, 03:17:49 pm »
Those are old cells, none high performance requirements. I need them to keep voltage for several hours (@7mA) when switched off the mains.. The max battery voltage will be around 12.25V. No rocket science here :)
Use at your own risk.

PS: the 78L12 limits the current to 100-150mA.
« Last Edit: March 18, 2023, 03:25:13 pm by imo »
 

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #6 on: March 18, 2023, 03:20:25 pm »
Also, why not use a bridge and a cap on the input? It will likely work as pictured (but check your linear regulator datasheet for dynamic characteristics), but I'd say it looks weird.

"AC adapter" in the schematics means "the DC 18-19V output wall adapter from an old notebook" or similar.
No "AC" at the 78L12 input, of course..
« Last Edit: March 18, 2023, 03:22:54 pm by imo »
 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #7 on: March 18, 2023, 03:26:42 pm »
"AC adapter" in the schematics means "the DC 18-19V output wall adapter from an old notebook" or similar.
No "AC" at the 78L12 input, of course..
Ah. Then D3 can be removed or replaced with a power resistor to drop some voltage and dissipate extra heat before the linear regulator, if D1 is present.
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 732
  • Country: us
Re: Trickle charging of 18650 li-ion cells
« Reply #8 on: March 18, 2023, 04:58:13 pm »
You can charge them any way you like, as long as two conditions are satisfied:

1) charge current must not exceed the allowed value (a rule of thumb being 0.5C normal charge, 1C fast charge);
2) voltages on the terminals of each cell must not exceed 4.20 V, or, for some cells, 4.10 V -- use this value to be safe.

Trickle charging is not a good idea.

https://batteryuniversity.com/article/bu-409-charging-lithium-ion

 

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #9 on: March 18, 2023, 05:02:43 pm »
Let's define what "tricke charge" is so we talk about the same thing.
 

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #10 on: March 19, 2023, 06:40:46 am »
Yes, the trickle charging is not recommended for Li-Ion (and Li-poly) as I wrote above.
So the subject of this thread is a little bit misleading..

On the other hand I've tried with a "first help" for my voltage reference this night - I've wired an old 12.23V wall adapter via a BAT54 (as the schottky was laying on my desk, I would use an 1A schottky next time..) to the battery (without disconnecting it), which had 11.48V. Over night it went up to 11.78V. This is not the trickle charging, however, as the voltage has been limited to 12.23V (minus some forward drop at the diode).

Long time back I charged a 4x18650 pack off my bench PSU set to 16.80V via a 2.2ohm 5W resistor and it worked fine (the 15y old battery is still in use, difference between the cells is <=5mV). That was not the "trickle charging" either, as the max voltage and current was limited, and the current went to almost zero when charged..

Trickle charging mean you feed some small current (like 0.05 of C) into the battery without any voltage limiting.
« Last Edit: March 19, 2023, 06:55:46 am by imo »
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 8090
  • Country: fi
Re: Trickle charging of 18650 li-ion cells
« Reply #11 on: March 19, 2023, 06:55:42 am »
Just avoid the term "trickle charging". It is not well defined in context of li-ion, and will just trigger non-technical people who haven't got the slightest idea what they are talking about to start posting links to the well known fake information heaven, the Battery University.

"float charging" is slightly better but that causes emotional reactions in some people, too. So maybe just describe the idea as "CC-CV non-terminated charging".

While manufacturers of cells usually suggest terminating the charging, that will only result in microcycling between maybe 90% and 100% SoC (depending on charge restart hysteresis), which definitely is no good for cell life.

Most important in your "float" application, IMHO, is to make sure your CV regulator cannot (easily) fail as short circuit or otherwise output too high voltage. Keep it to 4.1V/cell, or even better, 4.0V/cell if you can afford that ~20% loss of usable capacity.

Assuming the cells are not seriously bad (increased self-discharge), even a non-balanced pack is an option. I know people here don't like it, but for example, Robert Bosch had no issues I know of supplying completely no-BMS power tool packs for years. I have dismantled a few after End-of-Life and they were in perfect balance. No cell-level monitoring, no connections to taps at all. But if you go that way, I'd recommend leaving more room for both charge voltage, and discharge cutoff voltage (e.g., 4.1V and 3.0V instead of 4.2V and 2.5V), and check the cell-level voltages with a multimeter manually every few months.
« Last Edit: March 19, 2023, 06:57:36 am by Siwastaja »
 
The following users thanked this post: NiHaoMike

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #12 on: March 19, 2023, 07:18:32 am »
Yep, the "float charging" sounds good.
Basically, when the float charger consisting of:
a constant voltage source->(diode+)resistor->Li-Ion battery
where the voltage source is set to those say 4.0VxN and the resistor limits the max current to some safe value, is permanently wired to the battery, it charges the battery "permanently", but it is not a "trickle charger", imho :)
« Last Edit: March 19, 2023, 07:25:06 am by imo »
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 8090
  • Country: fi
Re: Trickle charging of 18650 li-ion cells
« Reply #13 on: March 19, 2023, 07:32:04 am »
Proper charging algorithm qualifies the cell voltage before starting. I.e., don't charge if below 2.0V or something like that. Variation of this is very low charging current (like C/100) and voltage qualification to see if the cell has increased leakage current which does not get fixed by slow charging (i.e., voltage does not rise as expected).

But some commercial products fail to do this to level I would be comfortable with, and happily start charging cells at 0V with reduced current which still is beyond what I would consider safe, so I don't know :-//

Your "always connected" solution fails to do this qualification, but the "kosher" chargers that just cycle the cell between 4.0V and 4.2V have the exact same shortcoming.

Also remember the temperature limitations to charging. Usually stated as a step function, charging disallowed below 0 degC, but better described as current derating (less charging current the colder it is).
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6136
  • Country: ro
Re: Trickle charging of 18650 li-ion cells
« Reply #14 on: March 19, 2023, 07:54:31 am »
The problem is the more they stay at higher voltages the more they suffer, and often ended up bulged when kept for months at top voltage (4.2V or so).  Or at least that's what happened with many of my phone batteries kept disconnected but charged.  Same with the IR light sensors, the ones plugged 24/7 into their wall adapter, bulged in 1-2 years.  The ones charged hen used on battery only, didn't bulged.

Then I've searched about storing fully charged Li-Ion, and some manufacturer specify that's a bad idea.  Also the typical shelf charge level (delivered from the factory) is about 30% of the nominal capacity.

Ideally you want to charge them to their max voltage (4.2V or so) only when you know you'll start discharging them immediately or very soon (soon as in days later, not months later).

You don't want to float them 24/7 at 4.2V, but it's OK to float them indefinitely at some 3.7V or so, and fully charge them at 4.2V only when you know you'll start discharging them very soon.
« Last Edit: March 19, 2023, 07:59:40 am by RoGeorge »
 
The following users thanked this post: ahbushnell

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #15 on: March 19, 2023, 08:09:44 am »
"float charging" is slightly better but that causes emotional reactions in some people, too. So maybe just describe the idea as "CC-CV non-terminated charging".
The problem with the term "float charging" is that it is somewhat misleading too, because no actual charging is taking place when the cell voltage (or the cells combined, if there's more than one) reach the open circuit voltage of the source. Current becomes essentially zero, save for the microscopic amount that equals the self-discharge current. Of course, this depends on the particular chemistry of the cells and the voltage setting of the charger, but I'm talking in the context of charging typical LiIon/LiPo cells to 4.10V and even 4.20V.

Most important in your "float" application, IMHO, is to make sure your CV regulator cannot (easily) fail as short circuit or otherwise output too high voltage. Keep it to 4.1V/cell, or even better, 4.0V/cell if you can afford that ~20% loss of usable capacity.
This.
There is no difference from the battery's point of view between it being stored disconnected or connected to a CC/CV charger set to the voltage equal to the battery's own open-ciruit voltage. Equal voltage = no current = no charging. Save for the self-discharge current remark, that is.

Choosing the right final voltage is a separate issue. The lower the cell voltage, the slower it degrades. But just how much, is a question that isn't easy to answer. Good information on this is not exactly easy to find. One study gives a number, taken from another article, where it was derived from modelling, of 0.04% capacity loss per cycle when charging to 100% SoC and 0.02%/cycle with 80% end-of-charge SoC. It doesn't tell anything about capacity loss over time when the battery is stored at any given SoC without cycling.
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 8090
  • Country: fi
Re: Trickle charging of 18650 li-ion cells
« Reply #16 on: March 19, 2023, 09:30:38 am »
Choosing the right final voltage is a separate issue. The lower the cell voltage, the slower it degrades. But just how much, is a question that isn't easy to answer. Good information on this is not exactly easy to find. One study gives a number, taken from another article, where it was derived from modelling, of 0.04% capacity loss per cycle when charging to 100% SoC and 0.02%/cycle with 80% end-of-charge SoC. It doesn't tell anything about capacity loss over time when the battery is stored at any given SoC without cycling.

Good information is hard to come by because the answer is, "it depends". There is no universal generalization; it depends on the small chemistry/manufacturing details.

One indeed needs to separate cycling damage from storage damage. The study I made measured storage damage between a few off-the-shelf cells and the differences between products were huge. (see https://www.eevblog.com/forum/chat/lithium-ion-battery-degeneration/msg3217888/#msg3217888 . One cell had room-temperature capacity fade when stored at 4.2V of 0%/year, while another had 5%/year.)

Pretty much the only generalization you can trust is that both storage and cycling damage is significantly reduced only at SoC < 60% or so. There are huge differences between products how the aging performance behaves when 100% vs. 80% is compared; some benefit significantly from such small drop, some not that much. (E.g., in my dataset, Samsung 29E capacity fade is 3.0%/year, 2.3%/year, 0.0%/year respectively when stored at 4.20V, 4.00V, 3.60V. And I remember a paper where some Panasonic off-the-shelf cell aged slightly worse when stored at 80% and not 100%.)

But cycling damage between 80% and 100% is probably more significant. That is also why I prefer the feared "trickle/float/whatever" taboo instead of the manufacturer-recommended microcycle-the-cell-to-death strategy.
 
The following users thanked this post: iMo, shapirus

Offline shapirus

  • Super Contributor
  • ***
  • Posts: 1080
  • Country: ua
Re: Trickle charging of 18650 li-ion cells
« Reply #17 on: March 19, 2023, 09:55:41 am »
You don't want to float them 24/7 at 4.2V
Once again, the key point here is "4.2V", not "float". It's the cell voltage that affects the rate of degradation rather than the fact of it being connected to a powered-on CC/CV source.

For example, here is a cell that has been sitting idle on a shelf for several weeks now. Will it make any difference for it if I connect it to a 4.2136V source now? Obviously it won't, and that will be exactly what is called "float charging".



The charger that it was charged with has 4.2373V between its terminals without load (except for the 10M input resistance of the DMM). So the cell lost only 0.0237V in several weeks. This means essentially zero current in case of "float charging" flowing from the charger to the cell after reaching the final voltage, even if set to 4.2V.
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 6136
  • Country: ro
Re: Trickle charging of 18650 li-ion cells
« Reply #18 on: March 19, 2023, 12:53:36 pm »
My point posting there was to say that apart from the current, even when the current is zero, the voltage itself damages the battery with time.

The bigger the internal voltage, the bigger the damage.

Offline paulca

  • Super Contributor
  • ***
  • Posts: 3989
  • Country: gb
Re: Trickle charging of 18650 li-ion cells
« Reply #19 on: March 22, 2023, 11:37:40 am »
You can charge them any way you like, as long as two conditions are satisfied:

1) charge current must not exceed the allowed value (a rule of thumb being 0.5C normal charge, 1C fast charge);
2) voltages on the terminals of each cell must not exceed 4.20 V, or, for some cells, 4.10 V -- use this value to be safe.

Trickle charging is not a good idea.

https://batteryuniversity.com/article/bu-409-charging-lithium-ion

"That" site again.  While it is full of useful information, it is also full of crap.  It quotes data for LiIon which nobody can find references for or the source of and it isn't repeated with modern cells.  Such studies as storage voltage, vs. storage temperature and their effects on degradation.

With Lithium batteries the layman information is designed to be taken literally to stop people from being "inventive" in how they charge them, to prevent them from burning their house down.  Like, for example, connecting an unregulator 12V wall wort to a 12V lithium ion battery pack they bought off Amazon = a near 100% risk of fire eventually.

However a lot of that "for the layman" information does not hold up to actual investigating research and test of batteries.  Or rather a lot of the "NEVER EVERs" turn out to be ... "it depends".  It's just for people not qualified to understand why it depends should take the laymans dos and donts as bible.

In terms of the OP.  Why not consider a solar panel.  Even if you don't use one, consider that nearly every battery+solar system these days uses Lithium or wants to be using Lithium.  A solar panel can be considered "trickle" charging about 60% of the time with the 40% during peak sun hours when it "spanks" the battery with as much current it can throw at it, then goes back to trickly charging with the panel directly across the battery.  No balancing necessary as long as you understand the chemistry and the "it depends" factors.  If you are using older LiIon cells in 3S config, you will either have to accept a much lower pack voltage or provide individual cell monitoring if not balancing to prevent cell over-voltage.  (or undervoltage for that matter).

Modern LiFePO4 cells are preferred for this kind of application as they have such a flat main curve and such excessive voltage margins on either side, the batteries tend to balance themselves either in absorption during boost charge, or with eddy currents at low SoC.  "It depends".
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline paulca

  • Super Contributor
  • ***
  • Posts: 3989
  • Country: gb
Re: Trickle charging of 18650 li-ion cells
« Reply #20 on: March 22, 2023, 11:44:08 am »
If you can get access inside the pack with a soldering iron, you could consider adding individual cell protection straps across them.

These can be found on many sites in several forms.  The ones you want are the ones that put a resistor across the cell when it hits 4.2V and only disconnects it (disconnecting the whole pack) at 4.3V.  This will help top balance the cells and as long as your regulated voltage is unlikely to exceed maybe 12.2V and the charge current is small enough such that the balance resistor can absorb it.  It will be fine.

These protection straps will also disconnect the battery, and thus the whole pack in the even they get drained under 3.0V per cell.
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #21 on: March 22, 2023, 11:56:33 am »
OP is not recommending trickle charging.
The other question for experts here is what happens with the 18650 cells when "float charged" for unlimited time, off a stable constant voltage source (with a reasonable current limit built in), via a diode, with the max voltage at the battery pack less than 12V (for the 3 cells pack, such max voltage per cell is <4V, say 11.9V), with constant load currents like 7mA.
« Last Edit: March 22, 2023, 12:06:20 pm by imo »
 

Offline paulca

  • Super Contributor
  • ***
  • Posts: 3989
  • Country: gb
Re: Trickle charging of 18650 li-ion cells
« Reply #22 on: March 22, 2023, 12:05:39 pm »
My solar charge controller floats the lithium pack.  It does nothing.

It charges the 4S pack to 14.40V  and keeps it there for up to 2 hours.  In most battery terms this is considered "boost" or "absorption" although those terms do differ.

Then it drops back to float voltage of 13.6V and the pack current drops to 0. 

Slowly the load will pull the battery down and my LiFEPO4 cells will get to 13.6V fairly quickly when they will start to "charge" but really its just the solar -> battery -> load through path.  It will do a "boost recon" every hour to see if the battery will take amply current at 14.40V again and if it will switch back to boost mode, if it won't it remains at float.

"Floating" them at 4.20V constantly is probably not recommended. 

Actually LiIon cells it's not really recommended to have them at a constant 100% SoC.  Not least because 100% SoC is temperature dependant as the voltage across the chemistry is temperature dependant.

A common gotcha in my early days of high power 20-30C lithium pouch cells.  Was to fast charge a battery from the car at the field while flying a helicopter, but have the wind pick up and you decide to head home.

However you have a 4S 3Ah pack which has been charged to 4.2V per cell at an ambient temperature of 3*C.  You take your stuff out of the car and set it in the garage, in the window, in direct sun.  Not long and the battery cells are now sitting at 25-35*C and their voltage will no longer be 4.2V per cell, but significantly higher.  Best practice is to keep them around "storage" voltage or about 30-50% SoC.
"What could possibly go wrong?"
Current Open Projects:  STM32F411RE+ESP32+TFT for home IoT (NoT) projects.  Child's advent xmas countdown toy.  Digital audio routing board.
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 8090
  • Country: fi
Re: Trickle charging of 18650 li-ion cells
« Reply #23 on: March 22, 2023, 12:15:53 pm »
The other question for experts here is what happens with the 18650 cells when "float charged" for unlimited time, off a stable constant voltage source (with a reasonable current limit built in),

Nothing happens. When the cell terminal voltage equals the supply voltage, no current is flowing. The diode you are suggesting pretty much eliminates any possibility of supply ripple voltage being an issue.
 
The following users thanked this post: iMo

Offline iMoTopic starter

  • Super Contributor
  • ***
  • Posts: 4627
  • Country: nr
  • It's important to try new things..
Re: Trickle charging of 18650 li-ion cells
« Reply #24 on: March 22, 2023, 12:29:43 pm »
The diode there is mainly for the situation the DC constant voltage power supply (hard wired to the battery via the diode) is switched off the mains (such the battery is not discharging itself into the power source when the power source is powered off). When the battery voltage goes down after a longer mains' power off, it will be charged up with the max current limited inside the DC constant voltage source (it will be couple of times a year). Otherwise float charged.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf