Float Charging Lithium-Ion Batteries at Lower Voltages

[mhz]

I understand from various sources that float charging a Li-Ion battery at full voltage (usual 4.2V case) is bad, degrading the battery lifetime and potentially causing an unsafe situation. For example, https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries: "Li-ion cannot absorb overcharge. When fully charged, the charge current must be cut off. A continuous trickle charge would cause plating of metallic lithium and compromise safety. To minimize stress, keep the lithium-ion battery at the peak cut-off as short as possible."

I also have seen in some sources that leaving a Li-Ion battery to float charge at a lower voltage (say 4.1V) is safe, however I haven't seen this stated at any references I'd consider authoritative. Does anyone have any references for or against this practice that come from a seemingly authoritative source like "battery university"? Ultimately I'd like to know if I can leave a 4.1V cell in CV stage indefinitely without worrying about causing it damage or creating an unsafe event. (Actually, my application is 13s5p pack, made of Samsung 25R cells).

For that matter, any general references (books, etc) that can help bridge the divide between practical use and the underlying chemistry would be appreciated.

[BravoV]

Although I am not sure if this is a real legit authoritative reference, from the video, at least to me, its very convincing to watch Prof. Jeff Dahn presentation.

Bottom line is, the "ideal" nitpicking spot is 4.03 volt (at 80% SOC), and anything above 4.1 will get worst, especially if we aim for cell life longevity.

This video is quite long > 1 hour, suggesting to watch from the beginning.

But don't worry, I marked the time to start the video at 1:07:05 at below link, watch and note carefully what he said.

https://youtu.be/pxP0Cu00sZs?t=1h7m5s

Hope this helps.

[edpalmer42]

Look for info on solar powerplants that use Lithium batteries or something like a Tesla Powerwall system that's used during AC failures.  AFAIK, these systems float their batteries rather than occasionally doing a charge cycle.

I've heard that the rule of thumb is to float the cells a bit above the voltage that they drift to when they're idle.  That voltage puts the least amount of stress on the cells.

Sorry, but I don't have any more definitive info.

Ed

[digsys]

I work with lithium EV packs for many years and the comments are pretty much spot on. They do NOT like "trickle" charge of any sort, especially above their
"resting" state ~3.7 - 3.8V. IF you know the the self-discharge rate, you can safely put that back, but that changes with time / temp etc, so not easy to find.
If the cells are in "reasonable" condition, they shouldn't need ANY trickle maintenance anyway. Button Li cells in RTCs for example hold charge for 10-20 yrs+

[jmaja]

I was designing a device that would work on a very small solar panel + 50 mAh lipo cell. Since I could not find a charging IC that would have Iq below 1 uA I planned to use a LDO setup for 4-4.1 V. I did find some references that supported constant voltage charging 0.1-0.2 V below max.

I decided to dump the solar panel and use nonchargable lithium coin. Can't remember where I found the reference for trickle charging. I think there were tests how cell life was affected at different trickle voltages.

[David Hess]

The trick is that like supercapacitors, static operating life is some inverse power function of the voltage while capacity is only proportional to voltage.  So for example raising the charge voltage by 0.1 volts might boost capacity 10% at the expense of decreases the operating life by several times.  Manufacturers pick the charge voltage based on intended operating life but nothing prevents you from using a lower voltage to sacrifice some capacity while increasing the operating life even more.

That is how Tesla for instance can sell the same physical battery with different capacity specifications depending on how much you pay.  The higher cost makes up for the greater likelyhood that the battery will wear out before the warranty has expired.

[Siwastaja]

Life expectancy starts going rapidly up when you choose something below about 70-80% (below about 4.0V). If you can float at around 60% (~ 3.8V), that would be great. Floating at 4.2V isn't catastrophic. Floating at 4.1V usually isn't much better, and with some cells, it has been shown to be slightly worse than at 4.2V. This is assuming negligible cycling effect (only few charge/discharge steps, so very smooth floating). If you have constant changes between charge/discharge, then going to 4.1V will indeed give you extra life compared to full charge.

Yes, you can permanently connect a CC-CV source if you can guarantee it behaves correctly. Charging timeout is a safety feature which protects for unsafe events in case your regulator feedback fails and gives, for example, 4.5V instead of 4.2V. Floating itself isn't a problem. Current doesn't flow into a cell when its open-circuit voltage (which is dependent on the State-of-Charge) is the same as your supply's output voltage.

A full 4.20V charge (typically defined at C/20 stopping) tends to equal to around 4.16V open-circuit voltage; at this voltage, no current flows (expect for what makes up for self-discharge, which is very small). If you float at 4.20V, you have a very slight overcharge (maybe around 101-102%), which could decrease the life expectancy slightly, but isn't a safety issue.

Source: own research. Not published, sorry.

Battery University is a fake science site. Their quality has increased, but it has no credibility whatsoever.

[David Hess]

With a lot of charge-discharge cycles, the cycling will shorten the operating life faster than charging to a higher voltage for higher capacity.

In float applications the voltage should be lower for maximum operating life.  The degradation mechanism is some power function of the voltage.  If you search online, then you can find the details.

In a supercapacitor, operating life decreases by 10 times for every increase of 0.4 volts or 30 degrees C.

[MadScientist]

The major factor in Li ion degradation is largely calendar  life time , ( they die just sitting there ) due to the constant parasitic action. After that high charge rates have a factor

Li doesn’t normally actually need float charging as the self discharge rate are very low so I’m not sure why you think you need float charging in the first place

If the application is long term standby , then Li is a poor choice because of my first paragraph

Low value CC charging subjects the Li to multiple micro charging cycles and studies have shown that any charging cycle in regards overall degradation. Is nearly as bad as a more complete charge. , so most li wait for a significant discharge point to be reached , then recharge and switch off

I’d examine the technical reasons why you think you need float charging in the first place 

[David Hess]

The major factor in Li ion degradation is largely calendar  life time , ( they die just sitting there ) due to the constant parasitic action. After that high charge rates have a factor

I agree.  But float charging them to the highest rated voltage for maximum capacity increases the degradation by some non-linear factor.

[KL27x]

People are idiots.
"Trickle charge" is a term derived from lead acid batteries.

If the float charge of a li ion cell reaches 4.2V (a good cell), and you charge it from 4.2V source, there is no appreciable current. No current, no trickle. That said, after some damage, a cell might no longer hold a float voltage of 4.2V. It could be lower. But battery university has a mish mash of information, much of which is nothing more than myth and completely incorrect. This is maybe halfway correct if you squint at it the right way.

When you trickle charge a lead acid cell it's because you are using a voltage that is higher than the float voltage to begin with. So after it is fully charged, there is still a small current that continues flowing into the battery which is greater than the self-discharge rate. Again, this is only possible because the psu/charger outputs a higher voltage than the float voltage of the battery. This is not typically the case with a li ion charger.

So when you trickle charge a lead acid battery, instead of pushing a high current into it and disconnecting it when it's done charging, you use a higher voltage than the float voltage and have some simple current limiting like a series resistor, so that the charge current is relatively low. And you don't disconnect it when it's done charging, because the current flow is relatively small, and with lead acid batteries there is no harm no foul.

It is completely true that you shouldn't trickle charge a li ion cell. To trickle charge a cell you'd need to use a voltage source that is higher than 4.2V. We already know that you can't do that. This is redundant information that has been twisted into something that is pretty much meaningless. By idiots.

[Siwastaja]

The major factor in Li ion degradation is largely calendar  life time , ( they die just sitting there ) due to the constant parasitic action. After that high charge rates have a factor

I agree.  But float charging them to the highest rated voltage for maximum capacity increases the degradation by some non-linear factor.

If negligible cycling happens, and floating causes damage mostly by calendar aging, then the function is indeed non-linear, and interesting. From what I have read and measured myself, it's clearly a piecewise function of two slopes:

Below about 75-80%, you have strongly increasing life expectancy, the lower you store it at. You don't need to go arbitrarily low; at around 50%, it's already so freaking good it's hard to say if there's anything more to gain.

But, contrary to intuition and common belief, over about 80%, strange things happen. The cells I have tested show very little difference at all (i.e., 80% and 100% are equally bad), and I have read a paper in which a Panasonic cell (IIRC) actually showed better calendar life at 100% compared to at 80%. I really don't know the mechanism behind this (haven't looked at it). Do note that this is based on actual production chemistries available now (visible on both LCO and NCA), but is not a fundamental physical law, so if you read this post after years, things may be different.

Now, between 80-100%, charging current does more damage (than at, say, between 60-80%), so if a lot of cycling happens, then limiting yourself to 80% will be a benefit. But for storage, or very low-ripple, low-cycling floating, 80% is not any better than 100%. So you can as well use the near-full capacity; or go even lower to really increase the life.

You need to make the distinction between calendar and cycle life anyway: for calendar aging, lower temperature helps (the lower the better), but for charge current induced damage (cycling), higher temperature helps (optimum can be near the rated maximum charge temperature, often 45 degC). (This is unsurprising; as you know, charging tends to be completely forbidden below 0degC. In reality, it's not a step function, and as there is no water anywhere to freeze, 0 degC is nothing magical, just a nice number where charging already produces "too much" damage.)

So when you combine both calendar and cycling damage (as you do in a practical float system), you'll have a combination of opposing constraints, and need to know which damage mode dominates. Even a guesstimate is much better than nothing.

For example, in a highly cycling system, I have measured that a heated battery (at 50 degC) does over 1500 cycles just fine, but a cooled (at 10 degC) battery dies in around 100 cycles. The heated use case was outside the manufacturer's specification; the cooled one was inside the specification, yet died early.

[edpalmer42]

How much does the chemistry of the cell affect this behaviour?  I've got a stash of LiFePO4 cells that I thought would work great for a 24V backup system.

For those who aren't familiar with them, LiFePO4 cells are still Lithium ion cells, but they have a lower voltage and lower power density than typical Lithium cells.  Their big advantage is that they're much more stable than their more volatile cousins.  It's almost impossible to get them to burst into flames.  They also seem to be more forgiving regarding charging and discharging.

Ed

[Siwastaja]

LFP cells are easy to get burst into flames, and tend to contain fewer safety mechanisms due to the "false sense of security" phenomenon caused by the safer cathode chemistry, and substandard manufacturers. Lack of shutdown separators, current interruption devices, and PTC fuses is alarming, and it will enter thermal runaway somewhere around 350degC anyway, although with less energy, still shooting the exact same flammable electrolyte out. So in some cases, they are more dangerous, sometimes less dangerous.


The voltage curve of the LFP chemistry is almost a straight line when compared to the classical LCO or modern NCA. This means, better regulation is needed to float it at any particular SoC, and it's still difficult to say what the exact SoC is, based on voltage. If you try to keep it, say, 70%, you are easily accidentally at 60% or 80%. Does this matter? Maybe not, just float it at 100%, it'll do fine.

LFP is quite good even at 100%, if you think about this the same way you think LCO/NCA - through the open-circuit voltage. What's "4.20V" on a modern NCA cell? The same for LFP is not 3.65V or 3.60V, the typical "charge cutoff voltage". If you keep an LFP indefinitely at 3.65V by floating, you'll overcharge it slightly, without getting much more capacity out of it. I can't offhand remember the exact 100% open-circuit voltage, but you can easily measure it by fully charging the cell, disconnecting it, then measuring the voltage after a few minutes (or hours, to be sure). For an NCA cell, it's almost 4.20V, but for LFP, it isn't even close to 3.65V, even if you charged it to 3.65V. So, I'd float an LFP cell at 100%, no problem, but just make sure you know what voltage 100% is at. It's somewhere around 3.40-3.45V IIRC, but I haven't done anything with LFP cells in years.

The best thing in LFP is that, by luck, the voltage range matches so that's it's fairly easy to replace lead acid packs: 2s for 6V, 4s for 12V, 8s for 24V. With LCO/NCA, can't have 3.5s to replace 12V lead-acid! The curve shape is also closer to lead-acid.

[digsys]

I would add an extra comment to Swijas Li charge / trickle comments above, which I very much agree with. We often get "experimental" or "new generation"
Li chemistry to trial, and everything changes !! In coming years, with all the new chemistry "tweaks" they are doing, these % figures will flap around all over
the place. Sometimes, you can't even be sure what batch chemistry you buy. Generally need to run general tests to check.
They are "guidelines" for now :-)

[BravoV]

But, contrary to intuition and common belief, over about 80%, strange things happen. The cells I have tested show very little difference at all (i.e., 80% and 100% are equally bad), and I have read a paper in which a Panasonic cell (IIRC) actually showed better calendar life at 100% compared to at 80%. I really don't know the mechanism behind this (haven't looked at it). Do note that this is based on actual production chemistries available now (visible on both LCO and NCA), but is not a fundamental physical law, so if you read this post after years, things may be different.

Now, between 80-100%, charging current does more damage (than at, say, between 60-80%), so if a lot of cycling happens, then limiting yourself to 80% will be a benefit. But for storage, or very low-ripple, low-cycling floating, 80% is not any better than 100%. So you can as well use the near-full capacity; or go even lower to really increase the life.

Interested at above note, mind elaborate which "exact" brand and type are you talking about ?

How about current popular 18650, say like Panasonic NCR18650GA, as I used these intensively.

[Siwastaja]

Interested at above note, mind elaborate which "exact" brand and type are you talking about ?

How about current popular 18650, say like Panasonic NCR18650GA, as I used these intensively.

Talking about basically all available cells from big brand manufacturers such as Panasonic, Samsung, LG, Sony; they seem to perform in a somewhat similar way. They are of LCO (now starting to get obsolete), NCA or NMC chemistry. None of these exhibit strong calendar life increases when going from 100% to 80%, as expected by some people.

My point is, there is no strong benefit in calendar lifetime between 80-100%, in any cell I have measured or read about, only relatively small and uncertain differences to either direction. Actual calendar life benefits start way below 80%, so if you need long calendar life, keep it low enough.

[SilverSolder]

Poking around in an older hybrid car, it was noticeable that the control system keeps the battery somewhere in the range of 35% to 55% charged under all circumstances.  (NiMH on that vehicle).

The 13 year old vehicle had >200,000 miles on it and the battery still performed like new... 

[thm_w]

...

Li doesn’t normally actually need float charging as the self discharge rate are very low so I’m not sure why you think you need float charging in the first place

If the application is long term standby , then Li is a poor choice because of my first paragraph

Low value CC charging subjects the Li to multiple micro charging cycles and studies have shown that any charging cycle in regards overall degradation. Is nearly as bad as a more complete charge. , so most li wait for a significant discharge point to be reached , then recharge and switch off

I’d examine the technical reasons why you think you need float charging in the first place

Do you have citation for this?

Sure any charging cycle matters for degradation, but a 10% dicharge and 10% charge ten times is better than a full 100% discharge and charge, in terms of cell wear. So you are telling me there is some point between 0% and 10% charge where the behavior reverses, and wear increases. Its possible, I'm just curious where that would be?

This study used various pulsed charges, and found almost no difference in capacity: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.876.2485&rep=rep1&type=pdf
But maybe the frequency is too low.

http://mail.lancaironline.net:81/Lists/lml/Message/56976-02-B/Li-Ion%20Battery%20Life.pdf